Annual Report 2003 Project SB-2 2 1 NDV. 2DD3 Conservation and Use of Tropical Genetic Resources Formerly know as: Assessing and Utilizing Agrobiodiversity through Biotechnology CIAT For Internal use only November, 2003 Project SB-2: Conservation and Use of Tropical Genetic Resources (formerly known as Assessing and Utilizing Agrobiodiversity through Biotechnology) PROJECT OVERVIEW PROJECf DESCRIPTION Objective: To preserve the Designated Collections and employ modem biotechnology to identify and use genetic diversity for broadening the genetic base and increasing the productivity of mandated and selected nonmandated crops. Outputs: l . Improved characterization of the genetic diversity of wild and cultivated species and associated organisms. 2. Genes and gene combinations used to broaden the genetic base. 3. Increase efficiency ofbreeding program using genomics tools 4. Mandated crops conserved and multiplied as per intemational standards. S. Germplasm available, restored, and safely duplicated. 6. Designated Collections made socially relevant. 7. Strengthen NARS for conservation and use of Neotropical plant genetic resources. 8. Conservation of Designated Collections linked with on-farm conservation efforts and protected areas. Milestones: 2004 High throughput screening of germplasm bank and breeding rnaterials implemented, using microarray technology. Al tolerance in Brachiaria characterized. Gene discovery for drought tolerance in bean for nitrification in brachiaria initiated. Marker-assisted selection for ACMV and whitefly resistance initiated. Transgenic rice resistant to a spectrum of fungal diseases. Development of insertion mutagenesis populatíon in rice, using Ac/Ds. Gene flow studies for bean and rice completed. Links with conservation efforts in protected areas and on farms established. Germplasm collections regenerated. lnitiation of DNA banks for core collections. Safe-duplication and restoration continued. Biosafety field testing of transgenic cassava initiated. 2005 Efficient transformation system devolved for cassava Bean with high iron and zinc tested and transferred to CIAT Africa program for bioavailability testing. Survey of cassava germplasm for beta carotene. SNP markers developed for bean and implemented for MAS. Targeted sequencing of cassava genome. lsogenic of QTL in rice developed and tested. Gene expression studies. Technology transfer for rapid propagation system to NARS. Testing of Ac/DS population for gene identification. 2006 Scalíng up of marker assited selection and transfonnation established for rice bean and cassava. High trough put screening for selected tropical fruits initiated Marker assisted selected for multiple traits implemented in beans, rice and cassava Target genes for drougt identified and tested in beans. High iron and zinc bean lines developed through marlcers assisted selection released for field testing. Beta carotene cassava tested in Colombia, Brazil and selected countires in Africa Users: CIAT and NARS partners (public and private) involved in germplasm conservation and crop genetic improvement and agrobiodiversity conservation; AROs from Des and LDCs, using CIA T technologíes. CoUaborators: IARCs (lPGRI through the Systemwide Genetic Resources Program, CIP, and liTA through root and tuber crop research, IFPRI through biofortification proposal and CATIE); NARS {CORPOICA, ICA, EMBRAPA, IDEA, INlAA, lNIFAP, UCR, INIAs); AROs (IRD, CIRAD, Danforth Center, CAMBIA, NCGR, and universities-Comell, Y ale, Clemson, Kansas State, Bath, Hannover, Rutgers, Ghent, Gembloux); biodiversity institutions (A von Humboldt, INBIO, SINCHI, Smithsonian); corporations and prívate organizations. CGIAR system Unkages: Saving Biodiversity (40%); Enhancement & Breeding (55%); Training (4%); Information (l %). CIA T project linkages: Inpws to SB-2: Germplasm accessions from the gene bank project. Segregating populations from crop productivity projects. Characterized insect and pathogen strains and populations from crop protection projects. GIS services from the Land Use Project. Outputs from SB-2: Management of Designated Collections (gene banks); genetic and molecular techniques for the gene ban.k. crop productivity, and soils (microbial) projects. ldentified genes and gene combinations for crop productivity and protection projects. Propagation and conservation methods and techniques for gene banks and crop productivity projects. lnterspecific hybrids and transgenic stocks for crop productivity and IPM projects. CIA T: SB-2 PROJECT LOG F'RAME (2004-2006) PROJECT: PROJECT MANAGER: CONSERVATION AND USE OF TROPICAL GENETIC REsOURCES JOETOHME Narra ti ve Summary Measurable Indicators Means of Verification Goal To contribute to the sustainable increase of productivity CIA T scientists and partners using biotechnology CIA T and NARS pubUcaúons. and quality of mandated and othcr priority crops, and the infom1atioo and tools in crop research. Statistics on agriculture and biodiversity. conservation of agrobiodiver¡ity in tropical countries. Genetic stocks available to key CIA T _Qartners. Purpose To conserve the genetic diversity and ensure lhat lnformatioo on diversity of wild and culúvated species. Publications, repons, and project proposals. characterized agrobiodiversity, improved crop geoetic Mapped economic genes and gene complexes. stocks, and modero molecular and cellular rnethods and lmproved genetic stocks, Unes, and populations. toob are used by CIAT and NARS scientists for i~viog, usiog. and cooserviog crop genetic resowt:es. Output 1 Geno mes chancterized of wild and cultivated species of Molecular iofonnation on diversity of rnaodated and Publications, repons, and project proposals. rnalldated and ooomandated crops and of associated noomandated crops species, and relatelasm and data. Number of COI'e collec:lions rwlliplied and shipped. Usen aslced fOI' novel germplasm and data. Output6 Oesipted Collec:tions made socially relevant. J..andrace diversity resiOI'ed to rarmen. Germplasm catalogs. IIIICmational collectin& possible. Fannen use oew vvieties. Plaot variety re¡istnlinn logs. Quan.ntioe maucn cleared. Breeders use novel genes. National catalogs. Output7 Strengthco NARS fOI' COOSUYation and use or Neocropical NARS eern1>lasm collec:lioos conserved. Country qucstionnaires. NARS aod oetworlcJ willin& to cooperate. planl geoetic resoun:es. Number of tn.inees tn.ined 11 CJA T. Coorses re¡istered. Number or universities aod NARS usiog tn.inins Distribution and sales or tniniog materials. materials. Output8 Conservation or Oesi¡nated Collec:lions linlced with OD· Number of case studies aod pilot in situ conservat.ion Project documentation. NARS interested in conservalion efforu. farm conservalion erroru aod protected are as. projects. Farmen ioterested in conservation effOI'ts. - ---- ----- ii i CONTENT Project SB-2: Assessing and Utilizing Agrobiodiversity Through Biotechnology 11 PROJECTOVERVIEW ii CIA T: SB-2 Project Log Frame (2004-2006) 111 OUTPUT I.Characterizing genomes of wild and cultivated species of mandated crops and associated organisms 1 Activity 1.1 Characterization of genetic diversity 1 1.1.1. Gene flow studies in the bean plant model 1 1.1.2 Studies of gene flow under field station conditions 4 1.1.3 Studies of gene flow with help of biochemical and molecular markers 5 1.1.4 Genetic diversity assessment of Caribbean common bean germplasm 8 1.1.5 Evaluation of microsatellite diversity in combined parental surveys 14 1.1.6 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local __ Cassava V arieties from Guatemala 17 1.1.7 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava V arieties from Ghana and Predictability of Heterosis ____ __ 24 1.1.8 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Uganda _________________ 32 1.1.9 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Sierra Leone ____ ___________ 39 1.1.10 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Cuba __________________ 42 1.1.11 Molecular characterization of rice and red rice using microsatellites and its relation with the morphological seed traits. _____________ 45 1.1.12 Assessment of combinatory ability between red rice and rice under greenhouse conditions 52 1.1.13 Assessment of gene flow from transgenic and non-transgenic rice into red rice under experimental field conditions 56 1.1.14 Genetic diversíty in the multípurpose shrub legume Flemingia macrophylla and Cratylia argentea ____________________ 62 i Activity 1.2 Identification and mapping of useful genes and gene combinations 76 1.2.1 Development of a genome-wide anchored microsatellite map for 76 common bean (Pizaseolus vulgaris L.) 76 1.2.2 Analysis of iron reductase as a mechanism for enhanced iron uptake in common beans 79 1.2.3 Tannin studies on parents and progeny of the DOR364 x G19833 population 82 1.2.4 Identification of QTLs for resistance to Thrips palmi in common bean_ 84 1.2.5 Development of SCAR and microsatellite markers for Apion resistance_ 86 1.2.6 Marker-assisted selection for BGYMV resistance in small-seeded beans 88 1.2.7 Marker assisted selectíon of Arcelin-derived bruchid resistance ____ 92 1.2.8 Generation means analysis of climbing abilíty in common bean crosses 96 1.2.9 Adaptation and use of SCAR markers for the BCMV resistance 102 1.2.10 1.2.11 1.2.12 1.2.13 1.2.14 1.2.15 1.2.16 1.2.17 1.2.18 1.2.19 gene bc-3 in the Andean bean breeding program _________ 102 Identification of an important QTL for symbiotic nitrogen fixation (SNF) in Mexican environments 108 Molecular Marker-Assisted Breeding for Resistance to the Cassava Mosaic Disease in Latín American Cassava Gene pools 111 Molecular Marker-Assisted and Farmer Participatory Improvement of Cassava Germp1asm for Farmer/Market 114 Preferred Traits in Tanzania _ ___ _______ ___ _ _ 114 Molecular Marker-Assisted Selection (MAS) for Breeding Early Root Bulkíng in Cassava 118 Genetíc Mapping of Beta-Carotene Content 121 Genetic Mapping of Dry Matter Content (DMC) 123 Genetic Mapping of Cyanogenic Potential (CNP) 127 Genetíc Mapping of Leaf Retention 130 Mining the Primary Gene Pool of Cassava: Introgression of High Root Protein from Accessions of Manihot esculenta sub spp Fabellifolia and Manihot Tristis into Cassava 132 Mining the Primary Gene Pool: Oreen Mites (CGM) Resistance Genes from Manihot esculenta sub spp Fabellifolia 139 ii 1.2.20 1.2.21 1.2.22 1.2.23 1.2.24 1.2.25 Depelopment of a novel approach for the analysis of diallel mating __ _ designs for better understanding the inheritance of quantitative traits _ 142 lntroduction of inbreeding in cassava ------------ 155 Development of a male-sterile Nípponbare population 158 Utilization of New Alleles from Wild Rice Species to Irnprove Cultivated Rice in Latin America ------------- 159 Identification of QTLs for yield and yield components in rice: Populations derived from backcrosses between the wild species ( Oryza barthii) and cultivated rice (Lemont). 164 Associating Hom Worm Resistance in 60444 (MNG11) with Introgression from M anihot Glaziovii 171 ili Acti vity 1.3 Development of molecular techniques for assessing genetic diversity and mapping useful genes 173 1.3.1 Development of strategy for abiotic stress project in CIA T ______ 173 1.3.2 Development of a molecular marker: DREB for drought tolerance in common bean 177 1.3.3 Dissection of a cluster of Resistance Gene Candidates (RGCs) associated with resistance to angular leaf spot (ALS) in common bean (III) 180 1.3.4 Identification of candidate genes for mineral uptake and storage in cDNA libraries from common beans 185 1.3.5 Generation and analysis of cassava ESTs: towards the identification of a unigeneset _______________________ 188 1.3.6 Identification of genornic regions responsible for conferring resistance to white fly (Aleurotrachelus socialis) in cassava 190 1.3.7 Gene expression profiling of cassava responses to Xanthomonas axonopodis pv.manihotis infection 197 1.3.8 Development of a Diversity Arra y Technology (DArT) Chip for Cassava _ 202 1.3.9 Diversity Array Technology (DarT) for potato diversity análisis 208 1.3.10 1.3.11 1.3.12 1.3.13 1.3.14 1.3.15 1.3.16 1.3.17 1.3.18 Microarray gene expression analysis as a functional genornics tool for postharvest physiological deterioration in cassava _________ 211 Development of molecular techniques for assessing genetic diveristy and mapping of useful genes 216 Positional Cloning of CMD2 the Gene that Confers High Leve! of Resistance to the Cassava Mosaic Disease (CMD) 221 Saturation of the Molecular Genetic Map of Cassava with PCR- based Markers:Progress on the Mapping of a New Set of 140 New SSR Markers 225 Study of gene expression during pistil development in apornictic and sexual Brachíaria spp. 227 Phenotypic characterization and seed multiplication of a collection of rice T- DNA insertional mutants 234 Identification of genes induced during the defense response of Brachiaria to the Spittlebug 245 Exploring surface charge density of root apice as potential factor contributing to alurninun resístanse of Brachiaria decumbes 251 Evaluating physiological components of acid soil adaptation in a population of Brachiaria ruziziensis x Brachiaria decumbens hybrids 253 i v 1.3.19 Identifying candidates genes whose expression is associated with aluminum resistance in Brachiaria 255 1.3.20 lsolating genes from root apices of Brachiaria decumbens that enhance Al resistance of yeast 258 V OUTPUT2 Genes and genes combínatíons made available for broadeníng the base of mandated and non mandated crops 260 2.1.1 Development of tepary x comrnon bean interspecific hybrids with improved competence to Agrobacterium mediated transformation 260 2.1.2 Progress in the Anti-Sense Mediated Silencing of the Granule Bound Starch Synthase I (GBSS D for the Production of Waxy Cassava Starch 264 2.1.3 Interspecific hybridization of comrnon and tepary bean for introgression of competence to Agrobacterium mediated transformation and insect resistance traits 268 2.1.4 Bioassays with transgenic Bt-cassava plants, cuJtivars 60444 and CM3306-4, to test for stem borer and hom worm resistance 276 2.1.5 Herbicide tolerance in cassava: Second generation of transgenic plants carrying the bar gene------------------- 281 2.1.6 RHBV (Rice Hoja Blanca Virus) Resistance mediated by RNA- Cross Protection in Transgenic Rice 282 2.1.7 Foreign genes as novel sources of resistance for fungal resistance 288 2.1.8 Use of red rice as potential trait source for comrnercial rice breeding 296 2.1.9 Field performance and fruit quality of in vitro propagated plants of Solanum quitoense (lulo) and their use as elite clone materials by farmers 300 vi Activity 2.2 Development of cellular and molecular techniques for the transfer of genes for broadening crop genetic base 306 2.2.1 Development of an In Vitre Protocol for the Production of Cassava Doubled- Haploids and its Use in Breeding 306 2.2.2 Wax.y cassava starch: Transgenic plantlets expressing gus in a vector that contains a GBSSI gene in sense orientation 31 O 2.2.3 Functional analysis of a Caffeic Acid 0-Metiltransferase gene from Brachiaria decumbens (BdCOMT) in transgenic rice as a model plant 312 2.2.4 Cassava propagation using low-cost in vitro propagation techniques and conservation of native varieties from southwestem Colombia _____ 316 2.2.5 Implementation of the encapsulation-dehydration cryopreservation method for the cassava core collection 320 2.2.6 In vitro systems ( conventional tissue culture and RITA®) to support CIA T research agenda 323 2.2.7 Embryo Rescue of Sexual Seeds from Breeding Populations for Molecular 2.2.8 2.2.9 2.2.10 Assisted Selection (MAS) of CMD Resistance 326 Dissemination of Improved Cassava V arieties as Tissue Culture plantlets _ 330 A Simple Method for the Rapid Multiplication of Clean Cassava Planting Material 333 Temporary Immersion System (RITA) for Anther Culture of Rice 337 ,,.;; Activity 2.3 Identification of points of genetic intervention and mechanism of plant stress interation 343 2.3.1 Exploring the genetic potential to improve micronutrient content of cassava_ 343 2.3.2 Stability of carotene content after altemative storage methodologies for overcoming the problem of Post-Harvest Physiological Deterioration (PPD) 343 2.3.3 Comparison of colorimetric and HPLC methods for measuring carotene content in cassava roots from 100 clones 347 2.3.4 Stability of carotene accumulation across different locations 350 2.3.5 Effect of home processing on carotenes content of cassava roots. 352 2.3.6 Inheritance of carotene content based on Sllines. 354 2.3.7 Evaluation of S 1 Farnilies for Waxy Mutants 356 2.3.8 Irradiation of Sexual Seeds for the Production of Waxy Cassava and other Mutants 360 2.3.9 Development of Populations Tolerant to Inbreeding Depression in Cassava_ 363 viii OUTPUT 3 Collaboration with public and private partners enhanced _______ 366 Activity 3.1 New Collaborative Arrangements and Networks ---------- 366 3.1.1 Update on HarvestPlus (Biofortification Challenge Program) 366 3.1.2 Biofortification: Linking The Strategic Breeding Work To Downstream Impact: Reaching And Engaging End-Users ------------ 367 3.1.3 Models ofFood Safety Assessment ofTransgenic Crops, Workshop funded by USAID and the Rockefeller Foundation, Washington DC, May 6-8,2003 _ 369 3.1.4 Updating CassavaDB, a ACEDB-type data base for Results of Genome Mapping 371 3.1.5 Report of the Second Tri-Annual Workshop of the Molecular Genetic Diversity Network of Cassava (MOLCAS) 372 3.1 .6 Progress in the Development of a Web Accessible Data Base for SSR Marker Assessment of Diversity of Local Cassava Varieties 374 Activity 3.2 Cassava Biotechnology Network's activities for 2003 _ _______ 376 3.2.1 Application of low-cost in vitro propagation techniques to conserve native varieties and produce quality cassava seed in_southwestem Colombia 377 3.2.2 Farmer participatory in vitro cleaning and multiplication of local and improved cassava varieties 383 3.2.3 Diagnosis of the use and production of cassava in Manabí province 385 3.2.4 Projects under CBN-LAC Small Grant Scheme 386 3.2.5 Genetic manipulation of proline biosynthesis in cassava aiming to increased tolerance to water stress 389 3.2.6 Development of protocols for genetic transformation of cassava genotypes from Northeast Brazil 392 3.2.7 Isolation of genes involved in the sugary phenotype in the storage root of cassava (Manihot esculenta Crantz) 394 3.2.8 Maintainance and characterization of cassava living collection at the Universíty of Brasília 397 3.2.9 Use of in vitro technology by small farmers to clean and preserve natíve cassava varieties in Southem Colombia's Andean Region 398 3.2.10 Rescue and production of high quality seed of local cassava and cocoyam varieties by biotechnology tools adapted to Cuban rural conditions 405 3.2.11 Application of new techniques to control sorne viral diseases in cassava (Manihot esculenta Crantz) in combination with massive propagation technique 407 3.2.12 Application of molecular techniques for genotypes differentiation of Manihot spp. 408 3.2.13 Establishing the contribution of Manihot leptophylla to the genetic constitution of cassava and the differentiation of sweet and bitter types 410 3.2.14 In situ conservation of cassava varieties cultivated by Kichwas from 'Alto Napo' 420 3.2.15 Cleaning of released varieties and recovery of cassava genetic resources in production area ofEcuador 421 3.2.16 Other CBN Activities 422 Activity 3.3 Database and Libraries ___________________ 426 3.3.1 3.3.2 Databases about distribution of wild relatives of crops --------- 426 Updating the database and maintaining the ceparium of the Biotechnology Research Unit ----------------------- 428 Activity 3.4 Training and Workshops ------------------ 432 3.4.1 3.4.2 National and international Collaboration -------------- 432 Training, Workshops, International and National Conferences for CIAT Personnel ________________________ 437 Actvity 3.5 Publications ---------------------- 442 3.5.1 3.5.2 3.5.3 Refereed Journals, Books ------------------- 442 Proceedings, Abstract and Others 446 Thesis 448 Activity 3.6 Projects ________________________ 450 3.6.1 3.6.2 3.6.3 Project approved or on going ----------------- 450 Project Subrnitted, in preparation and concept notes 451 Projects funded and their Donors (Oct 2002- Sept 2003) 452 X Activity 3.7 Project SB-2 Project ------ ------------- 456 3.7.1 Current SB-2 lnvestigators: Discipline, position and time fraction 456 3.7.2 Current Graduate Students 458 3.7.3 Undergraduate students (current) 459 Annex .. List of Acronyms and Abbreviations U sed in the Text OUTPUT1. Characterizing genomes of wild and cultivated species of mandated crops and associated organisms Activity 1.1 Characterization of genetic diversity 1.1.1. Gene flow studies in the bean plant model Special Project, supported by BMZ, Germany D.G. Debouck1, R. Araya2 1 SB-1 Project, 2University of Costa Rica Field work The field work is part of the monitoring of wild forms of common bean, genetically compatible with the cultivated form (landraces, commercial cultivars). lt includes: the mapping of all populations in one geographic area (either through collection consultations or direct visits), the spotting of observed cases of gene flow, and the study of the conditions by which gene flow is created and maintained. During this field work (December 2002- January 2003), we were interested in 1) searching more populations of wild common bean in the Valle Central of Costa Rica, 2) verifying the stability of wild-weed-crop complexes spotted in previous field works (1987, 1998), 3) finding more 'intermediate' materials that would deserve study by molecular markers (see section 3). The methodology followed was the one defined elsewhere to look for wild Phaseolus species (Debouck 1988). The results of the field work can be seen in the following table and figure. 1 Número Especie Fecha d/m/a Provincia, Distrito Sitio cercano Coordinates 2111 vu1gs 1511212002 San J os6, Aserrí Aserrí 84"07'W 9"52'N 1550 m 3106 vulgs 1311212002 Alajuela, Carrizal ChagUite 84°10'W 10"06' N 1510 m 3132 vulgs 1411212002 Alajuela, Zarcero Zarcero 84~3'W 10010' N 1610 m 3133 vu1gs 1411212002 Alajuela, Sabana Red. Sabana Redonda 84°14'W 10"07' N 1380 m 3134 vulg s 15/1212002 San Jos6, San Gabriel Tranquerillas 84"07'W 9"48'N 1500 m 3135 vu1gs 15/1212002 San Jos6, Tarbaca Chiroiees 84"06'W 9"48'N 1480 m 3136 vulgs 1511212002 San Jos6, San Miguel Sn Milll!el Desamp. 84"04'W 9"5I'N 1370 m 3137 vulgs 16/1212002 San Jost, San Antonio Bebedero 84°10'W 9"54'N 1600 m 3138 costar 16/1212002 San Jost, San Antonio Bebedero 84°IO'W 9°54' N 1700 m 3139 costar 16/1212002 San José, Vista de Mar Vista de Mar 83°58'W 9°58'N 1790 m 3140 vulg s 17/ 1212002 Cartago, San Rafael Parque lztarú 83°58'W 9°54'N 1750 m 3141 Ieptos 1711212002 Cartago, San Rafael Parque lztarú 83°58 'W 9"54'N 1640 m 3142 costar 18/1212002 Cartago, San Nicolás Río Taras 83°55'W 9"55' N 2000 m 3143 vulg s 18/1212002 Cartago, San Rafael Hda. Tres Ríos 83°59'W 9"54' N 1500 m 3144 costar 16/01/2003 Cartal!O, San Rafael Cerros Carpintera 83°59' W 9°54'N 1630 m 3145 xantho 15/01/2003 San José, San Antonio Bebedero 84°10'W 9"54'N 1650 m 3146 leptos 15/01/2003 San José, Aserrí Piedra de Aserrí 84"07'W 9"52'N 1550 m 3147 vulg s 15/01/2003 San José, Tarbaca El Tigre 84"06'W 9"49'N 1450 m 3148 vulg s,w 15/01/2003 San Jos6, San Miguel El Manzano 84"05'W 9"49' N 1370 m 3149 Sp. (X) 16/01/2003 Cartago, San Nicolás Quircot, sitio 31 83°56'W 9°54'N 1520 m 3150 vulg e 16/0112003 Cartal!O, San Nicolás Quircot, sitio 26 83°56'W 9°54'N 1560 m 3151 vulg w 16/01/2003 Cartal!O, San Nicolás Quircot, sitio 29 83°56'W 9°54'N 1510 m 3152 vulg e 16/01/2003 Cartago, San Nicolás Quircot, sitio 29 83°56 'W 9"54'N 1510 m 3153 vulgc 16/01/2003 Cartago, San Nicolás Quircot, sitio 29 83°56'W 9°54' N 151 0 m 3154 vulg w 17/01/2003 Cartago, San Nicolás Quircot, sitio 26 83°56'W 9°54' N 1560 m 3155 vulgs,w 17/0112003 Cartago, San Nicolás Quircot, sitio 26 83°56'W 9"54' N 1560 m 3156 vulgs,w 17/01/2003 Cartago, San Nicolás Quircot. sitio 30 83°56' W 9"54' N 1530 m 3157 Sp. (X) 17/01/2003 Cartago, San Nicolás QuircOt, sitio 30 83°56'W 9"54'N 1530 m 3158 Sp. (X) 17/01/2003 Cartago, San Nicolás QuircOt, sitio 30 83°56'W 9"54'N 1530 m 3159 xantho 13/01/2003 Alajuela, Carrizal ChagUite 84° IO' W 10"06'N 1510 m 2 • Tawna & Vill.19• O Prote rM:ASAJn ·-11> . ...-.. PNQIJal4 -· ..,. ..... JOIICI.Il ~.JOIQ.I9 ......... rNUDII• ,....,., \'llAOOI.AII.IlaAOOI """""""'· M-.ma lliiOCOVIIIC """""""'' -· llOOOOVDIA OIIOCOVIIIO - ~C.PNfNUo ,..,,, ~ •~u -· "" awo::ol.AIAM .......... """""""' V4b0 -ICA"'"" ..., .. ..., ... """'"' IWD"" IWDtJI IWDCII ,..,...., ,...,_,. """'"" """"'" ·--· 00141» """"'" ....... """'"' ..,_ QIWIJI '"""""" .......... ....... 0!11112 -INI aJStofAl. ...,_ ...,.,,. ........ I:IOf«. JI ..., .. IK»CCIVAlUU IW1IOH .............,_ ........ ........ ·~·~ RIDfiDO VAJ.aiO HOfC)C) VAU.In -· HOfCJO VAL.Ulot ....... IIDtCJO VAUJI Z1 ......,... .. .,,..,. .. ........ ....... Plt/:.N.NJI ..., - ""'""" CAT CAT CAT .,_ ... .,_ ... -... T .,.., T CAT T CAT T ......... T -- T ......... T ......... T ......... T ......... T ......... T ........ T ..... ... T -... -.. ... :=.":t. T ......... T ......... T -... T -... .,_ .,_ ... -.. .,_ ..... .,_ .. .,_ .. .,_ .. .,_ ... .,_ ... .,_ .. -... .,_ -... "'"' .,_.. T -.. T .,_ ..... .,_ .. .,_ .. .,_ .. ........ T ""' .. CAT .. :=.":t. r ,_ ... T ......... T ......... T .,_ .. T ..... ... T ,_ .... T ........ e .......... T ,.. ... T ....... T ......... T ......... T CIAT - .... .,_ .. T .,_ .. T ....... T CIAT T - r - .. ""' .. .,_.. CIAT .,..,. ... .... ..~ .... ..... .... .,_ .. -.. .... ..... .,_ .. .,_ .. -.. -.. .... ..... .,_ .. .,_ .. ..... .,_ ... ... .. .... ... -.. ... .,_ .. -.. .. .,_ .. .... -.. .... - .,_ ... -... .,_ -... -.. .,_ .. .,_ ... -... -.. .,_ .. .,_ ... .,_ .. .,_ .. -... .... Figure l. Dendrogram of relationsbips among Caribbean germplasm uncovered by microsatellite marker analysis. 13 1.1.5 Evaluation of microsatellite diversity in combined parental surveys MW Blair, MC Giralda, HF Buendia, E. Tovar SB-2 Project - CIA T Introduction The objective of this work was to characterize rnicrosatellite diversity in three sets of comrnon bean genotypes that are parents of recombinant inbred line populations at CIA T and include wild and cultivated germplasm of common beans, both Mesoamerican and Andean, which have been used as parents in the bean breeding program. Methodology The 44 genotypes used in this study are described in previous CIA T annual reports (2000, 2001) and were the parents of over a dozen mapping populations being studied at CIAT for various traits. The genotypes were grouped in 3 parental surveys that were carried out separately with common controls (DOR364, 019833) included in each survey. The genotypes were evaluated for allelic diversity with up to 150 rnicrosatellite markers (of which 65 were gene-based; 85 were genornic) depending on the survey to which they belonged (150 for survey 1, 148 for survey 11 and 97 for survey ID) as shown in Table l. The markers were amplified at different annealing temperatures according to the estímated melting temperatures of the primers. The amplification conditions are given in other parts of this annual report. Markers that did not amplify (6 in survey I, 14 in survey 11 and none in survey ill) were not considered further. The PCR products were resolved by electrophoresis for approximately one hour at 120 constant volts on silver-stained 4% polyacrylarnide gels. Microsatellite alleles were sized by comparison to the 1 O and 25 bp molecular weight standards (Promega). Null alleles were uncommon for both genornic and gene-based rnicrosatellites but were scored as rnissing bands. The discrirninating power (D) of each rnicrosatellite was calculated. Results and Discussion In all three parental surveys, the average number of alleles and discriminating power was higher for genomic rnicrosatellites (ranging from 3.7 to 5.4 alleles and 0.467 to 0 .578 discrirninating power when including monomorphic markers, and 4.4 to 6.3 alleles and 0.613 to 0.692 discrirninating power when excluding monomorphic markers) than for gene-based rnicrosatellites (ranging from 2.8 to 3.3 alleles and 0.370 to 0.481 discrirninating power when including monomorphic markers, and 3.2 to 4.1 alleles and 0.484 to 0.642 discrirninating power when excluding mono~orphic markers) (Table 1). The highest diversity was registered in parental survey 1, which contained a good mix of Andean and Mesoamerican cultivated genotypes as well as wild accessions, compared to the other two surveys. Parental survey 11 had the lowest diversity values, mainly because it was predorninantly made up of Mesoamerican genotypes only. Parental survey m 14 Table l. Average number of aUeles and discriminating power ( d) for genomic and genic microsatellites with and without monomorphic markers as evaluated in each of three parental surveys. Parental Panel Marker No.of Average No. Alleles Average Di se. Class Markers power (d) Total Poi y Mono NA w/o Mono w/ w/o w/ Mono Mono Mono Survey I geno mi e 85 66 13 6 6.29 5.42 0.692 0.578 genic 65 48 16 o 4.06 3.30 0.642 0.481 overall 150 114 29 6 5.35 4.47 0.671 0.535 Survey II genomic 83 64 14 5 4.42 3.81 0.623 0.511 genic 65 46 10 9 3.20 2.80 0.564 0.464 overaU 148 110 24 14 3.91 3.38 0.598 0.491 Survey III genomic 59 45 14 4.53 3.69 0.613 0.467 genic 38 29 9 3.83 3.16 0.484 0.370 overall 97 74 23 4.26 3.48 0.563 0.429 contained a mix of Andean and Mesoarnerican genotypes but no wild accessions. When allele number is plotted against discrimination power (d), the higher diversity of genomic versus gene-based microsatellites is evident (Figure 1). Sirnilarly, one can observe the different amount of diversity assessed in each of the three parental surveys. As shown in Figure 1, the discrimination power of each rnicrosatellite was positively correlated with the number of alleles produced at the locus (r=0.686 to 0.803). Future plans • Cross survey comparison of microsatellite diversity by correlating allele sizes found in the respective genotype panels. • Assembling of microsatellite fingerprint data into the AceDB database, BeanGenes that were described in last year' s annual report. 15 SURVEYI o 5 10 15 20 Number ol bandlng patt•rno per primer SURVEYII e: ~ 0.8 -~--~~t-l!il,_._....:...:-:~=---::::::~~=-~,:--l 0.. .§ 0.6 +--.~...a.:~...._-=-:;;,~;::--~-----~--1 j ~ 0.4 +---:-.....,-~:7-'"--~-=-==:....:--==----..,..:;;-.::-:-c=-=::--t ~ Q 02+---~~~~7---~~~~~~-=~~~~~~-; o 5 10 15 20 l'&omber <:A bandlng patterns per primer SURVEYIII 12 ~ 1.0 e. i 0.8 o a.. á 0.8 = • e e o.• 'r ~ e 02 0.0 o 5 10 15 20 Number of bandlng patlltrns per primer Figure l . Relationship between discrimination power (D) of each microsatellite and the number of alleles detected with each microsatellite in th.ree sets of parental genotypes evaluated with a total of 150 markers, of which 85 were genomic microsatellites 16 1.1.6 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Guatemala Cesar Azudia1 Luis Monte1 Charles Buitrago2, Daniel Debouck3, Martin Fregene2 1Facultad de Agronomía, Universidad de San Carlos de Guatemala, 2 SB-2 Project 3. SB-1 Project Introduction Two primary centers of diversity, one in South America and the other in Meso-America have been postulated for the genus Manihot (Roger and Appan 1973). Although severa! studies have demonstrated a likely South American origin for the cultivar (Allem, 1994; Fregene et.al 1994; Roa et al. 1997; Olsen and Schaal 1999), the diversity of cassava and its wild relati ves in Memo- America is great enough to suggest a second center in Meso- America. Besides, the potential of Meso-American diversity in cassava improvement has not been properly assessed. Three recent studies of genetic diversity in land races from South America and Meso-America (Chavariaga et. al. 1999; Fregene et. al. 2002; Raji et. al. unpublished data) have revealed unique alleles in land races from Guatemala at a frequency high enough to suggest a Meso American center of cassava diversity. The results of the three studies were based upon 6, 4, and 13 Guatemalan land races. The small sample size of the previous study could distort the allele frequencies and lead to wrong conclusions. A larger collection and SSR characterization of land races from Guatemala was therefore planned to confirm preliminary data of a Meso-American center of diversity and to secure the largely untapped diversity in Guatemala before it becomes extinct. In addition, a selection from the Guatemalan collection will be crossed to CIA T elite parents to evaluate the utility of the Meso-American diversity in cassava breeding. The present study was to confirm the high genetic differentiation between cassava land races from Guatemala and Nigeria, Brazil, and Colombia. If the uniqueness of the Guatemalan germplasm is confirmed, gene tic crosses to CIA T' s elite breeding lines will be made to test hybrid vigor and delineate heterotic pools. Plant materials are a collection of cassava from all over Guatemala anda representative group used in previous studies from Nigeria, Colombia and Brazil to confirm earlier results. It is hoped that results of the uniqueness and the utility of the Guatemalan germplasm will give collection and conservation of this germplasm in regions of Meso-America high priority (Azurdia and Gomez 2002) Methodology A collection of cassava land races was carried out all over Guatemala in May last year (Azurdia and Gomez 2002). A total of 128 accessions were collected in the departments of Baja Verapaz, Quiche, Huehuetenango, Alta Verapaz, San Marcos, Escuintla y Santa in Guatemala. See Guatemalan study on the MOLCAS web site 17 (http://www.ciat.cgiar.org/Molcas) for names of accessions. For comparison with results of previous studies, DNA from 6, 11 and 12 cassava land races from Nigeria, Colombia y Brazil respectively were included, making 4 sarnples for the analysis of genetic diversity and differentiation. DNA from the Guatemalan accessions was isolated at the Facultad de Agronomía, Universidad de San Carlos de Guatemala using a micro-prep protocol of the Dellarporta (1983) methodology and transferred to CIAT. DNA from the other accessions was obtained from previous studies at CIA T. A set of 36 SSR markers, carefully chosen to represent a bread coverage of the cassava genome with moderate to high polymorphism information content (PIC) and robust amplification, were used in this study. SSR markers, PCR arnplification, polyacrylamide gel eiectrophoresis, and siiver staining used in thís study have been described elsewhere (Fregene et al 2002). The allele data was captured using the program "Quantity One" (Bio- Rad Inc) and entered directly into EXCEL (Microsoft Inc) for statistical analysis. Statistical analysis on the raw SSR data include: genetic distance analysis using a of a distance matrix based upon 1-proportion of shared alleles (1-PSA), principal component analysis (PCA) and cluster analysis (UPGMA) of the distance matrix, and pararneters of genetic diversity and differentiation. Results A total of 33 SSR markers were analyzed in the 128 accessions from Guatemala that includes an accession of the wild relative M.aesculifolia. Unique alleles were observed in the accessions from Guatemala for the markers SSRY 12 (0.14), SSRY 20(0.383), SSRY 34(0.006), SSRY 38(0.063), SSRY 51(0.1), SSRY 59(0.014), SSRY 63(0.014), SSRY 69(0.023), SSRY 82(0.007), SSRY 103(0.24), SSRY 108(0.043), SSRY 135(0.13) y SSRY 147(0.013). In parenthesis are the frequencies of the observed alleles. The first and second principal components of the PCA, based on upon the genetic distance 1-proportion of shared alleles, are shown in Fig l. Accessions from Guatemala form two groups, one that clusters along with land races from Brazil, Nigeria and Colombia in a broad group and a second group that is clusters separately. The only sarnple from M. aesculifolia is located far away from both clusters. The results observed confirms previous observation of a high genetic differentiation of between certain groups of cassava land races from Guatemala and these from other parts of Latín America and Africa (Fregene et al. 2003) Assessment of genetic diversity was based on samples of cassava land races from the 4 countries, with an addition that accessions from Guatemala were divided into two groups G 1 and G2 based on clustering from PCA of genetic distances and UPGMA of FsT data. Table 1 surnmarizes the parameters of genetic diversity observed for accessions from the 5 samples. Genetic diversity, as assessed by the average gene diversity CHE) was high in the accessions analyzed 0.5422 ±0.2468. The population with the highest diversity was Colombia followed by the cluster G2 and that with the lowest was the cluster of the Guatemalan accessíons clustered separateiy from other accessions. Average number of alleies was 3.8+/-0.0358 for all accessions. Average number of alleles per locus was 18 highest in the cluster G2 of Guatemalan land races, 5.2, and the lowest in the Nigerian land races, 2.9. A break down of genetic diversity parameters by individual SSR markers can be seen in Table 2. Samples from · Guatemala Principal Component Analysis PC1 Manihot aesculifolia • Guatemala Btazil lx Colotrbia :X Nioeria Fig.l. Principal component analysis of genetic distances between cassava accessions from Guatemala, Brazil, Colombia and Nigeria. A UPGMA cluster analysis of the genetic distance data also produced 2 clusters of the Guatemalan accessions similar to that found with the PCA (data not shown). In addition, 2 sub groups were found within the group G 1 that clustered away from the majority of accessions. A UPGMA of a pair-wise analysis of genetic differentiation CFsT) again confmned the separation of a group of accessions from Guatemala (Fig 2) as observed with the PCA and UPGMA analysis of genetic distances. The geographical distribution of accessions in cluster G 1 can be observed in Figure 5. The distribution of accessions closely mirrors the distribution of 2 wild Manihot species in Gutemala, namely Manihot rhomboide and Manihot aesculifolia (Figure 3). The majority of accessions in sub group A are found in westem Guatemala and they overlap, as regards geographical origin, with Manihot rhomboide. On the other hand, genotypes from sub group B are found mostly in the Eastem part of the country together with natural populations of Manihot aesculifolia. 19 ~~----------------------------------------~ Cp2------, &azil- Colomlia Nigeria ------l 0.()4 0.08 0.12 Coeff!cient 0.16 0.20 Fig. 2: UPGMA tree of pair-wise FsT data calculated between samples from the four different countries. • o • , o • • Group A GroupB + M. aesculifolia .A M. rhomboidea. Fig 3: Distribution of accessions from groups A and B, of Gl and Manihot aesculifolia and Mahihot rhomboide in Guatemala . 20 The origins of highly differentiated samples of cassava germplasm from Guatemala can be explained by independent domestication events in populations of different Manihot species that yet exist or are now extinct. They can also be explained by an introgression from Manihot species in certain regions that overlap in geographical spread with cassava. Cassava is an allogamous crop and natural cross pollinate between cassava and populations of wild Manihot species has been demonstrated Wanyera et al (1993). The highly differentiated landraces from Guatemala may represent heterotic pools, like those for maize (Shull 1952). A principal reason for this study was to assess genetic diversity in cassava landraces as a first step to delineating heterotic pools for a more systematic improvement of combining ability via recurrent reciproca! selection (Keeratinijakal and Lamkey 1993). The heterotic pattems found in maize populations at the tum of the century is the basis of a very successful maize hybrid industry and has raised maize yields 500% since 1928 (Shull 1952; Tomes 1998). A high level of genetic differentiation, as revealed by molecular markers, was later found between these populations (Melchinger et al. 1990). lt is noteworthy that accessions from sub group A and B, for example accession 405, 478, 332, and 729 have excellent agronomic characteristics (data not shown) Table l. Intra-population and inter-population estimates of genetic diversity parameters of cassava and races from different agro-eclogies of Guatemala, Brazil, Colombia, Nigeria No. Average No. Average No. No. of Percent ofof of allles/Loc.HO HE HEc_p Po p. n de loci Poi. Loe. alleles/Loc. Poi. Loe. Poi. G1 24 33 28 84.8 2.8 3 0.6273 0.41 0.419 G2 74 33 33 100 5.2 5.2 0.5895 0.6066 0.6107 BRA 12 33 32 97 4 4.1 0.5562 0.5745 0.6013 COL 8 33 33 100 4.1 4.1 0.5912 0.6111 0.6555 NIG 4 33 31 93.9 2.9 3.1 0.648 0.5085 0.6067 mean 33 31.4 95.15 3.8 3.87 0.6024 0.5422 0.5786 std 6.28 0.98 0.91 0.0358 0.0845 0.0918 0.2468 Ho: observed heterozygosity; He: Average gene diversity ; Hec_p: Average gene heterozygosity corrected for small samples sizes 21 Table 2. Parameters of Genetic diversity, Ho, Hs, Ht, Dst, Gst and Gst' (correction for differences in sample size) by SSR locus Locus Ht Hs Dst Gst Dm Rst SSRY4 0.7596 0.6843 0.0753 0.0991 SSRY9 0.61 0.5767 0.0333 0.0546 SSRY12 0.7061 0.6427 0.0634 0.0897 SSRY19 0.7698 0.652 0. 1177 0.153 SSRY20 0.8235 0.7471 0.0764 0.0928 SSRY21 0.6691 0.6035 0.0655 0.0979 SSRY34 0.6554 0.4869 0. 1685 0.2572 SSRY38 0. 1665 0.1585 0.008 0.0477 SSRY51 0.7177 0.6324 0.0854 0.1189 SSRY59 0.7533 0.52 14 0.2319 0.3078 SSRY63 0.6556 0.5402 0. 11 55 0.1761 SSRY64 0.7151 0.5325 0.1826 0.2554 SSRY69 0.7428 0.6926 0.0502 0.0676 SSRY82 0.8154 0.6638 0.1517 0.186 SSRY100 0.7673 0.6648 0.1025 0.1336 SSRY102 0.2887 0.2534 0.0354 0.1225 SSRY103 0.5395 0.4988 0.0407 0.0755 SSRYl05 0.687 0.5074 0.1796 0.2615 SSRY106 0.5847 0.5 173 0.0674 0.1153 SSRY108 0.6 123 0.4013 0.21 1 0.3446 SSRY110 0.3628 0.3502 0.0126 0.0346 SSRY127 0.5545 0.5353 0.0192 0.0346 SSRY135 0.6644 0.6218 0.0426 0.0641 SSRY147 0.3403 0.3132 0.0271 0.0796 SSRY151 0.7632 0.7046 0.0587 0.0769 SSRY155 0.6983 0.6141 0.0843 0.1207 SSRY161 0.6889 0.5989 0.09 0.1306 RY164 0.7527 0.6476 0.!052 0.1397 RY169 0.6498 0.5982 0.0516 0.0794 SSRY171 0.3128 0.2801 0.0326 0.1044 SSRY179 0.5225 0.5018 0.0208 0.0397 SSRY180 0.6236 0.565 0.0587 0.0941 SSRY181 0.6305 0.5831 0.0473 0.0751 Mean 0.6244 0.5422 0.0822 0.1252 0.0822 0 .1516 Std 0.1619 0.1397 0.0589 0.0791 95% CI 0.5627 0.4793 0.064 0.1009 95% CI 0.6743 0.587 0.1035 0.1514 Ho Average observed heterozygosity with.in country Ht Total Heterozygosity in the entire data set Hs Gene diversity with.in country averaged over the entire data set Dst Average gene diversity between populations Gst Coefficient of gene differentiation 22 Conclusions A previous study of the assessment of genetic diversity of cassava land races in 14 South and Central American and African countries revealed a number of unique alleles in accessions from Guatemala and suggests a second center of diversity in Guatemala (Fregene et al 2003). Meso America is a center of diversity for many other food crops including common beans, maize, amongst others. This study shows unique alleles from Guatemala for a higher number of SSR markers and provides additional evidence for possible independent domestication of cassava in Meso America. However, an introgression with Manihot species that overlap with the geographical origins of these accessions makes it impossible to rule out introgression with these species. Further studies are required to clarify which is the most likely scenario via the collection and characterization of wild Manihot species the eastem and westem parts of Guatemala. Additional future activities include a diallel cross of theses excellent land races from Guatemala and other regions (initial experiments are described below in activity 8) . References Azurdia, C. y M. Gonzalez. 1986. Informe final del proyecto de recoleccion de algunos cultivos nativos de Guatemala. FAUSAC, ICfA, CIRF. Azurdia, C.;K Willearns; D. Williams; A. Jarvis (2003). Inventario de las especies silvestres parientes de los cultivos natives de Guatemala. FAUSAC, USDA, IPGRI. En preparación Chavarriaga-Aguirre, P., M. M. Maya, J. Tohme, M. C. Duque, C. Iglesias, M. W. Bonierbale, S. Kresovich, And G. Kochert, 1999 Using núcrosatellites, isozymes and AFLPs to evaluate genetic diversity and redundancy in the cassava core collection and to assess the usefulness of DNA-based markers to maintain gerrnplasm collections. Molecular Breeding 5: 263-273 Dellaporta SL, Wood J, Hicks JR (1983) A plant DNA núnipreparation: version Il. Plant Mol Biol Rep 1: 19- 21. Dje Y., Mheuertz, Clefebvre, X Vekemans. 1999. Assessment of genetic diversity within and among germplasm accessions in cultivated sorghum using núcrosatellite markers. Theor Appl Genet Fregene M., Suarez M., Mkumbira J. , Kulembeka H. , Ndedya E., Kulaya A. , Mitchel S. Gullberg U., Rosling H., Dixon A., Kresovich S. (2001) Simple Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta Crantz) Landraces: Genetic Structure in a Predorninatly Asexually Propagated Crop (Theor Appl Genet, in review). Fregene MA, Vargas J, Ikea J, Angel F, Tohme J. Asiedu RA, Akoroda MO, Roca WM (1994) Variability of chloroplast DNA and nuclear ribosomal DNA in cassava (Manihot esculenta Crantz) and its wild relatives. Theor Appl Genet 89: 719-727. Melchinger AE, Lee M, Lamkey KR, Woodman WL (1990) Genetic diversity for restrictionfragment length polymorphisms: relation to estimated genetic effects in maize inbreds. Crop Sci 30:1033-1040 Olsen K and Schaal B. 1999. Evidence on the origin of cassava: phylogeography of Manihot esculenta. Proceedings of the National Academy of Science 96: 5586-5591. 23 Roa, A.C.; M.M. Maya; M.C. Duque; J. Tohme; A.C. Allem; M.W. Bonierbale. 1997. AFLP analysis of relationship among cassava and other Manihot species. Theor Appl Genet 95: 741-750. Shull GF (1952) Beginnings of the heterosis concept. In: JW Gowen. Heterosis. Iowa Sate College Press, Ames, Iowa, pp 14-48 Tomes D (1998) Heterosis: performance stability, adaptability to changing technology and the foundation of agriculture as a business. In: K Lamkey, J E Staub. Concepts and breeding of heterosis in crop plants. CSSA Special Publication Number 25. Crop Science Society of America, Madison, Wl. Wanyera, N. 1993. Phylogeny of two Manihot species and their natural hybrids. Ph.D. thesis University of Ibadan, Nigeria. 1.1.7 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Ghana and Predictability of Heterosis Elizabeth Okai1, Dr John Otoo1 Charles Buitrago2, Martín Fregene2 Dr Alfred Dixon3 1Crop Research Institute, CRI, Kumasi, Ghana; 2CIA T ; 3IIT A Introduction Cassava is the number one provider of calories in Ghana. The Portuguese first introduced cassava from Brazil to Ghana during the 161h and 17th centuries (Jones, 1959). In the then Gold Coast, the Portuguese grew the crop around their trading ports, forts and castles. It was a principal food eaten by both the Portuguese and the slaves. By the second half of the 18th century, cassava had become the most widely grown crop of the people of the coastal plains (Adams 1957). The spread of cassava from the coast into the hinterlands was very slow. It reached Ashanti region, Brong Ahafo and the northem Ghana, mainly around Tamale in the 1930. Until the early 1980s,the Akans of the forest belt preferred plantain and cocoyams and sorghum and millet in the north. Cassava became firrnly established in most areas after the serious drought of 1982/83 when all other crops failed completely (Korang-Amoakoh,Cudjoe and Adams 1987). Cassava ranks first in the area under crop cultivation and utilization. It contributes 22% of the agricultura! gross domestic product AGDP compared to 5% for maize, 2% for rice, and 14% for cocoa (Al-Hassan, 1989;Dapaah, 1996). According to the Ghana Living Standards Survey (GLSS), 83% of 1.73 million sampled households were found engaged in cassava production. The spread of cassava into the upper west and upper east of Ghana is an indication of growing trend in cassava production through area expansion( MOA, 1990). In the traditional bush-fallow system, sorne cassava plants are allowed to grow during the fallow period, which is long enough to allow cassava to flower and set seeds. The usual out crossing habit of cassava leads to the production of numerous volunteer hybrids and a 24 heterozygous gene pool, which creates phenotypic cliversity. Desirable new hybrid combinations from volunteer seed are often selected by farmers and propagated. This process creates pools of new land races, which are adapted to the different agro-ecological zones of Ghana. Farmers have done selection for desirable traits for over 500 years. Hence the landraces possess higher frequencies of genes required for adaptation to biotic and abiotic stresses, food quality characteristics than unadapted materials. Vegetative propagation also leads to the accumulation of pest and cliseases and good varieties susceptible to these biotic stresses disappear. These factors lead to a fairly high turnover of varieties and have implications for gene pool structure of cassava in any center of diversity. Selection is one of the principal factors at work in cassava' s gene differentiation in Africa. High heterosis for yield components, starch, and number of roots have been observed in cassava, and hence considered a promising method of genetic improvement (Easwari Amma and Sheela, 1996). Heterotic groups identified in mai~e in the early 20th century (Shull et al 1953) have been the basis of a very successful hybrid seed industry. Objectives for the study were: The objective for this study is to assess the genetic cliversity and differentiation in Ghanaian landraces To detect heterotic pattems in the collection and between the Ghanaian collection and land races from other countries and regions. To generate hybrids between the Ghanaian land races and genotypes from putative heterotic groups and select together with farmers superior hybrids from the crosses. Methodology In January 2002 a collection of cassava land races from all the agro ecological zones in Ghana was done. A total of 45 villages visited during the collaborative study on cassava in Africa (COSCA) were visited. Another 28 villages, important for cassava production, were also visited. Farmers were assembled and asked to share information on cassava varieties grown by them, characteristics of their varieties, and reasons for keeping them. Farmers volunteered to give mature cassava stems, which were labeled. A total of 320 land races were collected. For the list of genotypes, passport data and characteristics please see the MOLCAS home page (http://www.ciat.cgiar.org/molcas) Fresh young leaf samples of the accessions were collected on ice and used for DNA extraction. An amount of 0.1g of the fresh young leaf was ground in liquid nitrogen and the DNA extracted using the Qiagen kit. The extraction was carried out in liTA, Ibadan, Nigeria. The DNA was carried in absolute ethanol to CIAT. DNA quantification was done using the fluorometer. The DNAs were diluted to 10ng/ul and used for SSR reactions. A sub-set of 36 SSR markers, two from each of the 18 linkage groups of the cassava geno me, was employed to obtain an es ti mate of genetic di versity and differentiation in the land races. PCR amplification, automated gel analysis and date collection were as descried by Fregene et al (2003). Genetic clistance, based upon the proportion of shared alleles 25 (PSA), obtained from the raw allele size data using the computer microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html). Distances between the accessions were subjected to principal component analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of relationship between the land races. Parameters of genetic diversity and differentiation were calculated from allele data using the computer packages GENSURVEY (Vekeman et al1997) and FSTAT (Goudet 1990). Results A total of 320 landraces were collected including 18 genotypes with yellow roots. Farmers who responded were predominantly women. Among the land races sorne were very early bulking with maturity at 3-9 months after planting. The various local names given suggest a lot of useful traits farmers had associated with the varieties. Woody stems were cut to 20-30cm sizes and planted in plastic pots in a nursery. These were sent to the field after 4weeks and planted in an irrigated field at the Ashiaman office of the Ghana Irrigation Authority. A copy of the collection was packaged and sent to liT A. Accessions were planted in single rows at 1m x 1m spacing with improved varieties as checks. Data from a total of 33 of 36 SSR loci, 3 markers gave poor quality data, was used to derive estímate of genetic diversity and differentiation genetic distances between individual genotypes. The average number of alleles for each locus was close to 5 and is similar to that found for a study of land races from Nigeria, Tanzania and 7 Neo-tropical countries (Table 1). The probability that 2 randomly selected alleles in a given accession are different, average gene diversity, was 0.5245±0.0045 and it is lower than that found in the found for the previous study. Average gene diversity was comparable across all regions with an exception of the central region and central savannah region. Genetic diversity parameters, including total heterozygosity (Ht) and genetic differentiation (Gst) ranged widely across markers. Genetic differentiation, as estimated by FsT (theta), was very low for samples between regions, overall 0.04, with the exception of sorne accessions from Northern Ghana that showed moderate to high genetic differentiation. (data not shown). The results found here support previous findings that agricultura! practices and the allogamous nature of cassava produces a large pool of volunteer seedlings that natural and human selection acts upon to maintain a high leve! of diversity and low differentiation (Doy le et al. 2001; Fregene et al. 2002). 26 Table 1 Intra-population and inter-population estima tes of genetic diversity parameters of cassava land races from different agro-eclogies of Ghana Population N #loe. #loc_P PLP K K_P HO_p HE_p HEc_p Ashanti 11 33 30 90.9 3.9 4.1 0.5285 0.5017 0.5262 Brong Ahafo 37 33 31 93.9 5.4 5.6 0.5012 0.5267 0.5339 Central 8 33 29 87.9 3.3 3.7 0.5082 0.4701 0.4999 Eastem 27 33 30 90.9 5.4 4.9 0.5322 0.5123 0.5223 Coastal Sav. 4 33 27 81.8 2.9 3.4 0.5404 0.467 0.5405 Greater Acera 10 33 29 87.9 3.7 4 0.5016 0.5065 0.5336 Volta 28 33 31 93.9 5.3 5.5 0.5133 0.5735 0.5839 Northem 1 109 33 31 93.9 6.9 7.2 0.5369 0.5779 0.5806 Northenll 53 33 31 93.9 6.3 6.5 0.5479 0.5851 0.5908 mean 90.57 4.69 4.98 0.5234 0.5245 0.5457 std deviation 4.13 1 0.36 1.31 0.0176 0.045 0.03"17 PLP: Percentage of polymorphic loci at the 5% leve! within accessions K: Mean number of alleles per locus within accessions K_P: Mean number of polymorphic alleles per locus within accessions Ho_p: observed heterozygosity HE-p: Average gene diversity HEc_p: Average gene heterozygosity corrected for small samples sizes Genetic d.istances between all pairs of individual accessions was calculated by the 1- proportion of shared alleles (1-PSA) and presented graphically by a principal coordinate analysis (PCA) (Figl). The PCl and PC2 accounted for 26% and 16% of the total variance respective! y. The PCA shows loose clustering of the land races by region but of note is the sub-structure of sorne land races from Northem Ghana. A similar sub-structure in accessions from Nigeria and from Tanzania was observed in earlier stud.ies. The presence of a defined sub-structure in the genetic relationship of cassava land races from Africa appears to be a common feature of cassava germplasm in a number of countries but it is yet to be understood the underlying factors for the groupings. UPGMA cluster analysis of FsT estímate of genetic differentiation arnongst land races was able to group the land races into loose clusters according to agro-ecologies, with a group of genotypes from the Northem region sub-structure being the most differentiated (Fig 2). (Fregene et a1.2002). At least 63 duplicates or closely related accessions were identified in the collection. 27 Table 2. Parameters of Genetic diversity, Ho, Hs, Ht, Dst, Gst and Gst' (correction for differences in sample size) by SSR locus LoeN ame Ho Hs Ht Dst Dst' Ht' Gst Gst' SSRY4 o 0.346 0.439 0.093 0.104 0.45 0.212 0.23 SSRY5 0.499 0.474 0.481 0.006 0.007 0.481 0.014 0.015 SSRY9 0.463 0.582 0.587 0.005 0.006 0.587 0.009 0.01 SSRY12 0.704 0.597 0.598 0.001 0.001 0.598 0.001 0.001 SSRY19 0.811 0.738 0.764 0.026 0.029 0.766 0.034 0.037 SSRY20 0.79 0.73 0.76 0.03 0.033 0.764 0.039 0.043 SSRY21 0.596 0.487 0.5 14 0.027 0.03 0.517 0.053 0.058 SSRY34 0.479 0.428 0.425 -0.004 -0.004 0.424 -0.009 -0.01 SSRY38 0.043 0.08 0.082 0.001 0.002 0.082 0.018 0.02 SSRY47 0.429 0.668 0.739 0.07 1 0.079 0.747 0.096 0. 106 SSRY51 0.79 0.694 0.75 1 0.057 0.063 0.757 0.076 0.084 SSRY52 0.752 0.603 0.616 0.014 0.015 0.618 0.022 0.025 SSRY59 0.152 0.639 0.701 0.062 0.069 0.708 0.089 0.098 SSRY63 0.445 0.484 0.519 0.036 0.04 0.523 0.069 0.076 SSRY64 0.725 0.67 0.689 0.02 0.022 0.691 0.028 0.031 SSRY69 0.557 0.552 0.568 0.016 0.018 0.57 0.028 0.031 SSRY82 0.846 0.84 0.858 0.018 0.02 0.86 0.021 0.024 SSRYlO 0.725 0.779 0.798 0.019 0.021 0.8 0.024 0.027 SSRY10 0.007 0.01 0.009 o o 0.009 -0.037 -0.041 SSRY10 0.804 0.76 0.764 0.004 0.004 0.764 0.005 0.005 SSRYlO 0.829 0.761 0.768 0.008 0.009 0.769 0.01 0.011 SSRYlO 0.422 0.361 0.373 0.012 0.014 0.375 0.033 0.036 SSRY11 0.279 0.272 0.274 0.002 0.002 0.274 0.008 0.009 SSRY12 0.814 0.629 0.653 0.024 0.026 0.656 0.036 0.04 SSRY14 0.08 0.083 0.086 0.002 0.003 0.086 0.029 0.032 SSRY15 0.689 0.79 0.806 0.015 0.017 0.807 0.019 0.021 SSRY15 0.072 0.595 0.632 0.037 0.041 0.636 0.059 0.065 SSRY16 0.528 0.658 0.668 0.01 0.01 1 0.669 0.014 0.016 SSRY16 0.258 0.316 0.321 0.004 0.005 0.321 0.013 0.015 SSRY17 0.544 0.555 0.591 0.037 0.041 0.596 0.062 0.068 SSRY17 0.845 0.726 0.776 0.05 0.056 0.781 0.065 0.071 SSRY18 0.672 0.532 0.535 0.003 0.004 0.536 0.006 0.007 SSRY18 0.605 0.705 0.761 0.055 0.062 0.767 0.073 0.08 Overall 0.523 0.55 0.573 0.023 0.026 0.575 0.04 0.045 Ho Average observed heterozygosity within country Ht Total Heterozygosity in the entire data set Hs Gene diversity within country averaged over the.entire data set Dst Average gene diversity between populations Gst Coefficient of gene differentiation. 28 o 0.. PCA of gen e ti e distan ces (PS A) PC1 XGtAccra e Northem Regial +Vota regia~ •EA -NS Fig. 1: Principal component analysis of genetic distances, based 1· proportion of shared aUeles, between cassava accessions from Ghana. One of the principal reasons for this study was to assess genetic diversity in cassava land races as a first step to delineating heterotic pools for a more systematic improvement of combining ability via recurrent reciprocal selection. Based upon clusterings obtained above, genotyopes representative of the clusters were selected as parents for a diallel experiment to search for heterotic pattems and established in a crossing block at the CRI experimental station in Wenchi. Genotypes from other parts of Latin America, for example Guatemala, that cluster away from African accessions were also shipped from CIAT to Ghana for the study of heterotic pattems. The Latin American genotypes were multiplied, hardened and planted in the same crossing block where the land races are currently planted. 29 Ashanti Central Brong Ahafo Eastem Coast Savannah Volta Greater Acera Northem III Northem I 1 1 - r- T .1 - - 0.02 0.06 Coefli:imt 0.11 Fig. 2: UPGMA tree of pair-wise F ST data calculated between samples from different regio os of Ghana. Conclusions A total of 320 land races were collected and established at the University of Legan and llT A, Ibadan. Genetic diversity and differentiation in the collection was assessed using SSR markers. Discovery of a sub group within the land races, as observed from PCA of genetic distances (proportion of shared alleles) and UPGMA of pairwise FST between the different regions, that may represent heterotic groups. To test heterotic patterns present within the collection or between the collection and Latín American accessions, selected genotypes were planted in a crossing block for making genetic crosses. References Adams, C.D. L957. Activities of Danish Botanists in Guinea 1738-1850. Transactions of the Historical Society of Ghana III. Part l. 30 0.13 Al-Hassan, R. 1989. Cassava in the Economy of Ghana. In : Status of Cassava research in Africa, COSCA working paper No. 3, Eds, F. I. Nweke, J. Lynam and C. Y. Prudencia, Intemational Institute of Tropical Agriculture, Ibadan, Nigeria. Dapaah, S. K. (1996). The way forward for accelerated agricultura! growth and development. A paper presented to the Govemment of Ghana on behalf of the Ministry of Food and Agriculture. 6pp. Fregene M., Suarez M., Mkumbira J., Kulembeka H., Ndedya E., Kulaya A., Mitchel S. Gullberg U., osling H., Dixon A., Kresovich S. (2003) Simple Sequence Repeat (SSR) Diversity of Cassava Manihot esculenta Crantz) Landraces: Genetic Diversity and Differentiation in a Predorninatly sexually Propagated Crop Theor Appl Genet 107:1083-1093 Goudet J, 1995. FSTAT (vers. 1.2): a computer program to calculate F-statistics. J. Hered. 86: 485-486 . . Jones, W. O. 1959. Manioc in Africa. Stanford University Press 1959. Korang-Amoakoh, S. , Cudjoe, R. A. and Adams, E. 1987. Biological Control of cassava pests in Ghana. Prospects for the integration of other strategies. In Ministry of Agriculture. 1990. Medium Term Agricultura! Development Programme (MTADP) 1991-2000: An agenda for sustained agricultura! growth and development 1991 - 2000. Ministry of Agriculture, Acera. National Agricultura} Research Strategic Plan, Ghana, 1994. Final Report. 181pp. National Cassava Task Force. 1996. Final Report on the promotion of cassava production, processing and marketing in Ghana. 39pp. with appendices. SAS Institute, Inc., 1995. JMP (version 3.1). SAS Inst. Inc., Cary, NC. Shull G.F. 1952. Beginnings of the heterosis concept. P 14-48. In: JW. Gowen (ed) Heterosis, Iowa Sate College Press, Ames. Vekemans X. Lefebvre C., 1997 On the evolution of heavy-metal tolerant populations in Armería marítima: evidence from allozyme variation and reproductive barriers. J. Evol. Biol., 10: 175-19 31 1.1.8 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Uganda E. Kizito1, U. Gullberg1; A. Bua2, J. Omara2, M. Egwang3; W. Castelblanco4, J. Gutierrez4, C. Buitrago4, M. Fregene4 1Swedish Agricultura! University, Uppsala; 2Cassava program, Namunlonge Agricultura! and Animal Research Institute, Kampala, Uganda; 3.Medical Biotech Laboratories, Kampala, U ganda; Introduction Cassava is grown mainly for its starchy roots and the leaves and forms a staple by an estimated half of Uganda's population (Bua, pers.comm, 2001). Cassava was introduced into Uganda after 1850 by Europeans and Arabs (Langlands, 1970). Because of its excellent adaptability to erratic rainfall and low fertility soils, it became a major dietary staple, a famine reserve crop and a source of cash to many small-scale farming comrnunities. The impact of a cassava mosaic disease (CMD) epidemic in Uganda in the late 80's and early 90's on genetic diversity grown by small farmers is thought to be significant. In a survey carried out in 2000 in Mukono, Soroti and Apac districts in Uganda, the impact of the CMD epidemic on cassava diversity was clearly seen via the loss of previously well-known varieties (Kizito and Gullberg 2002, unpublished data) . But it was also observed that additional genetic variability had arisen from the use of volunteer seedlings by sorne farmers in Mukono district. Traditional farming systems of slash and bum followed by 3-15 years of fallow have been known to encourage the allogamous nature of cassava producing a large pool of volunteer seedlings that natural and human selection acts on to produce new varieties which maintains a high diversity, for instance in Tanzania (Fregene et al, 2003). We assessed the genetic diversity and differentiation based on SSR markers of landraces from all over Uganda and a small subsection from Latín America and other African countries. The objectives of this study were to: assess the genetic diversity and differentiation of cultivars within and between different districts in Uganda; also to determine how the Uganda cassava diversity compares with the total genetic diversity of species within Africa and the cassava collection after the CMD epidemic. Methodology A total of 257 local cassava varieties were collected accessions September through to December 2002 in 17 districts that líe between latitudes N02° 121 and S 00° 441, longitudes E029° 561 and E034° 21 1, and altitudes of 4451ft and 2177ft above sea level in Uganda. Three counties on average in each district were selected at random and fields with mature crop were sampled from every 7-10 km along the roads that traversed each of the counties. In each farmer's field the different varieties were identified according to their 32 morphological characteristics as well as by the name given by the fanner. Where no single variety dominated, plants of the co-dominant varieties were sampled, labelled to be planted and maintained in NAARI. Another 20 accessions, representatives of diversity, were selected from Tanzania, and 18 accessions South America from a previous study of cassava accessions from Tanzania and the neo-tropics (Fregene et al. 2003). In addition to this, 20 from the Ghanaian germplasm bank, 20 from Nigeria and 20 from Guatemala was included. In all, 350 accessions were studied. DNA isolation was from young leaf tissue harvested by the CfAB method (Doyle & Doyle, 1987) at Med Biotech Laboratories, Kampala. A subset of 36 SSR markers with high polymorphism information content (PIC) routinely used for diversity analysis in cassava was source of markers, the 36 markers ha ve been chosen to represent a wide coverage of the genome. PCR amplification, automated gel analysis and date collection were as descried by Fregene et al (2003). Genetic distance, based upon the proportion of shared alleles (PSA), obtained from the raw allele size data using the computer microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html). Distances between the accessions were subjected to principal component analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of relationship between the land races. Parameters of genetic diversity and differentiation were calculated from allele data using the computer packages GENSURVEY (Vekeman et al 1997) and FSTAT (Goudet 1990). Table 1: Genetic diversity within groups of cassava landraces classified according to country of origin. Standard deviations (SD) were estimated by jackknifmg over loci (200 replications). H" Hn Dm and Gs: are given over loci and over groups (country populations). Population Sample No.of No.of Percent Mean no. Mean no. H., e H.d H..> . Size loci poi. Ofpolb. Alleles alleles/ Lod loci llocus (!O lb .locus UGANDA 198 35 33 94.3 5.2 5.4 0.5530 0 .5454 0.5468 COLOMBIA 5 35 33 94.3 3.3 3.4 0 .5081 0.5363 0.5963 BRASIL 3 34 33 97.1 2.8 2.8 0.5735 0.5069 0.6304 PERU 3 35 33 94.3 2.7 2.8 0.5810 0.5218 0 .6619 GUATEMALA! 7 35 33 94.3 2.5 2.6 0.5290 0.3908 0 .4219 GUATEMALA2 11 35 34 97.1 3.8 3.9 0.5274 0.5640 0 .5906 TANZANIA 19 35 32 91.4 3.9 4.1 0.5658 0.5386 0.5536 NlGERIA 20 35 33 94.3 3.9 4.0 0.5002 0.5002 0.5131 GHANA 19 35 33 94.3 4.2 4.4 0.5429 0 .5542 0.5694 Mean 94.59 3.59 3.71 0.5423 0.5176 0.5649 Std 1.70 0.86 0.89 0.0285 0.0519 0.0698 H, H, o. G. Mean 0.6305 0.5635 0.0670 0.1078 Std 0.1696 0.1606 0.0332 0.0502 95%Cl 0.5713 0.5083 0.0566 0.0916 99%CI 0.6827 0.6135 0 .0767 0.1235 • H,= total heterozygosity in the entire data set; H,= heterozygosity within country averaged over the en tire data set; 0 51= average gene diversity between populations; G .. = coefficient of gene differentiation. b poL =polymorphic "H..= average observed heterozygosity within country; dHc= average expected heterozygosity within country •He-p= average expected heterozygosity within country corrected for small sample sizes {Nei, 1978) 33 Results A total of 35 of 36 SSR Ioci was used to provide estimates for the genetic diversity and differentiation of 350 accessions of cassava accessions from Uganda and eight other countries, one elirninated for being monomorhphic. The number of alleles observed at each locus in the data set ranged from 2 to 12 alleles per locus over the 35 loci. The average gene diversity, He, for the entire 350 accessions was more than half 0.5649± 0.0698, average gene diversity of accessions from Uganda was 0.5530 (Table 1). However, only 1% (Gs1=0.0192± 0.0511) of within district variation in Uganda was dueto differentiation. On the other hand, 10% (Gs1=0.1078±0.0502) of the overall heterozygosity (H1=0.6305± 0.1696) in all the country accessions could be attributed to differentiation among the samples from both Africa and Latín America, revealing that germplasm within Uganda is at the moment quite uniform. When accessions from Uganda were divided according to districts, the least values for average gene diversity were observed in Lira and Luweero districts, 0.4011 andO. 4219 respectively, while Kasese stands out with the highest value of 0.6208. This affirms earlier findings of higher varietal diversity in the westem and southwestem districts of Uganda as opposed to those in the eastem districts (Otim-Nape et al, 2001). The relatively high level of genetic diversity observed on the whole in this study is unexpected considering Uganda has reported two major cassava mosaic disease (CMD) epidernics before. The most recent CMD epidernic being in the last 10-15 years, since 1988, affected drastically the cassava varietal composition and saw a decrease in area planted to cassava at its peak between 1990-1994 (Otim-Nape et al, 2001). This finding continues to demonstrate the fact of active involvement of Ugandan farmers in continuous testing and the adaptation of new planting materials to their unique situations. The importance of volunteer seedlings in the dynarnics of cassava diversity has been demonstrated in traditional farming systems of the Makushi Indians from Guyana (Elías et al, 2001). Genetic distance between accessions based on 1- proportion of shared alleles (1-PSA) was calculated and presented graphically by a principie coordinate analysis (Fig. 1). The PCl and PC2 components accounts for about 39% and 10% of the total variance respectively. The PCA reflects a loose separation between accessions from Africa and the Neotropics as has been shown in earlier studies (Fregene et al. 2003). No distinct substructure was found amongst the Ugandan accessions except for accessions from Nakasongola district. Sub- structures in diversity have been reported for earlier studies for accessions from Ghana, Tanzania and Nigeria. Genetic differentiation averaged over all locí estimated by Fs1 (theta) was 0.103±_0.009 (jackknifing) and 0.082± 0.126 calculated by bootstrapping at 99% confidence interval (table 4). This agrees with previous diversity studies in Tanzania (Fregene et al, 2003). Pair wise calculations of Fs1 (theta) over all loci between pairs of country landraces and U ganda al so revealed lower differentiation between African countries compared with Latín American countries, the lowest being between Ghana and Uganda (0.039) and the highest being between a group of accessions from Guatemala group and Nigeria (0.2631). 34 A dendogram of landraces for UPGMA of pair wise Fst estimates separates the African from Neotropical accessions with the group of accessions from Guatemala being the most genetically differentiated (Fig.2). These results agree with sorne previous studies on which a high differentiation has been observed among certain cassava groups cultivated in Guatemala and those in other parts of Latin America and África (Fregene et al, 2003; CIAT, 2003). Of particular interest to cassava breeding programs is the group {Gl) from Guatemala, Accessions from G 1 from Guatemala is a representation of the region East and South of Guatemala and it may represent a heterotic group b~ed on differences in allele frquncies. The phenomenon of heterosis or hybrid vigor is an important factor in improvement of heterozygous crops such as cassava, and in cases like the com, the pattems found in these populations at the beginning of the XX century have been the base of a very sucecssful industry of hybrid com, elevating productivity by more than 500% (Shull, 1952; Tomes, 1998). Conclusions Characterization of the cassava landraces of Uganda with 35 markers and estimation of genetic diversity and differentiation Assessment of the genetic relationship between different countries of Africa and Latín America. Evidence of high differentiation between sorne land races from Guatemala and the rest of the countries that may represent heterotic pools. Evidence of low impact in cassava diversity after of epidemic of CMD and low structure in cassava landraces of U ganda. 35 References CIAT (2003) Assessment of simple sequence repeat Diversity of cassava Iandraces in Africa y Latín America: Progress report of the Ghana, malawi, Uganda, Guatemala, and brazil country studies. Report of activities by the cassava molecular diversity network (MOLCAS) Doyle JJ and Doyle JL (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin. 19:11-15 Elias M, Penet L, Vindry P, Mckey D, Panaud O, Robert T (2001) Unmanaged sexual reproduction and the dynamics of genetic diversity of a vegetatively propagated crop plant, cassava (Manihot esculenta Crantz), in a traditional farming system. Molecular Ecology 10:1895-1907 Fregene MA, Suarez M. Mkumbira J, Kulembeka H., Ndedya E, Kulaya A, Mitchel S, Gullberg U, Rosling H, Dixon AGO, Dean R, and Kresovich S (2003) Simple sequence repeat marker diversity in cassava landraces: Genetic diversity and differentiation in an asexually propagated crop. Theor. Appl. Genet. 107:1083-1093 Goudet J. ( 1995) FST AT (Version 1.2): A computer program to calculate F-statistics. J. Hered 86:485-48 Langlands BW ( 1966) Cassava in Uganda 1860-1920. Uganda Journal 30:211-218. Mba REC, Stephenson P , Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel K, GaJe M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: towards an SSR-based molecular genetic map of cassava. Theor Appl Genet:21-31 Otim-Nape GW, Alicai T, Thresh M (2001) Changes in the incidence and severity of Cassava mosaic virus disease, varietal diversity and cassava production in Uganda. Ann. Appl. Biol 138:313-327. Rohlf FJ (1993) NTSYS-PC numerical taxonomy and multivariate analysis system. Version 1.8 Exeter Publ., Setauket, NY. Shull, G.F. 1952. Beginnings of the heterosis concept. In: JW Gowen. Heterosis. Iowa State College Press, Ames, Iowa. Pp 14-48 Tomes D (1998) Heterosis: performance stability, adaptability to changing technology and the foundation of agriculture as a business. In: Lamkey K, Staub J E (eds). Concepts and breeding of heterosis in crop plants. CSSA Special Publication Number 25. Crop Science Society of America, Madison, WL Vekemans X., Lefebvre C (1997) On the evolution of heavy-metal tolerant populations in Armeria marítima: evidence from allozyme variation and reproducive barriers. J. Evoi.Biol 10:175-191 38 1.1.9 Simple Sequence Repeat (SSR) Assessment of Gene tic Diversity of Local Cassava Varieties from Sierra Leone A.Dixon1, B. Raji1 J. Marin2, C. Ospina2, C. Buitrago2, M.Fregene 11ITA, Nigeria; 2CIAT Introduction Cassava is an important staple~g crop in Sierra Leone, second only to rice in importance. Civil strife in the recent past has .lead to an erosion of genetic resources and loss of sorne very valuable germplasm. A collection was conducted by the National Program in collaboration with liTA in the Southem and Eastem provinces where cassava is or more importance to the populace, to safe guard local varieties for the future. The germplasm collection from Sierra Leone is being held at llT A Ibadan pending when peace fully retums to the country and the collection can be retumed to the National root crop program. The collection was analyzed with SSR markers as part of MOLCAS efforts to characterize diversity found in local African varieties compared to what exists in Latin America. Methodology Forty villages in the Eastem and Southem province of Sierra Leone were visited in 2001 for the collection of cassava germplasm. In each village farmers were invited to share their most important varieties, between 3 and 4 varieties were collected from each village. The stakes of all accessions collected was established in the field at llT A. DNA was isolated from leaf tissues using a DNA isolation kit (QIAGEN Gmbh) at liTA and carried to CIAT for SSR analysis. A subset of 36 SSR markers with high polymorphism information content (PIC) routinely used for diversity analysis in cassava was source of markers. The markers have been chosen to represent a wide coverage of the genome. PCR amplification, automated gel analysis and date collection were as descried by Fregene et al (2003). Genetic distance, based upon the proportion of shared alleles (PSA), obtained from the raw allele size data using the computer microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html). Distances between the accessions were subjected to principal component analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of relationship between the land races. Parameters of genetic diversity and differentiation were calculated from allele data using the computer packages GENSURVEY (Vekeman et a11997) and FSTAT (Goudet 1990). Results A total of 127 local cassava varieties were collected in from 40 villages. Thirty three SSR markers were analyzed in 98 accessions from Sierra Leone, the remainder accessions were not available at the time of SSR analysis. The average number of alleles for each locus was close to 5 and is similar to that found for a study of land races from Nigeria, Tanzania and 7 Neo-tropical countries (Table 1). Genetic distances between all pairs of individual 39 accessions was calculated by the 1-proportion of shared alleles (1-PSA) and presented graphically by a principal coordinate analysis (PCA) (Fig1). The PCA shows a sub- structure in diversity of the local varieties as have been observed in other collections from Africa. The presence of a defined sub-structure in the genetic relationship of cassava land races from Africa appears to be a common feature of cassava germplasm in a number of countries but it is yet to be understood the underlying factors for the groupings. The sub structure observed in the PCA was the basis of further analysis of genetic diversity parameters carried out. Average number of alleles per SSR locus in the collection was roughly 4 and average gene diversity, was 0.5749±0.0690 it is comparable to that found in previous studies in severa! African countries. Genetic differentiation, as estimated by FsT (theta), between the groups ranged from 0.088 to 0.140 which is low to moderate differentiation. The results found here support previous findings that agricultura! practices and the allogamous nature of cassava produces a large pool of volunteer seedlings that natural and human selection acts upon to maintain a high level of diversity and low differentiation (Doyle et al. 2001; Fregene et al. 2002). PCA of Gene tic Dlatanc:ea from SSR Marker data of Caasava Acc:esslons from Sierra Leone • • ••• • • •• • • • .. ,. • • • • ~. • • • .. . • • • 2 • • • •• • • • • • • • • • •• • • • • -10 ~\ 5 10 • 15 20 • • • • • -2 • • • -4 ... ... ... .... .... -· 4 4 ¡.4 t 4 ·• ·10 Figure l. PCA of Genetic Distance based 1-poportion of shared alleled (PSA) of local varieties from the Southern and Eastem districts of Sierra Leooe 40 A dendogram of landraces for UPGMA of pair wise F51 estimates separates the African from Neotropical accessions with the group of accessions from Guatemala being the most genetically differentiated (Fig.2). These results agree with sorne previous studies on which a high differentiation has been observed among certain cassava groups cultivated in Guatemala and those in other parts of Latin America and África (Fregene et al, 2003; CIAT, 2003). Of particular interest to cassava breeding programs is the group (G1) from Guatemala, Accessions from G 1 from Guatemala is a representation of the region East and South of Guatemala and it may represent a heterotic group ba&ed on differences in allele frquncies. The phenomenon of heterosis or hybrid vigor is an important factor in improvement of heterozygous crops such as cassava, and in cases like the com, the pattems found in these populations at the beginníng of the XX century have been the base of a very sucecssful industry of hybrid corn, elevating productivity by more than 500% (Shull, 1952; Tomes, 1998). Conclusions Characterization of the cassava landraces of Uganda with 35 markers and estimation of genetic diversity and differentiation Assessment of the genetic relationship between different countries of Africa and Latin America. Evidence of high differentiation between sorne land races from Guatemala and the rest of the countries that may represent heterotic pools. Evidence of low impact in cassava diversity after of epidemic of CMD and low structure in cassava landraces of Uganda. 35 15 1 <> <> 10 -1 <> <> <> <> <><> <><> 1 <> UGANDA <> <>óo <> li. ~ + 5 -1 ~ 9 ~<>- <> ~ <> <> + I •TANZANIA <> + ~ ó ~A . o<> <> <> 'Y\W V ~ A ~ . • 1 1 oNIGERIA - 1 ~ o o ...... o -C\J l~¡~<>~ ;:;;; (h~~} -~...:'r:'~""ti!í). -'' . ,;. ,.-:~•: .. ~: •. · "' ..... n.t-.:ii ;&,~~: .• ,;! o COLOMBIA ü a_ -5 ~ ~ .. .:.,_,¡.~ .. " ... ,.M t\, •' :•} f~\ •• ..... ........ -. ~~ •:. oPERU 1.,4. ):GUATEMALA 1 -10 .,~~. ·->'~· i ) .. ". +GUATEMALA 2 -15 -20 -15 -10 -5 o 5 10 15 20 25 30 35 PC1 (39%) Figure 4. PCA of Genetic Distance (1-PSA) of local cassava varieties within the different districts in Uganda 36 U GANO A GIANA TANZANIA NIGERIA COLOMBIA - BRASIL - PERU G2 Gl 0.01 f-- f---- 0.07 0.14 Coefficient 0.20 Figure 2. Unweigbted pair group metbod witb aritbmetic averaging (UPGMA) dendogram of tbe pairwise ftxation index (Fst) between cassava landraces, grouped by country and by source. 37 f--- 0.26 References CIAT (2003) Assessment of simple sequence repeat Diversity of cassava landraces in Africa y Latín America: Progress report of the Ghana, malawi, Uganda, Guatemala, and brazil country studies. Report of activities by the cassava molecular diversity network (MOLCAS) Doyle JJ and Doyle JL (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin. 19:11-15 Elias M, Penet L, Vindry P, Mckey D, Panaud O, Robert T (2001) Unmanaged sexual reproduction and the dynamics of genetic diversity of a vegetatively propagated crop plant, cassava (Manihot esculenta Crantz), in a traditional farming system. Molecular Ecology 10:1895-1907 Fregene MA, Suarez M. Mkumbira J, Kulembeka H., Ndedya E, Kulaya A, Mitchel S, Gullberg U, Rosling H, Dixon AGO, Dean R, and Kresovich S (2003) Simple sequence repeat marker diversity in cassava landraces: Genetic diversity and differentiation in an asexually propagated crop. Theor. Appl. Genet. 107:1083-1093 Goudet J. (1995) FSTAT (Version 1.2): A computer program to calculate F-statistics. J . Hered 86:485-48 Langlands BW (1966) Cassava in Uganda 1860-1920. Uganda Journal 30:211-218. Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: towards an SSR-based molecular genetic map of cassava. Theor Appl Genet:21-31 Otim-Nape GW, Alicai T, Thresh M (2001) Changes in the incidence and severity of Cassava mosaic virus disease, varietal diversity and cassava production in Uganda. Ann. Appl. Biol 138:313-327. Rohlf FJ (1993) NTSYS-PC numerical taxonomy and multivariate analysis system. Version 1.8 Exeter Pub!., Setauket, NY. Shull, G.F. 1952. Beginnings of the heterosis concept. In: JW Gowen. Heterosis. Iowa State College Press, Ames, lowa. Pp 14-48 Tomes D (1998) Heterosis: performance stability, adaptability to changing technology and the foundation of agriculture as a business. In: Larnkey K, Staub J E (eds). Concepts and breeding of heterosis in crop plants. CSSA Special Publication Number 25. Crop Science Society of America, Madison, Wl. Vekemans X., Lefebvre C (1997) On the evolution of heavy-metal tolerant populations in Armería marítima: evidence from allozyme variation and reproducive barriers. J. Evol.Biol 10:175-191 38 Table 1 Genetic diversity within groups of cassava landraces classified according to country of origin. Standard deviations (SD) were estimated by jackknifmg over loci (200 replications). " " H., Dst, and G,: are given over loci and over groups (country populations). Population n lloc. lloc P PLP K K p HO_p HE....p HF.c_p Fls_p A 60 33 32 97.0 4.2 4.3 0.6425 0.5639 0.5686 -0.1391 B 9 33 33 100.0 4.3 4.3 0.6747 0.6100 0.6469 -0.0607 e 16 33 30 90.9 2.9 3.1 0.7355 0.4930 0.5092 -0.4542 mean 3 pop. 95.96 3.83 3.93 0.6843 0.5556 0.5749 -0.2180 std 4.63 0.80 0.72 0.0472 0.0589 0.0690 0.2083 • Hr total heterozygosity in the entire data set; Hs= heterozygosity within country averaged over the entire data set; D,r average gene diversity between populations; G,r coefficient óf gene differentiation. b poi. =polymorphic ~= average observed heterozygosity within country dHe= average expected heterozygosity within country ~-p= average expected heterozygosity within country corrected for small sample sizes (Nei, 1978) Figure l. Silver stained acrylamide gel picture of the Sierra Leonean accessions analyzed with SSR marker SSRY 82 Conclusions The SSR analysis of genetic diversity of local cassava varieties from Sierra Leone reveals the same pattem of high average gene diversity, low differentiation, anda pronounced sub- structure. Future perspectives include tracing the lineages of sorne of the lines in the 3 clusters found to get a better idea of the impact of gerrnplasm development/introduction and adoption in the country. 41 References Fregene MA, Suarez M. Mkumbira J, Kulembeka H., Ndedya E, Kulaya A, Mitchel S, Gullberg U, Rosling H. Dixon AGO, Dean R. and Kresovich S (2003) Simple sequence repeat marker diversity in cassava Iandraces: Genetic diversity and differentiation in an asexually propagated crop. Theor. Appl. Genet. 107:1083-1093 Goudet J. (1995) FSTAT (Version 1.2): A computer program to calculate F-statistics. J. Hered 86:485-48 Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: towards an SSR-based molecular genetic map of cassava. Theor Appl Genet:21-31 Rohlf FJ (1993) NTSYS-PC numerical taxonomy and multivariate analysis system. Version 1.8 Exeter Publ., Setauket, NY. Vekemans X., Lefebvre C (1997) On the evolution of heavy-metal tolerant populations in Armería marítima: evidence from allozyme variation and reproducive barriers. J. Evol.Bioll0:175-191 1.1.10 Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local Cassava Varieties from Cuba Yoel Beovides1, Edgar Barreri, Janneth P. Gutiérrez2, Charles Buitrago2, Jaime Marin2, Martin Fregene2 1 INIVIT, Cuba; 2 CIA T Funding: Cassava Biotechnology Network (CBN) Introduction Cassava is an important crop of modero tropical economies and an attractive one for millions off resource poor farmers found in the tropics (Best and Henry 1994). Recently a second center of diversity have been postulated in Central America based on SSR markers (Monte et al. 2003), in addition to the one in Brazil (Olsen and Schaal 1999). However, the potential of diversity in the second center, particularly in the Caribbean is not well documented. Recently SSR markers have been utilized to study the diversity of cassava from different countries (Fregene et. al. 2003). SSR markers are particularly attractive to study genetic diversity due to their abundance in plant genomes, high levels of polymorphisms and adaptability to automation. These studies revealed a high amount of diversity in accessions from several neotropical countries, a low leve} of genetic differentiation between country samples, with the exception of a group of accessions from Guatemala, and sub-structure in diversity of accessions from sorne African countries. 42 SSR markers can contribute to a better understanding of genetic diversity present in a collection of local cassava varieties held in Cuba to permit a more rational conservation and use of diversity on the island. We present here preliminary results of SSR study of genetic diversity of cassava from Cuba compared to a sub-set of accessions from Africa, South and Central America Methodology A total of 94 accessions were selected from a collection of cassava held at INIVIT in Cuba, selection criteria was the economic importance and origin in Cuba. A set of 54 clones from Africa and the Neotropics, 12 from Nigeria, 10 from Tanzania, 12 from Guatemala, and 20 from South America, representative of a large set of accessions from these countries used in previous SSR studies (Fregene et al 2003) were included for comparisons. A third set of 13 improved genotypes from CIAT with traits of agronomic interest were added. DNA from all accessions was obtained using the Dellaporta et al. method (1983). Concentration and quality of the DNA was checked by flourometry and agarose gel electrophoresis respectively. The DNA samples were diluted to a working concentration of 10ng/ul for subsequent PCR amplification. PCR amplification, automated gel analysis and date collection were as descried by Fregene et al (2003) and Mba et al. (2003). Statistical analysis to be conducted include calculations of pair-wise genetic distance, based upon the proportion of shared alleles (PSA), using the computer microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html). Distances between the accessions will be subjected to principal component analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of relationship between the land races. Other analysis are estimation of parameters of genetic diversity and differentiation, calculated from the raw SSR allele data using the computer packages GEN SUR VEY (Vekeman et al1997) and FSTAT (Goudet 1990). Results A total of 15 SSR markers have been analyzed until now in all accessions. A high number of alleles and level of polymorphisms have been observed in all SSR markers analyzed until date (Fig 1). Evaluation of the remainder 21 SSR markers is ongoing and will be completed by October. Also ongoing is reading of the gels and determination of allele sizes using the prograrn Quantity one. This raw SSR data will be used for subsequent analysis as described in the methodology. 43 Figure l. Polyacrilamide gel of a PCR amplification of cassava accessions from Cuba, Nigeria, Tanzariia, Guatemala, and South America using the SSR primer SSRYSl. Conclusions A SSR study of genetic diversity of cassava in Cuba has been initiated. The outcome of the study is expected to provide insights on the significance of the Caribbean region as a center of diversity for cassava and to guide rational conservation and plant improvement efforts. References Monte L. Azudia C, D. Debouck y M. Fregene. 2003. Simple Sequence Repeat (SSR) Marker Assesment of Genetic Diversity of Cassava Land Races from Guatemala. (in preparation). Fregene M, M. Suárez, J. Mkumbira, H. Kulembeka, E. Ndedya, A. Kulaya, S . Mitchel, U. Gullberg, A. G. O. Dixon, R. Dean y s. Kresovich. 2003. Simple sequence repeats marker diversity in cassava landraces: genetic diversity and differenciation in an asexually propagated crop. Theoretical and Applied Genetics 107: 1083-1093. Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg, U., Apel, K., Gale, M., Tohme, J. and Fregene, M. (2001) Simple Sequence Repeat (SSR) Markers Survey of the cassava (Manihot escu/enta Crantz) Geno me: Towards an SSSR-Bassed Molecular Genetic Map of Cassava. Theoretical and Applied Genetics. 102: 21-31. Olsen K and Schaal B. 1999. Evidence on the origin of cassava: phylogeography of Manihot esculenta. Proceedings of the National Academy of Science 96: 5586-5591. 44 1.1.11 Molecular characterization of rice and red rice using microsatellites and its relation with the morphological seed traits E. Gonzálezl, L. F. Foryl, J. J. Vásquezl, P. Ruizl, A. Mora2, J. Silva2, M. C. Duquel, 2, and z. Lentini_l,2. 1 SB2, 2 IP4. GTZ, Germany. Project No. 99.7860.2-001.00 Introduction The genus Oryza (AA genome) contains two cultivated species of rice, Oryza sativa L and Oryza glaberrima Steud and five wild species including Oryza ru.fipogon Griff., the ancestor of rice, and one of them (0. glumaepatula) native of Central and South America (Oka and Chang, 1961;Vaughan and Tomooka, 1999). Red rice (Oryza sativa f spontanea) is a weedy rice with red pericarp and dark-colored grains, commonly found in rice fields . It is the same species as the cultivated, having similar morphologicals traits at vegetative phase that makes difficult to dintinghuish them in the field at early life cycle, but at maturity they are taller, with profuse tillering, seed shattering and dormancy favoring its persistence in the field. Most published reports have classified red rice populations into two major groups based on hull color of mature seeds, which are strawhull and blackhull ecotypes (Smith et al., 1977; Sonnier, 1978 cited by Noldin, 1999). According to Langevin et al. (1990), the red rice can be grouped in ecotypes with characters alike cultivated rice or wild rice (Oka and Chang, 1961). Other researches indicate that red rice shows intermediate characteristics between wild rice O. ru.fipogon and cultivated indica or japonica varieties of Oryza sativa L. (Oka, 1988 cited by Bres- Patry et al. 2001). Another hypothesis is that weedy rice may evolve through the degeneratíon of domesticated rice, as weedy types of rice, where wild rice is not present (Vaughan et al. 2003). Several population studies have focused on the genetic structure of red rice. These studies have related groups of individuals with morphological seed traits such as color of awn and hulls. A molecular analysis of 26 red rice accessions collected in Uruguay showed three groups: The first containing seeds with blackhulls, purple apiculus and long awn and the second showing seeds awnless or short awn and greyed yellow color apiculus and hulls. The third group included commercial varieties analyzed and sorne red rice accessions similar to the commercial varieties (Rodriguez et al., 2001). Avozani et al. (2001), analyzed 36 red rice accessions with RAPDs markers and found six groups, the most remarkable were the first with only red rice accessions, mainly with awned seeds, and the fifth group including commercial varieties and sorne individuals similar to varieties in seed traits. Gealy et al. (2002) used rnicrosatellite markers to distinguish among red rice, rice cultivars, and red rice-cultivated rice hybrids derivatives (RC hybrids) and the cluster analysis suggested that there were three distinct genotypic groups. The first group consisted of awnless, strawhull red rice types. The second group had primarily awned, blackhull red rice. The third group was composed of RC hybrids and rice cultivars. Prelirninary work conducted by our research group (Gonzalez et al., 2002, and SB2 Annual 45 Report 2002) usíng 148 accessions of red rice collected in fanners fields in Colombia, 9 commercial Latín American rice varieties, sixteen homozygous Cica 8 transgenic lines, four hand-made hybrids between these transgenic lines and three varieties, and the AA wild species O. barthii, O. glaberrima, O. glumaepatula, and O.rufipogon, and 50 microsateUites indicated that is possíble to discriminate the diversity of red rice by microsatellites associated with plant morphological traits as found in other studies, but also associated with plant development (phenology) characteristics such flowering. Microsatellite cluster analyses discriminated awn from awnless red rice, and within those groups distinguished early to intermediate flowering types from late flowering. Sorne red rice variety types were clustered with commercial varieties, and sorne morphologically like wild species were clustered with O. rufipogon. This year we report results from a multiple correspondence analysis including five additional microsatellites markers. Materials and Methods Plant Material and Genetic Analysis using Microsatellite Markers. The plant material used and procedures followed for the molecular characterization using microsatellites were the same as previously described in SB-02 Annual Report 2002 (González et al., 2002). The plant material was divided in four groups: Group 1 represented by the 148 red rice accesions, Group 2 included the rice commercial varieties (Coprosem, Oryzica 1, Cimarron and Fedearroz 50), Group 3 was composed by O. ru.fipogon accession from Malasia; and Group 4 by Oryza glaberrima and two wild species Oryza barthii and O. glumaepatula (accession from Costa Rica). The PCR products were resolved on silver- stained polyacrylamide gels and microsatellites alleles were sized by comparison to the 10 and 25 bp molecular weight standards (Promega). Statistic analysis. Allelic frequencies were calculated for all materials analyzed. Pearson chi square test was used to evaluate the association between specific microsatellites alleles with black brown awn, apiculus and hulls (BBAAH). Two multiple correspondence analyses (MCA) was conducted. The first analysis only included the molecular markers data (MCA-M), and the second analysis included both the molecular and seed morphological data (MCA-MSM). The Pearson chi-square and MCA are tests apply to establish the significance of associations between categorical variables. The Pearson chi- square test is based on expected frequencies in a two-entry data set, whereas MCA is a modeling technique to analyze associations in multi-entry data set. All analyses were conducted using SAS software (SAS, 1989). Results and Discusión Distribution of alleles in four rice groups. The number of alleles per locus ranged from 5 to 12 (average 10.4 alleles per locus). Figure 1 showed a total of 146 alleles obtained in this study. Group 1, composed by the red rice accessions, contained 110 alleles of which 56 alleles (51%) were specific to this group and were not present in the Groups. These results indicated that the red rice population is highly diverse and contained the highest 46 number of specific alleles respect to the other Groups. In contrast to Groups 2 and 4, it is interesting to note that O. ru.fipogon (Group 3) do not have specific alleles sharing all of them with red rice, and sorne of them with either the varieties (Group 2) or the other wild species (Group 4). However, O. ru.fipogon is represented in this analysis by just one accession. It will be important to determine if this pattem is obtained when a broader range of accessions are included. (Figure 1). Association between specific microsatellites alleles with black brown awn, apiculus and hulls (BBMH). Pearson chi-square test showed a highly significant association (Chi- squares from 18.4 to 61.2~ p = 0.0001) between 7 alleles (2, 7, 17, 23, 53, 90 and 115 derived from 7 microsatellites markers) and BBAAH traits {Table 2). These markers are distributed in chromosomes 1,3,5,7 and 12 respectively {Table 2). These alleles were specific of red rice and./or O. ru.fipogon (Figure 1). A limited number of chromosomal regions (1, 3, 4, and 7) enclosing most of the genes/QTLs identified in a natural hybrid between ajaponica variety and red rice collected from a rice field was previously found to be associated with key morphological differences between red rice and rice (Bres-Patry et al. 2001). We found that 4 of the 7 alleles highly significantly associated with BBAAH traits were located in chromosomes 1, 3, 4 and 7, in addition of other two alleles present in chromosomes 5 and 6 not previously reported (Table 2). The Pearson test also showed a high association between the absence of allele number 95 and BBAAH traits (Chi-square 15.61, p = 0.0001, Table 2). This allele was found in the red rice and rice varieties groups (Figure 1). Multiple Correspondence Analysis (MCA). The multiple correspondence analyses using molecular data (MCA-M) generated five groups of which three grouped all red rice accessions (Figure 2A). The wild species O. barthii and O. glaberrima clustered together in the fourth group, and O. glumaepatula in the fifth group (Figure 2A). The three groups that enclosed all red rice accessions were analyzed in more detail and the analysis was complemented with seed morphological traits (MAC-MSM) (Figure 2B). First group (Group V) clustered 86 red rice individuals with the rice varieties (Figure 2B). Group V was composed by 58% of red rice accessions, including awnless or greyed yellow-awn red rice (43 and 56 % respectively), and 98 to 99 % red rice with greyed yellow hulls and apiculus alíke the varieties. Most of the red rice variety biotypes are in this group. Group OR clustered ten red rice (7%) with O. ru.fipogon (Figure 2B). This group included red rice with awn (91%) and with brown apículus (91%) and brown hulls (82 % ). Group R (Figure 2B) was composed by the remaining 51 red rice accessions (35%) that did not fall in neither of the other two groups, and was characterized by individuals with awnless (41 %), greyed yellow awn!hull (37%) or brown awnlhull (22%) seeds. Greyed yellow apiculus and hulls were registered in more than 63% of individuals, whereas the 22% of red rice showed brown hulls and apiculus. In the MCA-M analysis, variability could be explained by 17 microsatellites alleles. Of these 15 alleles are specific to O. barthii, O. glaberrima andO. glumaepatula, one allele is shared by O. ru.fipogon and red rice group, and the other allele is shared by red rice and 47 rice varieties (Figure 1). In MCA-MSM analysis, variability was explained by the presence of 16 microsatellite alleles and four seed morphological traits (black brown awn, apiculus and hulls, absence of greyed yellow apiculus). Of these 16 alleles, 11 alleles were derived from the MCA-M analysis (Figure 1) and the other five alleles from the Pearson chi-square analysis (Table 1). Conclusions Specific microsatellites alleles were identified distinguishing rice vanetles, red rice accessions and wild species. These alleles could be useful for studying red rice genetics, dispersion of red rice genotypes, degree of hybridization between red rice and cultivated rice, and genetic introgression and persistence of domesticated genes in red rice and wild species populations. The addition of seed morphological traits complemented the molecular analysis facilitating the discrimination of three main groups within the red rice population. One group included red rice similar to commercial varieties, another group alike O. ru.fipogon accession, and the third group with intermediate traits. Future Activities • Red rice types similar to O. ru.fipogon based upon the morphological and molecular characterization should be subjected to taxonomic classification to elucidate if they are introduced accessions of the Asían wild species. References Avozani, A., Barbosa, J., Maciel, J., Gaedka, L., Haa, L. and Concerc;ao. 2001. Caracterizac;ao genética de biotipos de arroz vennelho com marcadores moleculares-RAPD. II Congreso Brasilero de Arroz 1 rrigado. XXIV Reuniao la cultura do arroz irrigado. Porto Alegre-Río Grande do Su!. 20 a 23 de agosto 2001. Instituto Rio Grandense do Arroz-IRGA. Bres-Patry, C., Lorieux, M., Clément, G., Bangratz, M. and Ghesquiere, A. 2001. Heredity and genetic mapping of domestication-related traits in a temperate japonica weedy rice. Theor. Appl. Genet. 102: llS-126. Gealy, D., Tai, T . and Sneller, C. 2002. Identification of red rice, rice, and hybrid populations using rnicrosatellite markers. Weed Science. 50: 333-339. González E .• J. J. Vásquez, E. Corredor, L. F. Fory, A. Mora, J. Silva, M. C. Duque, and z. Lentini. 2002. Molecular characterization of rice and rice wild/ weedy relatives using microsatellites and their use to assess gene flow in the Neo-Tropics. Poster. The 7th Intemational Symposium on the Biosafety of Genetically Modified Organisms. Beijing, China. October 10-16, 2002. Langevin, S., Clay, K. and Grace, J. 1990. The incidence and effects of hybridization between cultivated rice and its related weed red rice (Oryza sativa L.) Evolution. 44: 1000 - 1008. Noldin, J., Chandler, J. and McCauley, G. 1999. Red rice (Oryza sativa) Biology. l. Characterization of red rice ecotypes. Weed Technology. 13:12-18. 48 Oka, H. and Chang, W. 1961. Hybrid swarms between wild and cultivated rice species Oryza perennis and Oryza sativa. Evolution 15: 418-430. Rodríguez, M., Vaughan, D., Tomooka, N., Kaga, A., Wang, X., Doi, K., Francis, M., Zorrilla, G. and Saldain, N . 2001. Análisis de la diversidad genética de arroz rojo en Uruguay utilizando la técnica marcadores moleculares AFLP's. Porto AJegre-Rio Grande do Sul. 20 a 23 de Agosto 2001. Instituto Rio Grandense do Arroz-IRGA. SAS, 1.1. 1989. SAS/STAT. User Guide, Version 6, Fourth edition, Volumen 2. Cary, N.C: S.A.S. Institute Inc. p 864. Vaughan, D., Morishima, H. and Kadowaki, K. 2003. Diversity in the Oryza genus. Current Opinion in Plant Biology. 6: 139-146. Vaughan, D.; and Tomooka. 1999. Varietal differentiation and evolution wild rice in Venezuela. Rice Genetics Newsletter. 16:15-17 49 2, 4, 10, 17, 18, 22, 25, 26, 27, 31, 32, 33, 37, 38, 39, 43, 48, 49, 50, 51, 52, 55, 58, 66, 69, 75, 76, 78, 79, 81, 82, 84, 85, 87, 89, 94, 96, 97, 98, 99, 104,106, 110, 112, 114, 117, 122, 123, 125, 129, 130, !31, 134, 136, 138, 145 Group 2. Varieties l . ll, 12, 14, 15, 20, 28, 29, 35, 40, 47, 56, 57. 60, 61, 70, 72, 73, 91, 92, 93, 101, 103, 107, 113, 119, 120, 132, 133, 135, 141 , 144, 146 Group 4. Wüd sps and O. glaberrima Figure l. Venn diagram sbowing the microsatellites alleles in eacb rice group. 50 A ¡ o. barthii 1 O.glab0.005) discriminated three groups: early flowering (red rice 1-21-3, 1-3-4, and 5-36-4, and the Purple line) with a mean value 92 to 95 day-after-sowing (DAS); intermediate flowering (red rice 5-38-5, 4-12-2, and 5- 48-2, and all the rice varieties and transgenic lines A3-49-60-12-3/Cica 8-2, A3-49-60-4- 5/FB007-19-1) with a mean value of 99 to 110 DAS); late flowering (the transgenic line A3-49-60-4-5/FB007-19-2) with a mean value of 123 DAS. About 71 % red rice accessions flowered earlier than the variety Cica 8 and the transgenic line A3-49-60-12- 3/Cica 8-2 (Table 1), and most red rice flowered earlier than and the transgenic line A3-49- 60-4-5/FB007-19-2. No significant difference was noted in the number of tillers between 58 the red rice, the transgenic lines and rice (Table 1). About 70% of the red rice was as tall as the transgenic line A3-49-60-12-3/Cica 8-2, and the Purple line, whereas the varieties Cica 8, Fedearroz 50 and Fedearroz Victoria 1 were taller than 70% of the red rice. Height differences seem not to be a critica! point to prevent gene flow. Song et al. (2003) detected gene flow between species of O. rufipogon and the rice commercial variety Minghui- 63, which differed in 130 cm height (Table 1). Based on these results the transgenic line A3- 49-60-12-3/Cica 8-2 was selected as one of the pollen donors for the gene flow studies. About 60% of plants from the line A3-49-60-12-3/Cica 8-2 showed gus expression, indicating that the gene was still segregating in this F3BC1 generation. In order to have enough plants for the gene flow studies, a total of 4011 plants were evaluated, and 2246 plants showed gus expression (56%). Flowering was synchronous between the red rice; the Purple line and the transgenic line A3-49-60-12-3/Cica 8-2. Most treatments involving the Purple line and red rice 1-21-3, 1-3-4, 5-36-4 and 5-38-5 overlapped in flowering. The highest synchrony in flowering was noted between red rice 4-12-2 and the Purple line, with 87% of plants with flowering overlap. In the case of the transgenic line, flowering overlap with red rice of 56 to 77 % plants was noted. Even though the red rice 5-38-5 flowered earlier than the transgenic line, there was overlap towards the end of the flowering cycle of the red rice. Wind velocity ranged from Oto 0,7 mis, the maximum mean value of 1.6 mis was registered from 10: 00 AM to 12 PM on cloudy days. Seeds from red rice, transgenic and Purple line were harvested keeping record of the plant location within each experimental plot. Future plans Progeny plants from the different treatments will be analyzed using specific microsatellites markers to identify hybrid plants, as well as by scoring gus expression and the presence of the NS3 and hph genes by PCR (when using the transgenic lineas pollen donor) or by the presence of anthocyanins in vegetative and reproductive tissues (when using the Purple lineas pollen donor). These analysis will not only give an estímate of rate of hybridization between the different experimental types, but al so the distan ce of gene flow, and will allow the optimization of an experimental approach to use molecular marker for tracing/ and monitoring genetic introgression from rice at large scale suitable for risk assessment in farmers fields and natural environments. ~ Hybrids plants will be used to study genetic introgression dynamics and persistence of domesticated genes 1 recipient population over time. The information generated will be used to define management practices allowing a safety deployment of transgenic rice in the tropics. References Langevin, S., Cley, K. and Grace, J. (1990). The incidence and effects of hybridization between cultivated rice and its related weed red rice (Oryza sativa L.). Evolution. 44:1000-1008. 59 Messeguer, J., Fogher, C., Guiderdoni, E., Marta, V., Catalá, M., Baldi, G. and Melé, E. (2001). Field assessments of gene flow from transgenic to cultivated rice (Oryza sativa L.) using a herbicide resistance gene as tracer marker. Theor. Appl. Genet. 103:1151-1159. Noldin, J., Yokoyama, S., Antunes, P. and Luzzardi, R. (2001). Taxa e sentido do cruzamento natural entre o arroz transgénico Resistente ao herbicida Glufosinato e o Arroz vérmelo En: A vaila~ao do Fluxo Genico entre Genotipos de Arroz Transgenico Cultivado e Arroz Vérmelho en: ll Congresso Brasileiro de Arroz Irrigado. XXIV Reunion da cultura do arroz irrigado ANAIS. Porto Alegre. Rio Grande do Sul. Song, Z ., Lu, B., Zhu, Y. and Chen, J. (2003). Gene flow from cultivated rice to the wild species Oryza ru.fipogon under experimental field conditions. New Phytologist. 157:657-665. Zhang, N., Linscombe, S. and Oard, J. (2003). Out-crossing frequency and genetic analysis of hybrids between transgenic glufosinate herbicide-resistant rice and the weed, red rice. Euphytica. 130: 35- 45. • Fagure 1. General view offield trials in (A) Square-plots assay, and (B) Concentric circle assay. 60 Table l. Mean values of trains evaluated in the field Plant • Hight to Tiller height highest Line /plant (cm) panicle Gtnotype1 se LcJ GW F15 (cm) Red rice 1-21-3 3 6 3 92f 19 a 102abede 96abed 1-3-4 6 6 5 95ef 19ab 102abede 9labed 4-12-2 3 8 3 105bed lOabcd 95bedef 94abcd 5-36-4 4 8 9 lOOdef !Sabed 94bcdef 80d 5-36-4 4 8 9 95ef 17abe 97bcdef 84cd 5-38-5 4 7 5 lOlcdef !Sabed 94cdef 90abcd 5-48-2 6 7 7 107bcd 16abc lOlbedef 98ab Transgenic 60-12-3/Cica 8-2 9lef (GUS+) 4 8 5 llOb !Sabed 85cd 60-12-3/Cica 8-2 86f (GUS-) 4 8 5 lllb 13abcd 80d 60-4-5/Fd.S0-19-1 7 8 1 98 def 8.7cd 104abcde 96abcd 60-4-5/Fd.S0-19-2 7 8 1 124 a 7.7d lO Sabed 93abed Rice Cica 8 4 8 5 105bcd 18ab lOSa be 95abcd varieties Purple 10 9 3 92f 12abcd 93def 87bcd Cimarron 5 8 3 lOlcdef 19ab 95bcdef 92abcd Coprosem 3 7 3 lOOcdef lObcd 103abcde 95abcd Fedearroz 50 7 8 3 109bc 12abcd lila 103ab Fedearroz 2000 7 8 1 lOOcdef 13abed 99bedef 98abc Fedearroz Victoria 1 5 7 1 104bcde 14abcd 107ab 104a Oryzica 1 3 6 5 99def 12abcd lOOabcde 95abcd Values followed by the same letter are not significantly different (p=0.05) Ryan-Einot-Gabriei-Welsch multiple range test. 1Genotypes used: 6 red rice types. Transgenic lines 60-12-3/Cica 8-2 lines with (+) or without (-) gus expression. 2 SC= Stem color, 3LC = Leaf color. Stem and leaf color range from 3= light green color, 4 -6= green color, 7-8= dark green color, 9= purple color, 10= dark purple color. 4GH = growth habit, scored as l=Erect, 3=Semi erect, S=Intermediate, 7= opened, 9=decumbent, 5 Days to 50% plants flowering. 61 1.1.14 Genetic diversity in the multipurpose shrub legume Flemingia macrophylla and Cratylia argentea M.S. Andersson1 R. Schultze-Kraft2, M. Peters3, J. Tohme3, L.H. Franco3, P. Ávila3, G. Gallego3, Belisario Hincapié3, G. Ramirez3, C.E. Lascano3 1University of Hohenheim; 2University of Hohenheim; 3CIAT Little variation in nutritive value among Cratylia accessions was observed. CIAT 18674, 22375, 22406, 22408 and 22409 had higher dry matter yields in the dry and wet seasons than CIAT cv Veranera (CIA T 18516118668). These accessions, collected in the states of Goias and Mato Grosso in Brazil, were selected for seed multiplication given that they had DM production higher than 3 tfha per cut, good seed production capacity and equal or superior digestibility and eructe protein content than Veranera. Three semierect accessions CIAT 18437 from Indonesia, and 21083 and 21090 from Thailand were superior in comparison to CIA T 17403 (from Thailand). They had digestibility values >48% and dry matter yields >2 tfha. More detailed analysis of two subsets showed that accessions with higher feed quality in terms of digestibility have lower fiber and condensed tannin contents than low-quality accessions. Por Cratylia argentea analysis of origin, agronomic morphological and molecular marker information did not identify correlations between the clusters obtained in the different approaches. In the case of Flemingia macrophylla, clustering obtained by molecular marker information correlated well with morphological information and grouped accessions according to their different growth types . Introduction The work of CIA T on shrub legumes emphasizes the development of materials to be utilized as feed supplement during extended dry seasons. Tropical shrub legumes of high quality for better soils are readily available, but germplasm with similar characteristics adapted to acid, infertile soils is scarce. Flemingia macrophylla and Cratylia argentea have shown promising results in such environments and hence work on these genera is part of the overall germplasm development strategy of the CIA T Forages team. C. argentea is increasingly adopted and utilized, particularly in the seasonally dry hillsides of Central America, and more recently, the Llanos Orientales de Colombia. However, most research and development is based on only few accessions and hence activities to acquire and test novel germplasm of C. argentea is of high priority. F. macrophylla also is a highly promising shrub legume with excellent adaptation to infertile soils. In contrast to C. argentea, whose adaptation is limited to an altitude below 1200 masl, F. macrophylla can successfully be grown up to altitudes of 2000 masl. 62 However, the potential utilization of F. macrophylla is so far Iimited by the poor quality and acceptability of the few evaluated accessions. The project aims to investigate the genetic diversity within collections ofF. macrophylla and C. argentea with three main objectives: 1) To identify new, superior forage genotypes based on conventional gennplasm characterization/evaluation procedures (morphological and agronomic traits, forage quality parameters, including IVDMD and tannin contents) 2) To optimize the use and management, including conservation, of the collections. For this, different approaches to identify core collections for each species were tested and compared based on: (a) genetic diversity assessment by agroriomic characterization/evaluation; (b) germplasm origin information; and (e) molecular markers (RAPDs). 3) To assist future gennplasm collections on methodology, geographical focus and genetic erosion hazards. Material and Methods Agronomic characterization and evaluation. Space-planted, single-row plots in RCB design with three replications were established in Quilichao in March 1999 (Cratylia argentea, 39 accessions) and March 2000 (Flemingia macrophylla, 73 accessions). Additionally two replications were sown for seed production and morphological observations. The following parameters were measured in the trials: vigor, height and diameter, regrowth, incidence of diseases, pests and mineral deficiencies, and dry matter yield during wet and dry seasons. For the analysis of nutritive value, crude protein content and in vitro dry matter digestibility (IVDMD) of the entire collections were analyzed. For the morphological evaluation, qualitative and quantitative parameters were measured, such as days to first flower, days to first seed, flower color, flowers per inflorescence, flowering intensity, pod pubescence, seeds per pod, seed color, Ieaf area, peduncle length, etc. For F. macrophylla, a more detailed analysis of nutritive value was conducted of a representative subset (25 accessions), which included high, intermediate and low nutritive value accessions. The groups were selected based on crude protein content and IVDMD. The chemical analysis comprised fiber (NDF, ADF, N-ADF), extractable and bound condensed tannin (ECT, BCf) content and astringency (protein binding capacity). Monomer composition of the extractable condensed tannin fraction (procyanidin:prodelphinidin:proelargonidin ratio = C:D:P) was determined with a high- performance liquid chromatography system (HPLC). Due to extremely variable results both between Iaboratory replicates and among field repetitions, statistical analysis was not possible. 63 In order to give at least an idea of the monomer composition of extractable condensed tannins in F. macrophylla accessions, results from only five accessions, which were consistent between duplicates and among repetitions, are reported. Additionally, another subset of 10 accessions (9 high-quality accessions (18437, 18438, 21083, 21090, 21092, 21241, 21580, 22082, 22327) and CIAT 17403) was sampled 4, 6 and 8 weeks after cutting, to investigate the effect of age on digestibility as well as on protein, fiber and condensed tannin content and astringency. Based on data referring to the morphological, agronomic and feed quality variation of all accessions a core collection will be created, using multivariate statistic tools (Principal Component Analysis and Cluster Analysis). Analysis of available origin infonnation. Based on ecogeographical information on origin of accessions, a core collection was created, hypothesizing that geographic distance and environmental differences are related to genetic diversity. The analysis was conducted with FloraMapTM, a GIS tool developed by CIAT, which allows the production of climate probability models using Principal Component Analysis (PCA) and Cluster Analysis. Genetic analysis by molecular markers (RAPDs).: Efforts made in genetic analysis showed that common manual DNA extraction methods did not work well with F. macrophylla and C. argentea. A modified protocol, which was used to extract DNA showed promising initial results. However, frequent degradation, contamination and partial digestion of DNA occurred, due to secondary plant compounds, probably polyphenols. In preliminary trials with a commercial extraction kit instead, the DNA purity was higher but partial digestion continued to be a severe problem. Various studies with amplified fragment length polymorphism markers (AFLPs), the method of choice, did not succeed and finally studies using this methodology could not be completed. Instead, random amplified polymorphic DNA (RAPD) markers, which do not require enzymatic digestion, were successfully employed. A total of 47 RAPD 10-mer primers (Operon Technologies, Alameda, CA, USA) were screened as single primers for the amplification of RAPD sequences. Primers with highest levels of polymorphisms were repeated to test for reproducibility and those that produced polymorphic, distinct and reproducible bands were chosen for RAPD analysis. Multiple Correspondence Analysis (MCA) was performed on a matrix created based on the presence (1) or absence (O) of amplified bands. Subsequently, cluster analysis was performed on the coordinates obtained by MCA. Dendrograms were generated using UPGMA method. Nei 's coefficient was used as estimator of similarity between accessions in order to generate between- and within-group similarity tables. Diversity was estimated using Nei's H and GST estimators. Data analysis and synthesis. Individual and combined data analyses of all generated information was carried out using multivariate statistics. We have applied principie component analysis in all data sets (agronomic, morphological, geographical and 64 molecular). In addition, cluster analysis was performed and the resulting clusters were compared to identify similarities. Results and Discussion Agronomic characterization and evaluation. Results from evaluations per season carried out for Cratylia argentea and Flemingia macrophylla indicated considerable phenotypic and agronomic variation in the collections studied. Data for C. argentea and F. macrophylla have been presented already in previous reports. For C. argentea, IVDMD varied between 59 and 69% and crude protein content between 18 and 24%. Mean dry matter production was 2.2 (range 0.8 to 5.2) t/ha and 1.93 (range 0 .6 to 3.3) tlha in the wet and dry season, respectively. Dry season yields were relatively high and confirm the good adaptation of C. argentea to dry conditions. There was a pronounced effect of season on sorne agronomic and quality traits. DM production was higher in the rainy season than in the dry season whereas ADF was higher in the dry than in the wet season. A season x genotype interaction was detected for IVDMD. The cluster analysis dendrogram (Ward's Method) was truncated at the 6-group level. The detailed agronomic characteristics of each group are listed in Table A. Group 4 was the agronomically most promising cluster. lt contained three accessions with the highest DM production (3.2 t/ha in the rainy and 2.4 tlha in the dry season), CP content, regrowth and plant diameter values. The highest dry matter yields (2.4 to 3.8 tlha) were recorded in accessions 22375, 22406 (Group 4), 18674, 22408 and 22409 (Group 6). Productivity of these accessions was higher than yields of the cultivar released in Costa Rica (cv. Veraniega) and Colombia (cv. Veranera) - an accession mixture of CIAT 18516/18668 (yield 1.9 and 3 tlha). In addition to the higher yield, these accessions (18674, 22375, 22406, 22408 and 22409) also had equal or superior digestibility values (65 to 69%) and crude protein content (20 to 24%) in comparison to CIAT 18516/18668 (IVDMD 64 to 67%, CP 21 to 24%). Based on high forage yield and good seed production potential we selected CIAT 18674, 22375, 22406, 22408 and 22409 for seed multiplication and regional testing (Table B). 65 TableA. Identification of Cratylia argentea accessions of agronomic interest. * rainy/dry season value Group 1 (averagellow* yields, low regrowth, high dry season digestibility, high CP, high ADF): CIAT 22382,22390,22392,22393,22394,22396,22399,22411 Group 2 (highllow yields, low regrowth, high dry season digestibility, average CP, high ADF): CIAT 18675, 22380,22383,22384,22386,22387,22391 Group 3 (averagellow yields, low regrowth, average digestibility, lower CP than group 1, high ADF): CIAT 18672,22376,22378,22381 Group 4 (very high/average yields, good regrowth, high digestibility, high CP, low ADF, low NDF): CIAT 22374,22375,22406 Group 5 (high/average yields, good regrowth, average digestibility, average CP, high ADF): CIAT 18676, 18957,22373, 22400, 22410, 22412 Group 6 (very highlaverage yields, good regrowth, high digestibility, high CP, low AFD, higher NDF than group 4): CIAT 18516, 18667, 18668, 18671, 18674,22379,22404,22407,22408,22409 Table B. Selected promising Cratylia argentea accessions and the two control accessions CIA T 18516/18668. Data of two evaluation cuts with 8 weeks of regrowth per season. IVDMD = in vitro dry matter digestibility, CP = crude protein. DM production IVDMDb (%) Accession number (tlha) • Rain y Dry Rain y Dry 18674 3.82 2.44 65 65 22375 3.12 2.41 65 66 22406 3.54 2.59 64 66 22408 3.25 2.45 69 67 22409 3.11 2.62 66 68 18516 (Control) 3.06 2.04 64 67 18668 (Control) 2.39 1.91 64 65 • Plant density 10 000 plantslha b Two-stage technique (Tilley & Terry 1963) e KjeldahJ nitrogen x 6.25 (AOAC 2003) d 15 months after sowing CP" (%) Rain y 20 21 21 21 22 21 21 Dry 23 23 22 22 24 24 23 Seed productiond (glplant) 153 255 152 153 97 18 110 In F. macrophylla, accessions evaluated differed in IVDMD, DM production, ECI', tannin extractability (ECI'/total Cf) and astringency (protein binding capacity) whereas CP and BCT showed only minor variability. IVDMD varied from 28 to 58% and crude protein content from 13 to 25%. Mean dry matter production was 2.08 tlha in the wet and 1.18 t/ha in the dry season. The chemical composition of 25 F. macrophylla accessions with contrasting digestibility varied greatly among accessions and in response to harvest season (Tables C and D). 66 Table C. In vilro digestibility, fiber and crude protein content of a representative subset of F. macrophylla. Data of one evaluation cut in the wet season and one in the dry season. n = no. of accessions evaluated, lVDMD =in vitro dry matter digestibility, CP = crude protein, NDF and ADF = neutral and acid detergent fiber, N-ADF = nitrogen bound to acid detergent fiber. lVDMDb(%) CY: (%) NDr <%> AD~(%) N-AD~ (%) Forage quality• Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry High (n=6) 52.8 48.0 22.4 21.6 33.0 33.8 20.3 23.7 11.1 11.2 Medium (n=12) 46.4 43.7 21.3 20.5 34.6 36.1 23 .7 24.9 12.0 11.2 Low (n=7) 42.4 40.1 20.5 20.2 34.6 35.9 23.1 24.4 11.5 12.0 Minimum 39.9 36.8 17.0 17.6 29.5 31.2 17.0 21.5 9.1 6.6 Maximum 56.2 51.3 24.4 23.6 39.3 39.8 27.6 29.2 15.4 16.9 Mean 46.8 43.7 20.9 20.7 34.2 35.6 22.7 24.5 11.7 11.5 a high: average IVDI\-ID ~48%. intermediate: ~43-47%. Low: <43% b Two-stage technique (Tilley & Terry 1963) e Kjeldahl nitrogen X 6.25 (AOAC 2003) d van Soest et al. 1991, Robbins et al. 1987 Total condensed tannin content ranged from 1.5 to 16.7% in the rainy season and from 1.8 to 22.4% in the dry season. Astringency ranged from 1.7 to 6.8 (PBE) in the rainy season and from 2.4 to 7.9 in the dry season. The acetone-extractable CT among accessions ranged from O to 19.4%, whereas the content of acetone-bound CT ranged from 1.3 to 3.3%. The ECT represented 0% of total condensed tannins in CIAT 21090 but 95% in CIAT 20616. Positive correlations were found between ECT and astringency (rrainy = 0.712, rdry = 0.721, P < 0.01). IVDMD was negatively correlated with ECT (rrainy =- 0.694, rdry==- 0.576, P < 0.01) and astringency (rrainy=- 0.632, rdry= 0.548, P < 0.01). 67 Table D. Condensed tannin content and composition in 25 F. macrophylla accessions with contrasting digestibility. Data of one evaluation cut in the wet season and one in the dry season. ECT::: acetone-extractable condensed tannins, BCT ::: acetone-bound condensed tannins, PBE ::: protein-binding entities, ND ::: not detectable. n.a. = not available. EC-r- (%) Wet 18437 4.2 18438 K 0.2 20065 ¡gh 7 7 Quality · 21083 0.1 21087 7.2 21090 ND --,....,------- 17403 4.3 20622 12.3 20744 11.5 20975 13.5 20976 Intermedia~.4 21092 Quality 6.6 21249 9.0 21529 9.3 21982 8.0 21992 7.8 22082 0.3 J 001 7. 1 17407 7.3 19457 8.5 20616 Low 5.9 20621 Quality 14.2 21241 10.0 21580 9.4 21990 12.4 High 3.9 Dry 7.6 1.6 9.7 0.0 6.6 ND 9.6 13.5 13.2 14.5 16.7 3.8 10.7 10.3 12.5 12.0 0.3 15.8 11.5 11.6 15.7 17.1 8.8 5.7 19.4 5.1 BCTa(%) Wet 2.33 1.76 1.30 1.36 1.57 1.57 2.12 3.27 1.91 2.59 1.85 1.53 2.18 2.90 1.87 2.19 1.86 2.55 2.13 2.76 2.00 2.54 2.16 1.79 2.30 1.6 Dry 2.65 2.02 2.24 1.81 1.98 1.99 2.43 1.20 1.75 2.01 1.33 2.21 2.59 3.32 2.90 2.37 1.96 1.73 2.25 2.65 0.84 0.96 2.1 1 2.07 2.98 2.1 Intermediate 8.6 11 .1 2.2 2.2 Low 9.7 12.8 2.2 2.0 Mean 7.9 10.3 2.1 2.1 • Butanol-HCl (Terrill et al. 1992, Barahona et al. 2003) b Radial diffusion assay (Hagerman 1987, Lareo et al. 1990) Extractability Astringencyb (ECTffota1CT %) (PBE) Wet 64.54 12.00 85.47 7.48 82.04 0.00 67.08 79.05 85.77 83.90 87.90 81.11 80.55 76.19 81.13 78.17 11.85 73.63 77.41 75.51 74.59 84.78 82.22 84.06 84.38 41.9 73.9 80.4 68.0 Dry 74.05 44.81 81.26 0.00 76.95 0.00 79.83 91.83 88.29 87.80 92.62 63.35 80.5 1 75.64 81.16 83.51 11.31 90.15 83.58 81.46 94.92 94.67 80.62 73.22 86.67 46.2 77.2 85.0 71.9 Wet 4.21 3.33 4.84 1.65 4.40 2.03 4.52 4.60 4.39 6.77 6.09 2.81 4.44 4.88 6.03 5.12 2.73 4.11 5.46 5.80 4.71 5.42 3.81 5.49 6.37 3.4 4.7 5.3 4.6 Dry 4.57 3.78 n.a. 2.39 6.44 2.53 4.99 5.85 5.79 6.33 5.33 3.55 5.70 5.15 7.70 7.91 2.88 5.58 5.33 6.74 5.97 5.88 6.13 5.07 6.95 3.9 5.6 6.0 5.4 Monomer composition of the extractable CT fraction in 5 accessions F. macrophylla was quite variable due to accession but not due to season of the year (Fig. A). In four accessions (CIAT 20621, 20744, 20975 and 20976) prodelphinidin made up more than half of the proanthocyanidins (range from 49 to 79% ). The second most important constituent 68 was propelargonidin, which ranged from 16 to 38%. Procyanidin was only present in small proportions (O to maximum 16%). It was interesting to observe that in CIAT 21092 propelargonidin represented 82% of total proanthocyanidins in the rainy season and 95% in the dry season. Procyanidin was absent and prodelphinidin was less than 20%. The five accessions for which we have reliable data on monomer composition of ECT are not representative of the entire Flemingia collection in terms of forage quality. However, four of them had very high ECT concentrations (13-17%) whereas CIAT 21092 presented relatively low ECf levels (7 and 4% in the rainy and dry season, respectively). The latter had an exceptionally high propelargonidin proportion but totally lacked cyanidin, which could indicate a relationship between monomer composition and forage quality. - Delflnidln c::::::J Cyanldln - Pelargonldln IDim Unkllown Fig. A. Monomer composition (procyanidin: prodelphinidin: propelargonidin ratios (C: D: P)) of the ECT fraction of five F. macrophylla accessions in rainy and dry season. Analysis of a subset of 10 high-quality accessions (including control) showed that forage quality varied over time. Pattems were different in the rainy and dry season for both the averaged values of the 10 accessions and for individual accessions (Fig. B). Correlations found in this analysis confirmed the negative correlations between IVDMD and ECT and IVDMD and astringency. Season had a large effect on IVDMD, DM production, plant height and diameter (higher in the rainy than in the dry season) and ADF, NDF, ECT and astringency (slightly higher in the dry season than in the rainy season). Extractability (percentage ECT of total CT) was relatively stable between harvest seasons (differences <10%). Only in six accessions (CIAT 17403, 18438, 20616, 20622, 21092 and J 001) differences up to 33% between 69 rain y and dry season were found. No genotype x season interactions were detected for DM production and regrowth. The accessions with the highest average in vitro dry matter digestibility were CIAT 18437, 18438, 21083, 21090, 21092 and 21241. The most productive accessions were CIAT 7184, 21090, 21241, 21248, 21249, 21519, 21529, 21580 and CPI 104890 with a total DM production >3.5 tlha in the rain y and >2 t/ha in the dry season. Arnong the materials superior to CIAT 17403 (digestibility 36%, DM production 1.5 tlha in the dry season) CIAT accessions 18437, 21083 and 21090 were identified for further testing as promising materials for dry season supplementation because they combined high digestibility with high productivity and low extractable condensed tannin content. These accessions had digestibility values > 48% and dry matter yields > 2 tlha. Extractable condensed tannin content was 4.2 and 7.5% in the rainy and dry season for CIAT 18437 and nil in CIAT 21083 and 21090. However, their low seed production in the site where they were evaluated can limit their value. 70 Ralny season Dry season 15 15 --- ECT(%) _....._ BCT(%) --- A.tringency (PBE) 10 10 5 ~ 5 ~ • • • ___. ... ... ... ... ... o o <4 weeks 6weeks 8weeks <4weeks 6weeks 8 weeks 60% 60% .... • • • • • 40% 40% 20% 'f' ... ... 20% 'f' ... ... --+- IVDMO __.,._ CP 0% 0% 4 6 8 4 6 8 60% 60% --+- AOF __.,._ NOF --+- N-AOF 40% 40% ~ ~ • • • • • • 20% 20% 0% L---~==~==~l._ _ __J .. 6 8 4 6 8 Fig. B. Variability in IVDMD, CP, fiber and tannin content of 10 Flemingia macrophyUa accessions after 4, 6 and 8 weeks of regrowth in the rainy and dry season. ECT, BCT = acetone-extractable and bound condensed tannin, PBE = protein binding entities, IVDMD =in vitro dry matter digestibility, CP = crude protein, NDF and ADF = neutral and acid detergent fiber, N-ADF = nitrogen bound to ADF The dendrogram (Ward' s Method) was truncated at the 7-group level, explaining 72% of variation. The detailed agronomic characteristics of each group are listed in Table E. Group 4 was the one that had the most promising accessions from an agronomic point of view. It contained eight accessions (7 semierect, 1 'tobacco') with the highest digestibility values of the collection (51% in the rain y and 47% in the dry season) and high DM production (2.6 t/ha in the rainy and 1.2 t/ha in the dry season). The three selected promising accessions CIAT 18437, 21083 and 21090 were contained in this cluster. 71 Table E. ldentification of Flemingia macrophylla accessions of agronomic interest. * rainyldry season val u e Group 1 (averagellow* digestibility, low yields, low regrowth, average CP, low vigor, low plant height): CIAT20973,20977,21080,21086,21994,22053,22087,22090 Group 2 ( averagellow digestibility, averagellow yields, high regrowth, average CP, low vigor, low plant height): CIAT 20979, 21079, 21990, 21993, 22285 Group 3 (high/Jow digestibility, low yields, average regrowth, highlaverage CP, low vigor, low plant height): CIAT 18048,20972,20976,20978,20980, 20982,21982,21991,21992, 21995,21996, 22327 Group 4 (high digestibility, highlaverage yields, good regrowth, highlaverage CP, high vigor, average plant height): CIAT 18437, 18438, 20975, 21083,21087, 21090, 21092, 22082 Group 5 (average digestibility, high yields, average regrowth, high CP, high vigor, high plant height): CIAT 801, 7184, 20622, 20625, 20626, 20631, 20744, 21241, 21248, 21249, 21519, 21529, 21580, C104890, 115146, J001 Group 6 (like group 5, but lower yields, lower vigor and lower plant height): CIAT 19453, 19454, 19797, 19798,19799,20065 Group 7 (low digestibility, highlaverage yields, average regrowth, high CP, high vigor, average plant height): CIAT 17400, 17403, 17404, 17405, 17407, 17409, 17411, 17412, 17413, 18440, 19457, 19800, 19801, 19824,20616,20617,20618,20621,20624 Genetic analysis by molecular markers (RAPDs): Out of 47 random primers tested, 9 were chosen that produced 171 RAPD bands ranging from 4 to 18 polymorphic bands per primer. Eight primers were selected for Flemingia macrophylla (DOl, D04, D15, 107, J04, J06, J07, Jl2), and six for Cratylia argentea (D15, G12, 107, J06, J07, J12) (Table F). Table F. Oligonucleotide primers employed in RAPD analysis, their sequence, number of bands obtained and percentage of polymorphic bands per species (% PBS). Primer code Sequence Number of bands PBS (%) (5' to 3') Polymorphic Monomorphic Flemingia macrophylla o 01 ACCGCGAAGG 7 1 o 04 TCTGGTGAGG 5 o 015 CATCCGTGCT 18 1 107 CAGCGACAAG 4 o J 04 CCGAACACGG 8 1 J 06 TCGTICCGCA 14 2 J 07 CCTCTCGACA 13 o J12 GTCCCGTGGT 6 2 Total 75 7 91.5 Cratylía argentea 015 CATCCGTGCT 10 4 G 12 CAGCTCACGA 17 1 107 CAGCGACAAG 17 4 J 06 TCGTICCGCA 13 1 J 07 CCTCTCGACA 8 3 J12 GTCCCGTGGT 9 2 Total 74 15 83.1 72 Clustering of 47 Cratylia argentea and 1 C. mollis (outgroup) accessions resulted in 5 groups (Fig. C), plus two genetically very distinct materials: "yacapani" (the only prostrate C. argentea accession) and C. mollis (data not shown here). Group 1 included 28 accessions and group 2 twelve. Group 3 comprised accessions 884 and CIAT 18668 and 22389. Group 4 was conformed of CIAT 22386 and 22387. Group 5 contained only CIAT 18674, one of the two agronomically most promising accessions. No correlation was found between the clustering according to RAPD polymorphisms and agronomic, morphological or geographical characteristics. ..... .... ___ .. f .... .... ~ y ;- O = Group 1 <::? = Group 2 4> = Group 3 O = Group4 1) = Group 5 /' ~;' ~~ Fig. C. Tridimensional representation of five groups (without outgroup and prostrate accession "yacapani") resulting from clustering (UPGMA) of 47 Cratylia argentea and 1 C. moUis accessions according to molecular marker information (RAPDs). Analysis of genetic diversity within accessions revealed high variability. Nei and Li similarity between groups often was as high or higher than within groups (Table G). This could indicate either seed contamination of accessions and/or outcrossing during multiplication in the field. Research on reproduction of C. argentea is urgently required to determine the rate and impact of outcrossing in this species. 73 Table G. Nei similarity within and between groups resulting from clustering (UPGMA) of 47 Cratylia argentea and 1 C. mollis accessions according to molecular marker information (RAPDs) Group N 1 2 3 4 5 6 7 Total 1 28 0.825 0.814 0.759 0.774 0.769 0.487 0.413 2 12 0.839 0.721 0.764 0.720 0.515 0.404 3 3 0.757 0.748 0.754 0.457 0.388 4 2 0.717 0.757 0.479 0.400 5 1 1.000 0.426 0.433 6 1 1.000 0.444 7 1 1.000 Total 48 0.776 Clustering of 111 Flemingia macrophylla and 2 F. paniculata (outgroup) accessions resulted in six groups (Fig. D), distinguishing well arnong the different morphotypes of this species, which have been described in the morphological evaluation (Photo, see also CIAT Annual Report 2002). 4 (s:l) -' 2 (e) l (out¡rouW S(p,sl,t) w 1 J • .. ·::: . ) Fig. D. Tridimensional representation of the groups resulting from clustering (UPGMA) of 111 Flemingia macrophylla and 2 F. paniculata accessions according to molecular marker information (RAPDs). e = erect, s1 and s2 = semi-erect 1 and 2, p = prostrate, t = "tobacco" morphotype. Group 1 included the two F. paniculata accessions. Group 2 was conformed by 55 of the 111 F. macrophylla accessions, which - with the exception of CIA T 20065 (prostrate) - 74 belonged to the erect growth type. Group 3 was composed of 23 semierect-1 and one "tobacco"-type accession and group 4 comprised 3 semierect-2 accessions. Group 5 included 8 prostrate, 14 semierect-1, 4 "tobacco" and 2 erect accessions and group 6 contained one semierect-2 and one semierect-1 accession. No correlation was found between the clustering based on RAPD polymorphisms and agronomic or geographical characteristics. On the other hand, RAPD analysis preved to b useful for the identification/distinction of the different F. macrophylla morphotype. It is suggested that the employment of the more powerful AFLP markers would detect higher polymorphisms wíthin the morphotypes of this species. Photo. Four F. macrophylla morphotypes: 1=erect, 2=semi-erect 2, 3=prostrate, 4='tobacco' 75 Activity 1.2 Identification and mapping of useful genes and gene combinations 1.2.1 Development of a genome-wide anchored microsatellite map for common bean (Phaseolus vulgaris L.) M.W. Blair1, F. Pedraza\ H.F. Buendia1, E. Gaitán-Solís1 S. E. Beebe2 P. Gepts3, J. Tohrne1 1SB-2 Project- CIAT; 2 IP-1, ~C-Davis Introduction Microsatellites are polymerase chain reaction (PCR) based markers that have been developed for a wide range of plant species, including many commercial crops. Among he grain legumes, microsatellite markers are now available for soybeans, chickpea, cowpeas, peanuts and more recently common beans. Microsatellite markers have been developed for cornmon beans from both non-coding (genomic) and coding (genic) sequences containing simple repeats. Our principal objective in this study was to map the both types of microsatellites in a single mapping population derived from the cross DOR364 x 019833 and to integrate this map with the genetic maps developed by Freyre et al. (1998) and Vallejos et al. (1992). Oiven the different sources of the microsatellites we compared the polymorphism rates of markers derived frorn genes versus those derived from random genomic sequences. Materials and Methods Populations and DNA extraction. Two populations of recombinant inbred lines (RILs) were used for this study: the first population 'was based on the cross DOR364 x 019833 (and heretofore will be referred to as the DO population). A total of 87 RILs were developed for this cross by a modified single seed descent from the F2 to the F9 generation. The plants within the F9 progeny row were bulked and used for subsequent genetic analysis. The second population was based on 91 RILs from the cross BAT93 x JaloEEP558 whose development and origins are described by Freyre et al. (1998). Total genomic DNA for each of the recombinant inbred lines in both populations was isolated from bulked leaf tissues of eight greenhouse-grown plants per line, using a CT AB extraction method. Source and development of markers. We used three sets of markers in this study: 1) genomic microsatellites developed in this laboratory by Oaitán-Solís et al. (2002); 2) gene- coding microsatellites developed by Yu et al. (1999, 2000) and 3) additional gene-coding and non-coding microsatellites from searches for SSR containing Phaseolus sequences deposited in the Oenbank database before July 15, 2001. SSRs were found using the SSR identification tool (SSRIT) that screens for all possible dimeric, trimeric and terameric repeats. Only sequences containing a mínimum of three tetra-nucleotide, four tri- nucleotide or five di-nucleotide motif repeats were used for primer design. Primers were designed using Primer 3.0 software to produce PCR amplification fragments that were on average 150 bp long, and PCR primers with consistent melting temperatures of 55°C or above and an average length of 20 nucleotides. Primer pairs were checked to make sure that they had similar rnelting temperatures and did not suffer from palindromes or end pairing. Microsatellite analysis. Polymorphisms between the mapping parents were determined on parental survey gels. Standard microsatellite PCR conditions were used throughout and the PCR reaction was carried out in 20 rnL final volumes. Gel staining and image capture are as described in more detail by Gaitan et aL (2002). The sizes of the parental alleles were estimated based on 10 bp and 25 bp molecular weight ladders. To determine the genotypes of the progenies, alleles were scored based on the parental bands that were amplified as controls along with the RIL individuals. Data Analysis. Segregation distortion was measured with a Chi-square test for an expected 1:1 ratio and segregation data was used to place the microsatellites on the established genetic maps for the DG and BJ populations (Beebe et al. 1998; Freyre et al. 1998). The DG map included 240 RFLP, RAPD, SCAR and AFLP markers described in Beebe et al. (1998) while the BJ map contained 141 markers as described in Freyre et al. (1998). The two maps were linked by cornrnon RFLP markers with the map described by Vallejos et al. (1992). Linkage analysis was conducted with the software application Mapmaker 2.0 using a mínimum LOD of 4.0. Results and Discussion A total of 150 cornrnon bean microsatellites were used in this study. Of these, 81 were anonymous genornic or non-coding microsatellites and 69 were gene-derived rnicrosatellites. In the Genbank searches, SSRs were found in a range of coding and non- coding sequences. The highest number of repeats detected in the simple sequence repeats was nine, while the average number of repeats was 5.5 among all the microsatellites identified. Genomic microsatellites hada significantly higher (P=0.02) average number of repeats than gene rnicrosatellites (6.2 versus 5.3) in unpaired t-test. In this study, di and tri-nucleotide motif containíng microsatellites did not have significantly different average number of repeats. All the microsatellite markers were screened for amplification products and polymorphism in the parents of the DG and BJ populations and no difference in band intensity between the cDNA and genomic derived microsatellites was observed. A majority of the microsatellites produced single bands for the parents. Polymorphism rates for the DG and BJ populations were 65.4 and 63.2% for the genomic microsatellites, and 46.3 and 46.2% for the genic rnicrosatellites, respectively. Overall the percentage of polymorphism between the parents of both populations was very similar: a total of 84 out of the 150 microsatellites tested for the parents of the DG population were polymorphíc (56.0%), while a total of 68 out of the 122 rnicrosate]Jjtes tested for the parents of the BJ population were polymorphic (55.7%). A total of 100 new microsatellite loci were placed on the two genetic maps (78 on the DG population and 22 on the BJ population) during this study. Microsatellite loci were found on each of the eleven chromosomes of the species and each chromosome was tagged with at least five or more microsatellite. Two chromosomes, B02D and B04B hada relatively greater number of microsatellites placed on them, with 17 and 13 markers respectively; while the average number of microsatellite loci per chromosome was 10. The total cumulative map length for the DG population was 1720 cM with an average chromosome length of 156.4 cM. The average distance between microsatellite loci in this map was 19.5 cM; however the distribution of loci was variable and several large gaps between microsatellites remained on the map. The largest gaps between microsatellite markers remained on chromosomes bOl, b07, b08 and bll. Among the markers that presented multiple bands, duplicate loci could be mapped for two markers. The gene-based microsatellites were better distributed than the genomic microsatellites and several clusters of genomic microsatellites were found on almost every chromosome except b06 and b08 which were the linkage groups with the fewest microsatellites. This study brings toa total of 115, the microsatellite loci located on the bean genetic map and provides coverage for every chromosome in the genome with from five to twenty markers each. Although the genomic distribution of microsatellite markers in this study tended to be random, sorne large gaps between microsatellites occurred and certain chromosomes contained more microsatellite loci than others. As single-locus markers, the microsatellites in this study were specific to a given place in the genome and this allowed them to be used for comparati ve mappíng across both the DG and BJ populations. This comparative mapping showed the consistency of microsatellite location on both populations: with all the microsatellites mapping to the same individual chromosome and equivalent map locations in each of the populations. Mapping in two populations was useful for placing microsatellites that were monomorphic in one or the other of the populations. Comparative mapping allowed us to determine the identity and orientation of each linkage group and to obtain a more accurate position for each of the microsatellites. The synteny between maps as reflected by the map order of jointly mapped markers was conserved providing further evidence that no major rearrangements have occurred in the common bean genome. The current set of microsatellite markers described in this study provides the basis for anchoring and aligning genetic maps one to each other based solely on PCR-based markers, something that previously was done with single-copy RFLP markers or by tentatively associating RAPD bands found in different populations. Therefore, the microsatellites make ideal second-generation markers for the whole genome analysis important for gene tagging and quantitative trait loci studies. The microsatellites mapped during the course of this research will also be invaluable for marker assisted selection because they are simple to analyze, specific for single genes of interest and diagnostic in most crosses dueto their high level of polymorphism. The mapped microsatellites can also provide a good set from which to chose markers for studies of genetic diversity in comrnon bean. 78 References Freyre R, Skroch PW, Geffory V, Adam-Blondon AF, Shirmohamadali A, Johnson WC, Llaca V, Nodari RO, Perlera PA, Tsai SM, Tohme J, Dron M, Nienhuis J, Vallejos CE, Gepts P (1998). Towards an integrated linkage map of common bean. 4 Development of a core linkage map and alignrnent of RFLP maps. Theor Appl Genet 97:847-856 Gaitán-Solís E, Duque MC, Edwards KJ, Tohme J (2002). Microsatellite Repeats in Common Bean (Phaseolus vulgaris): Isolation, Characterization, and Cross-Species Arnplification in Phaseolus ssp. Crop Sci 42:2128-2136 Vallejos CE, Sakiyama NE, Chase CD (1992). A molecular marker based linkage map of Phaseolus vulgaris L. Genetics 131: 733-740 Yu K, Park SJ, Poysa V (1999). Abundance and variation of microsatellite DNA sequences in beans (Phaseolus and Vigna). Genome 42:27-34 1.2.2 Analysis of iron reductase as a mechanism for enhanced iron uptake in common beans MW Blair1 and M Grusae 1SB-2 Project , CIA T; 2USDA-Houston, Baylor College of Medicine Introduction Nutritionai genornics is being used as part of the Biofortification Challenge Program to discover the basic mechanisms for mineral uptake and accumulation. As part of this program CIA T is collaborating with the Grusak lab at the USDA-Baylor College of Medicine to determine the genes in common beans that determine iron uptake and utilization. As part of the overall genomics approach, information from other well studied specíes such Medicago truncatula, peas and soybeans, as weii as other model species such as Arabidopsis thaliana which have extensive genetic and molecular resources are being used for gene discovery and functional analysis. The underlying concepts of this work are to take advantage of metabolic unity among plants to characterize gene function and to apply bioinformatics and molecular cloning approaches to identify potential orthologous genes. As a first example of this approach the Grusak lab is trying to dissect the importance of iron reductase in the accumulation of iron in beans by assaying iron reductase activity in roots and by cloning an ortholog of the gene from common bean based on similarity to the same gene already isolated from Medicago truncatula and from Pisum sativum (Pea). Iron reductase is a member of the protein super-family of flavocytochromes and functions to convert iron from an unavailable form (ferric, Fe3+) to an available form (ferrous, Fe2+) that can be readily absorbed by plants. The iron reductase protein (FRO) is located in roots and straddles the root cell membrane where it is active for iron reduction and was first isolated from A. thaliana and P. sativum. both of which are fairly efficient at extracting iron from the soil and serve as model species for enzyme activity. Methodology SCAR marker. the Grusak lab selected conserved primers for RT -PCR based on the Pisum iron reductase gene (PsFR01). At CIAT we tried using these primers for mapping of the gene as a SCAR marker. Reductase Assay. In the Grusak lab, seeds are genninated for 3-4 days then planted in a hydroponic system for 12 days of growth in various levels of iron concentration (eg. 2, 5 , 10 and 20 uM Fe). Iron reductase assay is conducted at the end of this period by removing the entire root systems of four plants and staining for reduced iron measured as umol Fe reduced/g FW/hr. Results and Discussion The SCAR primers produced multiple banding pattems that were not of the expected size range, suggesting that the conservation between common bean and peas for this gene is low, and that it will be difficult to clone via a direct PCR approach. Given this the Grusak lab has been screening a common bean leaf cDNA library which we made at CIAT and will begin next year to screen a set of two root leaf cDNA librarles, where the mRNA for iron reductase is more likely to be expressed. The iron reductase assay is producing interesting results that suggest that there are differences between parents of severa! mapping populations for their ability to reduce iron (Table 1). These differences are evident more at low Fe concentration than at high iron concentration and seem to be somewhat correlated with the seed iron status of the parents, for example in the cross of G21242 x G21078, the high iron parent G21242 has higher reductase activity than the low iron parent G21078. Many of the wild accession also had low reductase activity as well, except for the Colombian accession G24404, which in contrast had high reductase activity and therefore will be investigated further in future experiments. 1t was notable that significant difference exist between the parents of the DOR364 x G19833 mapping population ata range of hydroponic Fe concentrations (Figure 1) and that these results were consistent with the results observed in the first trial with multiple parents at 2 uM and 15 uM Fe concentrations (Table 1). More information on this trait will be reported next year when a set of recombinant inbred lines have been fully tested and the QTLs for this trait localized. 80 Future Steps • Evaluate Fe reductase activity in a greater number of parents of other populatíons and a range of Fe concentrations for each paren t. • Evaluate the full set of recombinant inbred lines from the DOR364 x G19833 and G21078 x G21242 crosses to determine inheritance and location of QTLs for Fe reductase activity. • Develop a DNA marker for Fe reductase activity either based on QTL mapping or cloning and mapping of orthologs of the Fe reductase gene. Table l. Commoo bean genotypes tested for root reductase activity at low (2 uM) and high (15 uM) Fe concentration. Geootype Status Origin Seed Fe Root Fe(III) Reductase Activity content umol Fe reduced/g FW/hr 2uMFe 15 uMFe G11350 Cult. Meso High 0.835 0.350 G11360 Cult. Meso Low 0.454 0.165 G19227A Cult. Meso ND 0.879 0.047 G19833 Cult. Andean High 1.054 0.123 Gl9839 Cult. Andean High 1.083 0.200 G19842 Wild Andeao ND 0.227 0.248 G21078 Cult. Aodean Low 0.049 0.307 G21212 Cult. Meso ND 1.089 0.482 G21242 Cult. Andean High 0.713 0.528 G21657 Cult. Andean ND 0.639 0.231 G24390 Wild Meso ND 0.187 0.668 G23585 Wild Andean ND 0.048 0.036 G24404 Wild Meso ND 1.039 0.705 G24423 Wild Andeao ND 0.180 0.342 c .::--:~ .e: 1.4 !~ 1.2 (/)O) 1-~ o"t:l 0.8 -5~ ~ -5 0.6 S~ 0.4 :::::...cn ~~ 0.2 .... o 8 E o EX:~ o 5 10 DOR364 15 20 25 1. 1 o. o. 5 10 Growth solution Fe concentration (f.iM) G19833 15 20 Figure l . Parental genotypes tested for root reductase activity over a raoge of Fe conceotrations. 25 1.2.3 Tannin studies on parents and progeny of the DOR364 x G 19833 population MW Blair1, G Caldas1 SE Beebe1-2 P. Avila3 1SB-2, CIAT 2IP-1, CIAT,3porages, CIAT Introduction Part of the effort to increase the nutritional quality of common bean has concentrated on increasing iron bioavailability, where bioavailability is the proportion of the consumed nutrient that is digested, absorbed and utilized by human beíngs. Bioavailability is determined by both food composition and the nutrient status of the consumer as well as a mix of promoters (such as sulfur ami no acids: methionine and cysteine, vitamin A or C and lipid content) and anti-nutritional factors (including fiber, lectins, phytates, polyphenolics and tannins, as well as Calcium and Manganese). Among the anti-nutrients, tannins are important because of their ability to interact with proteins and to chelate minerals which results in reductions in protein digestibility and mineral bioavailability. Tannins are derived from phenolic compounds and contribute to the coloring found in common bean seed coats. They can be divided into hydrolizable 1 soluble tannins (derived from Gallic acid) and condensed tannins 1 proanthocyanadins (derived from polymerized flavonoids), which are measurable by different techniques. Previous studies looking at overall tannin content m bean seed coat, using a water/methanol extraction found variability in the content of tannins in seed coats of different varieties of common beans. In this study our objective was to identify the genetic 82 variability for tannin in a segregating population and to collect preliminary evidence on the inheritance of soluble and insoluble tannin content in the seed coats of recombinant inbred lines from the cross DOR364 x 019833. In this study we applied a more accurate extraction technique and began to look at the inheritance of various fractions of tannin con ten t. Methodology Tannin extraction. Total tannin extraction and analysis followed the methods of Jones et al. (1976) and Terrill et al. (1992). Extraction involved the use of a mix of acetone/water/diethyl ether. Colorimetric tannin analysis was realized with a Butanol-HCl method which allows total condensed tanriins to be measured. A butanol-water (5%) mix was used as a blank. This method had been standardized for forage tannin analysis and has been used successfully for the analysis of tannins in sorghum·grain. Plant Material. Tannins were extracted from seed coats that had been peeled from common bean seed, dried at 60 C for one hour and ground into a fine powder to use in the analysís. An n-heptane treatment was used to facilitate seed peeling and consisted in 12 hours immersion in the n-heptane solution, followed by seed drying and hand-peeling. Different amounts of ground seed coat were used for the parents (30 g) than for the individual recombinant inbred lines (15 mg). This was done to obtain enough purified tannin from the parents to construct the concentration calibration curves used in the analysis of the pro gen y. Three replicates were used per seed coat sample for the analysis. Results and Discussion Soluble and insoluble tannins were successfully purified from DOR364 and G 19833, the parents of the population, to use for the detennination of a calibration curve for absorbance vs. concentration to use in estimating the amount of tannins in the progeny. The color of the dried tannin extract was darker and more reddish for the DOR364 tannins than for the 019833 tannins as would expected from their respective seed coat colors. When the progeny and parents were tested against the calibration curve, a range of seed coat tannin concentrations (expressed in percent) were observed (Figure 1). Soluble tannin was found to range from 13 to 41%, while insoluble tannin ranged from 1 to 8 %, depending on the progeny lines tested in the population. The parents had similar amounts of soluble tannins (both around 26%) while DOR364 contained more insoluble tannin (5.3 %) than did 019833 (2.9 %). The normal distribution of both soluble and insoluble tannins in the population suggests that both are inherited as quantitative traits. Conclusions and future plans We will use a QTL analysis approach to determine the genes involved in tannin content in common bean seed coats and use this information to devise a strategy for reducing specific fractions of tannins with the hope of increasing bioavailability of iron in beans. The strength with which we pursue the strategy of reducing tannins to increase iron bioavailability, must be counterbalanced by the evidence that sorne tannins have been shown to have beneficia! aspects as anti-oxidants and anti-carcinogens. ! :::1 "O ·s: :;; .S -o z Soluble Tannins 13-17 17·2 1 21 ·25 25-29 &3:1 ~37 37-41 Pucent.Jge (%) ~ :::1 "O ! "O .S -o o z 10 Insoluble Tannins 1-2 2-3 3-4 4-6 u 6-7 7 .. Pernnu ge (%) Figure l. Histogram showing the population distribution for soluble and insoluble tannin content among the recombinant inbred lines of the cross DOR364 x G19833. 1.2.4 Identification of QTLs for resistan ce to Thrips palmi in common bean A. Frei1; MW Blair; C Cardona3 ; SE Beebe 2-3 ; H Gu4 and S. Dorn4 1ETH, Zurich, 2SB-2, CIAT, 3IP-1, CIAT ; 4ETH, Zurich. lntroduction Thrips palmi ís a damaging insect pest of common bean and other dicotyledenous crops that was introduced from Asia (Java, Indonesia) into the Americas during the last decade. Starting in the Caribbean, (Cuba, Dominican Republic, Haiti and Puerto Rico) the species spread rapidly into the United States and northem South America (Brazil, Colombia, Ecuador and Venezuela). The greatest damage inflicted to common bean production in Colombia is seen in climbing bean varieties that are grown for the fresh market (including snap beans and Cargamanto dry beans). Sequential plantings, common in the production of snap beans is very conducive to heavy infestations of thrips and whiteflies, which are synergistic in the damage that they inflict. Misuse of insecticides also can lead to resurgence in thrips populations. Host plant resistance on the other hand is a promising component in an integrated cropping system to reduce damage by Thrips palmi Kamy. Therefore the objective of this research was to identify the stability of Thrips resistance across environments and the genes and QTLs controlling this resistance. 84 Methodology The BAT881 x 021212 population, consisting in 139 F5:7 generation RILs, was evaluated over three seasons at a field site in Pradera, Valle, Colombia. The details on phenotypic data collection are reported in CIAT- AR (2000). A genetic map was constructed based on the screening of 151 RAPD markers and 107 microsatellites as reported in CIAT-AR (2002). A linkage map was constructed in two phases; first with the RAPD markers only and then with both RAPD and microsatellite markers. The software package MAPMAKER3.0 (Lander et al., 1987), was used to generate the genetic map, whereby genetic mapping was done by: first, grouping markers at LOD > 5.0 and then, ordering them at LOD > 3.0 using three point analysis with a maximum inter-marker distance of 37.2 cM. Broad sense heritabilities (h2) on an entry-mean basis in seasons and across seasons were calculated for both resistance traits based on mean square ANOV A results, genotypic variance (cr2g), phenotypic variance (cr2p), genotype x season interactions (cr2gs), and error variance (cr2e). Arbitrary linear (orthogonal) contrasts were conducted, using the Scheffe's F-test, to compare among (1) the five most resistant Rll..s, (2) the five most susceptible RILs, (3) the BAT881 parent, (4) the G21212 parent and (5) the susceptible PV A 773 check (Statistix, 1998). Quantitative trait loci (QTL) were identified through single-point regression analysis (SPA) and interval mapping analysis (IM) with the software packages Qgene, and QTLCartographer Vl.21, respectively. In the SPA analysis, probability thresholds of 0.05, 0.01, and 0.001 were used. In the 1M analysis a LOD threshold of 2.5, a window size of 10 cM anda 2 cM walking step were used to determine the presence and location of QTLs and whether there was evidence for more than one QTL on linkage groups with multiple LOD peaks.which had been run on the progeny. Results and Discussion The cross BAT881 x 021212 was found to produce progeny showing transgressive segregation for thrips resistance. Correlations between damage and reproductive adaptation scores were significant in seasons, and significant correlations existed between seasons. Broad sense heritabilities were moderate ranging from 32.4 to 63.4% depending on the parameter and the season (Table 1). The genetic map constructed for the cross had 11 linkage groups, eight of which could be identified as homologous to the chromosomes of the integrated linkage map of common beans, three of which remained unidentified. The most important thrips resistance QTL was located on chromosome b06, linked to two microsatellites and one RAPD marker, explaining up to 27.7% of variance in SPA. This QTL was located at the same region as the bc-3 and Ur-4 resistance genes. Other minor QTLs for thrips resistance were found on chromosomes b02, b03, b08, and b09, sorne of which were located in regions of genes encoding for disease resistance. The identification and mapping of thrips resistance genes is one of the first studies on insect resistance QTLs in common beans and is expected to facilitate the development of resistant bean cultivars by using molecular marker assisted selection. Future Studies Continue mapping with additional microsatellites to achieve complete map coverage in the BAT881 x 021212 population Table l. Estimates or variance components and heritabilities among RILs grown in two seasons in Pradera, Colombia (1999B, 2000B) for Thrips palmi damage and reproductive adaptation (RA) scores (both measured on 1-9 scales). Pararneters WiJhin seasons V ariance components genotypic variance ( rl ~ phenotypic variance (o p). Heritability (h2) Across seasons Variance components genotypic variance ( rl ~ · phenotypic variance (o p), genotype x season interactions (o2gs) error variance ( o2) Heritability (h2) Darnage 1999B 0.36 0.58 0.619 Damage 0.32 0.50 0.03 1.01 0.634 RA 1999B 0.14 0.44 0.324 Darnage 2000B 0.51 0.97 0.526 RA 0.135 0.31 0.02 0.98 0.436 RA 2000B 0.32 0.67 0.478 1.2.5 Development of SCAR and microsatellite markers for Apion resistan ce M.W. Blair1, C. Cardona2, C. Muñoz2 G., H. F. Buendia1, R. Garza3 1SB-2 Project, CIAT 2 IP-01 Project- CIAT 3IN1FAP, Mexico Introduction This year we continued a project begun in 2001 to tag resistan ce to the bean pod weevil (Apion goodmani Wagner) which damages beans grown in Mexico and Central America. Resistance is controlled by two possible mechanisms - either antibiosis involving a hypersensitive response that encapsulates the oviposition site - or antixenosis that affects the preference of oviposition sites. Epistasis between two independent genes, Agr and Agm, has been suggested to control the hypersensitive response. The fact that a few genes control resistance may explain why it has been relatively easy to transfer resistance from Mexican landraces where it is found to new breeding lines with Central American grain types. The objectives of this research were to identify additional markers linked to the genes controlling resistance in the recombinant inbred line (Rll..) population derived from 86 Jamapa x 1117 and to try to identify the chromosornal position of the resistance QTLs identified so far. Methodology Parental Survey and Genetic Mapping. Genetic material consisted in a total of 104 F5 derived recombinant inbred lines (RILs) from the cross Jamapa x 1117, where Jamapa is a susceptible cultivar released in Mexico and 1117 is a resistant breeding line. Screening of susceptible and resistant bulks (of 4 lines each) has continued from last year with a total of over 150 microsatellite markers. A genetic map was constructed with the new dataset of 104 lines and all the polymorphic RAPD markers using the program Mapmaker. RAPD cloning and SCAR primer design. Two RAPD bands (W9-1300S; Z4-800R) that were polymorphic from last year's survey and whích were significantly associated with the resistance phenotype were selected for cloning. The RAPD bands were purified in 4% polyacrylamide gels whích were used directly for a second round PCR amplification. U pon confirming that a single band had been arnplified thís DNA was purified from a 1% Iow melting point agarose gel using a Wizard PCR prep purification system (Promega). The purified insert DNA was cloned into the PGEM-T easy vector system for further analysis. Several clones were picked per ligation reaction and their inserts sequenced using standard techniques, TI and Sp6 primers and an ABI377 DNA sequencer. Specific primers were designed for each unique sequence using Primer 3.0 software and these were tested for their ability to amplify SCAR products. Any monomorphic SCAR products were digested with frequent cutting restriction enzymes (Alul, Haem, Rsal and Sau3AI). Results and Discussion As in last year's results, most of the markers were Iinked to each other in five tight linkage groups representing chromosome bOl, B05, B07, B08 and Bll, each with four or more markers per Iinkage group. The most significant markers occurred on chromosome bOl including the two RAPDs targeted for this SCAR development, W9-1300S and Z4-800R. We will continue to refine the genetic map with new markers for the other four significant linkage groups. BLAST searches identified homologies for the two cloned RAPD bands. Severa} similar clones from each RAPD, showed that W9-1300S was derived from a retrotransposon, while Z4-800R was derived from an unknown Soybean gene. A total of 5 primer sets were designed for the two RAPD band sequences and these were tested on the population parents and on the bulks. AU the primer sets showed monomorphism as SCARs. When the PCR products were digested with the frequent cutting restriction enzymes, polymorphism was revealed for the fragrnent W9-1300-15. This potential CAPS (Cleaved amplified product) marker will be tested across the full population during the upcoming year. Future Plans • Test the new CAPS marker in the entire population and use the same procedure to determine if other SCARs developed from the cloned RAPD fragments can be converted into polymorphic markers. • QTL analysis will be carried out when phenotypic data is available for the entire set of recombinant inbred lines which is expected for later in 2003. 1.2.6 Marker-assisted selection for BGYMV resistance in small-seeded beans C. Quintero1, H. Terán2, S. Beebe2, J. Tohme1 1SB-2 Project; 2IP-1 Project Introduction Bean golden yellow mosaic virus (BGYMV) is a devastating disease of common beans (Phaseolus vulgaris L.) in Latín America. It was first observed in Brazil in the early 60s, and a decade later it became the main biotic constraint to bean production in Brazil, demonstrating its considerable epidemiological potential for this country, Central America, the Caribbean and Mex.ico (Morales, 1994). Although the disease is not found in Colombia, indirect selection through molecular markers can be accomplished successfully (Beebe et al., 2002). Since 1999, a marker-assisted selection (MAS) breeding scheme has been applied using the SCAR marker DOR21 linked to the bgm-1 gene. Breeding for BGYMV has been facilitated and accelerated with this strategy, which is beginning to be used in other countries such as Cuba (Rodríguez et al., 2002). Last year efforts were made to use a second marker for a QTL (quantitative trait locus), identifi.ed in collaboration with USDA-Puerto Rico and the University of Puerto Rico severa! years ago. This marker is also a SCAR (named W12) and was planned to be used as a large-scale tool, such as DOR21, in the identification of plants carrying the resistance genes for BGYMV. Materíals and methods Alkaline DNA ex.traction and PCR for the bgm-1 marker were performed for MAS purposes as described previously (Quintero et al., 2002). Visualization of amplified products was carried out as usual, except that samples were loaded three times per comb instead of two as done previously. Given that the PCR conditions for W12 SCAR used last year yielded either unamplified products or spurious bands that made the screening difficult and time consuming, changes in annealing temperatures of the oligonucleotide primers and different PCR profiles were 88 assayed first in a set of 27 bean varieties, and those that gave best results were applied for screening the F7 red- and black-seeded farnilies . Once PCR conditions were set, a multiplex assay including both markers was performed as follows: each reaction contained 5 Jll of the alkaline DNA diluted 1:1 in sterile water or 20 ng of pure DNA; 0.2 mM each dNTPs, 0.2 JlM each forward and reverse W12 primers, 0.1JlM each forward and reverse DOR21 primers, 10 mM Tris-HCI pH 8.8, 50 mM KCl, 2.5 mM MgCh and 1 unit of Taq polymerase for a total volume of 15Jll. PCR products were resolved in a 0.5X TBE agarose gel with ethidium brornide at a final concentration of 0.02 Jlg/ml. Presence or absence of both SCAR markers was scored. In addition, new primer sets were designed from the sequence of the amplified fragment provided by P. Miklas (USDA-ARS Prosser, WA). These new primer sets were also assayed with bean varieties known for the presence/absence of DOR21 andlor W12 SCARs. The assays were first performed with DNA extracted using the protocol of Afanador et al. (1993) and then with the alkaJine method routinely used forMAS . Results and discussion MAS for introducing BGYMV resistance in small-seeded beans continued as reported last year, and then F1 plants were screened for the presence of the bgm-1 marker. Twenty-four segregant populations (BGMV code 548 to 607), corresponding to F1 multiple crosses for drought stress and high iron content, were evaluated. Of 1794 individual plants, 722 had the bgm-1 marker in homogeneous state and 122 in heterogeneous state. These plants were selected and will be evaluated under drought stress next season. In addition, a set of F6-F7 derived farnilies under drought stress was screened for the presence of DOR21 and W12 SCAR markers. The marker for bgm-1 was detected in either the heterogeneous or homogenous state in 644 out of 1350 F7 farnilies tested. The W12 marker was detected in 585 of these farnilies (Table 1). Both markers occurred ata similar frequency (bgm-1, 43% and W 12, 48% ). Al so, the number of red- and black-seeded farnilies having both markers was comparable: 200 (20%) and 74 (22%), respectively. Table l. Families expressing or lackíng two SCAR markers for resistance to BGYMV in F6-F7 derived fa mili es. Marker Color Grou~ Heterogeneous1 Present Absent bgm-1 Red 55 407 556 Black 14 168 150 Sub total 69 575 706 WJ2 Red 478 540 Black 107 225 Subtotal 585 765 TOTAL 1The bgm-1 marker is co-dominant for the resistant and susceptible allele, which permits defining heterogeneous class, while the Wl2 marker is dominant and only presencelabsence classes are scored. Total 1018 332 1018 332 1350 A multiplex assay using DOR21 and the original W12 primers was conducted in four bean varieties and pure extracted DNA. Identical PCR products were obtained when comparing individual amplification of the SCARs and the multiplex (Fig. 1). Figure l. Multiplex using W12 and DOR21 markers for BGYMV resistance. Lane 1, W12 check; lane 2, DOR21 check showing susceptibility allele (upper) and resistance allele (lower); lane 3, lanes 4-6, Tío Canela 75, SAMl, DOR 364, EMP496, multiplex; lanes 7-10, W12 on tbe same set ofvarieties; lanes 11-14, DOR21 amplified on the same set ofvarieties. Using a multiplex assay would speed up the MAS process, but a more specific W12 marker would be needed. Among six sets of designed forward and reverse primers, two sets were selected. The first yielded even more spurious bands than the original W12 oligonucleotides, but the second was more specific and produced a single band at 614 bp (Fig. 2). 90 Figure 2.PCR amplification of shorter region W12 marker (614 bp). Lane 1-8, bean varieties whose DNA was extracted using the alkane method; Lanes 9-13, pure DNA checks, Tío CaneJa 75, DOR 364, G19833, SAM1 and EMP496. Lane 14, original W12 fragment (732 bp). These new W12 oligonucleotides were more specific, at least in the small set of varieties selected for this study. Since no amplification was obtained in 019833 in contrast with the original oligonucleotides, it would be useful to run this new set of primers in the mapping population DOR 364 x 019833, prior to applying it for large-scale screening. Conclusions and ongoing activities The bgm-1 gene is being introduced through MAS on small-seeded farnilies also tolerant to the main biotic and abiotic constraints to bean production. Attempts were made to bring W12 SCAR to a high-throughput screening strategy, comparable to that of DOR21. Changes in PCR profiles for the amplification of the original fragment yielded more confident results than last year assays and were started to be used. A multiplex of both SCARs available for BOYMV resistance was successfully run in a set of bean varieties with known performance for both SCARs. Troubles with spurious bands or no amplification seemed to be overcome as a more specific set of forward and reverse primers for W12 fragment was designed. New W12 primers should be tested in the mapping population DOR 364 x 019833. References Afanador Kafury, L.; Haley, S.D.; Kelly, J.D. 1993. Adoption of a "mini-prep" DNA extraction method for RAPD marker analysis in common bean (Phaseolus vulgaris L.). In: Bean Improvement Cooperative. Annual Report (USA) 36:10-11. Beebe, S.; Miklas, P.N.; Quintero, C.; Pedraza García, F.; Terán Santofimio, H.; Tohme M., J. 2002. Implementación de un sistema de alta capacidad para selección asistida por marcadores en frijol común. In: Congresso Nacional de Pesquisa de Feijao (Vi~osa, MG, Brazil, 7:2002). Universidade Federal de Vi~osa, Depto. de Fitotecnia, Vi~osa, BL. p. 45-49. (Abstracts) Quintero, C .; Terán, H.; Tovar, E.; Beebe, S. And Thome, J. 2002. Marker-assisted selection for BGM resistance in common beans. In: CIAT (Centro Internacional de Agricultura Tropical). 2002. SB-02 Project annual report. CIAT, Cali, CO. p. 103-104. Morales, F.J. 1994. El mosaico dorado del fríjol: Avances de investigación- Bean golden mosaic: Research advances. Centro Internacional de Agricultura Tropical, Cali, CO. 193 p. Rodríguez, 0 .; Faure Alvarez, B.; Alvares, Y.; Quintero, C.; Terán Santofirnio, H.; Beebe, S.; Arencibia, A. 2002. Selección de genotipos resistentes al virus del mosaico dorado en líneas y variedades de fríjol común (Phaseolus vulgaris L.). Phytopathology (USA) 92(6) (Supp):S 133. Acknowledgments We would like to thank Eliana Gaitán, BRU, CIAT for providing assistance in primer designing. 1.2.7 Marker assisted selection of Arcelin-derived bruchid resistance MW Blair1, S Prieto1 ; C Cardona2 1SB-2 Project, CIAT; 2 IP-1 Project, CIAT Introduction Last year we described the testing on a total of 63 genotypes (including 7 wild accessions of common bean that were the sources of the seven variants of arcelin known to exist; 28 advanced breeding lines from the bruchid resistance program (either RAZ or GG designations); and 28 bruchid-susceptible parents used in crosses with Arel or Are 5 containing lines) of seven rnicrosatellite markers that were linked to the Arcelin resistance gene which is the most effective resistance factor for the most common storage pests of common bean, narnely the Mexican bean weevil , Zabrotes subfasciatus (Boheman). This year, our objective was to replace the serological/ protein based selection of arcelin in our Red Mottled common bean breeding program with a genetic assay using two of the rnicrosatellites that were most closely linked to arcelin. We analyze the two markers for their ability to distinguish susceptible and resistant genotypes and individual alleles of the Arcelin gene and confirm these results with both protein assays, insect feeding tests and 92 replicated trials for the most resistant lines. The long-term objective of this work is to increase the efficiency of breeding for multiple constraint resistance and facilitate the pyramiding of bruchid resistance with other biotic and abiotic stress resistances. The conversion of the protein based selection of arcelin to a usable DNA marker obviates the need for arcelin-specific antibodies and protein electrophoresis and streamlines arcelin selection with the widespread use of other SCAR markers that we have also embarked on in our breeding prograrn. In addition to its compatibility with other types of DNA based markers, the advantages of using the microsatellites over the time-consuming protein based selection was they were amenable to high-throughput and fast analysis. Methodology Genetic materials and Phenotyping. A set of 796 F4 and Fs derived advanced lines from multiple crosses between RAZ lines and susceptible parents were used in the marker assisted selection experiments. Of these, 772 were Andean genotypes from the breeding program for Red Mottled beans and 24 were Mesoamerican genotypes from the breeding program for Small Red beans. All 796 lines were tested for bruchid resistance using a single replicate of 30 seeds infested with 6 pairs of Zabrotes subfasciatus (Boheman) in a small mesh covered clear plastic vial whose walls were covered with sandpaper (No. 150, rough side of sandpaper facing inwards) to avoid egg-laying on the plastic surface rather than the bean seed coat. Data was collected on number of eggs, number of emerged adults and percentage emergence. Genotypes with Oto 15% adult emergence were classified as highly resistant (HR), from 15 to 30% as resistant (R), from 30 to 50% as intermediate (I) and from 50 to 100% as susceptible (S). A replicated, confirmation test was conducted with the 60 best, highly resistant advanced lines as well as the 2 worst, susceptible lines and the check varieties, RAZ44 (resistant) and ICA Pijao (susceptible). Days to emergence and percentage of darnaged seed were measured along with the above pararneters using the same protocol. A total of 4 replicates were used in these trials. DNA extraction. Two DNA extraction techniques were attempted. One was a rapid, high- throughput "microprep" method based on alkaline lysis. The other was an ammonium acetate based "miniprep". Tissue was harvested in the greenhouse as leaf disks cut with a hole-puncher for the microprep or newly emerging trifoliates for the miniprep. The DNA from the miniprep was found to be more adequate for microsatellite amplifications than the DNA from the microprep so subsequent analysis were done with this techniques. Microsatellite markers. Two rnicrosatellite markers were used for the marker assisted selection scheme: Pv-AG004 (X04660) and Pv-ATCT001 (M68913). Microsatellite amplifications were conducted according to standard PCR protocols. Microsatellite markers were run at 1800 volts, 120 Wats and temperature of 45C for one to two hours on 4% polyacrlamide gels and silver-stained with a recirculating tank system. Alleles were identified as reported in last year's annual report for the 63 parental materials used in multiple crosses to generate the advanced lines tested this year. Additional Tests. In addition to the marker assisted selection screening of the 796 lines, two other tests were conducted: 1) an additional microsatellite evaluation on a blind sample of 102 genotypes for which known arcelin reactions and alleles had already been detennined but for which marker genotype was unknown and 2) a protein assay was conducted to confirm the presence or absence of arcelin in the 64 genotypes selected in for the replicated confirmation test described above. Protein extraction and arcelin determination. 0.15g of bean flour was dissolved in 250 ul of extraction buffer, vortexed and centrifuged 14,000rpm for 15 minutes. The supematant was transferred and mixed with 50ul of cracking buffer, which was vortexed, boiled for 5 minutes, allowed to cool and centrifuged before loading 5 ul onto a stacking polyacrylamide gel. Samples were run ata constant 150 volts until the sample passed into the running gel where a constant 25 mA was maintained. Protein gels were stained for 4 to 5 hours in 120 m1 of 0.25% Coomassie Blue R-250, then transferred to Destaining solutions 1 and n for approximately 4 to 5 hours. Results and Discussion The phenotyping results are reported in another section of the CIA T annual report, therefore here we will concentrate on the results of the genotyping that was carried out. Preliminary to the process of marker assisted selection we carried out a parental survey, in which we found the number of alleles and level of polymorphism present for each of the seven microsatellites evaluated. In this parental survey, two microsatellites were found to be more tightly associated with the Arcelin gene: of these the microsatellite Pv-AG004 (X04660) was more polymorphic than Pv-ATCT001 (M68913) presenting 6 alleles versus 2 alleles, respectively. For both markers, unique alleles were found for the parent that provided Arcelin 1 (012882) giving a diagnostic test for this widely used allele of the Arcelin gene. Although this marker produced multiple bands, the pattem of bands produced was diagnostic for the 012882 allele and for resistant genotypes derived from this source, producing bands that were 184, 195 and 207 bp long, while susceptible genotypes had bands that were 207 and 245 or 207 and 285 bp long. Furthermore for Pv- AG004 (X04660) there was also a unique allele for the parent that provided Arcelin 5 (002771) that was different from the Arcelin 1 associated allele. This banding pattem consisted of fragments that were 184, 203 and 207 bp long. Meanwhile, the other marker, Pv-ATCT001 (M68913), could not distinguish between susceptible genotypes and Arcelin 5 containing genotypes. This marker produced single amplification products were the resistant allele of Arcelin 5 was associated with the 190 bp band while the susceptible allele of Arcelin 5 was associated with the 195 anles with buffer 1 b: sarq>les with buffer la T f T: Phaseolus (control) r: sarq>les with buffer 5 g: sarq>les with buffer 5a g e d e: sarq>les with buffer 2 d: sarq>les with buffer 3 e: sarq>les with buffer 4 h h: sarq>les with buffer 6 i: sarq>les with buffer 6a j: sarq>les with buffer 5b e j Fig. l. Ore-dirnmsional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of total proteins from root of four cassava cultivars. In order to standardize the protocol in cassava different extraction buffers were evaluated. A group of 33 genotypes, 5 plants each, that showed high root protein in an analysis conducted in 2001 were requested from the genetic resource unit and sent to the green house for hardening (Table 5). Table S. List of cassava varieties with high protein cooteot in ao evaluation cooducted in 2001 that are to be re-evaluated again this year. Clone and protein (%) Clone and protein (%) Clone and protein (%) CM5620-3 8.31 MCOL2436 6.25 :MBRA 101 5.94 SM 1406-1 8.13 :MBRA26 6.25 MCOL219 5.94 MCOL689B 7.75 MCR 136 6.13 MGUA33 5.94 MCOL 1563 7.38 MGUA9 6.13 CM 7310-1 5.88 MGUA 76 6.94 MGUA91 6.06 MCOL678 5.88 MCR 142 6.63 MMEX108 6.06 MMEX95 5.81 CM 696-1 6.44 SM 629-6 6.00 MGUA 79 5.81 CM 3199-1 6.44 SM 673-1 6.00 :MBRA 300 5.75 SM 734-5 6.44 MCOL2532 6.00 MCOL2~59 5.75 MCR38 6.31 MGUA 19 6.00 :MBRA 1384 5.75 . MGUA86 6.31 CM 3236-3 5.94 MCOL2694 5.75 Conclusions This year, inter-specific hybrids from wild relatives with high root protein and cassava were evaluated and results reveal the consistency of the trait, amino acid profile and preliminary SDS-PAGE analysis was also conducted with the root proteins. Future perspectives include genetic back crosses, to cassava, of inter-specific hybrids with high protein and evaluation of putative high root protein cassava varieties References Callaban, F. E., Jenk.ins, J. N. Creech, R. G. and Lawrence, G.W. 1997. Changes in Cotton Root Proteins Correlated with Resistance to Root Knot Nematode Development. The Journal of Cotton Science 1: 38-47. Carvalho, L.J.C., Cascardo, J.C., Ferreira, M.A. and Loureiro, M.E. 1992. Studies on proteins and enzymes related to tuberization and starch biosynthesis in cassava roots. In CBN: 234-238. Gonzales, P.L., Gutiérrez, M.M. y Fuchs, M. 2001. Estandarizacit,tn de la metodolog¡a para caracterizar genotipos de algodt,tn (Gossypium sp.) mediante patrones electrofor,ticos de prote¡nas. Rev.Fac.Agron.(Maracay) 27:95-103 Hovenkamp-Hermelink. J.H.M , Jacobsen, E .• Ponstein, A.S., Visser, G.F., Vos-Scheperkeuter, G.H., Bijmolt, E.W., de Vries J.N., Witholt, B. and Feenstra, W. J. 1987.Isolation of an amylase-free starch mutant of the potato (Solanum tuberosum L.). T AG 75:217-221 Jasso, D., Romero, J., Rodríguez, R. and Angula J .L. 2002. Characterization of Proteins from Sunflower Leaves and Seeds: Relatiopnship of Biomass and Seed Yield. Reprinted from: Trends in new crops, and new uses. ASHS press. Laemmli, U. K. 1970. Cleavage of Structural Proteins during the Assembly of the Head of Bacteriophae T4. Nature Vol. 227: 680-685 138 Shewry, P.R., Clowes, A., Tatham, A.S. and Beeching, J. 1992. Oportunities for manipulating the amount and composition of proteins in cassava tuberous roots. First Intemational Scientific Meeting of the Cassava Biotechnology Network. CIAT, CBN. Or (Proceedings CBN: 251-254) 1.2.19 Mining the Primary Gene Pool: Green Mites (CGM) Resistance Genes from Manihot esculenta sub spp Fabellifolia N. Morante, J. Guerrero, A. Bellotti, J. Marin, C. Ospina, H. Ceballos, M. Fregene CIAT Introduction Following a very dry spell in January 2002 at CIA T Palmira and a subsequent heavy incidence of the green mites, 4 inter-specific hybrid families, CW68, CW65, CW67, and CW66, from the wild Manihot accession MFLA 437- 007 showed a very high leve! of resistance to the green mites with an almost equal number of susceptible(score of 3-4) and resistant (score of 1-2) genotypes. Bulk segregant analysis (BSA) was used to identify 4 SSR markers NS74; NS217; NS260; SSRY330 polymorphic in the bulks and individuals of the inter-specific families resistant to green mi tes (CIA T 2002). At the same time, an attempt was made to transfer the resistance observed in the inter-specific families to elite cassava parents. A total of 45 BC1 families were developed, planted in a seedling trial, and a preliminary evaluation for mites conducted last season. Selected individuals of these BC1 families were extensively crossed to C.MD resistant parents used forMAS at CIAT, to combine C.MD and CGM resistance in progenitors meant for Africa, to produce BC2 families. The BC1 progenies were also cloned and planted in a single row trial for the evaluation of resistance to green mites during the dry spell early next year. We describe here evaluation of the putative markers in the inter-specific families, and the development of BC2 families from selected BC1 individuals. Methodology The SSR markers polymorphic in the bulks were evaluated in the 4 F1 inter-specific families and a simple regression analysis conducted using Microsoft excel. The markers found to explain a significant part of phenotypic variance of CGM resistance in the analysis of the 4 F1 families were then analyzed in the parents of the 45 BC¡ families .. SSR analysis of the 4 markers was as described by Mba et al (2001). The BC1 families were also evaluated for resistance to mites, a rather preliminary evaluation since only one plant is available per genotype. Based upon the preliminary CGM resistance evaluation of the 45 BC1 a large number of putatively resistant BC1 progenies were crossed to C.MD resistant parents for the generation of BC2 families from which C.MD and CGM resistant lines can be selected from for the generation of parents for breeding in African gene pools. Table l . Sexual seeds of BC1 families produced that combines resistance to CGM and CMD resistance ITEM Father Mother lcooe !No. of Seeds 1 C-4 CW74- l Al 94 2 C-4 CW236-14 A2 102 3 C-4 CW235-8 A3 135 4 C-4 CW234-1 2 A4 105 5 C-4 CW234-19 A5 6 6 C-4 CW235-72 A6 17 7 C-6 CW235-2 A7 21 8 C-6 CW232-8 A8 11 9 C-6 CW213-l A9 3 10 C-6 CW234-17 Al O 6 11 C-19 CW234- 17 All 6 12 C-19 CW235-8 Al2 23 13 C-33 CW235-2 Al3 5 14 C-33 CW234-12 Al4 38 15 C-33 CW2 17-7 Al5 12 16 C-33 CW232-8 Al6 9 17 C-33 CW258-17 Al7 12 18 C-33 CW235-100 Al8 15 19 C-39 CW235-8 A l 9 8 20 C-39 CW257-25 A20 54 21 C-39 CW258- 17 A21 5 22 C-127 CW234-8 A22 27 23 C-127 CW234-17 A23 12 24 C-127 CW234-19 A24 25 25 C-127 CW234-32 A25 51 26 C-127 CW235-2 A26 9 27 C-127 CW235-8 A27 16 28 C-127 CW235-51 A28 2 29 C-127 CW235-72 A29 2 30 C-127 CW235-100 A30 14 31 C-127 CW258- 19 A31 16 32 C-243 CW234-17 A32 2 33 C-243 CW234-19 A33 5 34 C-243 CW235-2 A34 2 35 C-243 CW257-10 A35 10 36 CW219-3 C- 127 A36 1 TOTAL 881 140 Results Simple regression analysis of SSR markers NS74; NS217; NS260; SSRY330 in the 4 inter- specific families had coefficients (R2) of 46%, 30%, 30%, and 5% in the 4 families respectively. The surprisingly low regression coefficients for resistance that is apparently controlled by a major gene (CIA T 2002) might be due to the single year evaluation of these families for CGM resistance upon which the analysis was based. CGM incidence tends to have focal points therefore a large part of phenotypic variance in an evaluation is due to the environment. The best way to avoid this is to evaluate over 2 or 3 growing cycles. A second more in depth evaluation of these families is being carried out in Santander the Quilichao over two dry seasons, a period of 18 months September 2002 until March 2004. The preliminary evaluation of the BCl families is being repeated this year in a clonal observation trial before further regression analysis is carried out. An evaluation of the SSR markers associated with CGM resistance in the inter-specific hybrids in the parents of the BC1 families revealed that the bands associated with CGM resistance were not always polymorphic between the parents of the back cross populations. An effort was therefore initiated to identify more markers linked to CGM resistance, BSA, using additional SSR and RAPD markers is being used to identify additional markers for evaluation of resistance. Transfer of a gene to other crosses via MAS using a single marker associated with the gene is fraught with problems of Iinkage disequilibrum, based upon the frequency of that allele in the gene pool. Many more markers are required to eliminate the problems of linkage disequilibrum encountered in diverse genetic backgrounds. Individuals of the BCt families that appeared to possess resistance to CGM in focal points of great CGM damage in the seedling trial were crossed to CMD resistant parents to obtain recombinants that carry CMD and CGM resistance. Table 2 shows the number of seeds by families of the new crosses. These crosses have been established in vitro from embryo axes to enable it to be shared with collaborators in Africa once CMD and CGM resistance have been confirmed by MAS and phenotypic evaluation respective! y. Conclusions Markers associated with CGM resistance in bulks of 4 inter-specific families were evaluated in the entire families and rather low regression coefficients were found. This is most likely dueto the environmental effect in the phenotypic data. BC2 farnilies have been generated from BCl individuals towards an introgression of this resistance into elite cassava gene pools. References CIAT, (2002). Annual Report Project SB2, Assessing and Utilizing Agrobiodiversity through Biotechnology, CIAT, Cali, Colombia, pp 239-24l.Dellaporta SL Wood J, Hicks JR (1983) A plant DNA minipreparation: version 11. Plant Mol Biol Rep 1:19-21 Mba REC, Stephenson P, Edwards K, Melzer S, Mk:umbira J, Gullberg U, Apel K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: towards an SSR-based molecular geneúc map of cassava. Theor Appl Genet: 21-31 1.2.20 Depelopment of a novel approach for the analysis of diallel mating designs for better understanding the inheritance of quantitative traits Hemán Ceballos, Juan Carlos Pérez, Fernando Calle, Nelson Morante y Jorge lván Lenis CIAT As stated in the report from the previous year, three diallel mating designs have been evaluated for the three main target environments (sub-humid environments, acid soil savannas and mid-altitute valleys). Diallel trials were planted in 2001 and in the northem coast and acido soil savannas also in 2002. The relevance of these studies certainly líes on the information they will produce. However, it is important to emphasize that these are diallel studies that offer new altematives to extract information and requires development of new models for their analysis. In this section a brief description of this innovative approach will be described. This research will serve as thesis work for Ph.D students from Vietnam and Uganda (a female and a male, respectively), as well as for a M.S. thesis for one of the assistants working in the project. Theoretical model for analyzing ínter and intra-family vanatzon. The most important difference between these diallels and the traditional ones is the fact that thirty clones were used to represent each cross, and individual measurements were made for each individual genotype making up the full-sib cross. Most diallels will harvest all the individual genotypes for each cross ignoring the within-family variation. This is because the main target is identifying good families or parents. However in the case of cassava breeding the individual genotype (clone) is the central point of attention whereas in other crops such as maize, the main target is the family or F1-eross. lt is important to emphasize the relevance of this distinction because the current method of selection (particular! y in the first stage of selection, called Clonal EvaluationTrials) the cassva breeding project is trying to develop a more systematic evaluation process in which we select on the family basis (all the clones derived from a given parental line), and then the best clones from that family (the individual clone that will eventuall be released 8-10 years leater). One of the severa! questions these diallels aim at answering is what is the relative genetic variability between and within farnilies. This is obviously fundamental for a more scientific process of selection. In other words, should we focus in developing good families of clones or should we concentrate on developing farnilies of clones that have large gentic variability to be exploited. Table 1 illustrates the progress made in developing a quantitative genetic model for the analyses of the diallels. The information presented was exposed to Dr. José Miranda Branco Filho who is an eminent quantitative geneticist who co-authored with Arnell 142 Hallauer the most widely used book on the subject (Quantitative Genetics in Maize Breeding, lowa State University Press). The issue of the model and altemative ways of using the information is still underway and there are excellent perspectives that these analyses will in fact be a relevant contribution in the area of quantitative genetics. That is, the studies will hopefully not only be useful for cassava but also for other crops as well. Selections for the Mid-altitude valleys. Tables 2 and 3 present the results of the combined analysis of variance for the two diallel evaluations conducted in the mid altitude valleys environment and the values for GCA effects, respectively. One sriking result in this analysis is the significante of SCA effects, which are related to the dominance of heterosis present in the hybrid clones evaluated in the studies. In Table 3 a more precise estímate for GCA effects that the one presented previously in 2002 is presented. The information provided is a good example of the complexities involved in cassava breeding. The parent generating the best yielding progenies was MPER 183 (the highest positive GCA value) but it showed very deficient performances regarding Harvest Index and Dry Mtter Cntent (both with negative GCA estimates). On the other hand CM 6740-7 has an excellent GCA estímate for dry matter but intermediate performance regarding fresh root yield and harvest index. The clone SM 1741-1 has an good overall performance as a parent, particular! y in relation to dry matter content. Selections for the Sub-Humid Tropical Environment. Following the same criteria for the previous section Tables 4 and 5 present the relevant results from the diallel studies for the sub-humid environmtents. There are sorne interesting differences between the two data sets. The most striking one is that here SCA effects did not reach statistical siginificance (Table 4). This may be the result of the strong dry spell in this environment, which has a strong effect on dry matter content. For the other traits both GCA and SCA showed highly significant statistical differences. Table 5 summarizes the GCA effects for each of the nine parents involved in this diallell. As in the previous cases, it is obvious that there is no perfect parentalline. Good GCA for yield productivity is generally associated with a negative value for dry matter content and/or harvest index, and viceversa. It was disappointing to see the poor performance of SM 1565-17 regarding dry matter content (GCA =- 1.467). One pf the promising clones for this environment was SM 1411-5, whose breeding value based on the results in Table 5, is clearly highlighted. Table l. Theoretical model for the quantitative genetic analisis of diallel crosses in cassava. In addition to the usual variation among Fl crosses, within family variation has been included. Tbey were considered ftXed and random genetic effects, respectively. Source of variation Degr~s of freedom Mean squares expectations Locations (L) a-1 R~alocations l!{_r-1) Among F 1 crosses [p(p-1)/2]-1 S2 e + gS2 ~ + grS2 Amon2 Fl x L + graSL AmonR Fl OCA p-1 S2e + _gS2e + rS2sc;, tl + !ÍE_-~ S~+ raS2ACE + ra(p-2) S2ocA SCA p(p-3)/2 S2e + gS2~ + rS\cA x L + raS2scA Within F 1 crosses (p_{p~l)/2)(g-1) S¿ e+ rSL Wilhin FJ XL+ raS¿ Wilhin Fl Among Fl crosses x L (a-l)([p(p-1)/2]-l) S\ + gS2 ~ + grS2 Amon2 Fl x L GCAxL (a-1)(p-1) S2e + gS2e + rS\cA XL +·r(p_-2) s~A '-.1 SCAx L (a-l)(p(p-3)/2) S2e + gS2[ + rS2scA XL Error (a) a([p(p-1)/2]-1)(r-l) S2e + gS\ Within F 1 crosses x L ~1)/2)(g-1)(a- l) S¿ e+ rSLWilhin Fl XL Error (b) a(p(p-1)/2)(g-l)(r-ll___.__s''e -- -- - Where: a = number of locations p = number of progenitors in the diallel S2AmonaFI = Variation among averages for each F1 cross S2 GCA = general combining ability S2 ~ = experimental error type a S2 ........ x L = Ínteraction Of each effect With the environment r = number of replications within location g = number of clones within each F1 cross S2w.lhin Fl = Variation among clones within F 1 crosses S2scA == specific combining ability S2e =experimental error type b 144 1 1 Table 2. Combined analisis of variance for a diallel study conducted in the mid altitude valleys at Palmira and Jamundí (Valle del Cauca), Colombia. Mean square sfor yield (kg 1 plant), Harvest iodex aod dry matter content are presented ooly for the among crosses component of the study. Mean Squares Source of variation Fresh root yield Harves Index Dry matter df (kg 1 pl) (0-1) (%) Locations (L) 1 3.949 ns 0.496 * 563.608 Rep/ L 4 17.579 0.028 19.009 AmongF1s 35 3.125 ** 0.009 ** 9.685 GCA 8 5.760 * 0.028 ** 26.855 SCA 27 2.461 ** 0.004 ** 4.597 Among F1s x L 35 0.645 ns 0.001 ** 3.200 GCA x L 8 1.392 ** 0.002 ** 11.304 SCAxL 27 0.424 ns 0.001 * 0.799 Error 140 0.447 0.001 0.899 Table 3. General combining ability (GCA) effects from the diallel study in two mid-altitude valley environments. Nine progeoitors were involved in the diallel design. Fresh root yield Harvest Index Dry matter content CJon (kg¿pl) (0-1) (%) CM 6740-7 0.003 -0.009 0.608 SM 1219-9 0.341 0.024 -0.572 ~M 1278-2 -0.426 0.010 0.914 SM 1636-24 -0.314 -0.022 -0.464 SM 1673-10 -0.308 0.007 0.641 SM 1741 -1 0.052 0.037 1.069 ~C1 -0.313 0.015 -0.444 ~ECU72 0.341 -0.048 -1.079 [MEER 183 0.624 -0.014 -0.672 ist.Dev. Gi 0.172 0.0060 0.489 ~t. Dev. (GI- Gj) 0.257 0.0097 0.734 ** ** ns ** ** ** ns Table 4. Combined analisis of varíance for a diaUel study conducted in tbe sub-humid environments at Pitalito and Santo Tomás (Atlántico), Colombia. Mean square sfor yield (kg 1 plant), Harvest index and dry matter content are presented only for tbe among crosses component ofthe study. Mean Squares Source of variation Fresh root yield Harves Index Dry matter df (kg 1 pi) (0-1) (%) Locations (L) 1 3.341 0.258 * 118.418 Rep/L 4 0.732 0.022 21.699 AmongFls 35 l.l40 ** 0.009 ** 6.744 GCA 8 2.976 ** 0.021 * 20.974 SCA 27 0.596 ** 0.006 ** 2.528 Among F1s x L 35 0.366 ** 0.003 ** 2.356 GCAxL 8 0.797 ** 0.008 ** 5.642 SCAxL 27 0.238 0.002 * 1.383 Error 140 0.157 0.001 0.677 Table S. General combining ability (GCA) effects from the diaUel study in two sub-humid environments. Nine progenitors were involved in the diallel design. Fresh root yield Harvest Index Dry matter content ~Ion {kg/pl) (0-1) (%) !MTAI 8 0.004 -0.011 0.053 ~M 6754-8 -0.372 -0.001 0.241 Pi 8027-3 -0.126 -0.005 0.858 ~M 805-15 -0.438 -0.032 0.055 ~M 1565- 17 0.267 0.039 -1.467 ~M 1411-5 0.139 -0.019 0.919 ~M 1219-9 0.220 -0.010 -0.032 ~M 1657-12 0.020 0.010 -0.280 ~M 1665-2 0.286 0.028 -0.347 ~t.Dev. Gi 0.130 0.013 0.345 St. Dev. (GI- Gj) 0.195 0.019 0.518 Selections for the Acid-Soil Savannas Environment. Results from the diallels at CORPOICA - La Libertad for the acid soils environments (Table 6) indicate a strong interaction with the environment. In fact this makes sense because the two locations had sharp contrast regarding their soils. One had more stressful savanna conditions and the other had alluvial soils, which support excellent yields and not as severe disease development. The large interactions with the environment (error term for the main genetic 146 ... ** ns ** ** ** effects) resulted in sorne of them not being statistically significant. Individual location analysis would yield highly significant statistical differences. Table 6. Combined analisis of variance for a diallel study conducted in the acid soil savannas at CORPOICA La Libertad(Meta), Colombia. Mean square sfor yield (kg 1 plant), Harvest index and dry matter contentare presented only for the among crosses component of the study. Mean Squares Source of variation df Fresh root yield Harves Index Dry matter Locations (L) 1 172.011 ** 0.263 16.988 Rep/L 4 3.743 0.060 0.551 AmongF1s 44 0.631 0.015 * 0.076 OCA 9 1.462 0.048 0.190 SCA 35 0.417 0.007 0.047 Among F1s x L 44 0.476 ** 0.008 ** 0.042 GCAxL 9 1.244 ** 0.023 ** 0.102 SCAxL 35 0.279 0.005 0.026 Error 176 0.189 0.003 0.019 Tab1e 7. General combining ability (GCA) effects from the diallel study in the acid soil savannas environments. Ten progenitors were involved in the diallel design. Fresh root yield Harvest Index Dry matter content (:Ion (kg/pl) (0-1) (%) CM4574-7 0.184 -0.001 1.106 CM 6740-7 0.076 -0.008 0.182 CM 7033-3 -0.110 -0.006 -0.348 SM 1219-9 0.089 0.029 0.676 SM 1565-15 -0.073 -0.004 1.434 SM2058 - 2 0.078 0.006 -0.301 SM2219 - 11 0.276 0.041 0.593 HMC 1 -0.127 0.019 -0.032 MPER 183 -0.326 -0.077 -3.422 MTAI8 -0.067 0.000 0.114 St.Dev. Gi -0.067 0.000 0.114 S_t. Dev. (GI - Gj) 0.216 0.029 0.062 As expected CM 4574-7 hada good performance as a parent with positive GCA for yield and, particularly, for dry matter content (Table 7). However, better that CM 4574-7 was SM 2219-11 with positive GCA values for the three variables evaluated. ** * * ** ** lllustration of the kind of additional infonnation that this analysis could provide. Below several examples are given on the way the additional information based on the within Fl family variation could provide. The model is taking advantage of the fact that a single genotype can be vegetatively propagated. Therefore each genotype (clone) was planted in two (mid-altitude valleys) or three (acid soil savannas and sub-humid environments) locations, with three replications in each location. An error exists for the within family variation which is not ordinarily available in diallel studies. In other words, it is possible to estimate the genetic component of the within family variation. That information can be contrasted with the one obtained from the among-Fl families variation and if differences are large, then circunstancial evidence of the occurrence of strong epistatic effects can be produced. Moreover, individual analysis of each variable can help identifying interesting events that otherwise would go unnoticed. For instance, averages for each F1 family and standard deviations (within each F1 family) can be ploted. This information can be very useful and as far as it is known the kind of analysis has never been described in the literature. The basic idea is that parentallines that have many dominant alleles affecting a given trait will yield more homogeneous progenies than a progenitor with recessive alleles. This is the principie in which the graphic approach for diallel analysis (Hayman 1954a; 1954b; 1958; Mather and Jinks, 1971). The first example tums around the reaction to white flies evaluated in the diallel experiment evaluated in Jamundí (Valle del Cauca, Colombia). Figure 1 shows the result for the averages of the F1 crosses as well as the variation (standard deviation) among the 30 clones that made up each F1 cross. 148 1.3 1.2 e • • o ;; ca • -.::: 1.1 MECU7¡ SM 1278-2 • ca • > ~ .. e • .l! 1.0 MECU72 x SM 1671-1 • • .S .~ . • -J::. .. j MEC,72 x SM 1673-1 O •• ... 0.9 • • • o - MECU,xHMCl en e • • o -· 7; 0.8 MECU72 x CM674oe • • • > ~ "C "C ... ca "C e ca .. (1) MECU72 x SM1636-24 •• • 0.7 MECU7~ MPER 183 • 0.6 • MECU72 x CM1219-9 • 0.5 1.5 2.0 2.5 3.0 3.5 4.0 Famlly averages for reactlon to whlte files (1 =reslstant; 5=susceptlble) Figure l. A verages and standard deviations for the reaction to white llies in a diallel study evaluated in Jamundí, Valle del Cauca, Colombia. Data points for the 8 different progenies with MECU72 as common parent are identified. In Figure 1 the variation in the response to white flies attacks is clearly illustrated through the values along the horizontal axis. Lower scores meant more resistant phenotypes. Higher scores indicate susceptible reaction. It was very interesting to see the location of the progenies from MECU 72 who has been shown to have resistance (antibiosis) against the white flies. As expected most of the progenies from MECU 72 where at the left of the figure, indicating a resistant reaction to the insect. This is as much as a traditional diallel analysis could go. What is interesting and innovative is the possibility to analyze the 4.5 e: ~ ¡¡ -.: 1111 > ~ e 1111 -e: ~ j ... .2 e: o ;; 1111 > Gl 'O ... 1111 'O e: S C/) \.80 í j \ .40 f • ';: \.20 1 ! 1 \ .00 t i 0.80 .S ., o.eo • 1 • • o.40 L---=::::~~=----......------....-----------J 1.00 \ .50 2.00 2.50 3.00 3.50 Figure 2.A verages and standard deviations from a dialllel study at Santo Tomás (abo ve) and Ptialito 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 1.2 • vior • • ••• • • •• • 1.7 2.2 2.7 3.2 Famlly averages for thrlp score (1::reslstant; S::susceptlble) (below) in the Altántico Department for the reaction to thrips (l=resistant; S=susceptible). Dominant and recessive ''behavior'' does not refer to the traditional meaingin for dominance and recessiveness. Dominance refers either to a trait where non-additive effects are important or cases where the parental lines may have loci at the bomozygous dominant status. 150 variation around the standard deviations within families. Sorne crosses had very small standard deviations (MECU72 x CM1219-9) indicating a very uniform family. Other progenies (MECU72 x SM1278-2) have a much higher within familiy variation. This contrast is extremely interesting for understanding the genetic structure for relevant traits. Continuing with reactions to pests, Figure 2 illustrates the results from the diallels targeting the sub-humid conditions (different progenitors that those for mid-altitude valleys and acid soil savannas). These results are typical of the trends that can be obtained by putting together the averages for each family and the standard deviation wühin families. First, there is a clear consistency between the results of the two locations, with an upward trend at the right of the figures . The results agree with the expectations, clones with lower averages (resistant to mites) show a more uniform progenies (smaller standard deviations) than the susceptible ones at the right of the figures. The tendencies observed in these plots could be explained as an statistical artifact: families with larger values are likely to have larger standard deviations. However, as will be shown below, for other traits the opposite is observed. In Figure 3 the plots for dry matter content (%) from one of the locations for the diallel for the sub-humid (Santo Tomás, Atlántico Department) and acid soil savannas environment (CORPOICA-La Libertad, Meta Department) are presented. Results for the other respective locations yielded similar results. In this case families tended to be more uniform with higher averages. In other words the magnitude of the averages do not necessarily dictate the magnitude of the standard deviations within each family. Families with high dry matter content tended to be clearly more uniform. This could be envisioned, perhaps, as a result of the continuos selection for increased dry matter content. Most of the germplasm would have the key alleles for high dry matter, the exception being the opposite. When a progeny has an unusualJy Jow dry matter content, then the variability within family tends to increase. It is interesting to note that similar trends were observed for this variable in the three different ecosystems and for all the locations in which they were evaluated. There is a consistency in the way each variable appear in this kind of graphs, which is further suggest that they may be an interestíng new approach for analyzing genetic variability and identifying relevant cases where further studies are justified. Harvest index ís another variable that consistently showed the same tendency as dry matter content, with reduced variability within family at higher levels of the index. As in the case of dry matter content the long selection process for optimum harvest index (Kawano, 2003). Thisprocess could serve to explain why high harvest índices result in reduced variabily. In other words, high harvest index is the rule at least among the elite parental lines used in producing the diallel families. Finally root yield showed a different tendency with increased variability within family in those Fl crosses that showed the highest averages. As in the case of the previous variables, this trend was shared by the three different ecosystems targeted and for each of the locations within each ecosystem. Figure 4 illustrates the results for root yield (kg/plant) for one of the locations at each target environment. It is clear that a common trend exists, with families with low productivity showing a "dominance behavior" that does not necessarily mean that root yield is a recessive trait. It is clear from the information provided in Figures 3 and 4 that interesting differences are found among and within the families involved in the diallel studies. Is is also evident the consistency with which each variable responds to the analysis. Further study will be conducted by determining the identity of each family in the plos (as done for white flies) . Concluding remarks regarding diallel analysis. Our knowledge on the inheritance of agronomically relevant traits in cassava is still very limited. The most interesting result of the analyses on dialllels described above can be summarized as follows: Innovative approach from the quantitative genetics point of view. The studies will hopefully provide advances in this area of research with a more complete model for explaining genetic variability, model that can be related to actual data. It is interesting, therefore, that cassava may eventually come to provide tools for genetic analyses of other vegetatively propagated crops. The graphic deployment of data facilitates the identification of contrasting families deserving a more detailed study. That is the case, for example, illustrated in Figure 1 among the progenies from :MECU 72. In general the statistical and graphic analysis provide an ideal group of genotypes that can be used for molecular analysis. This is for instance, what CIAT is currently doing with a group of genotypes that showed a contrasting phenotype regarding leaf retention. 152 5.50 5.00 • • g • "' • • • i 4.50 > ' 1:- -- e • .:! ~ !: 4.00 • • = LDw dty INitter •• j - .. behevtour ' ! • • • ~ 3.50 • • . • • • -> • ~ • ~ • .. • ~ 3.00 • • S ., 2.50 2.00 22.00 23.00 24.00 25.00 26.00 27.00 28.00 29.00 A .. rage dty ... tter content 6.5 e: 6.0 !2 ; ;:: "' 5.5 > ~ • E • ... !! 5.0 e: 5 i 4.5 ... • 2 • 111 • • e: !2 4.0 • ; > • 111 • o 3.5 • • ... • "' • o e: ., ' !! 3.0 • • , 2.5 25.0 26.0 27.0 28.0 29.0 30.0 31.0 32.0 33.0 34.0 Family averages for dry matter content (%) Figure 3. A verages and standard deviations from a dialllel study at Santo Tomás (above) andA cid Soil Savannas (below) in the Altántico and Meta Departments, respectively. 30.00 35.0 u u 1 fu 1 1 r 1 ... ... ... u .. ..• ,..., .................. .,.....~ • .. • •• ~mr-----------------------------------------------~ om~----------------------------------------------~ '·'" Ul U5 Ul ___ ... _ ... _ ...,_ ... r • • t. 1 • J fu t • u u •• u 4 0 .. _. 1.0 S.l ... 70 ___ ... ___ ""'- Figure 4. A verages and standard deviations from a dialliel study at Pitalito (above), CORPOICA La Libertad (middle) and Palmira (below) in the sub-humid, acid soils and mid altitude valleys, respectively. 154 References Hayman, B.l. (1954a). The analysis of variance of diallel tables. Biometrics 10:235-244. Hayman, B.l. (1954b). The theory and analysis of diallel crosses. Genetics 39:789-809. Hayman, B.l. (1958). The theory and alaysis of diallel crosses:II. Genetics 43:63-85. Kawano, K. (2003). Thirty years of cassava breeding for productivity-Biological and social factors of success. Crop Sci. 43:1325-1335. Mather, K. and J.L. Jinks. 1971. Biometrical Genetics. Cornell University Press, Ithaca, NY. USA. 1.2.21 Introduction of inbreeding in cassava Nelson Morante, Reman Ceballos y Martin Fregene CIAT For many years cassava research at CIAT has been interested in introducing inbreeding in cassava. The advantages of inbreeding can be summarized as follows: Because no inbreeding is carried out, a sizable genetic load (undesirable or deleterious genes) is expected to prevent the crop fully achieving its actual yield potential. Since there are no clearly defined populations (quantitative genetics sense) allelic frequencies cannot be efficiently modified. Because the highly heterozygous nature of the crop, dominance effects are likely to play a very important role in the performance of materials being selected. The current scheme can exploit dominance effects because, once an elite clone is identified, it can be propagated vegetatively (therefore carrying along the dominance effects). However, it is the same elite clones that frequently are selected as progenitors for the production of new segregating material. In that case, the current procedure has a bias because the breeding value of these clones are unlikely to be well correlated with their performance per se, precisely because of the distorting effects of dominance. In other words, good clones are just found not designed. Production of recombinant seed is cumbersome in cassava. Only 0.6 viable seeds per pollination are produced. It takes about 18 months since a given cross is planned until an adequate amount of seed is produced. When a desirable trait is identified, it is very difficult to transfer it from one genotype to another (even if a single gene controlled the trait). The backcross scheme, one of the most cornmon, successful and powerful breeding schemes for cultivated crops, is not feasible in cassava, because of the constant heterozygous state used throughout the breeding process. The lack of inbreeding in cassava implies a very restricted genetic variability based on recessive traits. Commercially important mutants such as those found and exploited in maize (waxy, floury, high-quality-protein, sweet com, popcom, etc.) are not known in cassava. It is not clear if cassava has or not this kind of useful mutants, but it is obvious that if they existed the breeding scheme employed did not facilitate their identification since the heterozygous nature of the crop significantly reduced the chances of the homozygosity required for the expression of recessive traits. The use of totally inbred material (as parents in the production of hybrids which will then be propagated vegetatively) would greatly facilitate the exchange of germplasm among different cassava breeding projects. Currently this exchange takes place as in vitro plants, which is expensive, time-consuming, and by its very nature restricted to a few genotypes. Botanical seed of inbred genotypes will breed true, and therefore the genotype can be trasnfered in this way without the genetic segregation that occurs from non-inbred materials. This will effectively reduce the relative isolation in which cassava-breeding projects currently operate. CIAT has begun the production of inbred cassava material (currently at 50 and 75% homozygosity). All the elite clones identified through the cassava-breeding project are going to be used for recombination (to produce new segregating populations), but also will be self-pollinated to initiate an S2 recurrent selection process to: a) reduce the inbreeding depression in cassava e~ reduce genetic load); b) indentify useful recessive traits of commercial (i.e. waxy roots), nutritional (acyanogenesis) or agronomic (reduced póst harvest deterioration) relevance. Paralel to this a special project has been approved for the development of a protocol for the production of doubled-haploids from cassava anthers. Table 8 summarizes the results of transplanted S2 seedlings in the Palmira Experimental Station in August 2003. A large number of seedlings (expected average of homozygosity of 75%) were produced and transplanted. 156 Table 8. Results of S1 seedlings transplanted to the field. These materials are the result of a second consecutive self-pollination (average of75% homozygosity). Progenitor Seeds Seeds Seedlings % Farnily produced gerrninated trasplanted Gerrn. AM 247 CM 507-37 50 4 2 8.0 AM 262 CM 2772-3 80 27 21 33.8 AM 273 MCOL 72 71 37 36 52.1 AM 298 CM 6754-8 34 28 27 82.4 AM 320 MTAI8 444 340 340 76.6 AM 321 CG 1141-1 47 29 29 61.7 AM 322 CM 4365-3 49 29 29 59.2 AM 323 CM4574-7 14 8 8 57.1 AM 324 CM 4919-1 36 19 19 52.8 AM 326 CM 6921-3 36 30 30 83.3 AM 328 CM 7514-8 23 13 13 56.5 AM 329 SM 805-15 28 13 13 46.4 AM 330 SM 909-25 56 37 37 66.1 AM 331 SM 1219-9 520 339 339 65 .2 AM 332 SM 1411-5 25 11 11 44.0 AM 333 SM 1438-2 19 12 12 63.2 AM 334 SM 1460-1 519 378 378 72.8 AM 335 SM 1511-6 493 418 418 84.8 AM 336 SM 1565-15 434 334 334 77.0 AM 337 SM 1665-2 506 444 444 87.7 AM 338 SM 1669-5 440 355 355 80.7 AM 339 SM 1669-7 241 230 230 95.4 AM 340 SM 1741-1 362 215 215 59.4 AM 341 SM 1778-45 23 21 21 91.3 AM 342 MTAI 16 35 23 23 65.7 AM 343 CM 3306-4 62 32 32 51.6 TOTAL 4649 3426 3416 62.0 During the first semester of 2004 these 3461 plants will be harvested. In the process extensive evaluations will be performed in search of useful traits. This is could eventually be a tuming point for cassava research at CIA T. 1.2.22 Development of a male-sterile Nipponbare population Intemational Rice Functional Genomics Consortium Cesar P.Martinez, James Carabali, J. Barrero and J.Tohme SB-2 Project Funding: USDA and CIA T core Introduction An intemational consortium of geneticists, molecular biologists and information scientists from Yale University, Cold Spring Harbor Laboratories, Brookhaven National laboratory, and CIA T was assembled to address the following specific goals: To generate an extensive collection of rice lines, each containing an independent, dispersed insertion of a genetically-engineered Ds transposon; To determine the chromosomal position of each insertion by sequence of its flanking genomic DNA; To establish a database of that relates lines, sequences and phenotipic information; To publicly distribute mutant lines and associated informatics. By making use of this public information, research scientists worldwide can rapidly identify mutant alleles in genes of agronomic importance for functional genomic studies and crop improvement. Materials and Methods A major CIAT involvement in this project is to produce foundation seed of stable rice lines containing the Ds insertions. All experiments are carried out in Oryza sativa ssp japonica cv Nipponbare. To produce T1 seed, stock plants (provided by Yale University) will be crossed as males to wild type female plants in the CIA T nursery. Efficient outcrossing can be achieved using a male-sterile female line. Since male-sterility in Nipponbare was not presently available a backcross- breeding scheme shown in Figure 1 was used to introgress this trait into Nipponbare. A nuclear male-sterility allele (msms) found in IR.36 (provided by GSKhush from IRRI) was used as the donor parent. The Nipponbare male-sterile version will be used for the production of foundation seed from each transposition selection (T1 seed) produced in this project. A simplified crossing method described by Sarkarung(1991) was used. A seed sample from the IR.36 source segregating for male sterility was planted under field conditions in CIAT; male-sterile plants were phenotipically identified at flowering time and used as the female parent. F2 seed was harvested from F1 plants and grown in the field for the identification of male-sterile plants, which were backcrossed to Nipponbare to produce BC1F1 seed. A second BC to Nipponbare was done and the BC2F2 population was grown in the field to allow the identification of male-sterile plants, which were used to produce the BC3F1 seed, and subsequently the BC4Fl. This seed was planted again to produce the 158 BC4F2 generation to check for segregation of the sterility trait. Male-sterile plants very similar to Nipponbare were identified. However, sorne segregation in terms of plant height, tillering, flowering time, and presence of awns was observed. Therefore, another BC to Nipponbare was done to obtain a more uniform population. From here on, the male-sterile Nipponbare population will be maintained by growing it in isolation to allow open pollination. Just seed from the male-sterile plants will be harvested to maintain this population. Seed increase of several genetic stocks was also done during the course of this project, and sent to Y ale University for use in transformation experiments. Future plans: Field/greenhouse evaluations of transformed lines carrying stable Ds insertions. 1.2.23 Utilization of New Alleles from Wild Rice Species to lmprove Cultivated Rice in Latin America C.P. Martinez, J. Barrero, A. Almeida, M. C. Duque, O. Giralda. J. Silva, J. Tohme SB-2 Project Funding: CIAT core funds , Ministerio Agricultura Colombia Summary Statistical analysis on the performance of F8 ínter specific lines from the cross BG90-210. rufipogon over eleven locations in Latín America indicated no significant difference in grain yield between Bg90-2 and its progenies. However, sorne of the progenies yielded 15-24 % more than BG90-2. Stability analysis showed that all lines were stable across environments. Analysis of the molecular data showed that all lines had introgressions (2.6-23%) from O.rufipogon; more introgressions were found in chromosomes 2, 5, 7, 12, and 3. No correlation was found between number of introgressions and grain yield. Markers RM5 and RMl were found in sorne of the top yielder lines but 52% of markers detected in the F8 lines were not found in the F2 generation. Introduction In spite of the great impact made in rice production in Latín America (LAC) there is a need to increase it in a sustainable way to meet increasing demand. New alleles can provide genetic variability for crop enhancement. There is wide genetic variability available in rice, but limited use of this variability has been made. It has been shown (Xiao et al, 1998; Moneada et al, 2001; Tanksley and McCouch, 1997; Thompson et al, 2003) that the Oryza wild species represent a potential source of new alleles for improving the yield, quality and stress resistance of cultivated rice. These studies indicated that O.rufipogon possesses new alleles on chromosomes 1 and 2 with positive effect on yield and yield components. However, these studies were conducted on a limited scale since early segregating populations (BC2F2) and few replications (1-2 sites) were used. No data are available confirming that yield advantages detected in the BC2F2 generation is pass on through generations of selection in pedigree nurseries, nor over a wide range of environments. This report focuses on the performance and stability of advanced breeding lines derived from the cross Bg90-2/0.rufipogon across locations in LAC. Materials and Methods Twenty-five lines (BC2F8) from the cross Bg90-2/0. rufipogon, derived from the BC2F2 generation following the pedigree method were planted in replicated yield trials in eleven locations under irrigated conditions (seven sites in Colombia and one each in Argentina, Surinam, Uruguay and Venezuela). This work was done in close collaboration with key partners from the national rice programs and pri vate sector. Transplanting was done in CIA T whilst direct seeding was done elsewhere. A complete! y randomized design with three reps was used and crop management was based on recommended local agronomic practices. Varieties grown locally were used as checks. Data on main agronomic traits, including grain yield, was taken. A two-way analysis of variance was used for the analysis of grain yield, whilst a GEBEI package that implements appropiate clustering and ordination procedures and AMMI model were used in the analysis of the GxE data. DNA of young leaves from the parental genotypes and their progenies was extracted by the Dellaporta method (McCouch et al. 1988) modified for PCR assay by CIAT Biotechnology Unit. Subsequent molecular assays were performed using 76 SSRs Traits for yield and yield-related characters were associated with the 76 molecular markers using simple single point analysis. Results Data are presented in Figure 1 and Table l. Statistical analysis showed no significant difference in grain yield between Bg90-2 and its progeny over alllocations. Although none of the interspecific lines out yielded Bg90-2 in alllocations, severallines performed better than Bg90-2 in each location. Analysis of the GxE (data not shown) indicated that contrasting and distinct environments were included in these trials and that the GxE interaction was high (75%). This suggests that the performance of genotypes was dependent on the climatic/soil conditions in each location and that there was a good level of genetic variability present in this group of lines, which explains the better performance of sorne progenies under specific conditions. This is very important for breeding purposes since the genetic variability hidden in this population was only observed when the progenies were exposed to a diverse set of climatic/soil conditions found in different rice growing areas. Analysis of molecular data from the bulked seed sample of the BC2F8 Iines shows that all of them had introgressions derived from O.rufipogon (Table 1). The number of introgressions ranged from 2(2.6%) to 18 (23%). More introgressions were detected in chromosomes 2(14), 5(9) and 7,12,3(7); chromosomes 4 and 10 had two introgressions. There was no correlation between number of introgressions and grain yield (Figure 1). Markers RM5 and RM1, located on chromosome 1, were found in sorne of the highest yielding lines. 160 Molecular data from the BC2F2 generation (Annual Report 2001) were compared with that of BC2F8. Only 46% of the 76 SSRs were detected in both generations, suggesting that sorne markers were lost during the phenotypic selection carried out through the generation advance. This was expected. However, 52% of the markers detected in the BC2F8 lines were not detected earlier in the BC2F2 generation. This could be explained because plants sampled for molecular analysis and agronomic characterization were different. On the other hand, only 48% of markers detected in BC2F8 lines are somewhat associated with yield or yield components. Molecular data confirmed that all breeding lines had introgressions from O.rufipogon. This wild species is known to have a high level of genetic variability and is adapted to diverse climatic/soil conditions. Statistical analysis on grain yield and performance through eleven locations in LAC indicated that there were contrasting differences among them. Therefore, breeding lines were subjected to diverse biotic and abiotic conditions, including high disease incidence (Villavicencio, Saldaña, Jamundi), acid and infertile soils (Villavicencio, · Jamundi), cold stress (Argentina, Uruguay) and good climatic conditions (Monteria, Aceituno). There were excellent and poor environments for rice production. However, all lines did better than BG90-2, the improvedl recurrent parent; sorne of the lines (8,13,12, 1,15, 21, 9, 19, 3) yielded between 15-24% more than BG90- 2. The stability analysis, based on the method described by Eberhart and Russell indicated that all lines were stable across environments. Data suggest that the superior performance of inter specific breeding lines is due to favorable allele introgressions derived from O.rufipogon. 162 Table l. Graio yield (1/ha) and perceotage of iotrogression from O. rujipogon of advanced lioes from lhe Bg90-2/0. rujipogotl cross. Bulk sample Yield (tlha) Pedigree No. Loci 'I>O O .r Aceituno Armero CIAT Jamundf Monterfa Saldaña Villaviceocio Argentina Surinam Uruguay Venezuela O .r OI.CTI3941-ll-M-25-1-M-M 11 14.3 11.2 4.8 5.9 7 .5 11.2 6.5 6 .5 10.2 4.3 4.3 6.2 7. 02.CTI3941-II·M·25-4·M·M 11 14.3 10.6 5.1 5.2 7 .8 9.8 7.1 5.6 9.3 3.7 3.8 6.0 6: 03.CTI3941 -11 -M-25-5-M-M 1 O 13.0 11.5 5.1 6.0 6.8 11.0 7.5 6.6 10.1 3.5 2.6 5.4 6.' 07.CT13941-27-M-19-1-M-M 9 11.7 10.3 4.2 4.9 6.9 9.3 5.9 4.9 11.2 3.2 1.4 6.3 6.: 08.CT13946-26-M-S-3-M-M 9 11.7 11.7 4.9 s.s 5.3 9.9 7.9 6.1 12.0 3.2 6.7 8.4 7 .. 09.CTI3946-26-M-S-6-M-M 8 10.4 11.4 4.5 4.4 7.6 10.5 6.4 5.4 11.3 3.7 4.9 6.7 7.1 10.CT13956-29-M-14-1·M·M 18 23.4 10.9 4.2 5.5 7.0 10.4 5.8 6.0 11.5 3.8 3.7 7.5 6.' II.CT139S6-29-M-25-7-M-M 9 11.7 10.7 3.3 4.6 4.4 10.3 5.9 5.5 9.9 3.9 4.4 7.0 6.· 12.CTI3958-12-M-1-7-M-M 2 2.6 12.2 5.3 4.2 4.6 9.6 8.0 5.9 12.5 4.1 5.9 7.0 7.: 13.CTJ3958-13-M-17-5-M-M 6 7.8 11.3 5.0 5.0 7.1 11.9 7.7 6.1 10.2 . 3.7 6.2 6.6 7~ 14.CTI3958-13·M·2·l·M·M 13 16.9 11.2 4.6 5.5 4.3 9.8 6.6 4.7 10.7 3.9 3.9 7 .7 6: 15.CTI3958-13-M-2-3-M-M JO 13.0 11.4 3.9 5.8 6.8 10.9 6.4 4.4 10.6 3.6 5.9 7.5 7.1 16.CTJ39S8-13-M-2-4-M-M 9 11.7 11.3 3.7 5.6 6.0 9.7 6.2 4.1 10.3 4.0 4.5 6.2 6.: 17.CTI3958-13-M-7-5-M-M 12 15.6 11.9 4.8 5.3 4.9 10.0 7.7 S. l 8.8 4 .2 3.1 8.0 6.: 18.CTI3958-13-M-26-4-M-M 15 19.5 11.3 3.5 4 .9 5.3 9.7 6.5 4.9 .. 3.5 5.8 7.0 6.: 19.CTI3958-13-M·26·5-M-M 16 20.8 11.5 4 .9 5.5 6.2 8.9 6 .8 4 .4 9.4 3.5 5.1 7.2 6.t 20.CT13958-13-M-33-1-M-M 15 19.5 12.0 5.4 4 .9 5.3 8.8 6.3 5.3 9.5 3.6 3.8 6.6 6.: 21 .CTI3956-29-M-29-2-M-M 12 15.6 11.6 3.9 5.6 5.9 9.5 7.3 5.5 12.6 3.4 S.! 6.5 7.1 22.CTI39S6·29·M·8· 3·M·M 1 O 13.0 11.0 5.3 3.9 5 .3 10.2 7.4 5.4 8.9 2.8 3.2 6.4 6 .: 23.CTI3959-3-M- I0-4-M-M 8 10.4 11.3 5.6 3.9 5.5 9.1 7.3 5.0 11.6 3.7 4.2 5.8 6 .1 24.CT13959-3-M- I0-5-M-M 7 9.1 12.2 4.7 5.0 5.3 9.4 8.0 5.0 .. 4.2 5.1 7.4 6.1 9 11.7 11.1 4.6 4.8 4 .7 idi. ~"" ............. _.. ~ 3.8 " '·1 4 ,9 ' 27.Fedearroz50 .. .. 10.5 6.5 5.4 8.6 10.0 7.8 5.0 .. .. . . ·- 7: Overall Means .. .. 11.3 4.7 5.1 6.1 9.9 7.0 5.3 10.6 3.7 4.4 6.8 6.1 11\1 Future plans Molecular and statistical analysis of the experiment comparing performance of different generations (BC2F3, BC3F5 and BC3F7). Continue the development of NILs for marker assisted selection References Moneada, P. et al. 2001. Quantitative trait loci for yield and yield components in an Oryuz sativa x O. rufipogon BC2F2 population evaluated in an upland environrnent. Theor Appl Genet 102:41-52. Tanksley S.A., McCouch S.R. 1997. Seed banks and molecular maps: unlocking genetic potential from the wild. Science 277: 1063-1066. Thomson, M.J., T.H. Tai, A.M. McClung, X-H, Lai, M.E. Hinga, K.B. Lobos, Y. Xu, C. Martinez, and S.R. McCouch. 2003. Mapping quantitative trait loci for yield, yield components, and morphological traits in an advanced backcross population between Oryza rufipogon and the Oryza sativa cultivar Jefferson. Theor. Applied Genet 107:479-493. Xiao, J. et al. 1998. Identification of trait-improving quantitative trait loci alleles from a wild rice relative, Oryuz rufipogon. Genetics 150:899-909. 1.2.24 Identification of QTLs for yield and yield components in rice: Populations derived from backcrosses between the wild species (Oryza barthii) and cultivated rice (Lemont). A.Almeida1, O.X. Giraldo2, C.P. Martínez2' 3, J. Borrero3, M.C. Duque2•3, J. López, J.4 Tohme2 1University of Alberta, Edmonton, AB Canada.2·SB-2 Project -CIAT; 3IP-4 Project-CIAT 4Kansas State University, Kansas, USA. Funding: CIA T Abstract Preliminary analysis done on a BC3F2 population from the cross Lemont/O.barthii showed that certain regions associated with important agronomic traits (putative QTLs) were identified. Chromosome 1 carries QTLs associated with grain yield and plant height, as reported in the literature. Likewise, chromosomes 3 and 8 are important for determining panicle length, perhaps one of the most relevant traits found in advanced breeding Iines derived from the Lemont/O.barthii cross. Markers RM227 and RM114 on chromosome 3 are associated with a region derived from O.barthii. Markers RM184 and RM304 on chromosome 10 are associated with days to flower; however, this is a very complex trait as has been reported by others. Advanced lines with different flowering dates have been developed out of this cross. Further molecular work is needed to confirm severa! interesting regions detected in this analysis. 164 Introduction Rice is one of the most important food crops in the word. The 21 wild species and 2 cultivated species (Oryza sativa and O. glaberrima) represent wide genetic variability for rice breeding programs. Most of this variability has not been used so far although severa! studies have suggested that Oryza wild species possess new alleles for improving cultivated rice. The main objective of this study was to focus on progress made in identifying quantitative trait loci (QTL) associated with yield and yield components in a BC3F2 population derided from a cross between a recurrent parent (Lemont) and a donar parent (Oryza barthii). The results of applying AB-QTL analysis to 327 individuals characterized with 113 microsatellite markers are described. Marker data were used to identify QTLs associated with yield and yield components. Materials and methods Lemont, which was the first commercial variety developed in Texas (USA) with an excellent industrial quality, was used as the recurrent parent in this study. O. barthii (Accession # 104119), which served as donar parent, is a relative of O. sativa. There was very high sterility so three backcrosses to Lemont were done. BC3F2 plants were selected based on phenotype, díscardíng plants with hígh sterility and long awns. The best 327 individuals were selected for agronomic and molecular characterization (Fig. 1). DNA of young leaves was extracted from the parental genotypes and the 327 BC3F2 families, using the Dellaporta method (McCouch et al., 1988). The CIAT Biotechnology Research Unit modified this method for the PCR assay. A total of 113 microsatellite markers were used to evaluate the hybrid from the Lemont/0. barthii. Markers used in the evaluations and the QTL analyses were selected from the rice molecular framework linkage map (10-20 cM intervals throughout the genome) (Causse et al., 1994; Chen et al., 1997). l O. barthü ~ 0 Lemont ~ (Donor parent) ~ (Recurrent parent) [::::1 0 1 Lemont 1 ~ (Recurrent parent) BClFl 1 01.--L-em_o_n_t ...,, ~ - (Recurrent parent) ~01 Lemont 1 - ~ (R:current p~ ~0~ ~ BC3F2 ~ Figure l. Generation of BC3F2 population, derived from the Lemont and O. barthii cross. Results The population of 327 BC3F2 families was analyzed using a total of 113 microsatellite markers (SSRs), distributed at approximately 10-cM intervals throughout the genome. The order of the SSRs in the molecular map was defined by the Comell Published molecular rice map (Causse et al., 1994; Chen et al., 1997; Temnykh et al., 2000). In this experiment five agronomic traits including days to flower (dth), plant height (ph), panicle length (pi), percent sterility (ps) and yield per plant (yld) were evaluated. The association between phenotype and marker genotype was investigated using single-point analysis (SPA), interval mapping (IM) and composite interval mapping (CIM) using QTL Cartographer Vers. 1.17d software. A normality test showed that only the distribution of trait pi was normal (Fig. 2); thus it was indispensable for the correct use of aforementioned models to estímate significance thresholds with permutation tests. 166 0% _:~ ~ > 15 o. u barthii e: 11 ::::1 10 l tT Cl ... u.. 5 o 0% _:1 ~ ) ~ 1 ~ 15 o. ..... c-i e: Cll barthii g. 10 f u.. 5 o l ........., 50.8% ~ !f;nt C! 15 u e: 11 ::::1 10 tT Cl ... u.. 5 o ..... <'! C! "'l ~ ~ N N 25 20 ........ :::!t ~ >-u 15 e: 41 ::::1 10 tT 41 ... u.. 5 o 0% 25~ 20 ~ ~ o. > 15 u e: 41 ::::1 10 ! g-... u.. 5 o .., ...... ~ :: ~ Plant helght (ph) ce er183 and therefore we explored the applicability of our EST data set for identifying coding SNP (cSNP). Among the 1809 contigs, 285 contained four or more sequence reads with CM523-7 and 129 with :MJ>er183. 7.7% (CM523-7) and 17% (MPer183) of these contigs contained SNPs with at least two reads for each. The estimated cSNP frequency was 1 polymorphism per 121 bp. Conclusions The cassava EST data presented here is the first effort in the large scale sequencing of the cassava expressed genome and also in cataloguing cassava genes.A unigene set of 6046 sequences was identified and a putati ve function assigned to 31% of unigenes. A number of ESTs were found to be present only in the Xam challenged librarles. he EST data represent a valuable source for the identification of SNP and further mapping analysis. The EST resource will increase the density of gene markers on the cassava genetic map. 189 Future plans Infonnation obtained here will be used to develop microarray technology for further cassava gene expression studies. Based on this, developing new cassava varieties having high dry matter content and durable resistance to CBB will be the next challenge. 1.3.6 Identification of genomic regions responsible for conferring resistance to white fly (Aleurotrachelus socialis) in cassava A.Bohorquez, J. Vargas, A. Bellotti, B. Arias, D.F. Cortes, M.C. Duque, J. Tohme SB-2 Project Introduction The whitefly (Aleurotrachelus socialis) is one of the most serious pests and disease vectors that affect agricultura! production around the world. In cassava (Manihot esculenta Crantz), the whitefly causes from 70 to 80 percent economic losses. The most important source of resistance genes is the genotype M Ecu 72. Dueto the whitefly's importance as a pest, it ís necessary to understand the nature of genes that confer resistance to the whitefly in genotypes such as M Ecu 72. For this purpose F1 segregation and the genetic expression of the cross M Ecu 72 (resistant genotype) x a very susceptible genotype (M Col 2246) and molecular markers are being used. This will help accelerate selection of whitefly-resistant materials, as well as isolate resistance genes (R genes). Genetic and molecular studies (Richter and Ronald, 2000) have shown that the R genes are clustered in the genome of several species. They display an apparent multiallelic structure, or they group as genetically separate loci. Different genes determining resistance to insects and nematodes have been reported within the same cluster in tomatoes (Rossi et al., 1998). R genes are thought to be functionally and evolutionary related. The sequences of several R-gene clusters from rice, tomatoes and lettuce have now shed light on the molecular mechanisms leading to their evolution. As suggested by Lefebvre and Chevre (1995), the genes goveming quantitative resistance could share homologies with the clone's R genes, making the candidate gene approach feasible for the study of possible association between resistance gene analogs (RGAs) and quantitative trait loci (QTLs) controlling pest resistance. In this work M Ecu 72 and M Col 2246 were amplified with RGA primers designed by C. Lopez in cassava (pers. com.), to find putative loci related with whitefly resistance. An additional step toward a better understanding of the attack response of the whitefly to cassava was the establishment of a cDNA library, which was developed with a new, highly effective method known as differential subtraction chain (DSC). Using this approach, two mRNA populations, extracted from both resistant and susceptible genotypes, were exarnined to elucidate the differential gene expression between them. Materials and methods For this work an F1 cross (family CM 8996, 276 individuals) between M Ecu 72 (as the resistant parent) and M Col 2246 (as the susceptible parent), elite cassava cultivars from Ecuador and Colombia, respectively, was used. The parents and their offspring were evaluated in the field at two sites: Nataima (Tolima) and Santander de Quilichao (Cauca). The purpose of this evaluation was to identify gene segregation in the offspring and select the resistant and susceptible materials. Both parents were evaluated with 343 cassava SSRs (simple sequences repeat) (Mba et al., 2001) including 156 cDNA SSRs (Mba et al., submitted). AFLPs (Vos et al., 1995) are being used to find markers associated to resistance for mapping and ultimately cloning the resistant genes. Silver staining is being used to visualize the allelic segregation of the markers. Cassava RGA primers were done in the parentals, and the polymorphics were mapped in the Fl. For the isolation of expressed sequences, 21 forty-day-old plants were used, 7 of each genotype (M Ecu 72 and M Per 334 resistant and M Col 2246 susceptible). These plants were taken to the greenhouse, where they were infested with 300 whitefly adults per plant, for a population of 2100 whiteflies per cage. Leaves were collected at six different times for RNA extraction. For the differential subtraction chain (DSC), the follow strategy was used: Genotype M Ecu 72 was infested for use as the tester, while genotype M Col 2246 was used as the driver. At present the DSC technology is being performed according to Luo et al (1999). The representational difference analysis of cDNA was divided into severa! phases: Generation of a PCR amplicon, which is representative of the original mRNA from M Ecu 72 and M Col 2246. Subtractive hybridization of this amplicon M Ecu 72 (tester) and M Col 2246 (driver), during which amplified portions of differentially expressed genes are enriched and cornmon sequences are depleted Cloning and screening of the resulting products Figure 1. Silv~r-staioed polyac:rylamide ¡el showiJJ« comblnatioo ACA CIT ol AFLP ol botb pareots M Ecu 72 (resistaot), M Col 2246 (susceptible) aod 25 lodlviduals ol progeoy. Note tbe polymorphic baods 1 SO aod#S4' . 191 Results Field evaluation. The field evaluation showed high pressure being exerted by the pest in Nataima, where test materials had high damage rates; however, sorne materia1s had lower Ievels of damage in the evaluations. We can conclude that these genotypes show a resistance level similar to parental M Ecu 72. AFLP analysis. An analysis was made of 128 combinations of primers with both parentals (M Ecu 72 and M Col 2246) and both bulks of 10 whitefly-resistant DNA and 10 susceptible DNA. We obtained 53 polymorphic combinations, in which there were 425 polymorphic bands between the resistant and the susceptible. All combinations were amplified in the Fl (Fig. 1). Mapping. Approximately 55 of the SSRs evaluated were polymorphics in the parentals and were evaluated in the F1 (286 individuals). To construct the linkage map, 103 SSRs were analyzed, of which 71 were anchored. A genetic linkage map of cassava was constructed with 71 SSR markers segregating from the heterozygous female parent (M Ecu 72) of an intraspecific cross. The map consists of 19 linkage groups, which represent the haploid genome of cassava. These linkage groups spanned 550.2 cM, and the average marker density was 1 per 7.9 cM. The position of the 71 SSRs markers is shown on the framework (LOD == 25, theta = 25) of the molecular genetic map of cassava (Fig. 2). Map distances are shown in Kosambi map units. Of these markers, 26 (shown in green, Fig. 2) had been placed previously on the cassava framework map (Fregene et al., 1997); the other 45 SSRs are new. Of the 71 SSRs, 31 were cDNA sequences (Mba, in prep.), while the others were genomic DNA. Association between molecular markers and resistance. The molecular data are being analyzed using QTL packages (QTL cartographer Q gene) to determine linkages between the SSR markers and phenotypic characterization. Preliminary analysis (X2 at the 5% level) was done using SAS. Putative associations were found between 43 SSRs markers and the field phenotypic characterization (score 1.0 to 2.0 of the damage levels and populations). Fig. 2: Preliminary Cassava framework Map of MEcu-72 for Resistance to White Fly, consisting of SSRs. (Lod = 25 and theta = 25) L. ~- .... E •n-e ... roup Lin ~e Gro u p A Lin ~e Gro u p B Lin ~e Grou p C Lin ~- Gro u p F Lin M¡¡ e Gro u p G ~~"lll:l (IU~.s (&1 tC)l2 ~fZZII ~,.,. (11.2 .~ (l~ftlZ(IU.I (UX)IO (li1"fl1 (UtC)U (145.~ Lin~e Group 1 le Nftl ~« 1Ji11: ....... le Nftl FMC.ct.l ( 10 tq.Ot('ft) v-RY1011 (u tq.§ ('G) v.RY1all (11§ ~ CillJ95R'r.IXI (17.1~ Lin M¡¡ e Grou p J Lin~e Group O ~- IJill: (lllll'S'SRUCI FMC. ct.l (I!!)'S'SR YTS (14)95R- Lin~eGroup K Lin~eGroupl Lin~e Group ... (IS1)11SR- Lin~cGroup N Lin~cGroup o ....... le Nftl Lin~e Group P Lin M¡¡ e Gro u p Q Lin~crGroupR ~- IJill: ....... FMC. ct.1 le Nftl (11-I:IIJ~ fi-CJili)'IISRf:m ll-(lllll'IISRncl (1~'1"31 ~'m (11~nll CJlHqt.S (UtC)&~ (lt.ltCJ1.0 (UtC)iO ( 5.~ tC)S.S (UtC)S.5 (lofi)r:W1101 (15.1 iiJ j(4}'11SRal (l:n),..~'tl --~(lllf-'l:SI 1Cil)'IISROI (15.1 ~,,. Lin ~cr Gro u p S ....... le Nftl CD}I'!~"IS (llli)r:!:~Y%2:1 ecli~'n!JS p!) !~'tl'JI 193 Cassava RGAs. We obtained eight polymorphic ROA primers in the parentals (Fig. 3). To date, we have mapped three Bac primers in the Fl to find associations with yield QTLs. Polymorphism of the presence and absence bands between the parentals was found in Bac 9, 31, 35, 45, Contig 39 and Contig 43d and polymorphism of different bands in Bac 36 and the RT. . - .. M~ . . . 8a&5 BD Blc7 Baca a.r:t M -H Figure 3. DNAs of M Ecu 72 and M Col 2246 amplified with RGA primers. Differential subtraction: RNA extraction. RNA was isolated from young leaves of M Ecu 72 (E), M Per 334 (P) and M Col 2246 (M), colleted in the greenhouse. To isolate total RNA, the Rneasy Plant Mini Kit QIAGEW'-~ was used. Genomic DNA was removed prior to isolation of poly (At RNA with DNAse l . The SV Total Isolation System of Promegar"" was used. The generation of cDNA was done using poly A+ mRNA as the substrate, which was isolated using the protocol Oligotex mRNA Spin Column of QIAGEW'-~. First- strand cDNA synthesis and cDNA amplification were done using SMART PCR cDNA Synthesis kit" de Clontech™ (Fig. 4) Figure 4. M: A digested with Pst l. cDNAs amplified with kit SMARTN. The PCR products from the amplification of cDNA, were purified using QIAquick PCR Purification kit QIAGEWM. Then digestion ligation was done, where the cDNA was digested with Dpnll, and then adapters (Bami and Bamii) were ligated. Finally, the amplicon generation was done for the hybridization reactions of the subtraction. For "tester'' M Ecu 72 (E), 150 ng was obtained; and for "driver" M Col 2246 (C), 15 J!g. Ongoing activities • Saturation of linkage map of M Ecu 72, using AFLPs • Isolation, cloning, sequencing and mapping of AFLPs polymorphic bands between resístant and susceptible genotypes • Desígn of SCARs for marker-assisted selection • QTL analysis for whitefly resistance • Mapping of cassava RGA polymorphics (BACs Primers, Gene Resistance Primers) in Fl (276 genotypes) f\ • Isolation of expressed sequences during the defense response of M Ecu 72 to whítefly attack. In order to identify differentially expressed sequences, a new technology known as DNA chips or rnicroarray can sean a significant number of clones. Microarray-expression profiling will be used to identify putative early response regulatory and/or signaling genes and to test the function of selected candidate genes using reverse genetics. 195 References Arias B. 1995. Estudio sobre el comportamiento de la "mosca blanca" Aleurotrachellus socialis Sondar (Homoptera: Aleyrodidae) en diferentes genotipos de yuca, Manihot esculenta Crantz. Tesis maestría. Universidad Nacional de Colombia, Palmira. Fregene, M.; Angel, F.; Gomez, R.; Rodríguez, F.; Chavarriaga, P.; Roca, W.; Tohme, J. and Bonierbale, M. 1997. A molecular genetic map of cassava (Manihot esculenta Crantz). Theor. Appl. Genet. 95:431- 441. Lefebvre, V. and Chevre, A.M. 1995. Agronomie 15:3-19. Luo, J.H.; Puc, J.A.; Slosberg, E.D.; Yao, Y.; Bruce, J.N.; Wright, T.C.; Becich, M.J. and Parsons, R. 1999. Differential subtraction chain, a method for identifying differences in genomic DNA and mRNA. Nucleic Acids Research. 27(19) Mba, R.E.C.; Stephenson, P.; Edwards, K.; Melzer, S.; Mkumbira, J.; Gullberg, U.; Apel, K.; GaJe, M. ; Tohme, J. and Fregene, M. 2001. Simple sequence repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: Towards an SSR based molecular genetic map of cassava. Theor. Appl. Genet. 102:21-31. Richter, T.E. and Ronald, P .C. 2000. The evolution of disease resistance genes. Plant Mol. Biol. 42:195-204. Rossi, M., Goggin, F.L., Milligan, S .B., Kaloshian, 1., Ullman, D.E. and Williams, V. M. (1998). The nematode resistance gene Mi of tomato confers resistance against the potato aphid. Proc. Natl. Acad. Sci. USA. Vol. 95, pp 9750-9754. Vos, P.; Hogers, R.; Bleeker, M.; Reijans., van de Lee, T.; Homes, M.; Frijters, A.; Pot, J.; Peleman, J. ; Kuiper, M. and M. Zabeau. 1995. AFLP: A new technique for DNA fingerprinting. Nucleic Acids Research. 23(21): 4407-4414 • 1.3.7 Gene expression profiling of cassava responses to Xanthomonas axonopodis pv. manihotis infection 1Soto-Suárez, M., 1Restrepo, S., 2Lopez, C., 1Mosquera, 0 ., 1Tohme, J., and 2Verdier, V. Biotechnology Research Unit, CIAT1 A.A. 6713, Cali, Colombia, E-mail : s.restrepo@cgiar.org, Institut de Recherche pour le Developpement (IRD)2, Universite de Perpignan, Perpignan, France, E-mail : vverdier@univ-perp.fr. Introduction Cassava (Manihot esculenta) is currently the fourth energetic food in the world after rice, " maize, and wheat and feeds more than 1000 millions people. Cassava Bacteria! Blight (CBB) caused by Xanthomonas axonopodis p.v manihotis (Xam) is a destructive disease in the South America and Africa (Lozano, J.C. 1986) and yield losses range between 12 and 100%. The most suitable approach for controlling CBB is through host-plant resistance (Verdier, V. et al., 1998). Cytochemistry and biochemistry of plant defense responses to CBB have been studied (Kpémoua, K. et al., 1996). However, plant response to pathogen attack remains uncharacterized at the molecular and cellular level. In cassava, using the cDNA-AFLP technique Santaella et al., (2002) identified genes involved in the cassava- Xam interaction. Sequences were isolated and showed homologies with plant genes related with resistance genes, signal transduction pathway, disease and stress responses and senescence associated genes. At the moment high-troughput techniques, such as cDNA microarrays, provide a genome-wide description permiting the characterization of the expression profiling of thousands of genes in a single experiment. The objective of this research was to identify genes associated with cassava defense response to Xam combining subtractive library construction and cDNA microarrays. Materials and methods Plant material inoculation and cDNA synthesis. Young plants (4-week old plants) from two resistant varieties (MBRA 685 and SG 107-35) were inoculated by stem puncture with Xam strain CIO 151. Stem tissues were collected at 6, 12, 24, 48 and 72 hours post inoculation (pi), and 7 days pi. The controls were healthy non-inoculated plants and plants inoculated with sterile water. mRNA was isolated usin.fi.t Oligotex mRNA Midi kit (QIAGEN, CA). cDNA was synthesized using SMART PCR cDNA synthesis kit (CLONTECH, CA) from 400-500 ng of mRNA as starting material. cDNA subtractive library. To identify differentially expressed genes during pathogen attack subtractive hybridization was performed, using the Differential Subtraction Chain method. A pool of cDN A obtained from inoculated plants was used as "tes ter" and a pool of cDNA obtained from healthy non-inoculated plants and plants inoculated with sterile water was used as "driver" . 197 cDNA Microarrays preparation. Cassava clones from subtractive libraries, cDNA-AFLP analysis (Santaella et al., unpublished results) and other clones were collected, amplified by PCR and printed on glass slides. The resulting microarray contained 3072 elements with each cDNA printed eight times as replicates. For control pourposes, a set of spot controls from tomato, potato and cassava housekeepings genes, human genes and spiked controls were arrayed. Fluorescent probe preparation and Hybridization. Total RNA was isolated using SV total RNA isolation system cPromega Corp.) cDNA was synthesized using SMART™ PCR cDNA synthesis kit (CLONTECH, CA). Fluorescent-labeled probes were prepared using Klenow labelling (for indirect labelling) and microarray hybridization were performed with cDNA pool from 24-48 pi. v.s cDNA from healthy plants, using two duplicate slides with reverse labeling (dye-swap). Data Analysis. Spot intensities from scanned slides were quantified using the ArrayPro 4.0. software. Background correction was realized, only signals that were higher than two standard deviation from local background were considered for further data analysis. Intensity-dependent norrnalization LOESS was performed and the differentially expressed genes were detected using SAM (Significance Analysis of Microarrays). Results and Discussion Two different strategies were used to identify pathogen-induced defense genes, cDNA subtractive libraries and cDNA Microarrays. Using the subtractive libraries, 768 cDNA clones were isolated for each resistant variety. Of these, 110 randomly selected clones were sequenced and a homology search was conducted using the BLAST program. The sequence analysis showed that 16 cDNA clones shared homology with plant genes involved in defense responses (Table 1), 70 clones were either homologous to plant genes of unknown function or showed no homology, and the remaining 24 clones showed homology with other plant genes. A hypothetical model of gene expression changes that occur in cassava-Xam incompatible interaction was proposed (Figure 1). Ca 2+ signals. Specific signa! transduction pathways are activated. Calcium fluxes are an early event in signaling pathway and trigger plant defense responses. Previous studies revealed that the rapid accumulation of cytosolic Ca 2+ is necessary for the production of oxidative burst and PR genes activation (SUS06, SUS07). Clone SUS05 is homologous to a fungus-inducible Calmodulin Mlo family protein isolated recently from rice. Table l. Sixteen cDNA clones from subtractive libraries sbared homology with plant genes involved in defense responses Clone SUS01 SUS02 SUSO :S SUS04 SUSO S SUSOI SUS07 suso• SUS09 SUS10 SUS11 SUS12 SUS13 SUS14 SUS1S SUS11 Xam Hoghost homology Pectin meft)iesilnlae (lllia:IM- c.biiCUm). Calelase CAT1 ~ esculenfBJ. ~idediphotlphala kínate 3 (ndpk3) ~/dopsla flalí.lna) Glu~ (Riclnus ..,munl.sJ- Calmodulin [garden pea). Upid "-* prolein (upland cotiDn~ Bet v 1 allerven family prolllin Nlk)rin1ike prolein ¡Antbldopsla lhellanaJ. »,fogluCM endotnlns~)lase prolein ¡Antb/dops'.t flaliMNtJ- Zinc-lnger prolllln 1Atabldop$is t'IM.tnaJ. MV8 nnsaiplion lador - llke proleln ¡Atabldopsis llallenal- Slmiar ID eukat)olic tanslation lniialion fllelor 28 Pol)c¡biquiln (ubq10) ¡Atabldopsla~Ntlh!nal- Ublquíln-llke prolllln (UBQ12) ~idops/s tlallan&). MI~ melaloptoleinue ¡Antbldop.tls lll.ollanal- OnaJ protein ~-b111JIIensi4 . ... Slgnal tran-.luctlon :e -· c. 2· S4gnals SUS05 • E-valuo 4E·30 3&43 1E· 38 4E·38 4E-« 2E-14 2E-14 SE-16 9E--47 9E-57 2E-Q8 4E-12 SE-36 1E-31 SE-13 7E·2<4 Avr-RGenes SUS02 SUS03 SUS04 Oxidatlve Burst T ranslalion lnhib«or SUS12 ProtWo Oegnldalion SUS1l SUS14 SUS15 Figure 1. Hypothetical model of the cassava defense response to Xam infection. In green are the cDNA clones from subtractive librarles involved in defense responses. 199 Oxidative Burst, PR genes and HR. During infection, cassava produce a biphasic or polyphasic oxidative burst that is regulated by proteins such as Catalase and NDPKs (SUS02, SUS03), these genes can inhibit Reactive Oxigen Intermediates (ROis) in specific time-points over infection, Clone SUS02 encodes a protein with significant homology (80% identity) to Ngcatl, a gene that was previously reported to be expressed following pathogen infection and salicylic acid treatment (Yi, SY. et al. 1999). Clone SUS03 shares a homology (68% identity) with AtNDPK2 that appears to play a regulatory role in ROis- mediated MAPK signaling over biotic and abiotic stresses (Moon, H. et al., 2003). PR and other defense-related gene expression are probably regulated by transcription factors, such as MYB and Zinc Finger Protein (SUS lO, SUSll). Ca 2+ signals, ROis and PR genes can activate the HR, thus Xam may be confined to dead cells. This local response could trigger nonspecific resistance throughout plant (SAR). Other defense responses. Associated with oxidative burst, HR and SAR, a number of other active defense responses are activated. These responses are also effective against Xam. A clone homologous to a brassinosteroid-regulated gene Xyloglucan Endotransglycosylase (XET)(SUS09) was reported. Brassinosteroids (BRs) were found to induce disease resistance in plants. BRs increase the abundance of mRNA transcripts for wall-modifying proteins such as XET that incorporate new xyloglucan into the growing cell wall. The pathogen spread is then hindered by physical strengthening of cell walls. Translation inhibition and protein degradation appear to play a role in cassava defense response, SUS12, SUS13, SUS14 and SUS15 genes activity may be coupled to HR development. Clones homologous to genes involved in glycolisis, vesicular transport (data not showed) and traffic protein (SUS16) were also identified. In addition, clone SUS08 encodes a protein with homology to an ankyrin from Arabidopsis. Ankyrin repeats are involved in the protein-protein interaction and have been implicated to function in sorne defense signaling pathways such as Arabidopsis NPRI. cDNA Microarrays. cDNA Microarray enriched for genes involved in the cassava defense responses was constructed and hybridized toa cDNA pool from 24-48 pi. v.s cDNA from healthy plants. Data analysis revealed that sorne clones showed significant differential expression in response to pathogen attack (Figure 2). For up-regulated clones, the difference between experimental and control expression levels were up to 11-fold. These results indicate that sorne gene transcripts were induced between 24-48 hours after inoculation. lt would be interesting to evaluate the microarray with other different time points throughout the infection. Severa! differentially expressed genes revealed by the microarray such as Glutaredoxin (SUS04), Lipid Transfer Protein PR gene (SUS06), Zinc- Finger Protein transcription factor (SUSlO) Dormancy-associated protein (DOR) and Ubiquitin (SUS13) are associated with oxidative burst, defense signaling regulation and protein degradation in the plant-pathogen interactions. Other differentially expressed clones were homologous to Arabidopsis genes of unknown function (Ath) or showed no homology with sequences in the databases (NH), representing a new source of genes potentially involved in cassava defehse responses. Conclusions Functional genomic tools such as subtractive libraries and microarrays permitted to give a comprehensive overview of the molecular basis of the cassava defense response to the bacteria! blight pathogen. Many defense signal transduction pathways lead to responses like oxidative burst, cell wall modification, protein inhibition, protein degradation, metabolic changes and subsequent HR and SAR induction. The determination of genes that are up and down regulated at different times after the inoculation can help in understanding the defense mechanism of cassava to different pathogens. Ongoing Activities Confmn the differential expression of the characterized clones by Real-Time PCR analysis. Figure 2. Three colors images showing differentially expressed clones SUS04, SUS06, SUS! O, SUS13, Dormancy-associated protein (DOR) and Arabidopsis genes of unknown function (A.th) or without homology with sequences in the databases (NH). References Kpémoua, K. Boher, B. Nicole, M. Calatayud, P . Geiger, J.P. Cytochemistry of defense responses in cassava infected by Xanthomonas campestris pv. manihotis. Can. Journal of. Microbiol. Vol. 42: 1131-1143. 1996. Lozano, J.C. Cassava bacteria! blight: a manageable disease. Plant Disease. Vol. 70: 1089-1093. 1986. 201 Moons, A. Prinsen, E. Bauw, G. Van Montaguasi, M. Antagonistic Effects of Abscisic Acid and Jasmonates on Salt Stress-lnducible Transcripts in Rice Roots. The Plant Cell. Vol. 9: 2243-2259. 1997. Santaella, M. Suárez, E.C. González, C. L6pez, C. Mosquera, G. Restrepo, S . Badillo, A. Tohme, J. Verdier, V. CDNA-AFLP analysis of diferential gene expression in the cassava -Xanthomonas axonopodis p.v manihotis interaction. CIAT, Annual Report: 174-178. 2002. Verdier, V. Restrepo, S. Mosquera, G. Duque M.C. Gerstl, A. Laberry, R. Genetic and pathogenic variation of Xanthomonas axonopodis pv. manihotis in Venezuela. Plant Pathology. Vol. 47: 601-608. 1998. Yi, S.Y. Yu, S.H. Choi, D. Molecular cloning of a catalase cDNA from Nicotiana glutinosa L. and its repression by tobacco mosaic virus infection. Molecular Cells. Vol. 9. No. 3: 320-325. 1999. 1.3.8 Development of a Diversity Array Technology (DArT) Chip for Cassava Andrzej Killian1, Peter Wenzel1;Carmen deVicente2 Edgar Barrera3, Ana-Maria Correa3, Martín Fregene3 1(CAMBIA); 2(IPGRI); \CIAT) Introduction Genetic resources, mostly held by small farmers representa critical resource for the future productivity and stability of production of the crop. How to evaluate and use in a systematic manner the vast amount of variability present in cassava is still a challenge to most cassava breeding programs. Genotyping micro-array technologies offer the highest throughput available up to date. One of them diversity array technology, DArT ™ (CAMBIA), is sequence-independent (low-input) and allows the fingerprint of an individual's genome based on a high number of polymorphic sites spread over the genome. These screening procedures should allow testing of thousands of individual samples in a speedy manner. We describe here a proof of concept on using the DArT tool as a cost- effective way for measuring and characterizing genetic diversity of cassava germplasm. Methodology The project was initiated in early October 2002, with the shipment of cassava DNA samples from CIA T to CAMBIA. Plant materials used for the generation of the DArT chip was chosen to represent a broad as possible diversity of the cultivar, a few genotype of its wild progenitors and 2 wild species were included to capture a large number of polymorphic fragments. They include 14 accessions from Brazil, 14 from Colombia, 4 from Guatemala, 2 each from Nigeria, Cuba, and Ecuador, Peru and Thailand respectively. Others include, one accession each from Argentina, Bolivia, Costa rica, Fiji islands, Indonesia, Mexico, Panama, Venezuela, and USA. Six and 2 improved varleties were included from CIAT and liTA respectively. The wild species accessions were 29 of Manihot esculenta sub spp Flabellifolia, 7 of M. carthaginensis and 1 of M. walkerae. DNA isolation was according to Dellaporta et al. (1983) followed by two washes of phenollchloroform. A critical step in DArT is the complexity reduction step. Work at CA?vfBIA with several other plant genomes has shown that digestion with Pstl restriction enzyme (RE) in combination with more frequently cutting RE is an efficient method to reduce complexity. A preliminary experiment was therefore conducted to determine the best enzyme combinations, mixture of genomic DNA from twenty cassava genotypes was digested with Pstl, ligated to adapters and further digested with severa} frequently cutting RE (BstNI, Apol, Taql and Banll), followed by amplification with an adapter-specific primer. The cassava genomic Pstl fragments lacking the recognition site for the frequent cutting RE (BstNI, Apol and Taql) were individualized by transformation into E. coli, amplified from bacteria! colonies and micro-arrayed. From each of the librarles 760 clones were arrayed. Genomic representations prepared in the same way (RE digestion!ligation followed by amplification) from each of the twenty genotypes separately, were labelled with Cy3-dUTP and hybridized together with Cy5-dUTP-labelled reference DNA to these microarrays. Slide preparation, hybridizations, washes and scanning are as described by Andrzej et al. (2002). Images generated by the scanner were used to extract quantitative fluorescence signal data for each array feature using our proprietary software. Same software was used to binarlze the data (score as 0/1) for all slides. Binary scoring table was used to prepare the Hamming distance matrix and to obtain a dendogram. Líbrary expansion to obtain more polymorphic clones was carried out using the enzyme combinations Psti/BstNI, and Psti!Taql, and 80 DNA samples from CIAT. About 3000 clones were evaluated for polymorphism from the 2 librarles. The arrays were printed and DArT analysis carried out as described earlier. 203 BIWI27 BIW22 BRAaiO IIRA70 WCOL251 BRM03 T\IE117 OWBo-71 '---OWIII-11 ~~~ WG~:~2 ~ :~¡ OW30-3 OW1S5- 0W164- .__--OW1S1- ...._ _____ m.walicerJ¡ M. esculenta M. esculenta subsp. flabellifolia M. esculenta M. carthagenensis M. walkerae Figurel. Genetic relationship among the cassava accession used to identify polymorphic clones in the Psti/BstNI, Pstffaql and Pst/Apol array. The dendrogram was created using the distance table based on 296 polymorphisms and the UPGMA clustering algorithm Results Among the three libraries /Psti!BstNI, Pstl/ Apol, and Psti!Taql) tested in the preliminary experimnt, Psti!BstNI gave the largest number of polymorphic clones (132), followed by Taql (112) and Apol (69). In total, 313 candidate polymorphic clones were obtained in the initial experiment to determine the best enzyme combination. DArT analysis based on 296 polymorphic clones without a single missing data point was used to generate a binary matrix and obtain a dendrogram, based on Hammings distance, representing genetic relationship between the 20 samples (Fig. 1). The libray expansion with 80 clones yielded 440 polymorphic clones (14.3%), for the Psti!Taql array, consistent with the results obtained with the initial, smaller array (14.6% polymorphic clones). A dendrogram based upon the analysis of the 80 cassava samples with the polymorphic clones is presented in Figure 2. r-- - 1 1 rl r- ~ '- y rl iC "---1 ~ y ~ L-f. COL1522 MCOL2056 MCOL2061 COL72 ECU79 MGUA62 CUB56 TAI16 MGUA7 NGA19 NGA2 MCOLlSOS CM6740-7 CM523-7 COL667A PER333 TMEl COL2737 HMC53 COL1468 BRA890 BRA522 MCOL2516 PER183 TAIS C4 AMA28 AMA19 BRA403 BRA1439 BRA124 BRA461 BRA383 OWS0-7 OWSl-8 M.waiOOl OW161-1 OW164-5 Figure 2. Genetic relationship among the cassava accession used to identify polymorphic clones in the Pstl!faql array. The dendrogram was created using the distance table based on 440 polymorphis~nd the UPGMA clustering algorithm. 205 In the Psti/BstNI array 554 polymorphic clones (18.0%) were identifid, also consistent with the polymorphism frequency in the smaller Psti!BstNI array in the first project phase (17.2%). The dendrogram resulting from this array is shown in Figure 3. r---COL2737 ARGS MCOLlSOS TMEl BRA927 CM6740· 7 HMCS3 BRA403 BRA931 TAIS C4 BRA124 BRA1439 MCOL2516 NGA19 NGA2 .----AMA28 '---BRA890 '---PER333 .----BRAS22 .----AMA19 '---BRA829 Í--------------¡LL.._-_-_-_-=_~- CM523-7 OWS0-7 L...----- NGA2 r---MGUA62 1L---MGUA7 TAI16 COL1522 COL1672 COL72 ECU72 ECU79 CUB56 MCOL2061 MCOL2056 ,.---BRA383 '----BRA461 '--------------M.waiOOl OW164·5 OW161·1 Figure 3. Genetic relationships among the cassava accessions used to identify polymorphic clones (markers) in the Psti/BstNI array. The dendrogram was created using the distance table based on 554 polymorphisms and tbe UPGMA clustering algorithm. Topology and branch lengths of this dendrogram are biased due to the presence of significant numbers of polymorphic clones derived from repetitive sequences. There is a difference between the two dendrograms obtained with the Psti/BastNI and the Psti!faqi array, respectively. A thorough inspection of the data suggests that the BstNI array contains a higher proportion of clones derived from repetitive sequences than Psti!faqi array. Typing using repetitive sequences introduces a bias in genetic diversity analysis due to over-representation. The Psti!faqi array does not show a high proportion of clones with repeated sequences and can be used as a routine genotyping tool for genetic diversity analysis. Work is ongoing to analyze a larger sub set of accessions from the CIA T core collection with at least 800 polymorphic clones Conclusion and Perspectives This feasibility project has resulted in the design and validation of a reliable complexity reduction method that is uncovering several hundred polymorphic DArT markers in cassava germplasm. Because DArT markers can be scored in parallel in a single analysis, a high throughput, cost-effective whole genome genotyping is now available for determining high-density genome profiles of cassava. As this study has demonstrated, the availability of such a method has the potential to improve dramatically the recognition, conservation and exploitation of cassava genetic diversity. This platform is now available for research on genetic diversity and genetic mapping studies in cassava. With current improvements of DArT's throughput and its cost reduction it is feasible to molecularly characterize very large germplasm collections within a reasonable period of time and cost. References Kawano K. (2003). Thirty Years of Cassava Breeding: BiologicaJ, and Social Factors for Success. Crop Science 43:1325-1335. Kawano K. (2001). The role of improved cassava cultivars in generating income for better farm management In:Howeler R.H., and Tan S .L. (eds). Cassava' s Potential in the 2111 Century: Present Situation and Future Research and Development Neds. Proceedings of the Sixth Regional Workshop, held in Ho Chi Minh City, Vietnam. Feb 21-25,2000. CIAT-Bangkok, Thailans. Pp5- 15. 207 1.3.9 Diversity Array Technology (DarT) for potato diversity análisis D. Posso,1 E. Gaitan,1 M. Ghislain,2 J. Tohme1 1SB-2 Project; 2CIP, Peru Introduction Potatoes are the fourth rnost important crop in the world after wheat, rice and maize (FAO, 1995). The annual potato production is about half that of all roots and tubers grown worldwide. Potatoes are consumed by more than a billion people in the world, including 500 mi Ilion from developing countries. There are about 500 potato species, which represent series frorn diploid to hexapod genomes. There are two principal centers of diversity: one in Mexico and the other in Peru, Bolivia and NW Argentina (Hanneman, 1995). Genetic diversity of potatoes includes 4000 cultivars, conserved and maintained at CIP (Centro Internacional de la Papa) in Peru. There are 200 wild species that have sorne traits of interest due to their resistance to diseases, pests and extreme environmental and stress conditions. Genetic resources are the foundation for generating new varieties and are a reservoir of genetic stability and adaptability. They are important to deal with environmental changes and pest attacks. To know about diversity is essential in order to preserve and use it for improving current cultívars. Diversity studies traditionally ernploy molecular markers based on gel platforms such as isozymes, SSR, RFLP and AFLP. These techniques are important because they have generated a lot of information, but they are linked to sorne problems such as being labor-intensive, the gel is expensive, not rnuch information results from each experiment, sometimes it is difficult to resolve bands on gels, and with sorne manufacturer's products, it is necessary to know the DNA sequences. DArT is a technology employing microarrays, capable of solving these problems. It is based on the generation of DNA segment arrays with unknown sequences (panels) of two or more genomic DNA individuals. These segments are hybridized with fluorescent- labeled DNA probes (representations) of two individuals that can be used to create the panel (Jaccoud et al., 2001). In our case panels were created with six potato genotypes including two individuals of a potato genome map. This technology has sorne applications such as tracking changes in DNA methylation, fingerprinting, diversity studies and elucidating complexes mixtures of DNA (Jaccoud et al., 2001). Materials and methods A total of 96 potato genotypes from the CIP potato gennplasm collection were collected, and DNA extracted. Six of them were used to make diversity panels. Generation of diversity panels.Total DNA (100 ng) of six potato genotypes was digested with 2 units of Pstl or Msel enzyme in separate reactions. Digestion products were ligated to their specific adapter and were PCR amplified. The amplification product was column purified, viewed in a 1% agarose gel and quantified. Then 400 ng DNA from the PCR of each genotype was mixed and 10.5 ng was bonded to pGEM T easy vector and used to transfonn competent Escherichia coli strain DH5oc by electroporation. The positively transfonned cells (white colonies) were transferred onto a freezing medium, where they grew at 37°C ovemight. The culture was PCR-amplified usingTI and Sp6 universal primers. The PCR products were dried using an incubator chamber at 50°C to obtain 2f.LL of the PCR total initial volume, which was re-suspended with 23f.LL of a 3X SSC, 1.5M betaine spotting buffer and used to spot microscope slides. Generation of representation. DNA (100 ng) of six genotypes used for building diversity panels was digested with Msel enzyme, bonded to its own adapter and PCR-amplified. The product was then column purified and labeled with Cy3 or Cy5 dyes using an Amersham Megaprime Labeling Kit, except for the enzyme. The enzyme used was Kleenow fragments, exonuclease-free, polymerase l. Five enzyme unites were employed for each labeling reaction. Hybridization. All Vanderbilt protocols were used to make panel hybridization which involves three steps: microarrays prehybridization, washing and hybridization. For the prehybridization a buffer containing 5X SSC, O, 1% SDS and 1% BSA was used to submerge microarrays for 45 min at 56°C. After washing, slídes were air dried for 30 min and hybridized in a Clontech hybridization chamber for 17 h at 42°C with a solution containing representations. Scanning and image analysis. Microarrays were scanned in a Virtek Chip Reader scanner, using a 2.0 version software. Images were analyzed using VersArray Analyzer version 4.5 for Windows (BIORAD), which is being demonstrated at CIA T. Net intensity spot values were nonnalized by using a loess regression that adjusts spot signal values according to their position on the slide and between signal channels. New signa! values were then analyzed, detecting sequences present in each potato genotype. 209 Results Two genonúc libraries were constructed using 6 out of 96 potatoes genotypes. One of these was made using Msel digestion enzyme; the other with Pstl enzyme. Each library contained 768 clones. Ten panels for potato-Msei library were generated, including alllibrary clones and control potato clones from TIGR. Six of those panels were hybridized with representations generated with PCR, using primers without an additional selective nucleotide at the 3'end. No polymorphic clones were identified in the Msel panel using those representations. The other four panels were hybridized with representations generated with PCR using primers with an additional selective nucleotide, adenine or thynúne, and with a bulk of six different genotype representations generated with PCR using primers without an additional selective nucleotide. Those experiments are still in analysis. Ongoing work • Hybridize Msel panels with different genotypes representations obtained by using PCR primers with selective nucleotides • Evaluate Pstl library for polymorphic clones by hybridizing Pstl panels with different genotype representations • Hybridize all potato genotypes with panel of genonúc library with more polymorphism References FAO. 1995. Producción de papa a partir de semilla sexual : Manual técnico. CIP-INIA. Santiago de Chile. 72 p. Hanneman, R.E. 1995. Riesgos ambientales de las plantas transgénicas en centros de diversidad: La papa como modelo: Ecología y biología reproductiva de la papa: Las implicaciones ambientales de la dispersión génica y su potencial. Robert J. Frederick, Ivar Virgin y Eduardo Lindarte. Memorias del Taller Regional, Parque Nacional lguazú, Argentina. Biotechnology Advisory Cornmision - llCA. Jacfoud, D.; Peng, K.; Feinstein, D. and Kilian, A. 2001. Diversity arrays: A solid state technology for sequence inforrnation. Nucleic Acids Res. 29: e25. 1.3.10 Microarray gene expression analysis as a functional genomics tool for postharvest physiological deterioration in cassava Cortés, D.F1.; Reilly, K.;2 Beeching 1.;2 Tohme, 11• 1SB-2 Project ; 2Dept. of Biology and Biochernistry, Bath University, UK Introduction The rapid postharvest physiological deterioration (PPD) of cassava roots is a major constraint that impacts negatively on the Iivelihoods of poor fanners, processors and consumers alike. Extending the shelf life of cassava to one or two weeks is perceived as a goal that would be beneficia!, particularly with respect to contributing to the sustainable livelihoods of small-scale rural farmers, as well as poverty alleviation. Genomics and bioinformatics tools are increasing our knowledge of plant genome structure, organization and gene function. Novel technologies such as expressed sequences tags (ESTs) and cDNA microarrays are proving rapid ways to identify genes and to link sequence information to biological function. The objective of this project is to identify the full set of major genes involved in the postharvest physiological deterioration by exploiting the powerful high-throughput analysis of cDNA microarrays. This will not only help understand the problem but also provide the tools (clones) that could serve as components of gene constructs to modulate PPD. This study is an important step in making this goal achievable, utilizing a cDNA library from different time points during the PPD process in cassava roots. Results and discussion Genomic library construction.A genomic library consists of a collection of clones that together encompass the entire genome of cassava. A genomic library cloned onto Lambda DASH II/Bamlll (Stratagene) has been made at Bath from cv. M Col 22. With a titer of 105 cfu/ml, the library is ready for screening with cDNA clones differentially expressed during a postharvest deterioration time course and subsequent isolation of corresponding genes including their regulatory elements (promoters). Microarray experiments with PPD-related cDNA libraries. A total of 7680 clones from the early library (0, 6 and 12 h after harvest) and 3072 clones from the late library (24, 48 and 96 h after harvest) were processed. Slide spotting for the microarray experiments was 211 carried out using the Hitachi SPBIO 1.55 system available at CIAT with PL-lOOC-Poly L lysine slides (CEL Associates). To have slides containing 4 technical replicates of each spot and containing 4 replicates of a control plate containing control genes and target spike genes, the early library was distributed onto 5 slides and the late library onto 2 slides. The total number of slides used in each hybridization was 7 (early slides 1-5, late slides 1-2) as shown in Table l. Table l. Hybridization strategy for microarray experiments. Hybridizalioo ~~. Ti~P-rnn 1 0 VS 12 Early Early Early Early Late Late experiments h slide 2 slide 3 slide 4 slide 4 slide 1 slide 2 2 0 VS 24 Early Early Early Early Late Late h slide 2 slide 3 slide 4 slide 4 slide 1 slide 2 3 0 VS 48 -Early Early Early Early Late Late h slide 2 slide 3 slide 4 slide 4 slide 1 slide2 4 0 VS 72 .EMiy Early Early Early Late Late h slide 2 slide 3 slide 4 slide4 slide 1 slide2 5 0 VS 96 Early Early Early Early Late Late h slide 2 slide 3 slide 4 slide 4 slide 1 slide 2 Range 6 12 VS 24 Early Early Early Early Early Late Late -'· h slide 1 slide 2 slide 3 slide4 slide 4 slide 1 slide 2 II:AJIII:llliiii:U,_, 7 24 VS 48 Early Early Early Early Early Late Late h slide 1 slide 2 slide 3 slide4 slide 4 slide 1 slide2 8 48 VS 72 Early Early Early Early Early Late Late h slide 1 slide 2 slide 3 slide 4 slide 4 slide 1 slide 2 9 72 VS 96 Early Early Early Early Early Late Late h slide 1 slide2 slide 3 slide4 slide 4 slide 1 slide 2 The probes used were cONA probes prepared from RNA extracted at different time points from cassava roots of CM 2177-2, the same cultivar used to make the spotted cONA library. For all experiments the earliest time point was labeled with Cy3 and the latest with Cy5; thus upregulated genes appear red. The spike genes used were TIGR El-ES (Cy 3 + Cy 5, should appear yellow); TIGR E6+E7 (Cy3 only, should appear green); TIGR E8+E9 (Cy5 only, should appear red). Scanning was carried out using the VersArray Chipreader (Biorad) available at CIA T. Initial analysis of all these hybridizations was carried out at Bath University, using ArrayVision software. Both normalized and raw data files were saved (126 analysis files). Subsequent processing of these analysis files is in progress using Cluster and Treeview software and has been completed for the early slide 1 time-course experiments (shown in gray in Table 1). Table 2 shows a partiallist with 18 of 165 upregulated genes of interest identified to date. An initial example of hierarchical cluster analysis of these genes is shown in Figure l. Recomrnendations and future work • Analysis has now been completed, and clones of interest transferred to Bath for sequencing analysis. • Given the number of clones of interest, it would be very labor intensive to confirm expression profiles by Northems for this number of candidate genes. Thus it would be advisable to repeat the hybridization experiments at CIA T so that the microarray data could "stand alone" as expression data in any publications/presentations. • Other microarray experiments to identify genes with specific tissue activity could be carried out using different plant tissues as the source for RNA isolation. 213 Genes of Interest- PPD Time-Course Exoeriments (loe 2) Fold Uoreeulation Flaes T ,ihrarv Slirle Snot Lahel c:Jone Name ifl VS 12 lo VS 24 O vs4R o vs 72 o vs 96 Earlv librarv IEarlv slide 1 ~ 2 B8 (4) Earlv olate 1 88 lo.64 1.20 1.97 1.58 1 1 ~2G20(4) Earlv nlate 1 G20 0.43 10.11 1.63 1.01 1 ~ 2 M21 (4) Earlv nlate 1 M21 -0.13 0.38 1.22 1.08 1 ~ 3 A2 (4) Earlv nlate 2 A2 0.45 Ú.39 1.34 1.02 1 ~ 3 A4 (4) Earlv nlate 2 A4 1.45 1.68 2.00 0.86 1 1 1 ~ 3 D19 (4) Earlv nlate 2 D19 1.22 Ú.70 1.11 0.62 1 1 ~ 3 14 (4) Earlv nlate 2 14 -0.36 0.52 2.02 1.11 1 ~ 5 K14 (4) Earlv nlate 4 K14 1.00 1.00 2.23 1.51 1 1 1 IEarlv slide 2 o 2 C20 (4) Earlv nlate 5 C20 1.40 -0.08 0.82 0.52 1.34 1 o 2 D20 (4) Earlv nlate 5 D20 1.38 [07 0.68 0.44 1.34 1 o 2 J13 (4) Earlv nlate 5 .Tl3 0.45 -0.04 0.95 0.19 1.29 1 o 3 C2 (4) Earlv nlate 6 C2 0.66 Ü.43 1.68 0.40 2.26 1 o 3 18 (4) Earlv nlate 6 18 0.07 -0.07 1.08 0.68 1.47 1 o 3 K4 (4) Earlv nlate 6 K4 0.95 Ü.12 0.68 -0.26 0.99 1 o 3 L21 (4) Earlv nlate 6 L21 1.02 0.16 0.72 -0.14 1.10 1 o 4 E16 (4) Earlv olate 7 E16 0.25 0.39 1.06 0.47 1.40 1 o 4 F16 (4) Earlv olate 7 F16 0.49 Ü.28 1.05 0.22 1.40 1 o 4 N14 (4) Earlv olate 7 N14 0.5.5 0.14 1.18 -0.04 1.31 1 Table 2. Partiallist of clones of interest selected on the basis of upregulation > + 1 fold in 2 or more arrays, yellow indicates visually confirmed: Early library. 214 Sum 1 3 1 1 2 1 2 1 2 3 2 1 2 1 1 4 1 2 1 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2 215 Figure l. Initial hierarchical cluster analysis for clones of interest - Early library. 1.3.11 Development of molecular techniques for assessing genetic diveristy and mapping of useful genes Development of a Diversity Array Technology (DArT) Chip for Cassava Andrzej Killian1, Peter Wenzel1 ;Carmen de Vicente2 Edgar Barrera3, Ana-Maria Correa3, Martín Fregene3 1(CAMBIA); 2(1PGRI); 3(CIAT) Introduction Genetic resources, mostly held by small farmers represent a critica! resource for the future productivity and stability of production of the crop. How to evaluate and use in a systematic manner the vast amount of variability present in cassava is still a challenge to most cassava breeding programs. Genotyping rnicro-array technologies offer the highest throughput available up to date. One of them diversity array technology, DArT ™ (CAMBIA), is sequence-independent (low-input) and allows the fingerprint of an individual's genome based on a high number of polyrnorphic sites spread over the genome. These screening procedures should allow testing of thousands of individual samples in a speedy manner. We describe here a proof of concept on using the DArT tool as a cost- effective way for measuring and characterizing genetic diversity of cassava germplasm. Methodology The project was initiated in early October 2002, with the shipment of cassava DNA samples from CIA T to CAMBIA. Plant materials used for the generation of the DArT chip was chosen to represent a broad as possible diversity of the cultivar, a few genotype of its wild progenitors and 2 wild species were included to capture a Iarge number of polymorphic fragments. They include 14 accessions from Brazil, 14 from Colombia, 4 from Guatemala, 2 each from Nigeria, Cuba, and Ecuador, Peru and Thailand respectively. Others include, one accession each from Argentina, Bolivia, Costa rica, Fiji islands, Indonesia, Mexico, Panama, Venezuela, and USA. Six and 2 improved varieties were included from CIAT and liTA respectively. The wild species accessions were 29 of Manihot esculenta sub spp Flabellifolia, 7 of M. carthaginensis and 1 of M. walkerae. DNA isolation was according to Dellaporta et al. (1983) followed by two washes of phenoVchloroform. A critica! step in DArT is the complexity reduction step. Work at CAMBIA with severa! other plant genomes has shown that digestion with Pstl restriction enzyrne (RE) in combination with more frequently cutting RE is an efficient method to reduce complexity. A prelirninary experiment was therefore conducted to determine the best enzyme combinations, mixture of genornic DNA from twenty cassava genotypes was digested with Pstl, ligated to adapters and further digested with several frequently cutting RE (BstNI, Apol, Taql and Banii), followed by amplification with an adapter-specific primer. The cassava genomic Pstl fragments lacking the recognition site for the frequent cutting RE (BstNI, Apol and Taqn were individualized by transformation into E. coli, amplified from bacteria! colonies and micro-arrayed. From each of the libraries 760 clones were arrayed. Genomic representations prepared in the same way (RE digestionlligation followed by amplification) from each of the twenty genotypes separately, were labelled with Cy3-dUTP and hybridized together with CyS-dUTP-labelled reference DNA to these microarrays. Slide preparation, hybridizations, washes and scanning are as described by Andrzej et al. (2002). lmages generated by the scanner were used to extract quantitative fluorescence signal data for each array feature using our proprietary software. S ame software was used to binarize the data (score as 0/1) for all slides. Binary scoring table was used to prepare the Hamming distance matrix and to obtain a dendogram. Library expansion to obtain more polymorphic clones was carried out using the enzyme combinations Pstl!BstNI, and Pstlffaql, and 80 DNA samples from CIAT. About 3000 clones were evaluated for polymorphism from the 2libraries. The arrays were printed and DArT analysis carried out as described earlier. BRA927 IIRA522 BRAaiO BRA70 WCOL251 BRM03 TWE117 OWB0-71 L--OW81-II ~=~ WGUA7 ·1 :~1 OW30-3 OW185- 0W1&4- L---OW111- '-------m.walkerjJ M. esculenta M. esculenta subsp. flabellifolia M. esculenta M. carthagenensis M. wa/kerae Figurel. Genetic relationship among the cassava accession used to identify polymorphic clones in the Psti/BstNI, Psttraql and Pst/Apol array. The dendrogram was created using the distance table based on 296 polymorphisms and the UPGMA clustering algorithm Results Among the three libraries /Pstl!BstNI, Pstl/ A poi, and Psti!I'aqn tested in the preliminary experimnt, Pstl/BstNI gave the largest number of polymorphlc clones (132), followed by Taql (112) and Apol (69). In total, 313 candidate polymorphic clones were obtained in the initial experiment to determine the best enzyme combination. DArT analysis based on 296 polymorphic clones without a single missing data point was used to generate a binary 217 matrix and obtain a dendrogram, based on Hammings distance, representing genetic relationship between the 20 samples (Fig. 1). The libray expansion with 80 clones yielded 440 polymorphic clones (14.3%), for the Psti!faql array, consistent with the results obtained with the initial, smaller array (14.6% polymorphic clones). A dendrogram based upon the analysis of the 80 cassava samples with the polymorphic clones is presented in Figure 2. r-- - 1 1 r----1 .-- ~ '-- y rf rC y ~ --------{ -r L_____f COL1522 MCOL2056 MCOL2061 COL72 ECU79 MGUA62 CUB56 TAI16 MGUA7 NGA19 NGA2 MCOL1505 CM6740-7 CM523-7 COL667A PER333 TMEl COL2737 HMC53 COL1468 BRA890 BRA522 MCOL2516 PER183 TAIB C4 AMA28 AMA19 BRA403 BRA1439 BRA124 BRA461 BRA383 OWB0-7 OW81-8 M.wal001 OW161-1 OW164-5 Figure 2. Genetic relationship among the cassava accession used to identify polymorphic clones in the Pstl!faql array. The dendrogram was created using the distance table based on 440 polymorphisms and the UPGMA clustering algorithm. 219 In the Psti/BstNI array 554 polymorphic clones (18.0%) were identifid, also consistent with the polymorphism frequency in the smaller Psti/BstNI array in the first project phase (17.2%). The dendrogram resulting from this array is shown in Figure 3. ~--COL2737 ARGS MCOLlSOS TMEl BRA927 CM6740-7 HMCS3 BRA403 BRA931 TAIB C4 BRA124 BRA1439 MCOL2516 NGA19 NGA2 .----AMA28 '---- BRA890 '---- PER333 ....---- BRAS22 .----AMA19 '---- BRA829 ...----------------1 L..---CMS23-7 P-----OWB0-7 '-----NGA2 MGUA62 ~--MGUA7 TAI16 COL1522 COL1672 COL72 ECU72 ECU79 CUBS6 MCOL2061 MCOL2056 ....---- BRA383 '----BRA461 '---------------M.waiOOl OW164-S OW161-1 Figure 3. Genetic relationshlps among the cassava accessions used to identify polymorphic clones (markers) in the Pstl/BstNI array. The dendrogram was created using the distance table based on 554 polymorphisms and the UPGMA clustering algorithm. Topology and branch lengths of this dendrogram are biased due to the presence of significant numbers of polymorphlc clones derived from repetitive sequences. There is a difference between the two dendrograms obtained with the Psti/BastNI and the Pstl!Taql array, respectively. A thorough inspection of the data suggests that the BstNI array contains a higher proportion of clones derived from repetitive sequences than Pstl!Taql array. Typing using repetitive sequences introduces a bias in genetic diversity analysis dueto over-representation. The Pstl!Taql array does not show a high proportion of clones with repeated sequences and can be used as a routine genotyping tool for genetic diversity analysis. Work is ongoing to analyze a larger sub set of accessions from the CIAT core collection with at least 800 polymorphic clones Conclusion and Perspectives This feasibility project has resulted in the design and validation of a reliable complexity reduction method that is uncovering severa! hundred polymorphic DArT markers in cassava germplasm. Because DArT markers can be scored in parallel in a single analysis, a high throughput, cost-effective whole genome genotyping is now available for determining high-density genome profiles of cassava. As this study has demonstrated, the availability of such a method has the potential to improve dramatically the recognition, conservation and exploitation of cassava genetic diversity. This platform is now available for research on genetic diversity and genetic mapping studies in cassava. With current improvements of DArT's throughput and its cost reduction it is feasible to molecularly characterize very large germplasm collections within a reasonable period of time and cost. References Kawano K. (2003). Thirty Years of Cassava Breeding: Biological, and Social Factors for Success. Crop Science 43:1325-1335. Kawano K. (2001). The role of improved cassava cultivars in generating income for better farm management. In:Howeler R.H., and Tan S.L. (eds). Cassava' s Potential in the 21st Century: Present Situation and Future Research and Development Neds. Proceedings of the Sixth Regional Workshop, held in Ho Chi Minh City, Vietnam. Feb 21-25, 2000. CIAT-Bangkok, Thailans. Pp5- 15. 1.3.12 Positional Cloning of CMD2 the Gene that Confers High Level of Resistance to the Cassava Mosaic Disease (CMD) M. Moreno, Martín Fregene CIAT Introduction Previous work revealed that the SSR markers SSRY 28 and NS158 are the closest markers to the gene CMD2 that confers resistance to the cassava mosaic disease (CMD), and are located at distances of 9 and 3 cM respective! y (Akano et. al 2002; Zárate 2002, CIA T 221 2002). High resolution of the C:MD2 region of the genome was therefore initiated in this region. The experimental approach involves a search for recombinants between CMD2 and the above markers using a large full-sib population, followed by an analysis of the recombinants with thousands of readily assayed markers to identify additional markers more closely linked to the gene. This requires the use of severa! types of marker systems that can achieve whole genome screens with a reasonable leve! of effort. Molecular markers identified that are closest to gene CMD2 will be used to screen a BAC library for the construction of a BAC contig that transverses the region of CMD2 via successive steps of BAC end sequencing. mapping and more BAC library screening. Methodology The fine-mapping population was 1690 individuals from a cross between TME3, the source of C:MD2 and the improved variety TMS30572. The cross was evaluated in the 2002 growing season for CMD resistance in the field at liTA, Ibadan, under heavy natural pressure of the disease. DNA was isolated from the individuals of the cross, using the Dellaporta et al. (1983) .method, and diluted 10X in TE without quantification for molecular marker analysis was at CIAT. The population was evaluated with the 2 SSR markers according as described by Mba et al. 2001 and recombinants between the markers and CMD2 identified. DNA from 10 resistant recombinants and 10 susceptible recombinants were combined to form 2 bulks which were then evaluated with severa] markers system including AFLPs, ISTRs, RAPDs and SSRs in a modified bulk segregant analysis (BSA) method.(Michelmore et al. (1991). Evaluation with AFLP markers (Vos et al. 1995) was using a commercial ALFP (Invitrogen Life Technologies, Gaithersburg, MD) following the manufacturer's instructions.. All 64 possible combinations were used in the evaluation. For ISTRs (lnter Sequence Tagged Repeat), the method described by Rohde et al (1996) was used with all possible 64 combinations of the 8 F and 8 B universal retro-elements (retro-transposons) sequence primers. Evaluation with RAPD markers was using 768 commercial primers (Operon Technologies lnc, CA) anda modified Williams et al. (1990) protocol. PCR conditions were 1X PCR buffer, 2.5rnM of MgC12, 0.4rnM of DNTPs, 0.8uM of each primer, l.OU of enzyme taq polymerase, and 50 ng/ul of DNA template in a of 2oül. The amplification ~rogram was an initial denaturation cycle at 94°C for 5min; 35 cycles of 94°C for 30s, 36 C for 1 min, 72°C for 1 min 30s; and a final extension cycle of 72°C for 5min. The amplified fragments were separated on 1.5% agarose gelsand visualized by staining with Ethidium Bromide and exarnination under UV light. A set of 146 newly developed SSR markers (CIAT 2002) were also used to evaluate the recombinant bulks using PCR and PAGE electrophoresis conditions described by Mba et al. (2001). Markers that were polymorphic in the recombinant bulks were then analyzed in individuals of the bulks. A polymorphic RAPD fragment in the individuals of the recombinant bulk was cloned into pGEMT -easy (Promega inc, Madison) and sent for sequencing at the University of Iowa sequencing facility. Primers were designed from the sequences and are being used as SCAR marker forMAS . Results The evaluation of the fine-mapping population with the SSR markers SSRY28 and NS158 allowed the identification 112 recombinant individuals. The evaluation of the resistant and susceptible recombinant bulks with AFLP, SSR, ISTR produced severa! candidate markers that were polymorphic in the bulks but the polymorphism was not consistent when individuals of the bulks were analyzed separately (opened bulks) . However, analysis with RAPD markers produced 2 polymorphic candidate markers, AC- 15 and RME-1 that remained consistent in the individuals of the bulks (Fig 1). Evaluation of the two markers in the entire fine map progeny revealed that AC-15 is at least 2cM from CMD2, while RME-1 is less than 1cM from the gene. The polymorphic fragment in RME-1, a 800bp fragment was cloned into the pGEMT -easy and sequenced. Homology comparison between the sequence of the RAPD band and sequences in public database using BLAST (www.ncbi.nlm.nih.gov) revealed the sequence is similar to the minor caspid protein of bacteriophage T3 suggesting that the fragment is single copy gene. This is being verified at the moment via southern hybridization to total cassava DNA. SCAR primers have been designed from the sequence for the use of the RAPD marker in MAS since it is closer than to the gene than NS 158, the closest marker to date to CtviD2. The abo ve result provides a molecular marker closely linked to CMD2 that can be used to screen a BAC library BAC of cassava for the construction of BAC contigs which is the next stage of the positional cloning of the CMD resistance gene. 223 RB · se RPSPIRBSB Figure l. Ethidium bromide stained agrarose gel of individuals from the recombinant bulks evaluated with the RAPD marker RME-1. A fragment at around 800bp (arrow) can be observed in the resistant parent (RP), and resistant bulks (RB) that is absent in the susceptible parent (SP) and susceptible bulk (SB) Conclusions A high resolution map with 4 markers, one at less than lcM has been constructed around the genome region of CMD2. The cloning of this gene is now proceeding to the next stage of BAC library screening and the construction of a BAC contig around CMD2. References Akano, A; Barrier, E; Dixon, A.G.O; Fregene, M. 2002.Genetic mapping of to dominant gene conferring resistance to cassava mosaic wished. Theorical and Applied Genetics. 105:521-525. CIAT 2002. Annual Report Project SB2, Assessing and Utilizing Agrobiodiversity through Biotechnology, CIAT, Cali, Colombia, 250pp. Hayden, M.J & Sharp, P.J. 2001. Targeted development of informative microsatellite (SSR) markers. Nucleic Acids Research. 29: 8e44. Mba, Rec; Stephenson, P; Edwards, K; Melzer, S; Nkumbira, J; Gullberg, U; Apel, K; Gale, M; Tohme, J.; Fregene, M. 2001. Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: Towards to molecular SSR-based genetic map of cassava. Theorical and Applied Genetics. 102:21-31. Michelmore, R. W; Paran, 1; Kesseli, R. V. 1991. Identitication of markers linked to disease resistance genes by bulked segregant analysis: a rapid method to detect markers in specific genomic regions by using segregating populations. Proceeding of National Acadernic of Science of The United States of America. 88: 9828-9832. Rohde, And 1996. Inver sequence-tagged repeat (ISTR) analysis: Two novel and universal for PCR- based technique genome analysis in the plant and animal kingdom. Journal Genetics and Breed.50:249-261. Williams, J.G.K; Kubelik, A.R; Livak. K.J; Rafalski, J.A; Tingey, S.V. 1990. DNA polymorphisms amplified by arbitrary primers plows useful ace genetic markers. Nucleic Acids Research. 18: 6531-6535. Zarate, L.A. 2002. Genetic Mapeo of a population F1 of cultivated yucca (Manihot escu/enta Crantz) using Microsatélites. Thesis of predegree. University of the Tolima. !bagué. Colombia. 1.3.13 Saturation of the Molecular Genetic Map of Cassava with PCR- based Markers: Progress on the Mapping of a New Set of 140 New SSR Markers L. Cano, J. Gutierrez, M. Fregene CIAT Introducción The wide-spread utility of the molecular genetic framework map of cassava published six years ago (Fregene et al., 1997) was delayed by the preponderance of the RFLP's markers used in constructing that map. Over the past 3 years, efforts have been geared to the development and genetic mapping of SSR markers to make markers on the cassava map more accessible to cassava researchers worldwide especially in the NARS of Africa, Asia, Latín America and the Caribbean. An initial effort led to the mapping of 77 markers (Mba et al., 200, CIAT 2001), other efforts have led to mapping of 57 SSRs (Zarate 2002) and 45 SSRs from cDNA sequences (Garcia 2002). We describe here the screening of another 140 SSR markers, identified in BAC end sequences, in the parents of the mapping population (Niogeria-2 and CM2177-2) and a preliminary report of the mapping of 26 polymorphic markers. Methodology Primers were earlier designed from a total of 141 BAC end sequences found to contain SSR motifs (CIAT 2002). A Dellaporta et al., (1983,) modified protocol was used to extract DNA extraction from the parents and the progeny of the mapping population. DNA quantification was done using the Dyna Quant TM.200 Fluorometer. Hoefer Pharmacia Biotech. PCR amplifications were carried out in 25 J..Ll reactions containing 50 ng of DNA, buffer lX, 2 mM of MgCh, 0.2 mM of each d.NTPs , 0.2 mM of each primer and 0.25 U of Taq polymerase. The PCR profile was changed for sorne markers, a reduction of the annealing temperature, to achieve amplification. The products of amplification were electrophoresed on 6% polyacrylamide gels and visualized by silver staining. For the survey of the parents, the following samples were included: Nigeria-2, CM2177-2, Kl50, TME3, TMS30555, CMD resistant and susceptible bulks. SSR markers that showed polymorphism in the survey of the parents, i.e. having a unique allele in either or both 225 parents of the mapping population, i.e. polymorphic, were used to screen the 150 progenies of the mapping population. Results A total of 35 polymorphic markers were found in the survey of the parents (Fig1). Of this 26 markers have been evaluated in the F1 progeny (Fig 2). The level of is about 25% polymorphism and it is much lower than that found for SSR markers developed from an enriched library (60%) and that from a cDNA library (about 40%). The low level of polymorphism found here could be due to the source of the markers, BAC ends, or other unknown reasons. NS 1141 NS 1040 NS 1072 NS 1019 NS 1035 NS 1008 NS 1099 Figure!. Silver stained gel polyacrilamide showing PCR analysis of seven markers NS used to survey the 2 parents of the mapping population, including another genotypes : the progeny klSO, TME3 and resistant (R) and susceptible (S) bulks. Each marker have changes in the annealing temperature (left to right : 52, 45, 52, 60, 60, 53, 55°C, respectively. ) The polymorphism between in tbe male and female genotypes can be observed. Figure 2. Silver stained gel polyacrilamide showing PCR analysis of NS 1019 marker in the progeny F1 (cross Nigeria-2 x CM 2177-2). The segregation of dates for this marker in a ratio 1:1, presence: absence of the unique parental allele (67n9 for the fernale and 76170 for the rnale framework respectively) can be observed. Conclusions A new set of 140 SSR markers identified from more than 2000 BAC end sequences of the white fly resistant variety MECU72 was evaluated in the parents of the mapping population and 35 polymorphic markers identified, of this 26 have been evaluated in the F1 pro gen y. References Fregene M, Angel F, Gomez R, Rodríguez F, Chavarriaga P, Roca W, Tohme J, Bonierbale M (1997) A molecular genetic map of cassava. Theor Appl Genet 95: 431-441. Garcia, T .. 2002. Mapeo genético de una población F1 de yuca cultivada (Manihot esculenta Crantz) utilizando Microsatélites que proviene de cDNA. Tesis de pregrado. Universidad del Valle, Cali. Colombia. Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg, U., Apel, K., GaJe, M., Tohme, J. and Fregene, M. (2001). Simple Sequence Repeat (SSR) Markers Survey of the cassava (Manihot esculenta Crantz) Genome: Towards an SSSR-Bassed Molecular Genetic Map of Cassava. Theoretical and Applied Genetics. 102: 21-31. Zarate, L.A. 2002. Mapeo genético de una población F1 de yuca cultivada (Manihot esculenta Crantz) utilizando Microsatélites. Tesis de pregrado. Universidad del Tolima. !bagué. Colombia. 1.3.14 Study of gene expression during pistil development in apomictic and sexual Brachiaria spp. D. Bemal, D. Cortes, J. Tohme SB-2 Project Introduction Apomixis is a reproductive mode through seeds that contain an embryo produced without the fusion of the egg and sperm cells; therefore it allows the fixation of any genotype, regardless of its complexity, through seeds. Perpetuating improved hybrids in this way could result in potential socioeconomic benefits that would challenge those of the Oreen Revolution (Vielle-Calzada, 1996). Currently, the most accepted point of view is that apomixis and sexuality are closely interrelated development pathways, where the former can be considered a deregulation of the sexual process in both time and space (Tucker, 2003). Sexuality is the reproduction mode in the 460 or so angiosperm families. There are several types of apomixis, of which one or more are present in at least 33 of the angiosperm families. Apomixis has been found to be a derived anomaly of polyphyletic origin (Carman, 1997). If just one or a few alterations do indeed change a sexual 227 development pathway to an apomictic one, then its polyphyletic origin could be understood from the standpoint of having arisen independently several times from different sexual ancestors (Spillane, 2001). Perhaps depending on the specific time and/or space deregulation presented, a particular type of apomixis appeared in each origin event. This project seeks to determine if there are different transcripts between apomictic and sexual plants; and if that is so, to typify them. Characterization of differences in the transcriptome of sexual and apornictic plants should give us insight into the mechanisms involved in the deregulation of the sexual reproductive developmental pathway, which end up transforming it in toan apomictic reproductive one. Methodology Using 13 plants from an interspecific Brachiaria hybrid population produced by the Tropical Pastures Program at CIAT, gene expression was compared among closely related apomictic and sexual genotypes in order to identify genes specific to the apomictic development. cDNA synthesis. Pistils are the organs containing the female gametophyte, where genes regulating the reproductive developmental pathway are assumed to be expressed. Pistils take around 18 days to develop, from their differentiation until they are ready for pollination at floret anthesis (Dusi, 1999). Obtaining clean pistils from Brachiaria plants is time consuming due to their small size and protected location within the floret; however it is worth the effort as the pistil accounts for only 0.001% of the floret's weight, which means that by extracting RNA only from pistils, one eliminates a large set of uninteresting transcripts. The 13 plants were classified as apomictic or sexual using the whole plant progeny test. About 300 pistils were dissected out of each plant, comprising the reproductive development from the 3rd to the 20th developmental day. cDNA was synthesized from mRNA fractions extracted from these pistils, from several individual plants (7 apomictic and 6 sexual plants), and two full-length cDNA librarles were constructed from the apomictic and sexual bulks as reported earlier (Cortes, 2002). Subtractive hybridization. Any phenotypic variation found between two biological samples is due in part to their transcriptome variation. Consequently, mRNA sequences found in greater abundance or found only in one of two mRNA fractions coming from different biological samples are particularly interesting to investigators as they could be responsible for their differences. If we are interested in understanding a particular biological condition such as stress responses or differentiation mechanisms, we must isolate those transcripts present only in that specific condition. Subtractive hybridization is a molecular method designed to isolate those target mRNA sequences. The biological sample that contains them is called the tester, and the biological sample that does not have the condition but is similar to the biological sample is called the driver. For instance, if we were studying the response to heat shocks, we would use heat-shocked plants as tester and non heat-shocked plants as driver. To identify differentially expressed genes between apomictic and sexual pístils, three different subtractive hybridization methods were used as each one has different advantages and disadvantages. The methods were subtractive suppression hybridization (SSH), mirror orientation selection (MOS) and differential subtraction chaín (DSC). In each method the apomictic and sexual cDNA bulks used to build the two full- length cDNA libraries, were used both as tester and driver in arder to isolate the cDNA sequences found only in the apomictic cDNA bulk (forward subtraction) and those found only in the sexual cDNA bulk (backward subtraction). The subtracted cONA ultimately obtaíned with each method was used to construct subtractive libraries of 384 clones each. Microarrays. cONA microarrays are a group of DNA probes spotted on to a salid substrate, usually a microscope glass slide. There may be around 10,000 spots or slightly more per slide. Hybridization of the microarrays with two complex cDNA fractions, each one labeled with different fluorochromes, allows the researcher to monitor the relative leve! of expression of the 10,000 probes simultaneously between the two complex cDNA fractions, by comparing the intensity of the fluorescence emitted by the fluorochromes at each DNA spot. Subtractive hybridization methods are not absolutely efficient, and subtractive libraries do contaín clones that are not differentially expressed between the two biological samples. Hence, microarrays were used to identify those clones from the subtractive libraries that are indeed differentially expressed between apomictic and sexual plants. Six hybridizations were performed between apomictic and sexual cDNA fractions. One hybridization was performed between two apomictic cONA fractions obtaíned from different apomictic species, Brachiaria decumbens vs B. brizantha. This hybridization was used as negative control, assuming that those sequences that differ between these two species are not involved in apomixes, since both of them are apomictic species. The hybridizations were performed as follows (each arrow represents a hybridization; the two~ headed arrows show that a repetition of each hybridization was done swapping the fluorochrome used to label each cDNA fraction.): l. Apomictic cDNA bulk 2. B brizantha cONA 3. B. decumbens cDNA 4. B. decumbens cONA fiim~ cONA bulk B. ruziziensis cDNA 'lllllft-=::::j~t Sexual cDNA bulk ~ B. ruziziensis cDNA 4 l Sexual bulk cDNA B. ruziziensis cDNA 4CE--_,;;~.., B. ruziziensis cDNA 229 After hybridization, each rrúcroarray was scanned with VersArrayTM ChipReader to obtain a 16-bit ~ged image file (TIF) for each fluorochrome, which was analyzed using Versarray Analyzer Software. Normalization between the two images was necessary to make biologically valid comparisons, thus the Loess intensity-dependent normalization method was applied as proposed by Yang et al. (2002). The program SAM (significance analysis of rrúcroarrays) from Stanford University was used to deterrrúne which spots were differentially expressed in each hybridization, choosing those whose ratio between the intensity of the fluorochromes is different from 1, keeping the false discovery rate below 7%. Those clones found to be differentially expressed in at least one repetition of the six hybridizations between apomictic and sexual cDNA, and were not found differentially expressed in both repetitions of the hybridization between the apomictic cDNA fractions, are considered candidates responsible for the differences between apomictic and ·sexual reproductive development. Subsequently, these differentially expressed clones were sequenced the ABI PRISMTM BigDyeTM Terminator Cycle Sequencing Ready Reaction Kit, with sorne modifications introduced by the CIA T Biotechnology lab, and the automated ABI PRISM 377 Sequencer (Perkin Elmer). Sequencing analysis software was used to obtain sequences, which were semi automatically cleaned (removal of ambiguous bases, adaptors and vector) using SequencherTM 4.1 . Cleaned sequences were compared against the All nonredundant GenBank CDS translations+PDB+SwissProt+PIR database, and against the GenBank non- mouse and nonhuman EST entries. Results and díscussíon Six subtractive cDNA librarles were obtained, doing a forward and reverse subtraction with three subtractive hybridization methos. Each one with 384 clones: l. Apomictíc cDNA minus sexual cDNA using SSH 2. Sexual cDNA minus apomictic cDNA using SSH 3. Apomictic cDNA minus sexual cDNA using MOS 4. Sexual cDNA minus apomictic cDNA using MOS 5. Apomictic cDNA minus sexual cDNA using DSC 6. Sexual cDNA minus apomictic cDNA using DSC The subtractive libraries 1, 3 and 5 are enriched in fragments found only, or more abundant in the apomictic pistil cDNA bulk. Each one of them was obtained using one of the subtractive hybridization methods aforementioned: SSH, MOS and DSC. The other three subtractive librarles, 2, 4 and 6 are enriched in fragments found only, or more abundant in the sexual pistil cDNA bulk. Each one was also obtained using SSH, MOS and DSC. Thirty cDNA microarrays containing 10,752 spotted probes each were constructed. Every cDNA microarray contains 4 repetitions of the 2,304 clones from the 6 subtractive libraries, and sorne other control probe spots used to analyze the microarrays properly. Based on these analyses, 65 out of the 2,304 clones from the subtractive librarles were found to be more abundant in the apomictic cDNA fraction than in the sexual one in all six hybridizations performed between apomictic and sexual cDNA, and they were not found to be in different quantities among B. decumbens and B. brizantha cDNA, which are both apomictic species. Figure 1 shows the distribution of the 65 clones among the 6 subtractive librarles. No clone was found less abundant in the apomictic cDNA fraction than in the sexual cDNA in the 6 hybridizations performed. Distribution of apomictic abundant clones among subtractive libraries 35.00 -r--------------------, 30.00 25.00 20.00 15.00 10.00 5.00 0.00 SSH MOS DSC Subtractive hybrldlzation method a Apo enriched library • Sxl enriched libra Figure l. Distribution of 65 clones found in bigher quantities in apomictic than in sexual cDNA according to microarray analysis among 6 subtractive libraries. Finding sorne clones at higher levels in the apomictic cDNA fraction and none at higher levels in the sexual one supports the hypothesis that apomixis is a deregulation of the sexual pathway, where this deregulation is controlled by the expression (and not lack of expression) of certain genes only in the female gametophytes with apomictic reproductive development. Finding clones at higher levels in the apomictic cDNA fraction from the subtractive library where the sexual cDNA was used as tester (i.e., apomictic cDNA was subtracted from the sexual cDNA fraction) is contradictory. A plausible explanation for this contradiction is that since apomictic plants are usually facultative apomictics (meaning they produce both apomictic and sexual seeds whereas sexual plants only produce sexual seeds), it is Iikely that the apomictic cDNA fraction contained sexual sequences in addition to the apomictic sequences. lf this is the case, when we take the apomictic cDNA fraction we are really taking apomictic and sexual sequences, and when we take the sexual cDNA we are taking only sexual sequences. Then, when we subtract the sexual cDNA fraction from the apomictic cDNA fraction, we end up with those sequences that belong only to the apomictic cDNA fraction. On the other hand, when we perform the subtractive hybridization backwards (i.e., subtracting the apomictic cDNA fraction from the sexual 231 cONA fraction), what we are really doing is taking sexual sequences and subtracting from them both apomictic and sexual sequences. Therefore, this subtractive hybridization is inefficient, because most likely there are no sequences found in the sexual sequences and not found in the apomictic plus sexual sequences. For that reason, this subtractive hybridization leaves us with random clones. Sorne of these random clones are found only, or at higher levels in the apomictic cONA fraction according to the microarray analysis. Another explanation could be that there are indeed no clones found only, or at higher levels in the sexual cDNA, so this backward subtractive hybridization would also be inefficient, also leaving us with random clones, and sorne of those clones are found only, or at higher levels in the apomictic cONA fraction according to the microarray analysis. However, just as among those random clones, there are clones found more abundantly in the apomictic fraction, there should be clones found more abundantly in the sexual fraction, hence the fact of not having found any clones only or more abundantly in the sexual cDNA fraction according to the microarray analysis, supports the second explanation, that is that there are indeed no clones found only, or more abundantly in the sexual fraction. According to the microarray analysis, there are 65 clones differentially expressed between apomictic and sexual plants, all of them found at higher levels in the apomictic fraction; 47 clones have been sequenced so far. Earlier, 150 clones were sequenced to assess the redundancy and kind of sequences we had obtained in the subtractive libraries. Currently, we have 197 sequences, of which only about 12% are redundant, meaning there is a big possibility that sorne sequences from the subtractive libraries were not cloned. The comparison of the 47 sequences against the All nonredundant GenBank CDS translations+PDB+SwissProt+PIR database yielded either random hits or homologies with hypothetical, putative, unknown or known proteins, distributed as shown in Figure 2a. The sequences homologues to putative, hypothetical and known proteins are distributed among functional categories as shown in Figure 2b. Homology type m Hypothetical •unknown O Putative O Protein •Random hit Functional category m Energetic • Other function OUnknown O DNA binding • RNA binding m Transduction Figure 2a. Figure 2b. Distribution of the apomictic abundant clones among homology types and functional categories. As Figure 2a shows, most of the hits are wíth real proteins (light aquamarine), around 12% show random hits, and the other hits were with sequences that are classified as proteins based on prediction with suitable software or on homologies with known proteins or ESTs. Therefore we might have sorne new sequences, putting together those that had random hits and those that gave hits with unknown proteins. Looking at the sequences with hits, which tell us something about their function (Fig. 2b), we have sorne sequences that probably control the idenúty of cells, being them DNA or RNA binding proteins or transduction factors. We also have a big portion of unknown function proteins, and a big portion of energetic or other function proteins such as transport or proteosome proteins. Given that apomixis might be a deregulation of sexuality in space or time, these sequences which code for transcription, translation or transduction factors are very interesting, since they regulate the fate of cells. A more in-depth analysis of their functions is being carried on. Future actions • More fragments from the subtractive libraries should be cloned, as the redundancy found in the sequenced fragments is very low. • The 65 clones classified as more abundant in the apomictic cDNA fraction by the microarray analysis will be used to find the full-length mRNA in the full-length cDNA libraries constructed earlier. • The full-length cDNA libraries will be amplified with clones from pistils less than 3 days old. • A subtraction library and microarray analysis should be done with this new full- length cDNA from pistils less than 3 days old. • A hundred and fifty clones that are different between the two apomictic species will be sequenced to gain basic knowledge about the evolutionary paths that are being taken by two closely related apomictic species. References Carman, J .G. 1997. Asynchronous expression of duplicate genes in angiosperms may cause apomixis, bispory, tetraspory and polyembryony. Biol. J. Linn. Soc. 61:51-94. Cortes, D., Bemal, D. & Tohme, J. 2002. Study of gene expression during embryo sac development in Brachiaria sp. In: CIAT (Centro Internacional de Agricultura Tropical). Assessing and utilizing agrobiodiversity through biotechnology. Annual report 2002, Project SB-2. Cali, CO. p. 230-231. 233 Dusi, D. M. A. & Willemse, M.T.M. 1999. Apomixis in Brachiaria decumbens Satpf: gametophytic development and reproductive calendar. Acta Biol. Cracov. Bot. 41:151-162. Spillane, C., Vielle-Calzada, J. P. & Grossniklaus, U. 2001. AP02001: A Sexy Apomixier in Como. The Plant Cell, Vol. 13, 1477-1480. Tucker, M. P., Araujo, A. C., Paech, N. A., Hecht, V., Ed. Schrnidt, D. L., Rossell, J . B., de Vries, S. C. & Koultonw, A. M. G. 2003. Sexual and apomictic reproduction in Hieracium subgenus Pilosella are closely interrelated developmental pathways. Plant Celll5:1524-1537. Vielle-Calzada, J. P., Crane, C. F. & Stelly, D. M. 1996. Apomixis: The asexual revolution. Science 274: 1322-1323. Yang, Y. H., Didoit, S., Luu, P., Lin, D.M., Peng, V., Ngai, J. & Speed, T. P. 2002. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucl. Acids. Res. Vol. 30: el5 1.3.15 Phenotypic characterization and seed multiplication of a collection of rice T -DNA insertional mutants M. Lorieux 1-2 and J. Lozano2 1SB-2 Project; 2IP-4 Project Project funded by the Génoplante consortium Introduction In the framework of its work plan for functional analysis of cereal genomes, the Génoplante consortium decided to construct a rice T -DNA insertional mutagenesis collection. Rice was chosen because of its small genome and because of all genomic resources available for this species (ESTs, genetic maps, complete sequence, etc.). The lines are produced in Cirad laboratories, and grown in Cirad and IRD greenhouses, in Montpellier, France. The present work carried out at CIA T as a collaboration with Génoplante consists in (i) a systematical phenotypic evaluation of the mutant collection, with production of an associated phenotypic database, and (ii) the multiplication of seeds for the entire collection, for later distribution to alllaboratories interested in rice functional genomics. What is a T-DNA mutant collection ? A T-DNA mutant collection is a library of lines obtained from transformation by Agrobacterium tumefaciens, using a specific construct called the T-DNA insert. If the T-DNA is inserted in a gene or its regulatory region, the insertion event results in alteration or di~ruption of the functionality of this gene, which can even be completely silenced. Each line contains one ore several insertions, resulting in the silencing of one of several genes in the line. Typically, dozens of thousands are produced. As the transformation events can be considered to be uniformly dispersed along the genome, one may expect that an important fraction of the genes contain an insertion. Additionally, the insertion sites can be individually characterized in sequencing the genomic region flanking the T-DNA insertion site. We then dispose of a Flanking Sequence Tags (FST) database. Hence, in the case of completely sequenced genomes like Arabidopsis or rice, the insertion(s) of each line can be placed on the annotated genome through sequence comparison and thus related to a specific gene or gene region. Utility of a T-DNA mutant collection. A T-DNA mutant collection is a powerful tool for discovering gene functions. Indeed, the disruption of a given gene or its promoter and/or transcription factors may result in alteration of the corresponding phenotype in relation to the wild phenotype. This phenotype can be observed at different levels - morphologic, developmental, physiologic, etc. As the entire collection gives access to genotypic- phenotypic association at the whole genome level, the scientific studies based on such a material fall into the field of functional genomics. Basically, there are two main approaches for discovering gene functions using a such library: • the collection can be used for forward genetics screens through systematical evaluation of all available Tlm families for a given trait. The mutant lines showing phenotypic variation in comparison to the wild phenotype can then be later analyzed for their T-DNA inserts (location on the genome, annotation on the sequence containing the insert) and association between the observed phenotype and the affected gene may be possibly established. • the other approach is called reverse genetics. One desire to test hypotheses about candidate sequences suspected to have a role in the control of the trait under study. Reverse genetics screens allow identification of the mutant(s) disrupted in a given (gene) sequence based on (i) PCR based amplification product in 2D or 3D DNA pools, (ii) hybridization signals on flanking regions spotted on medium density filters or (iii) by homology search in the FST database, supposing that the entire collection has been sequenced at least for one of the two flanking regions of each T -DNA insert. The identified mutants can then be screened for the targeted trait to evidence genotypic- phenotypic associations. The main advantage of the reverse genetics approach is that it saves large-scale phenotypic screenings. lt is specially interesting for time-consuming and/or high-cost phenotypic screenings. Characteristics of the Génoplante T-DNA rice mutant collection. The collection is based on the cultivar Nipponbare (Oryza sativa L. ssp. Japonica), because (i) it is relatively easily transformed, and (ii) its genome was used as matrix for the rice genome sequencing project. The lines are generated through Agrobacterium-mediated transformation through an efficient protocol (Sallaud et al., 2003). The objective is to ensure a primary coverage of the rice genome with enhancer trap T-DNA insertion sites (n = 40,000 lines). Individual and parallel characterization of insertion sites by systematic sequencing of the genomic 235 region flanking the left border of the T-DNA is carried out. Later, the enhancer trap T- DNA will be equipped with a non autonomous Ds element which can be further mobilized in trans for creating mutants alleles in genes in the vicinity of T-DNA insertion sites. The lines are also be individual! y characterized by GUS assays. All production and phenotypic (including GUS assays) data will be gathered from the different Génoplante partners (including the present work) using automated entry in a phenotype database software. A sequence database software will integrate finished flanking sequence data, survey against known DNA sequences as well as a functionally oriented representation of their location in the rice genome. All these data will be integrated in the future public database Oryza Tag Line. General features of T-DNA integration in this collection include (i) 1 to 4 copies were integrated at 1 to 2 loci , (ii) we observed 30-40% of single-copy transformation events (TE), (iii) the T-DNA segregated according toa 3:1 ratio in 95% of single copy TE, as well as 50% of multiple copy TE, (iv) a preferential integration in the gene-rich fraction of the genome was observed. Phenotyping of the rice mutant collection. Here, we present one of the several projects carried out on the Génoplante rice mutant collection. In its first phase, it consisted the following main activities: • to carry out screenhouse and field phenotyping of a first set of 5,000 lines, • to produce a mutant phenotypic database, • to produce T2 entire panicles for further detailed analyzes of grain filling, • to multiply T2 seeds to constitute a rice T-DNA mutant stock center for future distribution and collaboration with partners. Materials and Methods Screenhouse. Five thousand TO plants were produced at Cirad and grown in Cirad and IRD glasshouses in Montpellier, France. Twenty-five T1 seeds per TO plant were received at CIA T and were sawn in a screenhouse. Sawing was carried out in four batches of 1,250 lines, with about three weeks delay between two batches. Seeds were pre-treated by heat for three days at 50 oc to brake dormancy. The nurseries were realized in plastic trays with a mixture of CIAT (67 %) and Santander de Quilichao (33 %) soils. Germination was counted at ten days after sawing (DAS). The first phenotypic observations were carried out at 18-20 DAS, with counting of the number of individuals presenting the mutant phenotype. A list of possible phenotypic traits was established from data mmmg of several rice phenotypic databases (www.gramene.org, www.grs.nig.ac.jp/rice/oryzabase, www.irri.org/genomics), and was used as a guide for observations. An English-Spanish-French lexical of botanical and agronomic terms was established to facilitate phenotype identification. Field. A 2 hectares field was set up following the requirements of the ICA (Instituto Colombiano Agropecuario). In particular, the en tire field was covered by nets to avoid damage and seed dissemination by birds. Plantlets were transplanted at 25 DAS. A basic fertilization was applied, composed of Mono-Ammonium Phosphate, Iron Sulfate, Potassium Chloride and micro-elements. Irrigation was applied two times a week. Controllines ofNipponbare cv. were planted each 10 T-DNA lines in order to facilitate the comparison with wild phenotype. Phenotypic analyses were carried out at different ages, using the list of possible traits as a guide. A first round of observation was run on approximately 45 days-old plants. A second round was run at flowering time, while the ultimate observation was done at maturity. This permitted to maximize the chances to detect phenotypic variations, as various traits could be observed at one of these stages only. Moreover, this permitted to follow the evolution of a suspected phenotype at early stage and possibly confirm or invalidate it. Harvest and storage conditions. Harvest was organized as followed: - Three panicles per plant of each line were collected. This material is to be sent to Biogemma, a private partner of the Génoplante consortium. Biogemma will conduct phenotypic analyses on this material in the framework of a project on functional genomics of grain filling . Seeds from these panicles will later be sent to Cirad to constitute the main mutant stock center. - The seeds from three to six panicles - depending on the level of sterility- were collected from each plant of each line. This will constitute the replicate of the mutant stock center hosted by CIAT. - When possible, for each line showing a mutant phenotype, all remaining panicles from one plant with the phenotype were collected in order to facilitate segregation studies based on wild x mutant crosses. Results Legal aspects. All issues relative to legal aspects, i.e. importation permit, quarantine and authorization for field tri al of transgenic plants were fixed. 237 Gennination rates. Germination rate of controls (Nipponbare cv.) was close to 100 %. For T-DNA lines, the mean germination rate was around 61 %. A trial without heat treatment showed a mean germination rate of about 35% (Figure 1). Growth. Rice plants generally grown normally. The Nipponbare variety showed a normal phenotype, however, due to short-days conditions inducing a short cycle, plants showed slight reduction in height and tillering than under long-days conditions. Cycle. Nipponbare is a photosensitive cultivar. Thus, under short-days conditions (about twelve hours of daylight per day at Cali), its cycle is reduced in length. At 60 DAS, 50 % of the p1ants had flower, while the 90% of flowered plants was attained at 64 DAS. The complete cycle of Nipponbare (sawing to maturation) under our conditions was around 93 days. Diseases. Plant health was generally very good and close to optimum. Only very localized attacks of Hydrelia and Hoja Blanca Virus were observed. Also, zinc deficiency symptoms were observed in various locations of the field. Figure l. Evolution of the % of flowering in the Nipponbare control Fertility/Seed set. Seed set of the controls was good. The mean fertility was around 81 %. Mutant phenotypes. In the screenhouse, about 18% of the lines showed phenotype variation in comparison to the wild type. This is much more than currently observed in other mutant collections, where about 3 to 5 % of mutants are identified by visual phenotypic screening. T -DNA insertion is probably not responsible for all the variation observed. Indeed, it is well known that other sources of mutation like the Tos 17 retrotransposon are positively activated by in vitro culture of rice. Moreover, discrepancies in germination dates and seed quality, mainly due to the growth conditions of the TO plants, may be responsible for apparent mutations, notably Growth-Retardate (GR), Tillering, Height. Also, redundancy of phenotypes was frequently observed between two or more lines. We think that, for lines showing close serial numbers, this could result from the same transformation event. It will be necessary to verify if those lines proceed from the same callus, thus bearing the same transformation event. The rate of lines which showed any type of mutant phenotype was 18 %, which should be, as in the screenhouse and for the same reasons, an upper bound. Also, we chose to include even doubtful data as it is preferable to eliminate false-positive data after more detailed analyses for a specific trait than to miss real data. As it is impossible to describe all the data here in details, we chose to present only a few examples. · Numerous lines showed chlorotic or albino plantlets, with associated deficiency in leaf development (Figure 2a, b). General abnormal development was also frequently observed (Figure 2c, d). The most common phenotype included severa! types of dwarfism more or less pronounced (Figure 3a). Sorne lines also showed a mutation for increased size. Figure 3b shows an example combined with awn spikelets. Several lines showed reduced or increased tillering. Notably, a phenotype with very high tillering was observed for severallines (Figure 4b). The Growth-Retardate phenotype was frequent, with associated reduced tillering and height, and late flowering (Figure 4b). Database set up. A local database of all data relative to growth conditions, germination, flowering, and phenotypic observations was set up. This database is used as a working tool to facilitate data entering and compilation. It also can be used for data browsing as it permit to display information by mutant bar code number or CIA T number. This database also displays photos of the mutants (Figure 5). From this database, a flat datafile will be extracted in order to fill the Génoplante database which will be later made publicly available. Conclusion The overall process of seed multiplication and phenotypic analysis worked very well. The timetable was respected, and valuable phenotypic data were produced. This first trial should constitute a good basis for conducting a larger project. 239 Future plans To extend the phenotypic analysis and seed multiplication to the entire collection (35,000 lines). References e Sallaud, D Meynard, J van Boxtel, e Gay, M Bes, J P Brizard, P Larmande, D Ortega, M Raynal, M Portefaix, P B F Ouwerkerk, S Rueb, M Delseny, E Guiderdoni. 2003. Highly efficíent production and characterization of T-DNA plants for rice (Oryza sativa L.) functional genomics. Theoretical and Applied Genetics 106:1396-1408. Figure 2. Examples of mutant phenotypes observed in the rice T -DNA collection in the screenhouse. a: Chlorotic, b: Albino, e and d: Development defects a b e d 241 Figure 3. Examples of mutant phenotypes observed in the rice T -DNA collection. a: dwarf, b: increased size associated with spikelet awning a b Figure 4. Examples of mutant phenotypes observed in the rice T-ONA collection. a: growth retardate, b: high tillering a b 243 mutant database 1.3.16 Identification of genes induced during the defense response of Brachiaria to the Spittlebug C. Romero, l. F. Acosta & J. Tohme SB-2 Project Introduction The molecular basis of plant defense responses to insects is a challenging area whose understanding should make feasible the use of natural immunity in economically important plants. Although molecular biology has recently been incorporated in the exploration of these defense mechanisms, it has been mainly lirnited to studying the interaction between dicotyledonous plants and herbivorous chewing insects. Herein we focus in the defense responses of a monocot that exhibits resistance to a xylem-sucking insect, an interaction that is poorly understood at the molecular level given the peculiarity of this feeding habitat. One of the possible approaches to get closer to such a system without previous molecular data is the characterization of transcriptional changes during the plant response to the insect attack. In this work we show the isolation of differentially expressed sequences in the resistant Brachiaria hybrid CIA T 36062 when challenged with Aeneolamia varia nymphs. This was achieved by a subtractive hybridization technique and a rigorous sequence analysis to identify putative functions of the isolated transcripts. Sequencing analysis of -240 clones from the subtractive library revealed that they corresponded to 74 unique expressed genes. Putative functions were assigned to 41 transcripts through sequence sirnilarity searches and the predicted proteins were classified in eight functional groups. These include three biosynthetic pathways of important plant signaling hormones, cell signaling, transcriptional regulation, cell wall modification and the homeostasis of the plant during the water stress caused by the insect. Finally, we found two putative effector proteins that may be con tribute to the antibiotic action of the resistant plants on the insect. Methodology The subtractive hybridization was performed as described in the 2002 Annual Report, where we also showed the results of two 96-well pilot libraries (Annual Report, 2002). Consecutively, we decided to construct two additional libraries of 384 wells in order to expand the coverage of the subtractive product and to avoid the RNA ribosomal artifacts. Four bands corresponding to rRNA generated in the cDNA synthesis were identified by size when the product of the subtraction was run in an acrylarnide gel. The rest of the smear observed in this gel was excised and cloned to create the new libraries. Sirnilarity searches were performed in the GenBank using the BLASTX algorithm. The matching sequences were searched in annotated databases such as T AIR (The Arabidopsis 245 Research Institute), GRAMENE, SWISS-PROT, and ENZYME in arder to detennine their putative functions. More specific information was obtained in secondary databases as InterPro (lntegrated Resource of Protein Families), Pfarn (Protein Families Database), PRINTS (Protein Motif Fingerprint Database), AraCyc (AraCyc: Arabidopsis thaliana Biochemical Pathways), and CDD (Conserved Domain Database, at NCBI). In these databases we found functiona1 information such as precise biochemical roles, metabolic pathways and redundant proteins (other proteins with the same function). Furthermore, we found structural information such as protein motifs and domains contained in the predicted Brachiaria proteins. Finally, in sorne cases we performed two pair alignments of arnino acid sequences using the pair BLAST algorithm to confirm the structural relationship between elements isolated from Brachiaria and the proteins previously reported in resistance studies in other species. A macroarray experiment was carried out in arder to test the differential expression of the isolated transcripts. The clones of the four libraries were arrayed using a 384-well pin replicator on duplicated nylon membranes and grown on LB-agar medium ovemight. The bacterial colonies were denaturalized and the free DNA was UV-crosslinked. These filters were hybridized with radioactivity labeled cDNA from infested plants and from non- infested plants and exposed to autoradiograph films. Results The cloning strategy to avoid rRNA sequences in the new libraries -which have been partially screened at this point- was successful since the proportion of these artifacts decreased from 40% to 5%. Bringing back together the results of last year's pilot libraries, a total of 240 clones yielded readable sequences, which corresponded to 74 unique transcripts. No match was found for 26 (35%) of them in the GenBank; these may constitute a reservoir of new genes, absents in model species. Seven more sequences were not considered further (e-values over 10·\ Eighteen out of the remaining 41 sequences are similar to genes that have been shown to be part of defense responses in other plant-insect or plant-pathogen interactions (see Table). These sequences can be divided in six main groups according to their presumed roles in defense (see Figure) Hormone Biosynthesis Pathways. The first group contains sequences coding for putative enzymes that catalyze important steps in the biosynthesis of oxylipins, brassinosteroids and ethylene. Oxylipins are the most important hormones in the systemic response to wounding and insects. Plants deprived of Phospholipase (Tumer et aL, 2002), Fatty acid desaturase (Martin et al. , 1999) and Lypoxigenase (Bell et al., 1995) are notable to produce Jasmonic Acid (JA) and are incompetent to activate defense mechanisms such as proteinase inhibitors, defensins and thionins. Brassinosteroids have been mainly studied in cell division and plant development but recently they were assigned a role as systemic defense hormones in responses to virus, bacteria and fungi in both dicots and monocots (Nak.ashita et al., 2003). Although the role of Ethylene is widely recognized in plant-pathogen interactions, its effects can differ in different situations. The emission of ethylene has been proved as a mechanism of communication between organisms that attracts natural enemies of herbivores and causes unwounded leaves to initiate ethylene biosynthesis as a positive feedback loop (Arimura et al., 2002) Table: Putative functions assigned to the transcripts isolated. Sequences related with defense responses in other systems are shown in bold Putativa Function E vafua Scora 0-Malhyftransfarasa 3.00E-89 329 Hypothaticaf Protain 1 1.00E-71 269 Ornfthfne carbamoyltransferase (OeTase) 5.00E-67 233 eystelne Protelnase 7.00E-63 241 Phosphollpase e (A) 8.00E-58 213 Hypothaticaf Protein 3 1.00E-52 206 dTDP-glucosa 4-6-dahydratasa 4.00E-53 206 Chlorophyll a/b-binding protain (CAB) S.OOE-50 196 Phosphollpase e (B) 1.00E-49 195 Omega-3 fatty acld desatura 4.00E-47 187 eBL·Interactlng proteln klnase (etPK) 1.00E-41 168 Tubulin afpha 2.00E-41 168 eaffeoyl eoenzyme A 3-0-Methyltransferase 41.00E-39 161 MADS Box 1.00E-37 148 Gfutathione S-conjugata ABC transportar 3.00E-35 147 Xyloglucan endotransglycolase 2.00E-34 145 Lypoxlgenase A (LOX) 3.00E-34 145 Lypoxlgenase B (LOX} 3.00E-34 144 Unknown protein 3 (TMS1d) 2.00E·31 132 Fructose bisphosphata aldofasa~ 1.00E-27 122 Ca dapandant mitochondiral carriar protein 7.00E-2n 15 NAO+ dapendent isocitrata dahydroganasa sul1l.OOB2511 E Saquance associated to Pi2 (RG64 RFLP ma~E-24115 Devalopmentaf Protain 3.00E-221 04 Elongation factor 1-afpha 6.00E-211 OC Unknown protein 2 2.00E-20 98 Pur-alpha 1 E 3.00E-17 88 SCAREeROW 3.00E-17 88 Putativa raplication protainE 4.00E-16 84 Extracallufar fipasa 3 (A) 2.00E-15 55 Cold acclimation protain 8.00E-14 77 60s ribosomaf protain L 13 9.00E-14 76 Hypothatical Protain 6 ?.OOE-12 71 S.adenosylmethlonfne synthetaseE 2.00E-11 69 Hypotheticaf Protain 4 2.00E-1 O 66 eaffelc acld 0-methyltransferase 1.00E-()9 62 Carbonic Anhydrasa 3.00E-Q914E Metalloprotease 1.00E-o8 57 Unknown protain 1 1.00E-07 56 Sterofd 22-alpha-hydroxylase (cytochrome ~...()7 56 Extracallular lipasa 3t(B) 2.00E-07 57 Cell signaling. The presence of a putative CIP kinase can be explained by two different processes: a) this protein may interact with CBL, a plant calcium sensor, in the signaling cascade activated under a spittlebug attack. Ca+ signaling in responses to wounding and pathogens has been well documented. b) CIP kinases contain a SNFl domain, suggesting a possible role in gene regulation controlled by cytoplasmic carbohydrate concentration (sugar sensing), an activity that may be related with the rapid change in metabolism that should occur to supply the energy requirements of the defense responses (Rolland et al., 2002). Transcriptional Regulation. SCARECROW is a transcription factor that is rapidly induced upon perception of the elicitor N-acetylchitooligosaccharide. Moreover, its mRNA is induced in rice upon fungal infection but not in the presence of bacteria! pathogens (Da y et al., 2003). Both evidences suggest that this. transcription factor may be involved in 247 responses to enemies that contain chitin, which could be a way to regulate defenses against a broad but specific range of organisms (fungi and arthropods). <.t ... ~ o a: .... <( z ~ a: o :r Lypoxigenase t t ~ JA & oxylipinas 2~ Hydroxylase t r y BR SAM synthetase -r i' V Ethylene Chitm? r!l z :::¡ < z ~ <.t .... .... Ul o <.0 a: w a: ~ 111 Ul ~ <.0 :z Ul .... Ul e w" ~g Cl- ct: ~tt (.)UI 1 Signaling TI Ca~ CBL -interadíng protein kinase Cell wall modifications Caffeoyl Coenzyme A 3-0-Melhy/trans/erase melhvltransferase Xyloglucan endotransg~¡co&ase {Cartlohydrates¡ )t.. Caroohydrat~ metabolísm diDP~Iucose 4-6- .rl~~thvrlr.:tbt~tll Fructose bisphos- phale aldo&ase NAO+ dependan! isod- lral.e dehydt09enase (sug <.O r!l zw '"' ...... o .1. 1 o o o o. ~o ~1 • 11 o o o •A·~!'t>o • o \ 'lo 0 ° o o o 1 o ' ~ Cb ~·· •• o '· 1 o o ••• 1 o o ;. o o a o •' 1 .. ' o 1 o 1 1 -0.2 +---~-....-~-----,_........c....._,....~-+--J -0.2 0.0 0.2 0.4 0.6 0.8 Root growth vigor ( • RL in Al-free treatment) [m; log scale] Figure l. Segregation of root growth vigor (x axis) and Al resistance (y axis), both measured based on root elongation (RL = root length) in a B. ruziziensis x B. decumbens hybrid population. Al resistance was quantified by adjusting RL of plants in the Al treatment for their RL when grown without Al (see text for further details). Data points representing parents are highlighted with large black symbols. Error bars in the lower right comer denote the average SE of the hybrid population. Because root growth vigor varies among hybrids, the inhibitory effect of Al on root length (RL) is not easy to measure, and RL in the Al treatment has to be properly referenced. We tested two approaches. First, we calculated relative root length values by comparing - for each pair of stem cuttings distributed among the two treatments - RL in the Al treatment vs. RL in the treatment without Al (the logarithm of the ratio was used for this purpose). The resulting "Al resistance index" was not correlated with root growth vigor. We therefore also tested adjusting RL values in the Al treatment for RL values in the Al-free treatment using linear regression. The two approaches were compared using the degree of correlation among three parameters describing different effects of Al toxicity as a criterion. These were: (i) inhibition of root elongation (see above), (ii) lateral swelling of roots (resulting in a greater root diameter, RD), and production of a greater number or short laterals (resulting in a higher specific root tip number, sRT = number of root tips per unit dry weight). The RD and sRT parameters had been computed based on the two approaches described for RL. The three parameters were more tightly correlated in the second approach, thus suggesting superior data quality. The latter was also corroborated by the fact that the two parents were closer to the two extremes of the distribution. We therefore used the linear regression approach to present the data in Figure l . Bdec Parents + hybrtds In ascendlng order of RL values Figure 2. Comparison of three parameters that reflect Al resistance: root Iength (RL), average root diameter (RD), and number of root tips per gram or root dry weight (sRT = specific number of root apices). The parameters were measured with rooted stem cuttings in an Al· toxic solution. They were adjusted for harvest mean, the effect of the dry weight of stem cuttings used, and root growth vigor measured with another set of stem cuttings in an Al-free reference treatment (see x axis in Figure 1). Data points representing parents are highligbted with large black symbols. Figure 2 displays the relationship among the three different Al-toxicity parameters used above. Genotypes more severely affected by Al toxicity have smaller root systems (shorter RL) that are made up of thicker roots (greater RD) and are more heavily branched (more root tips per unit dry weight, sRT). lt may be possible to increase the robustness of detecting Al-resistant genotypes by computing the principal component of these three parameters to create "a composite Al resistance index" that captures the various Al- induced changes in root growth and architecture in a single parameter. References Causton DR, Venus JC 1981. The Biometry ofPiant Growth. Edward Arnold Publishers, London, UK. 307p. 1.3.19 ldentifying candidates genes whose expression is associated with aluminum resistance in Brachiaria A. Aranr,o1, D. F. Cortes1, G. Gallego\ P. Wenzl2, I.M. Rao1, M. Ishitani1 and J. Tohme1 1CIA T; CAMBIA -Australia Introduction Last year we reported on the identification of individuals of a Brachiaria ruziziensis x Brachiaria decumbens hybrid population with contrasting degrees of Al resistance. This 255 suggested the feasibility of isolating candidate Al-resistance genes from root apices based on a comparison of gene expression pattems between an Al-resistant and an Al-sensitive bulk of hybrids. This year we pursued this approach. Materials and Methods Rooted stem cuttings of B. decumbens, B. ruziziensis and a group of hybrids with contrasting Al resistance levels were cultivated as described in ''Evaluating traits associated with acid soil adaptation in a population of Brachiaria ruziziensis x Brachiaria decumbens hybrids". Root apices collected at various harvests were pooled to create 6 samples for RNA extraction: two Al-resistant bulks of hybrids (grown with +/-Al), an Al- sensitive bulk of hybrids (grown with Al), two B. decumbens samples (grown with +/-Al) anda B. ruziziensis sample (grown with Al). Total RNA was isolated, mRNA was captured with magnetic beads, and ds-cDNA was synthesized using anchored oligo(dT) primer/adapters. To minimize cross-hybridization among genes belonging to the same family, 3' -UTRs were amplified by suppressive PCR after simultaneous digestion of cDNAs with Msel and Mspl followed by adapter ligation. The resulting amplicons were then mixed in various combinations and subjected to three rounds of differential subtraction chain (DSC) (Luo et al., 1999). Results and Discussion Figure 1 displays the 3' -UTR amplicons obtained from the six mRNA samples. Control experiments confirmed that only 3'-UTR fragments and no interna! gene fragments had been amplified (i.e. suppressive PCR was successful). In addition, sequencing of random 3' -UTRs fragments identified homologies to 3' -regions of known genes, including one coding for a root-specific protein. As a result of the double digestion with two 4-bp cutters, the fragments are small. This should have minimized cross hybridization among different genes during the subsequent subtractive hybridization steps. The 3'UTR amplicons shown in Figure 1 were mixed in various tester-driver combinations to enrich differentially expressed genes using the DCS approach (Table 1). Table l. Combinations of tester and driver cDNA samples used for subtractive hibridization. Tester Driver Combinatio Genotypes Treatment Genotypes Treatment n Al-resistant F1 bulk 200 OM Al3+ Al-sensitive F1 200 DM buJk Al3+ 2 Al-resistant F1 bulk 200 OM Al3+ Al-resistant F1 bulk no Al3+ 3 B. decumbens 200 DM Al3+ B. ruziziensis 200 OM Al)+ 4 B. decumbens 200 OM Al3+ B. decumbens no All+ 11501 1700 1093 514 bp 1 2 3 4 S 6 Figure l. 3 'UTR amplicons obtained from six cDNA samples: 1, Al-resistant F1 bulk grown at 200 J.1M Al3+; 2, Al- resistant F1 bulk grown without AI3+; 3, Al-sensitive F1 bulk grown at 200 J.1M A¡J+; 4, B. decumbens grown at 200 J..LM Al3+; 5, B. decumbens grown without A13+; 6, B. ruziziensis grown at 200 J.1M Al3+. The 3' -UTR fragments remammg after three rounds of subtractíve hybridization were separated on a polyacrylamide gel. Distinct bands were visible in sorne combinations of tester and driver amplicons (Figure 2). This appears to suggest that the DSC procedure may have been successful. 330 240 180 140 bp Figure 2. 3-UTR fragments remaining after three rounds of DSC. The numbers refer to the four combinations of tester and driver amplicons listed in Table l. 257 Bands will be excised from the gel, cloned and sequenced. In addition, random fragments from the second round of subtraction will be cloned and sequenced. Depending on their redundancy they will be used to fabricate cDNA microarrays for screening for differentially expressed genes. References Luo JH, Puc JA, Slosber ED, Yao Y, Bruce YN, Wright Jr TC, Becich MJ, Parsons R 1999. Nuc/eic Acids Research 27: e24. 1.3.20 Isolating genes from root apices of Brachiaria decumbens that enhance Al resistance of yeast P. Wenzl1, E. Gaitán2, I.M. Rao2 and J. Tohme2 1 CAMBIA- Australia; 2CIAT Introduction Sorne genes that when transfonned into plants increase their resistance to Al, were originally identified based on their ability to enhance Al resistance of yeast. We used this approach to identify candidate Al-resistance genes in root apices of B. decumbens grown in the presence of Al. Materials and Methods Total RNA was extracted from root apices of rooted B. decumbens stem cuttings grown in a solution containing 200 OM CaCl2, 200 OM AlCl3 (pH 4.2). After capture of mRNA with magnetic beads, ds-cDNA was synthesized, ligated to adapters, size-fractionated and PCR-amplified. PCR products from the > 2 kb and < 2 kb fractions were ligated separately to linearized p YES2 plasmid and transfonned into E. coli. The libraries obtained were amplified on pi ates. Plasmids extracted were mixed at a 1: 1 ratio and re-transfonned into yeast. Transfonnants were plated on a medium containing enough Al to arrest growth of yeast cells transfonned with empty plasmid. Plasmids from the most quickly growing colonies were isolated, re-transfonned into E. coli, and extracted for further characterization. Results and Discussion Approximately 100 yeast colonies were obtained from two million cfu plated on the Al medium (= 0.005 %). The plasmids isolated from 48 well-growing colonies were digested with a mixture of restriction enzymes to identify clones that had been isolated more than once, thus minimizing the chance of selecting false positives for further analysis. This fingerprinting experiment identified nine clones that had been isolated at least twice from different colonies. Work is underway to sequence and characterize inserts from these clones. 259 lo ~b 11 OUTPUT 2 Genes and genes combinations made available for broadening the base of mandated and non mandated crops Activity 2.1 Transfer of gene and gene combinations using cellular and molecular techniques 2.1.1 Development of tepary x common bean interspecific hybrids with improved competence to Agrobacterium mediated transformation Alvaro Mejía Jiménez, Leonardo Galindo, Catalina Oviedo and Joe Tohme SB-2 Project Introduction Among the possible ways for the genetic transformation of plants, the transformation with Agrobacterium as vector, offers unique advantages. The common bean (Phaseolus vulgaris) has been recalcitrant to Agrobacterium mediated transformation. Of the Phaseolus species only the tepary bean (P. acutifolius) has been transformed through the use of this bacterium. Two methodologies have been used for achieving it: the inoculation of green nodular callus tissues (Dillen et al., 1997) or of meristems of mature seeds (Agrobacterium Mediated Mature Seed Meristem Transformation or AMMSM-Transformation; see Annual Reports 2000-2002). The AM:MSM-Transformation method does not require previous induction of callus tissue, and thus allows the screening of large populations of genotypes with less effort than that of Dillen et al. In the past 3 years we have screened more than 50 cultivated and wild genotypes of common and tepary bean for competence to AM:MSM-transformation with no success. Results from other series of transformation experiments, made with tepary bean genotypes and hybrids, suggested that nuclear genes of tepary bean accessions are involved in conferring competence to AM:MSM-transformation (see annual report of2001). In 2000 we started an interspecific crossing program using as parents a tepary bean genotype with competence to AMMSM-transformation and other common and tepary bean genotypes which show good response to in vitro culture and plant regeneration The objective of this crossing program is to breed AMMSM-transformable hybrid lines between these species (or any other Phaseolus species which can cqntribute to the transformation process) that can be used for the fast and easy transfer of transgenes to common bean cultivars by sexual crossing. As strategy for combining different traits involved in tissue culture response, plant regeneration or AMMSM-Transformation in one genotype, we are using the double congruity backcross (DCBC) methodology (Mejia - Jiménez et al. 2002). During 2003 we advanced in identifying DCBC hybrids, mainly with the cytoplasm of tepary bean, with improved competence to AM:MSM-Transforrnation. Methodology Mature seeds of advanced DCBC hybrids with common and tepary bean cytoplasm were used as explants in the transforrnation experiments. The DCBC hybrids corresponded to the generations: V-DCBC6A. V-DCBC6s. V-DCBC1s. V-DCBCsc. V-DCBCso, V- DCBCsE. with P. vulgaris cytoplasm and the generations, A-DCBCso. A-DCBC10A, A- DCBCJOs, A-DCBCHx:, A-DCBCt2B with P. acutifolius cytoplasm (see table DCBCHybrids: http://gene3.ciat.cgiar.org/blast/docs/DCBCHybrids.pdf). The binary Agrobacterium strains AGLl/pCambia1305-1 AGLl/pCambia3200 (http://www.cambia.org) or the strain C58Cl/ pTARC-Blb (Dillen et al. 1997) were used. The transforrnation methodology was described in the annual report of 200 l. Results Screening of hybrid populations for competence to AMMSM-Transfonnation. A total of 168 DCBC hybrid lines with cytoplasm of P. vulgaris (V-DCBC) and 197 lines with cytoplasm of P. acutifolius (A-DCBC; an average of 37 explants per line), were screened between October 2002 and September 2003 for AMMSM-transforrnation. Friable and meristematic calli (m-calli) were recovered after selection in 26 (7.1 %) of the screened lines. Initially only the recovered friable calli expressed the GUS-intron gene, which is driven by the p35S prometer. In severa! of the selected lines that have been screened for GUS expression the selected m- calli developed other cellular types such as friable calli (for example in severa! calli selected from the lines GKAX-11B(L)2 F4 and GKAX-13B F3 ). Due to this and to the fact that during selection conditions in vitro, we applied (by using a temporary immersion system) strong and uniforrn selective conditions, we believe that most of the tissues resisting selection are genetically transforrned and that the lack of GUS expression in the m-calli is caused by other factors (for example the inactivity of the 35S prometer in this tissue type). Since we can only regenerate plants from the m-calli and not from friable calli , we are measuring the forrnation of m-calli after transforrnation and selection to detect genotypes with transforrnation capabilities. Twenty (10.1 %) A-DCBC lines produced m-calli after selection, while only 6 (3.5%) of the V-DCBC lines produced this same tissue type. However the m-calli of all the V -DCBC lines and of several A-DCBC lines died within 3 months after selection. Thus not only the selection of m-calli seems to be important for the recovery of transgenic plants through the AMMSM-transformation methodology. It is necessary to be able to induce further growth and bud differentiation in the selected tissues. Only 6 (1.6%) of the lines produced m-calli which could be maintained for longer than 3 months and that could be induced to differentíate buds. Within these lines there are two with M-calli survival efficiencies of 21.7 and 35%. That is more than three fold the survival rate achieved with tepary bean hybrids (8.3%). Although from the m-calli of the V -DCBC lines no bud differentiation or plant regeneration could be achieved, the fact that they could be recovered after selection, is considered already an improvement over the results achieved routinely with comrnon bean cultivars and wild genotypes, which yield no m-callus after selection (Annual report 2001). Plants are being regenerated from the antibiotic resistant m-calli which differentiated buds (Table 2, last column) through micrografting. Table 1 Overall results of the AMMSM-transformation experiments of DCBC-hybrid lines with cytoplasm of P.acutifoüus and P. vulgaris, peñormed between October 2002 and September 2003 Cytoplasmic Number of Hybrid Average Lines % of Lines % ofLines Background of the Lines Screened Explants number of Producing Producing with Hybrids Explants per M-Calli M- Calli surviving M- Line calli P. vuls.aris 168 5018 29.8 6 3.5 o P. acutj]_olius 197 8809 44.7 19 9.6 3.0 TOTAL 365 13827 37.8 25 6.8 1.6 262 Tab1e2. DCBC bybrid lines wbich yielded meristematic (regenerab1e) calli after AMMSM-transrormation and efficiency or recovery and survival or antibiotic resistant meristematic calli. Hybrid Codel Cyto- Agrob. Leve! of Level of # of Geneúcin/ Recovery # M-calli % ofM-calli Generation1 plasm2 Strain1 Transicnt GUS M-Callus Hygromicin Efficicncy Surviving S urvi val and Expressioo• lnduction' Resistant of and Diffcren- M-Calli6 M-Calli Differen- úatioo9 %' úatin 1 NGPNMNG2-4 F, A +++ 47190 6 12.7 NGPNMNG2-2-I F. A + 18/41 5.5 NGPNMNG2-1-3F. A +++ 32/38 3.1 GKAX-1 8(8) F, A ++ 37/47 12 32.4 GKAX-18(L) F, A ++ 27132 6 22.2 3.1 GKAX-IAF A +++ 45/48 14 31.1 GKAX-28~q F, A +++ 30135 10 33.3 GKAX-2A(q F1 A +++ 19120 6 31.5 GKAX-38~L) F1 A + 21/33 6 28.5 GKAX-138 F, A +++ 60188 15 25 1.1 GKAX-IIAF1 A +++ 40140 12 30 GKAX-11 8~L)2 F. A 1 +++ 34134 15 44.1 12 35.2 XKA-8-1 F1 A 111111 +++ 28/46 9 19.5 XKA-8-2 F1 A 111111 ++ 29/32 1 3.1 ZXTGG-25-1 F, A + 25128 3 12 2 7.1 ZXX-13-2 F, A ++ 45/46 10 21.7 10 21.7 GKX-6-2A F, A 111111 +++ 13/13 3 23.1 GKX-6-3A F1 A 111111 +++ 28/41 2 7.1 GKX-II -2A F1 A 1111 + 37/43 4 9.3 8WG-20~ A 1 +++ 51158 5 8.7 4 6.9 AT7-2A F, V 1 +++ 33137 5 15.1 ATI-28 F1 V 111111 +++ 19/19 1 5.2 T7KT-90-40A F1 V 1 +++ 18/18 3 16.6 TZTB-12 ~ V 1111 ++ 32/37 12 37.5 TZTB-14 ~ V 1111 + 18123 4 22.2 TZTB-21 F2 V 1 28128 3.5 1.For understanding the pedigree of the lines see Table DCBC Hybrids http://gene3.ciat.cgiar.org/blast/inicio.htrn 2. Cytoplasm of the Hybrids; A = P. acutifolius; V = P. vulgaris 3. Agrobacterium strains: 1 = C58C 1 pT ARC; 11 = AGLI pC3200; lil = AGLI pC 1305-1 4. After three days of co-culture three explants were scored for GUS expression. A + or a - score was given for each of the explants expressing or not GUS in the cotyledonary node or apical meristem. 5. Number of explants forming M-calli/number of explants innoculated. 6. After 1 month (two subcultures) in media containing 50 mg/1 geneticin or 30 mg/1 hygromycin. 7. Number of M-calli selected x 100 1 Number of explants innoculated. 8. M-calli with differentiating buds which survived for more than two months after selection. 9. Selected m-calli surviving and differentiaring buds three months after selection. 2.1.2 Progress in the Anti-Sense Mediated Silencing of the Granule Bound Starch Synthase 1 (GBSS 1) for the Production of Waxy Cassava Starch Gina J. Puentes P., Edgar Barrera S., Martin Fregene., Paul Chavarriaga., Chikelu Mba CIAT Introduction Higher incomes from cassava in the developing world where the crop is generally found will require the industrialization of the crop and the development of novel industrial products from cassava. There are severa! novel products that can be produced from cassava, including modified starches such as 100% amylopectin or 100% amylose starches, from the down regulation of the granule-bound starch synthethase (GBSS) gene or the starch-branching enzyme (SBE) gene. Industrial application of either pure amylopectin or pure amylose starches, such as the production of high-value biodegradable polymers from pure amylose starches or the use of 100% amylopectin in thickeners, pastes and glues, have a market with unlimited growth potential. Biotechnology can play a very important role in the production of the abo ve products in cassava. With funds from the Colombian Ministry or Agriculture and Rural Development, a project was been initiated to genetically engineer industrial cassava varieties for the production of wax.y starch using an anti-sense and sense construct of the GBSSI gene. GBSS catalyses the conversion of ADP-glucose to amylose through the linkage of an ADP glucose to a preexisting glucan chain. Antisense disruption of the GBSSI gene has been employed to create potato transformants with 70-100% amylopectin via the down-regulation of the GBSSI gene (Salehuzzaman et al.,1993) and the disruption sense in sweet potato of the gene GBSS (Kimura et al. , 2001). Methodology Isolation of a cassava GBSS cDNA clone. More than 87,000 clones of a cassava root and leaf cDNA library cloned in the vector pCMV SPORT (GffiCO BRL Inc., USA) was gridded onto high-density filters (Mba et al., 2000 unpublished data). The library was screened using a potato GBSS cDNA clones, a gift from Dr.Christine Gerhardt (Max. Planck Institute, Cologne, Germany). The potato GBSS gene was labeled with e2P] dATP by random primer labeling and hybridized ovemight to the cDNA filters according to standard protocols for Southem hybridization used in cassava (Fregene et al., 1997). The filters were washed twice with 2X SSC +0.1% SDS at 60°C for 5 min, and autoradiography was al -80 °C using 2 intensifying screens. Construction of transformation cassettes. Primers were designed from published sequences of a full-length cassava cDNA of the GBSSI gene (Salehuzzaman et al., 1993). 264 BamiD and Xbal restriction enzyme recognition sites were incorporated in 5'end of the primers to enable sub-cloning of the cDNA clone in the sense and antisense orientation into the multiple -cloning site (MSC) of the vector pRTlOl. The primers were used to amplify the cDNA clone obtained above, and the PCR product was cleaned using the QIAGEN PCR Clean Up Kit (QIAGEN Inc., Los Angeles, CA) and digested with the appropriate enzymes. A 2.1Kb BamiD /Xbal fragment was subcloned in the sense and antisense between the 35S prometer and the 35S polyadenylated tenninator region of vector pRTlOl, a gift from Dr. Ryohei Terauchi, Iwate Biotechnology Research Center, Kitakami, Japan. The 35S prometer, GBSSI gene in pRTlOl was liberated using the restriction enzyme Pstl, separated on a agarose gel, eluted and cloned into the Psti site of the binary vector pCAMBIA 1305.2 having the GUSPlusR and HPT reporter genes. Transfonnation by Agrobacterium tumefaciens, of varieties. Friable Embryogenic Callus of the cassava genotype TMS60444, Mcol.2215 y CM 3306-4 was transfonned via Agrobacterium tumefaciens with the GBSSI gene in antisense-sense orientation, mediated technology CIA T. Results Three GBSS cDNA clones obtained form screening the cassava library were sequenced, and one was found to be a complete cDNA clone. The cDNA clone has the ATG start codon 81 bp down stream from the beginning of the cDNA sequence and a stop codon about 100 bp from the poly-A tail. PCR amplification with the designed primers yielded a fragment about 2.1 kb in size that corresponds to the full-length GBSSI cDNA clone (Figure 1). APstl 1 2 3 4 5 6 7 8 2100 bp A mnlific.ation Figure l. PCR amplification of the GBSSI cONA clone using primers to introduce restriction enzyme sites at the ends of the gene. Tbe first lane on the right is molecular weigbt marker Lambda DNA, digested with Pstl, the next three lanes are PCR of the gene GBSSI antisense and next are PCR of tbe gene GBSSI sense. The resulting PCR fragment, digested with BarniD and Xbal restriction enzyme, was cloned into the MCS of pRTlOl. Next, the GBSSI gene, promoter and terminator sequences, excised with Pstl and the resulting fragments separated from the vector fragment (sizes 2.8 and 2.7 kb) by electrophoresis was cloned into the Pstl site of pCAMBIA (Figure 2). These are the constructs that were used in the Agrobacterium- mediated transfonnation. Two genetic constructions (Figure 3) with the GBSSI gene in anti-sense and sense orientation in the vector pCAMBIA 1305.2 were made to achieve silencing of the gene. Antisense band of approx 2800bp clonad in the vector pCAMBIA 1305.2 J...Pstl 1 2 Sense band of approx 2800bp clonad in the vector pCAMBIA 1305.2 Band of approx 2700bp corresponding to the vector pRT1 01 Figure 2. Pstl digested of pRTlOl plasmid containing the cassava GBSSI gene in anti-sense-sense orientation. The fragments of about 2.8 and 2.7 Kb in size represent the GBSSI gene, tlanked by the 355 promoter and the polyadenylation terminator sequence, and the rest of the pRTlOl plasmid respectively. 266 (A) (B) HPT Pstl · ~ Pstl ..-2100pb--.. Pstl 2800 pb _ __,.,._,: ...... 2100pb-. Pstl Poly-A gen GB~I señal Antisentido CaM( '-<1111411..-- 2800 pb ----~.-.-: Figure 3. Gene constructs of GBSS in sense (A) and anti-sense (B) in the binary vector pCAMBIA 1305.2 The constructs were transforrned into friable embryogenic Callus (FEC) of the model transforrnation genotype MNGll via Agrobacterium tumefaciens. Results of GUS transitory assay revealed a successful incorporation of the gene (Fig. 4). ~ "'· ~ ~ ·~~ !.~ ···· :a ~ ./ f' ¿ .1=\:;, ;(\ 't ~1 ~¡,}/ V - '-· D Figure 4. Positive test of GUS in Cotyledonary embryos of variety TMS 60444 with the GBSSI sense gene constrauct in the vector pCAMBIA 1305.2 Conclusion We have successfully transformed full-length sense and anti-sense constructs of the OBSSI gene into the model cassava transformation variety :MNO 11. Transformed calli are being regenerated following which molecular and biochemical tests will be conducted to test stable expression of the gene and the eventually the waxy phenotype. The project was carried out as an undergraduate project for the Colombian undergraduate student project, Oina Jazbleidi Puentes P. of the Universidad Nacional de Colombia -Sede Palmira. References CIAT (Centro Internacional de Agricultura Tropical). 2003. Annual Report Project SB2, Assessing and Utilizing Agrobiodiversity through Biotecnology. Cali, CO. p.202-205 Fregene, M.; Angel F.; Gomez, R.; Rodriguez, F.; Chavarriaga, P,; Roca, W.; Tohme, J., Bonierbale, M. 1997. A molecular genetics map of cassava (Manihot esculenta Crantz) Theor Appl Genet 95:431-441. Kimura, T., M. Ontani., T. Noda., O. Ideta., T . Shimada & A. Saito. 2001. Absence of amylose in sweet potato (lpomea batatas (L) Lam) following the introduction of granule-bound starch synthase 1 cDNA Plant cell report 20: 663-666. Salehuzzaman, S .N.I.M.; Jacobsen, E.; Visser, R.G.F. 1993. Isolation and characterization of a cDNA- encoding granule-bound starch synthase in cassava (Manihot esculenta Crantz) and its antisense expresion in potato. Plant Mol Biol23:947-962. 2.1.3 Interspeciflc hybridization of common and tepary bean for introgression of competence to Agrobacterium mediated transformation and insect resistance traits Alvaro Mejía Jiménez1, Leonardo Galindo\ Arturo Criollo', Jose Flower Valor2, Juan Miguel Bueno2, César Cardona2 and Joe Tohme1 1SB-2 Project- CIAT~ 2IP-l Project Introduction The tepary bean (Phaseolus acutifolius A. gray) possesses severa! traits that are important to common bean breeding, such as resistance to drought (accessions 040020 and 040023), bruchids (A. obtectus and Z. subfaciatus; 040199), leaf hopper (Empoasca kraemeri; 040019 and 040036) and bacteria! blight (Xanthomonas campestris pv phaseoli; 040020), among others. Also one genotype of tepary bean has been dentified as competent to Agrobacterium mediated transformation (the wild accession Nl576; see report on bean transformation). 268 Crossing of common with tepary bean has been traditionally difficult. Using facilitator genotypes of both species, and applying recurrent and congruity backcrossing (CBC; Haghighi and Ascher, 1988), we could produce a large number of fertile hybrids (Mejía- Jiménez et al. 1994), from which common bean lines with high levels of resistance to bacteria! blight were developed (Singh and Muñoz, 1999). However using these two backcross strategies, no fertile progeny could be obtained from hybrids involving severa! of the abo ve mentioned tepary bean genotypes. Using a modification of the CBC methodology, we called Double Congruity Backcrosses (DCBC), we could develop fertile interspecific hybrids involving several tepary bean accessions that were formerly impossible to cross (Annua1 reports 2001, and 2002). In 2002 the project IP-1 reported the identification of drought tolerant lines after screening of DCBC hybrids (Beebe et al. 2002). During 2003 our backcross program through DCBC continued, and from the fertile progenies developed during this or previous years, Iines with improved competence to Agrobacterium mediated transformation or resistance to Empoasca or bruchids were identified. Methodology Double congruity backcross hybridization on the cytoplasms of common or tepary bean (V-DCBC and A-DCBC hybrids) were started in 1999 using advanced congruity backcross hybrids, developed in the past, with cytoplasm of common bean (Mejía-Jiménez et al., 1994), as described before (Mejía Jiménez et al., 2000; annual report project SB-ll). This backcross strategy consists of backcrossing altemately hybrids developed in a P. acutifolius or P. vulgaris cytoplasmic background, with the most advanced DCBC hybrids generated in the other backgrounds and viceversa. This is repeated severa! times to increase recombination between both genomes and pyrarnid desired traits in the hybrids. Initially only P. vulgaris and P. acutifolius genotypes were included in the DCBC, but later genotypes of different Phaseolus species were also incorporated (see table DCBC hybrids) Embryo rescue was applied when necessary to recover viable hybrids from aborting embryos (Mejía-Jiménez et al., 1994). Results Identification of DCBC-hybrid Iines with improved response to in vitro culture and plant regeneration or Agrobacterium mediated genetic transformation 'The inter-specific hybridization between common and tepary bean genotypes is being used for the development of hybrids with improved competence to AMMSM- transformation (Agrobacterium Mediated Mature Seed Meristem-transformation; see report on genetic transformation of Phaseolus beans). Three hundred and sixty five DCBC hybrid lines were screened during 2003. The production of antibiotic resistant calli of a singular morphological type, from which plants can be regenerated (meristematic calli or m-calli), was used as a parameter for selecting genotypes with superior performance. Six lines with the cytoplasm of tepary bean yielded antibiotic resistant m-calli which are potentially transgenic. In two of the lines (OKAX-llB(L) 37.2%, and ZXX-13-2 21.7%; see also Table 2 of the transformation report) the recovery efficiency of antibiotic resistant m-callus was higher than in tepary bean intraspecific hybrids 040065 x NI576 (8.3%; Annual report 2001). This indicates that probably alleles involved in the response to the AMMSM?- transformation methodology are being accumulated through the DCBC process which combines the screening for superior genotypes, in both cytoplasmic backgrounds, with the further backcrossing of the selected genotypes found. Three lines with meristematid calli survival and differentiation efficiencies ranging from 7 to 35%, and two V -DCBC lines which yielded antibiotic resistant m-callus after selection, were chosen as parentals for the next generation of DCBC hybrids (see below). It will be tested if the lines, which show the highest percentages of m-calli survival and differentiation (Table 2 of the tansformation report), can be used as bridge for the production of transgenic common bean cultivars. Introgression of Empoasca resistance to common bean through DCBC After a field screening performed by the IP-1 project, at least seven common bean lines derived from DCBC crosses were identified as Empoasca resistant (lines DCBC7A. -7C and _ 80; see table DCBC hybrids: http://gene3.ciat.cgiar.org/blast/inicio.htm). All these hybrids include in its pedigree the genotype of P. acutifolius 040036 and one of them the genotype 040019, which are resistant to Empoasca (table 1). These accessions of tepary bean may be the source of the detected resistance. In the past 10 years several common x tepary hybrid lines produced using a different backcross strategy, like recurrent or congruity backcrosses, were evaluated in the field for Empoasca resistance with no success in identifying resistant genotypes. This is an additional proof of the usefulness of the DCBC-strategy to allow introgression of traits from alien germplasm (last year the IP-1 project reported the identification of drought tolerant lines derived also from DCBC hybrids). The resistant common bean lines are being crossed to other Empoasca resistant A-DCBC hybrid lines (table 2), to pyramid different, possibly available, resistance alleles and increase the levels of resistance airead y found. Resistance to the bean weevil (A. obtectus) found in A-DCBC hybrids. Several populations of A-DCBC and V -DCBC hybrids ha ve been screened by the IP-1 project for resistance to the bruchid A. obtectus. Five lines with tepary bean cytoplasm have shown to be resistant (results were confirmed through three individual infestations with the insect; (hybrids A- DCBC with codes ONV AV-2 F3, OVV-1 F3, OKX-6 F2, OKA-12 Fz, ZXX-5 F2). All the 270 resistant seeds were planted in the greenhouse and the plants obtained are being used to generate other DCBC hybrids with common and tepary bean cytoplasm (table 2). It is not clear which is the origin of the resistance allele(s) in the hybrids. Development of new fertile DCBC hybrid progenies of P. vulgaris or P. acutifolius cytoplasms, with parental hybrid lines selected for its response to AMMSM- transformatíon or for resistance to ínsect pests. During 2003 new DCBCs were started, or completed, using as parental hybrid lines selected for its good response to in vitro culture, callus induction or competence to Agrobacterium transformation (V-DCBC hybrids with codes TZTB-12 and AT7-2A; and A-DCBC hybrids with codes GKAX-11B(L)2, ZXX-13-2 and BWG-20), resístance to Empoasca kraemeri (V-DCBC hybrids with codes A36Y-42, EMPZ-2, -8, -9, TZTE-20B and -71B) orA. obtectus (lines with code GKA-12, GKX-6, GVV-1, GNV AV-2). These lines were or are being crossed to the most advanced DCBC hybrid lines available of the same cytoplasm. Morphological traits are being used to verify the hybridization (Table 2) The generated fertile hybrids will further be screened for improved expression of the desired traits. Table: l. Pedigree of the interspecific DCBC hybrids with P. vulgaris cytoplasm, which have been found to be resistant to Empoasca in field experiments (see also IP-1 Annual Report). Highlighted in yellow are the resistant parental lines, and also tbe A-DCBC bybrid G36NGP-3, derived from the cross G40036 x NGPNMNG, which may be the principal source of the resistance found in the hybrids. To understand the pedigrees of the hybrids, see also the table DCBC hybrids: http://gene3.ciat.cgiar.org/blast/inicio.htm). 1'-ode of the Hybrid Pedigree ~6Y-42 T6 KBNKN-1 X NGGNV-2 ¡EMPZ-2 and EMPZ-3 A99Y-90 KBNKNI X G99NGPZ IEMPZ-5 A36Y-42 T6X G36NGP3 ¡EMPZ-8, EMPZ-9 Al9Y-103 KBNKN3 X Gl9NGP-3 G40019 x NGPNMNG ¡rz;TE-208 TZT-3 TZT-4 X G36NGP-3 G40036 x NGPNMNG xZXTGS21-9 ZXTG6 X G36NGP·3 X ZXTGS21 -11 ZXTG 8 X G36NGP-3 xZXTGS49-8 ZXTGS X G36NGP-3 X EMPZ-3 A99Y-90 xEMPZ-3 A99Y-90 ZXTGS21-9 ZXTG-6 X G36NGP-3 ZXTGS21-9 ZXTG-6 X G36NGP-3 Table 2. New DCBC hybrid progenies developed during 2003. DCBC= Double congruity backcross hybrids (for complete pedigree see table DCBC hybrids:http://gene3.ciat.cgiar .orglblastf'I.DÍcio.htm); V- DCBC= DCBC with the cytoplasm of common bean (common bean as female parent); A-DCBC= DCBC with the cytoplasm of tepary bean. Highlighted in yellow are the lines selected from Agrobacterium mediated transformation experiments, in green are the lines selected as Empoasca resistant and in red are the lines resistant to A. obtectus. Cytoplasm of P. vulgaris Female Parent/ Hybrid/ Code Code V-DCBC7c X TZ .L V-DCBCac TZT V-DCBCac X TZT .L V-DCBC80 TZTA, ~. TZTK v-¿~-8o X !f.i.&¡H , , • TZTA, .L TZTE, TZT V-DCBC1E TZTTZ V-DCBCac X TZTA, TZTE. TZTU .L V-DCB4A TZTEB.TZTAB V-DCB4 X TZTEB, TZTAB J, V-DCBC10 R OF P. acutifolius A-DCBCIOA X ZXTG J, A-DCBC10c ZXTGS A-DCBC10 X ZXTG J, A-DCBC11 ZXTGT ZXTGU A-DCBC11 X ZXTGU, ZXTGT .!. A-DCBC12A BW A-DCBC12A X BW .!. A·DCBC12B BWZ A-DCBC12A X BW ¡ A· DCBCIJA BWT A-DCBCIJA X BWT ¡ A-DCBC14A BWTZB [Verificai ton ofthe hybrid Male Parent/ Code V-DCBC6 Female parent self-sterile TIK. TK V-DCBC6A ACT, AT7, ~ UAC, Female parent self-sterile UQUQ TZTU, V -DCBC7A. -8A. -8o TZT, TZTU, TZTE. A-DCBC12A,.;_L2B sw.~Jm.m V-DCBCac. .8E TZT, TZTTZ V-DCBC6.8B T7K, TK, UAC A-DCBCIOA.IOB ZXTG, Z99ZX, GKXZG 272 TZTA, jHypocotyl pigmentation and flower color jHypocotyl pigmentation and flower color; ~ybrids became self-sterile IFemale. parent self-sterile pigmentation and flower color from the male parent; primary Jeafs the large petiole from P. vulgaris; hybrids self-sterile pigrnentation and flower color from the male parent; primary leafs the large petiole from P. vulgaris in the became self-sterile Conclusions A-DCBC lines have been identified which show an increased cornpetence to Agrobacterium rnediated transformation. These lines rnay have different levels of introgression of cornrnon bean and rnay be used to transfer transgenes to this species through sexual crossing. DCBCs allowed the introgression, to cornrnon bean, of tepary bean genes involved in Empoasca resistance. Also, lines with tepary bean cytoplasrn resistant to the bruchid A. obtectus were found arnong the progeny of the screened DCBC hybrids. Future plans • To continue with the DCBC strategy to produce more advanced DCBC hybrid generations • To perform additional DCBCs with hybrids that have already shown cornpetence for Agrobacterium rnediated transformation, Empoasca, orA. obtectus resistance. • To study the introgression in the hybrid populations of DNA fragments frorn the species involved in the DCBCs using AFLP techniques. References Haghighi, KR. and PD. Ascher. Fertile, intermediate hybrids between Phaseolus vulgaris and P. acutifolius from congruity backcrossing. Sexual Plant Reproduction 1 (1988): 51-58. Mejía-Jiménez, A., Galindo, L.F., Hassa, A., Jacobsen, H.J. and Roca, W.M. Genetic transformation of Phaseolus beans. Annual report 2000. Biotechnology Research Unit-CIAT Mejía-Jiménez, A., Muñoz, C. Jacobsen, H.J., Roca, W.M. and Singh, S.P. 1994. Interspecific hybridization between common bean and tepary bean: Increased hybrid embryo growth, fertility, and efficiency of hybridization through recurrent and congruity backcrossing. Theor. Appl. Genet. 88: 324-331. Singh, Shree P. and Carlos G. Muñoz. Resistance to Common Bacteria) Blight among Phaseolus Species and Comrnon Bean Improvement. Crop Science 39 (1999): 80-89. Recovery of transformed M-callus after transformation with binary Agrobacterium strains. The AMMSM-transformation methodology was developed using the hypervirulent Agrobacterium-strain LB A4404pTOK 233 that was developed through cointegration (Hiei et al. 1994). An important achievement of 2003 is the recovery of transgenic (GUS expressing) regenerable M-callus after transformation with the binary Agrobacterium strains AGLl and C58Cl (Table 2). Genes of agronomic interest can be cloned faster and easier in binary plasmids than in the cointegrate ones. Co-transformation with unlinked transgenes. Co-transformation, the transformation of plant cells with unlinked transgenes (which can be localized in the same or different Agrobacterium strains), followed by the separation of the transgenes in the progeny of the transformed plants, is a methodology which has been used for the production of antibiotic- and marker-free transgenic plants. With sorne of the DCBC lines we have been performing co-transformation experiments by transforming with the GUS-Intron gene and the BAR gene in different plasmids and Agrobacterium strains (see Table 2). The fact that GUS expressing calli was obtained after selection with the BAR gene in these experiments, indicates that both transgenes were co-transformed. It should be tested if they segregate independently in the progeny of the transgenic plants recovered from this kind of experiments. Development of a methodology for large scale sterilization of mature Phaseolus seeds with chlorine vapors. The application of the AMMSM-Transformation methodology requires the preparation of a large number of explants from mature seeds (embryos without one cotyledon). This is a time-consuming and a contamination-prone process, because the embryos have to be excised from the pods and seed coats under sterile conditions. During 2003, and based on a methodology developed for sterilizing seeds of Arabidopsis thaliana (Clough and Bent; http://plantpath.wisc.edu/-afb/vapster.html ) we designed an apparatus and developed a methodology for a simple and efficient sterilization of high quantities of seeds of Phaseolus beans. The methodology uses 5% sodium hypochlorite and 4 N HCL to generate Ch vapors, which in 3 hours completely sterilizes more than 90% of the highly contaminated Phaeolus seeds samples (seeds mixed with soil). Development of new populations of DCBC hybrids involving as parental genotypes (hybrids) with increased competence to AMMSM-Transformation. Three A-DCBC and two V -DCBC hybrid lines that yielded antibiotic resistant meristematic calli after transformation were chosen for developing new hybrid lines (highlighted in yellow, table 2). The selected lines were airead y crossed to the most advanced DCBC lines of the same cytoplasm, and fertile hybrids have been developed. The screening of the new DCBC- Hybrid populations for AMMSM-Transformation is in progress. 274 Conclusion As judged by the recovery of GUS expressing and antibiotic resistant regenerable calli after transformation, it is likely that competence to Agrobacterium mediated transformation has been transferred to various advanced P. acutifolius x P. vulgaris - DCBC hybrids with P. acutifolius cytoplasm. Sorne of these show rather high transformation efficiendes. These hybrids could be used as bridge to transfer transgenes to common bean cultivars through sexual crossing. However, to reduce the number of backcrosses that may be necessary for achieving this, transformable hybrids with common bean cytoplasm should be developed. Future Plans • Develop pure lines with transformation competence, from the already identified AMMSM-transformations competent F3 or F4 populations • Further screen the hybrid progenies developed this year for competence to AMMSM-transformation. • Continue with the co-transforrnation experiments (with marker genes and genes with agronomic importance in different strains or plasmids) in arder to develop a methodology to produce transgenic common bean plants free of antibiotic markers. • Start co-transfonnation experiments with genes of agronornic importance. References Dillen W, De Clercq J, Goossens A, Van Montagu y M, Angenon G. Agrobacterium-mediated transformation of Phaseolus acutifolius A. Gray. Theoretical and Applied Genetics 1997; 94 (2): 151-158. Hiei Y., Ohta S ., Komari T. and Kumashiro T. Efficient transformation of rice (Oryza satíva L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T -DNA. The Plant Joumal 1994; 6(2):271-282. Mejia-Jimenez, A., H. J. Jacobsen and W. M. Roca (1998). Development of an in vitro regeneration system ín common bean suítable for genetic transformation. EUCARPIA International symposium on breeding of protein and oil crops, Pontevedra - Spain 1-4 April. Mejía-Jiménez, A., Muñoz, C. Jacobsen, H.J., Roca, W.M. and Singh, S.P. 1994. Interspecific hybridization between common bean and tepary bean: Increased hybrid embryo growth, fertility, and efficiency of hybridization through recurrent and congruity backcrossing. Theor. Appl. Genet. 88: 324-331. 2.1.4 Bioassays with transgenic Bt-cassava plants, cultivars 60444 and CM3306-4, to test for stem borer and horn worm resistance 1D López, 2CJ Herrera, 1P Chavarriaga, 1J Tohme 1SB-2 Project; CIAT; 2IP-3 Project, CIAT Introduction One altemative method to obtain cassava cultivars resistant to insects is through transgenesis. We developed transgenic cassava plants, from three cultivars, that carry one gene for insect resistance. These plants were produced to test the efficacy of this crylAb gene (Bt gene) to control insect pests like the cassava stem borer (Chilomima clarkei, Lepidoptera) and the cassava hom worm (Erinnyis ello, Lepidoptera). We present a summary of the main findings after carrying out bioassays with transgenic plants and these insects. Materials and Methods Transgenic plants of cassava cultivars 60444 (MNigll) and CM3306-4 were obtained vía Agrobacterium-mediated transformation, for the former, and the particle gun for the latter. Bioassays for the stem borer ( Chilomima clarkei) and the cassava homworm (Erinnyis ello) were set in biosafety greenhouses, or in captivity rooms of the IPM program. We infested at least six young cassava plants per transgenic line with Chilomima larvae of second or third instars, one larvae per plant, and let them penetrate the stems and grow for 15 days inside it. All transgenic lines were derived from cultivar 60444. Controls included non transgenic plants of 60444 and CMC-40, a cultivar used to feed insects in captivity. Larvae were weighted at the beginning and end of the test, and the average weight increase was estimated and statistically analyzed. To test resistance to cassava hom worm, we fed 20 first instar larvae with transgenic leaves of lines derived from cultivars 60444 and CM3306-4. As controls we used non-transgenic leaves from cultivar 60444. Unfortunately there were not leaves available from non transgenic CM3306-4 that could serve as proper control. We opted for CMC40, a cultivar routinely used to feed the hom worm in captivity. Therefore, the results presented for CM3306-4 are preliminary and need confirmation. We measured larvae weight increase for 11 days and analyzed differences between transgenic lines and their non transgenic controls. 276 Results The main findings for the stem borer bioassays can be summarized as follows: No differences in average weight increase were detected between transgenic and non transgenic lines derived from 60444. In fact, larvae fed on non transgenic 60444 lines also lost weigh, if compared with those fed on control CMC-40 (Table 1; Figure 1). There were no statistically significant differences between the two transgenic lines 55 and 92 and the non transgenic control of same cultivar. Developmental stage of larvae seemed critical in detennining the effect of transgenic plants on their diet. Plants seem to express very little insecticida} CRY1Ab protein (Ladino et al 2002), so larvae on second and third instar are already too strong to be affected by this amount protein expressed in transgenic plants Sorne of the larvae fed on transgenic line 80 increased their weigh as much as did those of controls fed on CMC40, suggesting that this line may just represent an escape or a chimera. The bioassays with the steam borer, although limited by the number and developmental stage of available larvae (the insect is not found in the Valle del Cauca), indicated that larvae loose weight when fed on leaves from transgenic and non transgenic plants of cultivar 60444. If there was an effect of the low expressing cry1Ab gene, it will be surely masked by the varietal effect. Feeding on non transgenic control from cultivar CMC-40 produced much healthier larvae, with higher average weigh increase. Figure l. Chilomima clarkei larvae fed for 15 days on transgenic {LSS, L80, L92) and non transgenic plants (Control) of cassava cultivar 60444. Larvae fed on a non transgenic control of cultivar CMC-40 (CMC) is also presented. Table l. Average weigh increase of two weeks old Chilomúna larvae fed for 15 days on transgenic and non transgenic cassava plants. Lines labeled with the same letter on the first column do not sbow statistically significant differences. No difference could be detected between transgenic lines 55 and 92 and their respective non transgenic control. Qualification Average weight increase No of observations Line A 0.11280 10 CMC (control) A B 0.08350 10 L80 B 0.04000 8 1..55 B 0.03425 8 TMS (control) B 0.03183 6 L92 There was a significant difference between larvae fed with non transgenic CMC40 and non transgenic 60444. L55, L80, L92 and TMS (control) are derived from cultivar 60444. CMC (control) is derived from CMC-40 (= 0,05) Por the cassava hom worm bioassays, we evaluated the average daily weight increase of 20 first instar larvae during 11 days. The results are pictured in Figure 2. Up to the 8th day larvae fed with non transgenic CMC40 increased weight beyond 3 g, while those fed with the other lines did not exceed 0,5 g. After day eight, larvae fed on CMC40 entered pre- pupa stage and lost weight, so weighting stopped at day 10. Comparing the behavior of hom worms feeding on CMC40 and lea Negrita, at day 11 we eounted 14 pupa alive and 6 dead larvae fed with CMC40 (30% mortality), while there were no pupa and 18 1arvae died (90% mortality) when fed on transgenic lea Negrita. From the latter's, only three larvae survived to day 11 reaching less than 2 g final weight (Figure 3). The assays indicated that there was a trend of larvae to develop slower (smaller daily weight increments) and die if they fed on transgenic plants of L55 and lea Negrita, or non transgenic 60444., while they developed stronger and faster until the majority reaehed the state of pupa if fed with no-transgenie CMC40. As stated before, results obtained with the hom worm are preliminary and will be repeated to eonfirm observations. This time proper lea Negrita non transgenie eontrols will be tested to confirm if transgenic lines effeetively proteet from hom worm attaeks. 278 Average daily weight increase 3.5 1\ 1 1 \ 3 ,/ \ 1 1 1/ 1 / \ -- CMC40 2.5 1 / \ 1 ·· · ·lea Negrita 1 1 - · · Control 60444 ./ 1 \ -Unea 55 1 1 1 . 2 Cl 1.5 / \... . --· // .. .. ./ . . . ~~· ··· ··--· -· ........ -- -..... . ...... . - 0.5 o 1 2 3 4 5 6 7 8 9 10 11 Da y Figure 2. Average daily weigbt increase of cassava boro worm larvae feeding on transgenic and non transgenic fines carrying a crylAb gene. (CMC40 and Control 60444 are non transgenic; lea Negrita and LSS are transgenic). Figure 3. Developmental stages reached by cassava boro worms fed on transgenic and non transgenic lines. The tbree larvae on the left are tbe only survivors of 20 larvae fed with leaves of transgenic lea Negrita pJants. The four on the right are examples of the 15 out of 20 larvae, cbanged to pupas, that survived when fed with non transgenic line CMC40. Conclusion and Future Activities Transgenic and non transgenic plants derived from cultivar 60444 seem to delay, or even prevent, the normal development of larvae from the stem borer and the horn worm. The effect seemed stronger on the latter where mortality reached 90%. This may be due to higher concentration of the "insecticida! molecule" in the leaves than in stems. It is very unlikely that this effect comes from the crylAb transgene, at least in L55 where gene expression seems low. This finding opens the door to explore new insect resistance sources in cassava germplasm that contains genes from wild relatives. This is the case of MNigll (60444), a variety developed in Africa from crosses with Manihot glaziovii to obtain resistance to CMD. Transgenic lea Negrita also seemed to delay the development and kili the horn worm. However, these results need replícation to confirm the observations since the proper control plants were not available in the first assays. A second set of bioassays will be performed again towards the end of 2003. lf transgenic cassava lines that protect against the stem borer an 80% of relative humidity and 30°C) to simulate symptom development. Plants were evaluated 15 days and 25 days after inoculation according to IRRI evaluation scale and modified by the CIAT Rice Project (F. Correa, personal communication, Project IP4). The non-transgenic Cica 8 and Palmar varieties were used as susceptible control, and breeding line CT14534- 12M-3-4M, derived from the inter-specific cross Oryzica 3/0. rufipogon was used as interrnediate resistance control. In addition to the PAP-transgenic rice lines, transgenic lines A3-49-60-12-3-1-31, and derived crosses A3-49-60-12-3/Cica 8-2 and A3-49-60-4- 5/FB007-19 containing the N-protein gene and with hypersensitive reaction as the mechanism underlined viral resistance were also evaluated. Five plants of each line were analyzed for the presence of the transgene by PCR. RHBV resistance assays. lndependent transgenic events containing the PAPy123 gene were evaluated for resistance to RHBV in the greenhouse. At least, 10 T2 progeny per each of 22 transgenic events were inoculated with RHBV at 10 days or 15 days after planting using 4 viruliferous insects per plant. Disease evaluations were conducted using scale from Oto 9, were O refers tono d.isease symptoms and 9 ind.icates more than 90% Ieaf area affected by the RHBV d.isease. Results and Díscussion. From 83% to 100% of plants analyzed showed integration of the PAPy123 gene accord.ing to Southem blot analysis, yield.ing a total of 70 transgenic plants obtained from 371 callus originally Agro-infected (Table 1). The transformation efficiency varied from 11% for Fedearroz 2000 to 54% for Cimarron respectively (Table 1). The molecular analyses showed the integration of single and multiple copies of the PAPy123. All the Fedearroz 2000 plants showed single and non-rearranged gene copies (Figure 1). Most plants from Cimarron and CT11275 also showed simple integration pattems. But Nipponbare showed multiple copies, sorne of them of higher or lower molecular weight than the expected size ind.icating rearrangements in sorne of these copies, includ.ing possible gene fragmentation and deletion (Figure 1). These plants will be evaluated for agronomic traits, and selected lines will be advanced to T2-T3 generations and plants with stable inheritance and expression of P AP gene will be screened for sheath blight resistance. Transgenic plants of Cica 8 and Palmar were evaluated for resistance to Rhizoctonia for two consecutive generations (T2 and T3) using intermediate and hyper virulent strains (Table 2). About 50 T2 lines were identified showing intermed.iate level of resistance, sister plants of these lines were also evaluated for RHBV, and T3 self cross-derived progeny plants from the resistant ones were evaluated for sheath blight in the following year. T2 plants derived from transgenic Cica 8-446 8-1, Cica 8-446-14-6, Palmar-446 4-1, Palmar-446 23-1, and Palmar-446-39-4 showed intermediate resistance to Rhizoctonia. This resistance was inherited ínto T3 plants from the Cica 8-446-14-6-18 and Palmar-446- 39-4-7 lines showíng intermediate resistance to both strains of fungi (Table 2). These plants showed a significant reduction in d.isease reaction respect to non-transgenic control. In add.itíon, line Palmar-446-39-4 showed resistance to RHBV and the derived T3 line Palmar-446-39-4-7 showed a reduced leaf area affected by the pathogen respect to the cross Oryzica 3/0. rufipogon when inoculated with the hyper virulent strain 2399-1 (Table 2, and Figure 2A). Southem blot analysis of genomic DNA ind.icated that 100% of these plants contained simple integration pattem of non-rearranged copy of the PAPy123 gene (Figure 2B). These plants showed a fertílity of 93-95%. P AP protein seems to actívate the protein expression of host genes associated with hypersensitive response (HR) and defense-related signal transduction pathway in the absence of pathogen infection and elevated Ievels of salicylic acid (Zoubenko et al., 1997; Wang et al., 1998; Smimov et al., 1997). Add.itionally, PAP inhibits local Iesion formation to a number of different viruses, including both DNA and RNA viruses (Chen et al., 1993), conferring a resístant mechanism against both mechanically and aphid-transmitted virus. For this reason besides fungal resistance, it is interesting to evaluate the activity of the PAPy123 gene against RHBV and other rice viruses. Because PAP also confers resistance to a bread spectrum of fungal pathogen, it will be interesting to evaluate these transgenic plants for resistance to other pathogens as well. The transgenic Cica 8 line A3-49-60-4- 5/FB007-19 containing the RHBV-N gene for RHBV resistance, showed highest level of 290 tolerance to Rhizoctonia. The RHBV-N gene appears to confer resistance to this virus by RNA mediated cross protection where hypersensitive reaction mechanism is involved (Lentini et al., 2002). It could be possible that RNA transcripts from the RHBV-N gene could function as elicitors for the plant defense system before or during pathogen infection. This hypothesis requires more systematic research in order to determine the reproducibility of this phenomena and elucidate the resistance mechanism in vol ve. Future Plans • To study the mechanism of action of the gene PAPyl23 in resistant transgenic Cica 8 and Palmar • To evaluate the possible effect of the gene PAPy123 on Tagosodes orizicolus, the planthopper vector of RHBV • To explain why sorne lines are resistant only to one strain of the fungus • To study protein expression and gene integration in the resistant lines • To evaluate the tolerance to Rice Stripe Necrosis Virus under greenhouse condition • To multiply seed (T3) resistant to R. solani and evaluate resistance for Rhizoctonia, and other pathogens such as Sarocladium, and Helminthosporium under field conditions Reference Aron, G and Irvin, J. 1980. Inhibition of herpes simplex virus multiplication by the pokeweed antiviral protein. Antirnicrobial Agents Chemotherapy 17: 1032-1033. Chen, Z., White, R., Antoniw, J., Lin, Q. 1991. Effect of pokeweed antiviral protein (PAP) on the infection of plant viruses. Plant Pathology 40: 612-620. Correa, F., Prado, G., Aricapa, G y Escobar, F. 2000. Caracterización de la resistencia genética a Rhizoctonia solani (Añublo de la vaina). Desarrollo de métodos de evaluación e inoculación. Annual Report. Project IP-4 Lentini Z., Lozano I, Tabares E., Fory L., Domínguez J., Cuervo M .• Calvert L. 2002. Expression and inheritance of hypersensitive resistance to rice hoja blanca virus mediated by the viral nucleocapsid protein gene in transgenic rice. Theoretical and Applied Genetics 106: 1018-1026. Published online: 14 December 2002. McCouch SR, Kochert G, Yu H, Wan Z, Khush G, Coffman W and Tanksley. 1988. Molecular mapping of rice chromosomes. Theoretical and Applied Genetics 76: 815-829. Martinez, C.P., Borrero, J., Almeida, M., Duque, M.C. F, Correa. Delgado, D., Aricapa, G., Prado, G. Silva, J. and Tohme, J. 2002. Utilization of new alleles from wild rice species to improve cultivated rice in Latin America. Annual Report. Project IP-4. Sambrook, J. Fritisch, E. and Maniatis, T . 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor. NY: Cold Spring Harbor Laboratory Press. Smimov, S., Shulaev, V. and Turner, N. 1997. Expression of pokeweed antiviral protein in transgenic plants induces virus resistance in grafted wild-type plants independently of salicylic acid accumulation and pathogenesis-related protein synthesis Tabares, E.; L. Fory; M.A.Santana; T. Agrono; G. Delgado; C.Ordoñez; N. Tumer; C. Dorado; M. Duque; J. Silva; Z. Lentini. 2002. Foreign genes as novel sources of resistance for rice fungal disease. In: Annual Report SB-2 p: 261-265 Wang, M.B., Upadhyaya, N.M., Bretell, R.I.S. and Waterhouse, P.M. 1997. Intron mediated improvement of a selectable marker gene for plant transformation using Agrobacterium tumefactions. J. Genetics and Breed 51: 325-334. Wang, P., Zoubenko, Tumer, N. 1998. Reduced toxicity and broad-spectrum resistance to viral and fungal infection in transgenic plants expressing pokeweed antiviral protein II. Plant Molecular Biology. 38:957-964 Ussery, M., Irvin, J. and Hardesty, B. 1977. Inhibition of polivirus replication by a plant antiviral peptide. Annals of the New York Academy Science. 284:431-440. Zarling, J. Moran, P. Haffar, 0 ., Sias, J ., Rchman, K., Spina, C., Myers, D., Kuebelbeck, V., Ledbetter, J. and Uckun, F.l990. Nature, 347: 92-95. Zoubenko, 0 ., Uckum F., Hur, Y., Chet, l. Tumer, N. 1997. Plant resistance to fungal infection induced by nontoxic pokeweed antiviral protein mutants. Nature/Biotechnology 15: 992-996. Table l. Transformation efficiency of three Indica varieties and one ]aponica variety using Agrobacterium tumefaciens (NT446) Plants Plants Genotype lR Callus Plants 2RP Analyzed by 3plants S+ (%) Southern s• (%) Indica Cimarron 3 52 28 53.8 27 27 100 CT11275 2 166 27 16.2 17 17 100 Fedearroz 2000 2 44 6 13.6 6 S 83 Japonica· Nipponbare 3 109 23 21.1 23 21 91 ~ (%) 53.8 16.3 11.4 19.3 1 Number of replicates. 2Regenerated plants. 3S+ = Southern positive. 4TE = Transformation efficiency 292 Fedearroz 2000 CT11275 Figure l. Southern blots analysis of PAPy123 gene in T0 plants. (A) Lanes 1-23, DNA from Nipponbare, Line 24 non-transgenic control. Lines 25-31, DNA of Fedearroz 2000. Lines 32-33, non- transgenic control. (B) Lanes 3-23 DNA of CT1275. Line 24 non-transgenic control. Table 2. Evaluations for resistance to Rhizoctonia and RHBV under glasshouse conditions of T1 and T3 progeny plants derived from six transgenic Unes that carrying PAPy123 gene %LAF %LAF Rhizoctonia Rhizoctonia (2002)1 (2003)3 2399- T3 1953- T1 RHBV 1953-2 1 Plan 2 2399-1 Line Gene Plant 1 1<26 1<31 t 1<40 1<60 PAPy12 Cica 8-446- 3 9 17 88 11 52 57 12 72 8-1 13 53 55 Nd 29 22 16 63 55 17 53 65 14-6 18 38 52 19 46 67 Palmar-446- S 23 60 11 45 50 12 51 68 4-1 5 26 26 6 41 73 12 55 58 23-2 16 53 58 17 50 81 18 38 77 3 23 46 7 48 48 14 58 68 39-4 Non- t~ansgenic Cica8 Non e 9 37 46 NA 57 64 Non-transgenic Palmar 7 26 39 51 92 Non-transgenic Fedearroz 2000 NA 2 41 56 Nd Nd Non-transgenic CT14534-12-M -3-4-M 26 23 48 60 RHBV- A3-49-60-12-3-57 -3 N 1 63 95 Nd Nd A3-49-60-12-3-1-31-9 6 29 25 62 73 A3-49-60-12 3/ Cica 8-2 NA 5 Nd Nd 61 57 A3-49-60-4-S/ Fedearroz 50-19 4 Nd Nd 33 55 1 Mean score value often plants evaluated for RHBV per each T1 1ine. 1 ud 3 Mean values of percentage leaf area affected (LAF). Reactions to two strains of Rhizoctonia (1953-2, intermediate virulence; 2999-1, hyper virulent). Two evaluations on two consecutive generations. (1) al u es refer to the percentage of LAF indicating intermediate resistance. 1Means of five plants per each T 1 Une and ~eans of ftfteen plants per each T 3 294 A B Figure 2. (A) Disease reaction of plants inoculated with hyper virulent Rhizoctonia strain 2399-1. Left non-transgenic CT14534-12-M-3-4-M derived from interspecific cross Oryzica 3/ O. rufipogon. Right, transgenic Palmar 39-4 plants carrying the PAPy123 gene. Labels refer to disease prossure (1) high (2) medium and (3} low. (B) Southern blot of transgenic plants carrying the PAP123y gene. Lanes 1-14, transgenic plants from lines 39-4, 39·9 and 39-10 respectively. NT, non-transgenic control. 2.1.8 Use of red rice as potential trait so urce for commercial rice breeding L. F. Fory1, E. Conedor2, P. Ruiz1 , J. J. Vásquez1, A. Mora3, E. GonzáJez1, T. Agrono3, C. Ordoñez3, E. Bolañoz3, J. Silva1, and M. C. Duque1, and Z. Lentini1,3 1 SB2, 2 Fedearroz, 3IP4 Introduction Recent studies on genetic diversity of Latin American rice commercial varieties had shown the need for broadening the genetic base of breeding materials. Main targets are increase productivity, stability across environments, increase weed competition, and durable resistance to various diseases and pests. Latin American wild Oryza species are new sources of genetic diversity not found in the cultigens (Buso et al., 1998). These species represent a rich, largely untapped source of resistance genes to biotic and abiotic stresses. Work by CIAT rice breeding program has demonstrated increased evidence that O. rufipogon and O. glaberrima harbor genes of interest for the genetic improvement of cultivated rice (Martinez et al., 2002). However, other studies had shown that inter-specific hybrids in rice might be prompt to both F1 sterility and later-generation hybrid breakdown (Oka, 1988: Burke and Amold, 2001). Red rice (Oryza sativaf spontanea) is a weedy rice with having similar morphologicals traits as cultivated rice at vegetative phase but generally they have profuse tillering, and vigorous growth and other plant traits make this weed highly competitive respect to rice. According to Langevin et al. (1990), the red rice can be grouped in ecotypes with characters alike cultivated rice or wild rice (Oka and Chang, 1961). Other researches indicate that red rice shows intermediate characteristics between wild rice O. rufipogon and cultivated indica or japonica varieties of Oryza sativa L. (Oka, 1988 cited by Bres-Patry et al. 2001). Another hypothesis is that weedy rice may evolve through the degeneration of domesticated rice, as weedy types of rice, where wild rice is not present (Vaughan et al. 2003). Our work using microssateliites markers had indicated that red rice accessions collected from farmer' s field in Colombia (Huila and Tolima) showed a genetic diversity not present in either rice commercial varieties, and the accessions of O. rufipogon, O. glumaepatula, O. barthii and O. galberrima analyzed (Gonzalez et al., 2003 in this report Output 1). Thus red rice could a potential source to broaden the genetic base of rice varieties. Materials and Methods Selection of red rice accessions. Red rice materials were pre-selected based on previous phenotypic, phenological and molecular characterization (Gonzalez et al., 2002). Priority selection was given to accessions with resistance to RHBV, higher tiller number, yield potential and high re-growth capacity upon harvest. Selection criteria also included materials genetically diverse identified using the principal coordinate and component analyses, and microsatellite molecular characterization. 296 RHBV resistance assays. Prelirninary evaluations were conducted on a total of 141 accessions of red rice. Second generation of self-progeny seeds of the original plants collected in farmers fields were planted in a randornized plot design with 3 replications. Thirteen commercial rice varieties with known reaction to RHBV and one accession of O. rufipogon were used as control. Materials were inoculated at 15 days old with two insects per plant from a Togasodes colony of 80% virulence. Plants were evaluated following the IRRI scale at 15, 30 and 45 days after inoculation. Results and Discussion Of the 141 red rice accessions evaluated 83% were susceptible, and 24 red rice materials showed intermediare resistance in the field evaluations, with scores ranging from 5 to 6. Except for Fedearroz 2000 (score 3) and the transgenic line A3-49-60-12-3-3-57-74 (score 1) that were resistant to the virus, all the other varieties andO. rufipogon were intermediare (Oryzica 1 and Fedearroz Victoria 1, scores 5-6) or susceptible (scores 7-9) to RHBV. Thirty seven percent of the red rice with intermediare level of resistance was derived from field plots where Oryzica 1 has been cropped for at least 2 years in a row, and the other 37% from plots with Fedearroz 50. The other accessions carne from plots cultivated with either Coprosem 1 or Cimarron. At this point, it is not clear the source of resistance to RHBV present in these red rice materials, which needs more analysis. A total of 54 red rice materials were selected by having a significant higher number of tillers respect the commercial varieties. These red rice materials showed a range in tiller number from 22 to 30 tillers per plant, whereas the varieties showed from 11 to 24 tillers per plant. Sirnilarly the number of tillers that re-grew after harvest in red rice varied up to 29 tillers per plant, while the best variety (Fedearroz 50) showed a maximum of 29 tillers per plant. A total 59 red rice materials was selected based on its yield potential. These plants showed from 2 to 39 gram per plant, while the commercial varieties (Cimarron, Coprosem, Oryzica 1 and Fedearroz 50) showed a production from 10 to 25 g per plant. Other materials were selected to include the diversity found in the red rice population based on the analysis of qualitative and quantitative traits, as well as on the rnicrosateUite markers profile and resemblance with commercial varieties (17 materials) or wild species O. rufipogon (15 red rice). Selected materials will be planted in the field this semester for further characterization and selection, to initiate a breeding scheme (Figure 1). In addition to evaluations for agronomic traits, materials will be evaluated for RHBV resistance, milling and cooking quality traits. Selected plants will be crossed with Fedearroz 50, Fedearroz Victoria 1 and three FI...AR lines, segregating populations will be evaluated and selected F2 plants will be processed through anther culture to develop fix lines to initiate yield potential and regional evaluation earlier. Future Plans To initiate the field characterization of agronomic traits of the selected red rice To conduct molecular and taxonomic characterization of individuals similar to O. rufipogon by means of AFLP To initiate the crossíng program between red rice and selected commercial varieties To advance backcrossed populations and select best candidates for the generation of doubled haploids through anther culture To identify quantitative traits loci (QTLs) associated with yield increase and/or out performance in populations derived from crosses with red rice and commercial varieties Reference Bres-Patry, C., Lorieux, M., Clément, G., Bangratz, M. and Ghesquiere, A. 2001. Heredity and genetic mapping of domestication-related traits in a temperate japonica weedy rice. Theor. Appl. Genet. 102: 118-126. Buso, G., Rangel, P. and Ferreira, M. 1998. Analysis of genetic variability of South American wild rice populations Oryza glumaepatula) with isozyme and RAPD markers. Molecular Ecology 7. 107- 117. Burke, J. and Arnold, M. 2001. Genetics and the fitness ofhybrids. Annu. Rev. Genet. 35: 31-52. González, E., Fory, L., Vásquez, J., Ruiz, P., Mora, A., Silva, J., Duque, M., Corredor, E. and Z, Lentini. 2002. Molecular characterization of rice and wildl weedy relatives by microsatellites and their use to assess gene flow in the Neo-Tropics._Annual Report Project SB-2. Langevin, S., Clay, K. and Grace, J. 1990. The incidence and effects of hybridization between cultivated rice and its related weed red rice (Oryza saliva L.). Evolution 44: 1000-1008. Martínez, C.P., Barrero, J., Almeida, A., M., Duque, M.C., F, Correa, Delgado, D., Aricapa, G., Prado, G. Silva, J. and Tohme, J. 2002. Utilization of new alleles from wild rice species to improve cultivated rice in Latín America. Annual Report. Project IP-4. Olea, H. 1998. Origin of cultivated rice. In Developments of crop science. JSSP/Elsevier. Tokyo. 254 p. Oka , H. and Chang, W. 1961. Hybrid swarms between wild and cultivated rice species Oryza perennis and Oryza sativa. Evolution 15: 418-430. Vaughan, D., Morishima, H. and Kadowaki, K. 2003. Diversity in the Oryza genus. Current Opinion in Plant Biology. 6: 139-146. 298 First selection Second selection 20 materials j Phenological selection Crosses with Fedearroz 50 Fedearroz Victoria 1 FLAR-1 FLAR-2 FLAR-3 BCI Advance (anther culture) BC2 BC6 Figure l. Breeding scheme to incorporate red rice traits into commercial rice varieties 2.1.9 Field performance and fruit quality of in vitro propagated plants of Solanum quitoense (lulo) and their use as elite clone materials by farmers J. J. Ruiz1, V. Segovia1, E. Tabares1, F. Hincapie2, J. Cock3, F. Parra4, and Z. Lentini1 1SB2, 2IPRA, 3IP6, 4Corpoica-Popayán Introduction A large number of fruits of Andean origin have great potential to become premium products for local and export markets with a high economic retum for the farmers. Solanum quitoense, locally known as lulo in Colombia and as naranjilla in other countries, is among these fruits. This species is native from Colombia and Ecuador, and it is normally cultivated between 700 and 2000 meters above sea level. Sorne of the main attributes of this fruit include its high level of vitamin C, and the sub-shrubby perennial growth amenable for cultivation in hillsides and inter-cropping, aiding soil conservation practices. Recently in Colombia, naranjilla changed from being a fruit of local fresh consumption to become an important industrial fruit for juice and yogurt products, increasing its market value. A major constraint for the rapid adoption of naranjilla by the local farmers is the limited availability of elite germplasm free of pathogens from clonal propagation. The high level of heterozygosity of this species is reflected in the high segregation of traits by its multiplication through botanical seeds. Rapid multiplication of high quality planting materials is of paramount importance. One of the main objectives of this project is to develop a protocol for in vitro propagation of naranjilla with application for conservation and rapid multiplication of elite clones free of pathogens. This protocol would facilitate the conservation and multiplication of high quality of planting farmer' s material aiding to control one of the major constraints of this crop. In the previous two years we reported the development of an efficient protocol for the maintenance and propagation in vitro of clones selected from farmers field, for the plant regeneration from tissue cultures, and the preliminary evaluation of those plants in the field. This year we report the agronomic performance of those materials in the field, including quality traits of the fruit, as well the progress made to evaluate jointly with farmers the application of this technique at larger scale. Materials and Methods Plant material. High quality and elite clones provided by the Andean Fruit Center (Centro Frutícola Andino- CEPA) were used. This collection includes naranjilla with or without thoms commonly grown by farmers. The plants were propagated in vitro or plants were regenerated following procedures as reported by Segovia et al . in SB2 Annual Report 2002. Plant and fruit evaluations. A small-scale field tri al was conducted at 1700 m o ver sea level and a mean temperature of 22C to compare the growth, development and agronomic performance of regenerated plants respect to in vitro propagated clones. Fruits produced 300 during a period of five months were harvested, counted, weighted, and classified by size according to standard scale. Premium size: > 5.5.cm diameter. Commercial size: from 3.5 cm to 5.5 cm in diameter. No commercial: < 3.5 cm in diameter. The fruit maturation process (days from harvest to ripe to rotten) was evaluated using the Munsell color scale. Fruit quality trait analyses were conducted at the Valle University (Cali). These analyses included Brix grade at 20C, relative humidity, acidity percentage, vitamin C, sugar content, pH. A sensorial panel analysis was conducted to evaluate the fruits for its appearance, color, aroma and flavor as fresh fruit or processed in juice, and compared with those available in the supermarket. In vitro introduction of plants from greenhouse or field. Greenhouse or field grown plants were tested to select the best protocol for the in vitro introduction of elite farmer's clones from the field. In order to establish plants in the greenhouse, the recovery of plants from axillary buds using shoot stakes was tested. Clones stakes of 20 cm in length from adventitious shoots with 2-3 axillary buds were taken from selected plants in the field and disinfected first with 3 ml/1 of fungicide propamocarb HCL (Pervicur) for 5 min. The basal section of the stake containing the Iower two buds were either: a) Cultured in water until root formation; b) Soaked in a solution of 10 mgll NAA for 3 days and then transferred to water until root formation; ore) Potted directly in sterilized mix of soil, sand and sugar cane plant residues (Franco et al., 2002). Treatments (a) and (b) were aerated with a fish tank pump system. In order to establish plants in vitro, apical or axillary meristems from plants grown in the greenhouse or field were tested. Explants were surface sterilized and cultured in vitro on medium A instead of medium 17N (Segovia et al., 2002. SB2 Annual Report). Riphampicine (antibiotic) 100 mg/1 was added to the medium to control bacteria! infection from field explants, since the presence of trichomes in the tissues prevented thoroughly disinfections. Evaluation of the technology with farmers. A total of 20 farmers from two regions of Colombia (Cauca y Huila) with commercial production of lulo were selected. Farmers were chosen based on the years of experience cultivating lulo, the number of lulo plants currently grown and their interest to participate in this initiative. A strategy was planned jointly with the farmers to evaluate the potential commercial use of in vitro generated plants as an altemative of planting material clean of pathogens. This activity was conducted m collaboration with IPRA, the tropical fruit project of CIA T, and Corpoica-Popayán. Results and Discussion Last year we reported no differences in plant growth and development between regenerated and in vitro propagated plants. Likewise, no significant differences were noted between the regenerated and in vitro propagated plants respect to the total number and total weight of fruits produced (Figure lA). However, the clones with thoms (T) showed higher disease and pest susceptibility that significantly reduced fruit production (Figure lA). The average productivity in 5 months of harvest indicated a mean number of fruits per plant of 142 for materials without thoms (G) and 28 fruits for T (Figure lA), and a total weight of fruits per plant of 5.58 kg for G and 1,57 kg for T (Figure lB). About 80% of the fruits harvested from these clones showed a commercial size (from 3.5 cm to 5.5 cm in diameter). The rest of the fruits were classified as 15% small fruits ( < 3.5 cm in diameter) for the G plants, and 10% prernium size (> 5.5 cm in diameter) for the T materials (Figure 2). Commercial production of lulo may yield per year about 135 fruits/plant and 9 kg fruits/plant when cultured at a density of 3,000-plants/ hectare (3 m2 per plant, ccn. In this experiment plants were also planted at a distance of about 3 m2 per plan t. Of the plants evaluated four were identified because of its high yield potential. These plants produced 9 kg fruits/plant during the Y2 year experimental harvest time. These plants were selected; seeds and stakes of these plants were collected, and grown in the greenhouse and in vitro. In relation to fruit quality, no significant differences were noted between the in vitro propagated clones, the regenerated plants and the commercial fruits bought in the supermarket for most chemical traits evaluated. The exception was the content of reduced sugar, which was lower in the supermarket fruit possibly indicating a longer post-harvest time respect to the experimental fruits at the moment the analyses were conducted (Table 1). The two sensorial analyses conducted gave different results. The first panel with 5 panelists shows a higher acceptance for the experimental materials for all traits, but the panel with 11 panelists indicated a higher acceptance for the supermarket fruit in relation to flavor and aroma, and higher acceptance for the experimental materials for color and physical appearance. This analysis seemed to be highly subjective, thus it may need to be conducted with a larger number of panelists to elucidate clearer preferences. No differences were noted in shelf life between the different materials. Attempts to establish plants in the greenhouse from field grown materials indicated that the highest percentage of stakes with new shoots was obtained when the stakes were planted directly into soil in the greenhouse. About 50% of the stakes had healthy looking plantlets 1 month after planting (Figure 3). Explants derived from clones with thoms responded poorly, perhaps due to the weak stage of the donor materials since they were highly affected in the field. Once plants were established in the greenhouse stakes from greenhouse-grown plants produced profuse roots when treated with aerated water, which ease the m~s clona! propagation of plants in the greenhouse. A high percentage of contamination was obtained when meristems were introduced in vitro directly from field grown plants. Explant survival (elongated shoots with roots) was increased from 40% to 60% when the antibiotic was added to the culture medium. No contamination was noted and at least 80% explant survival was obtained when meristems derived from greenhouse grown materials. Based on these results, successful in vitro introductions of elite materials selected in farmers fields can be achieved by establishing first clonally propagated plants in the greenhouse, and use these plants as donors for the in vitro culture of meristems. In the case a direct introduction from the field into in vitro conditions is needed, a more efficient surface sterilization protocol needs to be revised. A process was initiated to test with farmers the suitability of using in vitro propagated plants as planting materials. The potential advantage on the in vitro source is the supply of free pathogens homogenous plants maintaining the selected traits of the elite materials. 302 Farmers were selected from two si tes with commercial production of lulo (Pescador, Cauca, and Tierradentro Cauca-Huila). These sites include small, medium, and large-size production fanns. The farmers have between 2 years to 9 year of experience cropping lulo, and 200-5,200 plant-size farms. Farmers attended a workshop at CIAT with the objective to define jointly a strategy to evaluate the potential commercial use of the in vitro plants. The advantages and lirnitations of using in vitro grown plants were discussed. The farmers evaluated the plants at the CIAT lulo experimental plot. The criteria were established for the collaboration. Farmers are highly enthusiastic with the project, and a teamwork approach will be implemented. For each site, two nursery plots will be established. Farmers will select the best two plants currently grown in their fields based on productivity, fruit quality, and disease/ pest resistance. These plants will clonally propagated in the field, and explants will be established in the greenhouse at CIA T. Meristems from these plants will be introduced in vitro. Pathogen-free in vitro propagated plants will be grown in the nursery plots jointly with the field-clonally propagated plants, and seedlings derived from the seeds of the selected plants (standard propagation mode used by the farmers). The in vitro, clonal-propagated and seed-propagated plants will be compared throughout the production season. The nursery plots could also be used for the evaluation of new germplasm, one of the main needs identify by the farmers, in addition to assistance for an integrated crop and disease/pest management approaches. Corpoica could play a key role in these aspects, reason why CIA T is trying to strengthen their linkage in the project. Farmers selected their best elite materials and currently the plants are being clonally propagated in the field. Following, plants will be brought to CIA T to initiate their introduction in vitro. Field evaluations will be conducted first semester of 2004. References CCI (Corporación Colombia Internacional): www.cci.org.co/publicaciones/TropicoffROPIC005.htm CIAT, 2002. The development of methodology for in vitro multiplication, plant regeneration and genetic transformation of naranjilla (lulo). SB2 Annual Report. p: 304-306. George, Edwin F. 1993. Plant propagation by tissue culture. The technology. Part l. 2nd. Edition. Great Britain. p: 1361 Franco, G.; Berna!, J.; Giraldo, M. 2002. El cultivo del lulo. Manual técnico. Asofrucol, Corpoica y F N F H. Manizales. 103 pp Segovia V. 2002. Optimización de la regeneración de lulo Solanum quitoense orientada a la transformación genética de plantas. Universidad Internacional de Andalucía. Sede Santa Maria de la Rabida- Huelva. España. 74 p. A Mean fruits per plant GC GR TC TR Plant type B Mean fruit weight per plant GC GR TC TR Plant type Figure l. Fruit production in the field of in vitro propagated (C) and regenerated (R) plants derived from materials with thorns (T) or wi!hout thorns (G) Figure 2. (A) Plant without thorns. (B) Fruits showing commercial and non-commercial sizes. (C) Dark green flesh that showed high acceptance by panelist in the quality trait assay Table l. Chemical quality trait analysis of fruits from regenerated plants witbout thorns (GR), in vitro propagated plants witbout thorns (GC), regenerated plants witb tborns (TR), in vitro propagated witb tborns (TC), and bougbt in supermarket. Trait GR GC TR TC Supermarket Brix grade (at 20 °C) 11.5 11.6 11.0 10.0 10.5 Acidity (%) 2.2 2.2 2.7 2.3 2.9 Humidity (%) 86.8 85.2 79.8 87 82.4 Reduced sugar (%) 4.6 5.5 5.0 4.9 0.7 Total sugar (%) 3.1 3.4 2.7 3.1 2.7 PH 3.2 3.2 3.2 3.3 3.2 Vitamin C (mg/100g) 38.2 37.3 48.8 64.2 39.1 60 50 ~ '-" !3 o 40 o ..e (/J ..e - .... 30 ~ (/J ~ ~ S 20 (/J -= ~ -=... 10 o Water NAA Asofrucol Figure 3. Percentage of stakes with new sboots derived from field-grown plants and treated in the greenhouse. {G) Plants without thorns, gJabrous. (T) Plants witb thorns. Activity 2.2 Development of cellular and molecular techniques for the transfer of genes for broadening crop genetic base 2.2.1 Development of an In Vitro Protocol for the Production of Cassava Doubled-Haploids and its Use in Breeding E. Tabares1, C. Olaya2, N. Morante3, and z. Lentini1 1SB2, 2Virology, 3IP3 Introduction Cassava is an important crop for marginal agriculture areas in Africa, Latín America and Asia. It produces well under marginal conditions and is identified as a famine reserve in subsistence farming. The globalization of world economy offers cassava new opportunities for becoming an even more important source of raw materials for different industries. To achieve this, cassava productivity must increase steadily and reliably based on an efficient breeding approach. Increased genetic variability will be required to achieve diversified end uses of cassava products. This implies not only access to existing variability but effective tools for proper evaluation and screening, as well as the generation of new genetic variability for several key traits. However, traditional cassava breeding has not changed in the last 30 years and there is a need for a more dynamic breeding system so that cassava can maintain its competitiveness respect to other commodities. Introducing inbreeding in cassava genetic improvement has many advantages, which will facilitate and expedite the generation of diversified improved breeding lines. But developing inbred lines through self-pollinations would require in cassava about 9-12 years. Rapid and complete homozygous can be reached by using in vitro haploid technology. The aim of this project is to develop a protocol to induce androgenesis in cassava for the generation of doubled haploids. Doubled haploids of cassava will regenerate from cultured anthers or microspores of elite ·varieties via embryogenesis. An oyerview of the results with out- crossing species prompt to inbreeding depression likewise cassava, indicates that an efficient technology for the production of DH in cassava could be developed and spite of inbreeding depression, doubled haploid lines can still be useful for uncovering desirable genetic variability that is masked in highly heterozygous plants (Andersen et al., 1990, Paillard et al., 1996, Zhang et al., 2002). Previous preliminary work conducted by Roca and Iglesias (1993) reported the induction of sporadic callus derived from in vitro culture of isolated microspores. Although these results were not reproducible, it is a proof of concept indicating that it is possible to induce cassava cultures of isolated microspores to divide in vitro. Thus a more s'ystematic analysis of the factors affecting androgenesis in cassava as proposed in this project, may allow the development of a reproducible protocol for the generation of DH in this species. In this first report we present a preliminary study directed to the identification of optimal microspore across different genotypes and environments, the optimization of environmental conditions for shipment of flower buds for establishment of collaboration with ARis, and the cytogenetic analysis of various clones. Materials and Methods Clones gown at CIA T experimental Station or Santander de Quilichao of varieties with contrast flowering cycle (early, intermediate and late) were used: AM244-31, MCol-1505, CM4574-7, MCo11468, AM312-103, CM523-7, MPer-183 and AM523-7; preference was given to clones with profuse flowering. lmmature flower buds were collected when plants were about 8-12 month-old. Based on previous work on cassava flower biology conducted at Ciat by Roca et al. (1989), flower buds were collected when they were between 0.8 mm and 2.5 mm is diameter and fixed in a solution of 3:1 ethanol: glacial acetic acid with 0.5% ferric chloride for 24 hr for deterrnining the stage of rnicrospore development according to flower bud size. The number of rnicrospores per anther was deterrnined, and cytogenetic analyses were conducted to identify the different stages throughout the meiosis, the number of chromosomes and bivalents (ploidy level), identify chromosomal aberrations, as well as the development from cytokinesis to mature pollen grain. Two treatments were tested to evaluate optimal conditions for flower bud shipment. Flower buds were surface sterilized with ethanol 70% for 3 rnin followed by chlorox 20%-30% for 10-15 rnin, and three washes with distilled sterile water. Sterilized flower buds were either transferred in a Falcon 50 ml tube containing sterile water at 24-26 C (room temperature), or in zip-plug plastic bags 4 C (refrigerator). Pollen viability was measured through one week, time expected for the shipment to last. Results and Discussion The cassava pollen is generally from 122 to 148 J.l.m in size, which is large compared to other flowering plants. Prelirninary work suggests that most varieties show rnicrospore in the tetrad stage (Figure lB) of development when flower buds are between 0.8 and 1 mm in diameter (Figure 2). Flower buds of 1.2-1.5 mm in diameter contain rnicrospores at the uninucleate stage (Figure lC), whereas those of 1.6-1.8 mm in diameter are at the binucleate (Figures ID and 2). Most flower buds contain mature pollen grains when they reach 2.0-2.2 mm in diameter (Figures IF-G, and 2). This association between flower bud size and rnicrospore developmental stages appears to change with each peak of flowering. Early varieties seem to show a faster development of pollen grains in smaller flower buds (AM 244-31, Figure 2). Prelirninary data suggest that about 100-200 rnicrospores are produced per anther = 1,000- 2,000 per flower bud. However, a variation in the number of rnicrospores/ anther was found between anthers from different flowers of the same inflorescence and between different inflorescences. The intermediate flowering genotype CM523-7, produced about twice rnicrospores per anther (200 rnicrospores/ anther) than the late flowering material MPer-183 (100 rnicrospores/ anther). This figure needs to be checked in a broader range of genotypes across different environments. If this low amount of pollen grains is corroborated, it will be an important bottleneck for isolating large number of microspores required for isolated microspore culture, in such a case the pollen shed system should be considered. In addition, MPer-183 was more affected by the different environmental conditions between the two sites, producing about half of the microspores/anthers (about 80 microspores/anther) when grown at Ciat (satine soils) respect to Quilichao (acid soils, 166 microspores/anthers). The intermediate clone CM523-7 seemed to be less affected by the contrasting conditions in the two sites, producing 189 microspores/anthers and 224 microspores/anthers at the two field sites CIAT and Quilichao, respectively. Based on agronomic performance, CM523-7 is a clone known to have a broader adaptation to contrasting environments respect to MPer-183. This preliminary data suggests that there is a strong genotype-environmental interaction affecting the production of microspores. This interaction seems not only to affect the yield of microspores/ anthers but also the stage of development. The intermediate clone CM523-7 showed a tendency of a faster development of pollen grains in smaller flower buds when grown at Quilichao respect to CIAT. The cytogenetic study indicated that about 95% of pollen mother cells analyzed for MCOL 1505 showed 36 small and similarly shaped chromosomes and 18 bivalent pairing at meiosis (Figure 1H). Late chromosome condensation and chromosome adherence was noted in sorne cases atprophase l. Three nucleolar chromosomes were noted at prophase 1 and 11, and telophase 11 and l. Cytokinesis showed tetrahedral arrangement (Figure 1A-B). Cassava is a highly heterozygous species. Few complete cytogenetic studies are available in cassava. Studies with molecular markers suggest disomic inheritance for sorne loci with evidence of gene duplication (Carvahlo et al. 1999 and 2002~ Jennings and Iglesias, 2002). Although three nucleolar chromosomes are noticed as for true diploids, duplication of sorne chromosomes is also present suggesting segmenta! allotetraploidy. 308 Figure l. Cassava microspore Delopmental stages within the anther. (A) Cytokinesis. (B) Tetrad, tetrahedral arrangement. (C) Early uninucleate. (D) Early binucleate. (E) Late binucleate. (F) Mature poDen grain. (G) PoDen dimorphism, fluorescent viability assay. (H) Thirty six chromosomes at anaphase 1, and 18 bivalents at rnetaphase l. Figure 2. Microspore stages of developmen bud size for different genotypes grown in the field Poli en Binucleate Uninucleate Tetrad 0.8 - 1.0 1.2 - 1.5 1.6 -1.8 2.0. 2.2 Flower bud diameter (mm) •AM244-31 •MPER183 •CM523-7 •MCOL 1505 oMCOL 1468 Future Activities • Input from breeders will be used to gather information on flowering, environmental adaptation and genetic diversity to select the best candidate experimental genotypes. Genotypes with profuse, continuous flowers , synchrony of microspore development, slow maturation process to pollen grain, ideally with known in vitro response, and desirable breeding traits are available, will be used as experimental genotypes to develop a protocol for efficient induction of androgenesis in vitro. References Andersen S.B., l. Christiansen, B. Farestveit. 1990. Carrot (Daucus carota L.): In vitro production of haploids and field trials. In: Y.P.S. Bajaj (Ed.). Biotechnology in Agriculture and Forestry. Vol 12.Springer-Verlag. Berlin. P: 393-402 Carvahlo R., M. Guerra, P.C. Carvahlo. 1999. Occurrence of spontaneous triploidy in Manihot esculenta Crantz. Jap Mendel Soc 64: 137-140 Carvahlo R., M. Guerra. 2002. Cytogenetics of Manihot esculenta Crantz (cassava) and eight related species. Hereditas 136: 159-168 Paillard M., P. Lashermes, and V. Pétiard. 1996. Construction of a molecular linkage map in coffee. TAG 93:41-47 Roca, W.M. and K.Moman; R. Chavez and M.L.Marin. 1989. Cassava._jn_ Working document. Biotechnology Research Unit. Annual Report 1988. No Wd 48. CIAT. Pp 8-22. Roca W.M., C. Iglesias. 1993. Isolation and in vitro culture of cassava immature pollen and zygotic embryos. CIAT Annual Report. Biotechnology Research Unit. P: 175-179 Zhang J., W. Guo, T. Zhang. 2002. Molecular Iinkage map of allotetraploidy cotton (Gossypium hirsutum LX Gossypium babadense L.) with a haploid population. TAG 105: 1166-1174 2.2.2 Waxy cassava starch: Transgenic plantlets expressing gus in a vector that contains a GBSSI gene in sense orientation 1G. Puentes, 2JJ Ladino, 2D Lopez, 1E Barrera, 1M Fregene, 2P Chavarriaga and 2J Tohme (1) Cassava Breeding, CIAT (2) SB-2 Project; CIAT Introduction Cassava starch is a polymer composed of linear (amylose) and ramified (amylopectin) glucose uníts. Starch is a raw material used to produce glues, textiles, paper, chemicals and 310 animal feed components among other products. Sorne of these require a 100% amylopectin-containing starch. One way to obtain it in cassava is by silencing the gene GBSSI (granule bound starch synthase), involved in amylose synthesis via transgenesis. We summarize the latest achievements on producing amylose free cassava. Materials and Methods The GBSSI gene was isolated from a cDNA cassava library, using a potato probe, to make vectors containing sense and antisense versions of the gene (pCAMBIA1305.2-S and pCAMBIA1305.2-A, respectively, in Agll A. tumefaciens strain), which were transformed into friable embryogenic callus (FEC) of cassava cultivars 60444 and MCol2215. Putative transgenic tissues were subject to selection with hygromycin. The transformation vector contained a gusA gene that allowed for early scoring and follow up of transformation events. Results The first transgenic lines of cultivar 60444 are being regenerated (Figure lA) from in vitro cultures. As expected, 60444 responded faster than Mcol2215 to selection and regeneration, although putative transgenic somatic embryos of MCol2215 are now on embryo maturation medium, without selection, to produce plants (not shown). Sorne plantlets of 60444 were rooted on media with 10 mg/1 hygromycin and expressed gus in what seems to be a chimeric pattem (Figure lB). lt is still too early to say if plants have silenced GBSSI expression. Figure l. (A) Transgenic FEC lines expressing gus, transformed with sense GBSSI, all from cultivar 60444. (B) Transgenic leaf of in vitro plant with a chimeric gus expression pattem. There are two ways to calculate transformation efficiency with cultivar 60444: There are 10 independent transgenic plants, seven gus positive and three gus negative, that regenerated plants. The initial amount of FEC that was transformed was 8.2 g of FEC; it gives 1,2 independent trasngenic lines/g of FEC The same 10 independent transgenic lines were recovered from 100 FEC clusters. Assuming that each cluster represents an explant, transformation efficiency, in percentage, is 10/100 = 0,10 (10% ). The fonner way of calculating transfonnation efficiency has been adopted as the standard for transfonnation in cassava. Future Activities Regenerate plants and molecular confinnation of presence of transgenes Test for silencing of GBSSI at molecular level (RT-PCR) Move plants to biosafety greenhouse, make them produce storage roots and test for starch composition 2.2.3 Functional analysis of a Caffeic Acid 0-Metiltransferase gene from Brachiaria decumbens (BdCOMT) in transgenic rice as a model plant C.P. Flórez1, E. Tabares1, G. Delgado2, L.F. Fory1, Z. Lentini1'2 •sa2. 2IP4 Introduction Brachiaria species are important forage grasses in the tropical lowlands of America, Asia, Africa and Australia. B. decumbens cv Basilisk is one of the most extensively cultivated Brachiaria species. It shows adaptation to acid soils, rapid growth and provides good soil coverage, and forage yield (Miles and Do Valle, 1996). Tropical grasses have high content of cell wall constituents, and this generally correlates negatively to intake, digestibility and animal performance. The major anti-quality components of grass cell walls are lígnin and phenol monomers, which esterifies with cell wall polysaccharides (Smith and He, 2000). Lignin composition, particularly the relative proportion of syringyl (S) and guaíacyl (G) units in the lignin polymer, and the nature of the covalent linkages between lignin and other polymers are also important determinants of the leve! of digestibility (Heath et al., 1998). Genetic variation for quality is present in grass species and traditional breeding programs have been using that variability to improve forage quality. Although progress has been made, it has been slow and limited. Developments in genetic engineering now offer additional ways to improve forage quality. Methods are available to down-regulate lignin biosynthesis genes to reduce lignin content and/or change its type to improve forage digestibility. The lignin biosynthesis pathway and severa! key enzymes are reasonably well understood (Smith and He, 2000). Caffeic acid 0-methyltransferase (COMT) ís a key enzyme in the lignin biosynthesis pathway. Therefore the modification of lignin content and/or composition by genetic manipulation would be of great economic interest, if consider that a 1% in crease in mean 312 live weight gains per animal (Casler and Vogel, 1999). We had reported the cloning and characterization of a caffeic acid 0-methyltransferase gene from Brachiaria decumbens (BdCOMT) (Florez et al., 2003). The objective of this part of the research was to finalize the characterization of this gene by functional analysis of its expression in transgenic rice plants used as model species. Materials and Methods Rice genetic transformation. Mature embryos derived embryogenic callus from indica (cv Palmar) andjaponica (Nipponbare) varieties, where used targets. Agrobacterium mediated transformation was conducted using the Agll strain containing the constructs pC010MT- 2, pC010MT-4, pC05.20MT-l and pC05.20MT-2 previously generated in the laboratory (Florez et al., 2002, SB2 Annual report p. 254-259). These constructs contain the BdCOMT gene in sense or antisense orientations. Transgenic rice was according Tabares et al., 1999. Southem Blot Analysis: Genomic DNA from rice young leaves was isolated following the methodology McCouch et al. (1988). Brachiaria DNA (15 l-lg) were digested with the restriction enzymes Hind m. The fragments were separated on 1% agarose gels and transferred to Hybond W membranes according to the manufacturer's ínstructions (Amershan). Probe consisted of the open reading frame of BdCOMT prepared by PCR. DNA probes were random primer labeled using Megaprime DNA Labeling System kit (Amersham) and e2P] dATP. The hybridization was carried out overnight at 55°C following the manufacturer' s recomrnendations. Northem Analysis: 15 }!g of total RNA from rice leaves was separated on 1% formaldehyde gels and transferred to Hybond N+ (Amersham) membranes according to the manufacturer' s instructions. The pro be consisted of the open reading frame of BdCOMT prepared by PCR. DNA probes were random primer-labeled using Megaprime DNA Labeling System kit (Amersham) and [32p] dATP. Hybridization and washing were carried out following the manufacturer' s recomrnendations. Histochemical reactions: Leaves sections were sliced by hand and stained according to Weisner and Maüle reactions. For Maüle staining, leaf sections were immersed for 5 min in 1% Kmn04 , rinsed and distained in 30% HCI, washed and mounted in concentrated ~OH. Por the Weisner reaction, leaf sections were incubated for 2 min in phloroglucinol solution (2% in ethanol/water 95/5 v/v), and then mounted in 50% HCI. Bright field photographs were taken between 20 min after staining procedure. Plant evaluations: Six agronornic traits of transgenic rice plants were evaluated in the greenhouse. These characteristics included: (1) days to flowering, (2) height, (3) number of tillers, (4) fertility, (5) weight of 100 grains, and (6) total grain weight per plant. Results and Discussion Southem analysis indicated that the BdCOMf gene was introduced in sense and antisense orientations in the rice genome of cultivars Palmar (indica) and Nipponbare (japonica). Total of 321 embryogenic hygromycin resistance callus were transferred to regeneration medium containing 50 mg r1 hygromycin. Between 8-75% of the callus regenerated plants. A total of 125 To plants were transferred to the greenhouse and Southem blot analysis confirmed the integration of BdCOMf gene in 92 - 100% of the plants analyzed. These results suggest transformation efficiency between 10- 74% for Palmar and between 15- 56% for Nipponbare cultivar for different experiments. The BdCOMf expression was measured using histochemical analysis through Weisner and Maüle reactions. Weisner stain in known to react with cinnamaldehyde residues in lignin and the color intensity reflects the totallignin content. No difference was detected with this stain between control and transgenic plants, indicating no significant variation in lignin quantity. In contrast, the Maüle reaction, which is specific for free S units in lignin, revealed variations in staining intensity from pale yellow (inhibited expression) to red (over expresion). Sorne plants displayed a bright red color in their vascular tissue suggesting putative over expression of the BdCOMf gene in these plants. A good correlation between Northem and histochemical analysis was found. High RNA transcript levels were found in these over expressing plants, while no signal was detected in untransformed controls and a very weak signal or no signal was detected in the most plants showing pale yellow color (Figure 1). Agronomic characterization of the transgenic plants indicated sorne variation between transgenic and non-transgenic controls, mainly in the transgenic lines from Nipponbare. High levels of sterility were noted in transgenic Nipponbare plants. This type of phenotypic variation was also noted on non-transgenic regenerated plants, which did not contain the transgenes as indicated by the Southem blots. Transposons and retrotransposons have been found to induce mutation during in vitro culture. In Nipponbare, three retrotransposons (ToslO, Tos17 and Tosl9) are activated during in vitro culture and inactivated during in planta growth (Hiroshima et al., 1996). Previous analyses of expression of other COMf genes in other plants have not found effect on plant growth and development in contrast to the 0-methyltransferase gene CoA-3, which has been demonstrated to affect plant growth (Picon et al., 2001). Thus the phenotypic variation found in transgenic BdCOMf Nipponbare plants seems not to be associated with the presencel expression of the transgene. Future Plans The quantification of the lignin content and composition in transgenic BdCOMf rice will be conducted in order to complement these analyses. 314 Figure l. (1) Northern analyses, showing different levels of BdCOMT transcripL (2) Maüle Reaction. {A) Non-transgenic Palmar control; (B) Transgenic Palmar line 100; (C) Transgenic Palmar line 82; (D) Transgenic Nipponbare line 10; (E) Transgenic Nipponbare line 110 (inhibited expression); (F) Transgenic Palmar line 37; (G) Transgenic Palmar line 62 (over expression). References Casler, M.D. & Vogel, K.P. 1999. Accomplíshments and impact from breeding for increased forage nutritional value. Crop Sci. 39: 12-20. Florez, C.P.; Emmerling, M.; Fory, L.F.; Spangenberg, G. & Lentini, Z. 2003. Isolation and characterization of a caffeic acid 0-methyl transferase from signa! grass. Molecular Breeding of Forage and Turf, Third Intemational Symposium, May 18-22, Dalias-Texas, USA. Poster # 136, p. 131. Heath, R.; Huxley, H.; Stone, B. & Spangenberg, G. 1998. cDNA cloning and differential expression of three caffeic acid 0-methyltransferase homologue from perennial ryegrass (Lolium perenne). J. Plant Physiol. 153: 649-657 Hiroshima, H., Sugimoto K., Otsuki Y., Tsugawa H., Kanda M. 1996. Retrotransposons of rice involved in mutations induced by tissue culture. Proc Natl Acad Sci USA 93: 7783-7788 Pican, G., Chabannes M., Lapierre C., Pollet B., Ruel K., Joseleau J .P., Boudet A., Legrand M. 2001. Plant Physiol 126: 145-155 Smith, R.L. and He, X. 2000. Genetic engineering for the improvement of forage grass quality. Proc. 55th Southem and Pasture Crop Improvement Conference, Raleigh. p 12-14. Miles, J. and Valle, C. do. 1996. Brachiaria: Biology, Agronomy and Improvement. CIAT. P 164-167. 2.2.4 Cassava propagation using Iow-cost in vitro propagation techniques and conservation of na ti ve varieties from southwestern Colombia R.H. Escobar\ L. Muñoz1, C.M. Hemández2, G. Ospina3, J. Restrepo3, J. Tohme1 and W.M. Roca4 1SB-2- Project, CIAT; 1Rural Community of Santa Ana, Cauca; Jpundación para la Investigación y Desarrollo Agrícola-FIDAR; 4CIP-Perú Introduction Low-cost propagation methods could support decentralized seed systems. Today this project is in a position to distribute clean materials (certified FSD-free material) for small fanners to use in their next planting cycle (2004-A). Our materials could also be used for reestablishing cassava plantations in the Cauca area. 316 Materials and methods Materials being grown by farmers were collected from Buenos Aires, Caldono, Morales and Santander de Quilichao (Cauca Province). A total of 27 materials were planted in two local seed banks (Pescador 1500 m alt. and Piendamó 1700 m alt.). Stakes were collected from each si te and taken to CIA T HQ in Palmira to initiate meristem culture; and 19 morphological descriptors were evaluated (Fukuda and Guevara, 1998). Younger leaves, to be used in the AFLP analysis, were collected from the CIAT greenhouse for DNA extraction by the Dellaporta method. A total of 6061 plants from 6 clones (M Bra 383, CM 523-7, M Col 1522, HMC-1, CM 6740-7 and M Per 183) were evaluated for frog-skin disease (FSD) by ICA officials. A new plot of cassava plant materials was introduced to the propagation scheme to reactivate a rural in vitro laboratory located in San Rafael. Results and discussion The materials collected were Algodona Gigante, Algodona Rápida, Chiroza, Verde, Algodona Grande, Varita, Pate Pava, Bajuna Pequeña, Correita, Independencia Mejorada, Bajuna Armenia, Totoqueña, Falsa Chiroza, Sata, Chiroza Roja, Blanquita, Cáscara Roja, Parroquiana, Amarilla, Algodona Pequeña, Bajuna, Bajuna Grande, Panameña, Sauce, Regional Morada, SM 5080-1 and Yuca Blanca. Two seed banks were established using materials collected in Pescador (20 materials) and Piendamó (17 materials, sorne of them replanted). When the materials from Pescador were harvested, the farmers expressed interest in Totoqueña (4.1 kg avg), Algodona rápida (2.76 kg avg) and Algodona grande (2.6 kg avg) (Fig. 1-A). The average per total harvested area (840m2) was 1.8 kg with a total production per area 1650 kg after 14 mo. Materials from Piendamó will be harvested in November. Good DNA quality was obtained using the Dellaporta method (Fig. 1-B). Quantification and dilutions were initiated in order to start running the AFLP's. We expect to find diversity among the Algodona materials (farmers preferred these materials by its good starch quality). At present they are very interested in obtaining material from Rojita, but it has not been collected yet although sorne farmers are growing it. A total of 614 meristems were placed in in vitro conditions and will be used as a source of explants for in vitro thennotherapy. ICA functionaries certified the San Rafael plot as free of FSD-diseased plants. Those materials were planted last years as part of CBN-PRGA supported activities (Escobar et al., 2002). Sorne materials (HMC-1, CM 6740-7, M Bra 383, CM 523-7) have good potential (Fig. 2A-B; Table 1). A rapid propagation scheme will be implemented using these materials as the initial source. Figure 1. (A) Total DNA using Dellaporta method. (B) Totoqueña material with best average root production after 14 months. Figure 2. (A) Panorama of plot with clean seed material coming from in vitro conditions located in San Rafael, Cauca. (B) Harvested material certified as free of FSD by ICA officials. Table l. Average production of 6 clones produced by farmers using in vitro, low-cost technology Clone CM523-7 MC 6740-7 HMC-1 M Bra 383 M Col1522 MPer 183 % Commercial Roots 52.2 54.6 67.4 60.4 25.2 65.7 Average (kglplant.) 3.67 4.15 4.57 3.42 1.25 4.4 SD 0.20 0 .07 0.93 0.27 0.65 0.049 Pr>F (<0.0001); R2 =0.9256; CV= 12.07; average among clones=3,57 Kg/plant. 318 ANOVA Grouping a a a a b a Five plants per accession (M Col 1468, M Col 1522, M Per 183, CG 402-11, HMC-1) were give to FIDAR by the GRU after Material Transfer Agreement signature; and 3 plants per accession were maintained by the SB-2 project to increase the number of plants. The remaining plants can be used for farmers' production. After 4-5 propagation cycles, a large part of the plants must be sent to the end-users to be incorporated into in vitro rural activities. A few plants per clone will be maintained as backups. Conclusions The use of in vitro materials could be used initially as part of a decentralized seed system. This would permit certifying the quality and quantity of material per cycle. In vitro systems could be use for seed releases or renewing materials that are showing decreases in root production. Materials from San Rafael will give strong support to cassava systems in Cauca. Future activities • Running the AFLP fingerprinting. • Implement two thermo therapy cycles and then material will be tested with Secundina (M Col 2063) for FSD certification • Clean certified materials will be propagated for the rurallaboratory. A second plot will be planted to establish a clean seed bank. • Implement rapid propagation scheme using certified material planted in San Rafael. References Dellaporta Escobar, R.H.; Hemadez, C.; Ospina, G.; Restrepo, J.; Tohme, J.; Roca, W.M. 2002. Cassava propagation using low-cost in vitro propagation techniques and conservation of native varieties from Southwestern Colombia. In : Annual Report, Project SB-02: Assesíng and utilizing agrobiodiversity through biotechnology. CIAT (Centro Internacional de Agricultura Tropical), Cali, CO. p. 238-241. Fukuda, W.; Guevara, C. 1998. Descriptores morfológicas e agronómicos para a caracetezaylio de mandioca (Manihot esculenta Crantz). Cruz das Almas: EMBRAP A-CNPMF, BR. 38 p. (Documento 78) 2.2.5 lmplementation of the encapsulation-dehydration cryopreservation method for the cassava core collection R.H. Escobar1, N.C. Manrique1, A. Rios1, G. Mafla2, D. Debouck2, J. Tohrne1 1SB-2 Project, CIAT; 2SB-1 Project, CIAT Introduction Cryopreservation techniques can be divided into classical and new. CIAT has developed altematives for both kinds of techniques (Escobar et al ., 1997, Escobar et al. , 2000). New techniques based on encapsulation-dehydration, which are simple and rapid, would be useful for preserving large germplasm collections such as that of cassava. Not too many steps are involved as in the classic methods. Using these techniques with a wider number of clones will give a general idea about how safe this conservation method is in the long term with respect to repeatability and consistency after freezing. Materials and methods The encapsulation-dehydration method was implemented using materials from the in vitro bank. These materials were propagated using 4E medium (Roca, 1984) for the bud/node explants. When we tried scaling up, we considered increasing the number of beads per container (recycled baby food jars). Normally each jar with 50 g of silica gel holds 20 beads; we tested 40 beads per jar. To determine if a clone is amenable to being cryopreserved, it has to form shoots from at least 30% of the beads after freezing. This is considered the threshold (Escobar et al., 2001). Results and discussion To date we have received 447 cassava clones, which are being propagated to increase them up to 100-150 plants/clone to produce enough new shoots (2-3 months old and without cuttings) to use in the cryopreservation process. During the last 6 months we had a problem with dust mites infecting the production and lost 187 clones (42% of the material received). This attack made it necessary to increase propagation activities, reducing the number of clones put under cryopreservation. From the material received, we cryopreserved 348 clones (78% of material received, which corresponds to 55.2% of the core collection). Of the total clones cryopreserved so far, 68% have surpassed the threshold value (Table 1). Of the 145 clones tested in 2002-2003, 43% reached the threshold (Table 2). 320 Table 1. Summary of response of 348 cassava clones cryopreserved from the core coUection. Group response* Highest (>70%) Intermediate (30-70%) Lowest ( <30%) No. of clones tested No. of Clones 91 146 111 348 %Response 26.15 41.95 31.90 *Represented as shoot formation after freezing in liquid nitrogen Table 2: Response of cassava clones cryopreserved during 2002-2003. Group response* Highest (>70%) lntermediate (30-70%) Lowest (<30%) No. of clones tested No. of Clones 16 46 83 145 %Response 11.03 31.72 57.25 *Represented as shoot formation after freezing in liquid nitrogen The 40 beads in the baby food jar did not dehydrate sufficiently or homogeneously; thus encapsulated shoots did not survive freezing. In contrast, when the 40 beads were placed in a petri dish, they reached similar hurnidity levels to the 20 beads in a baby food jar (Fig. 1). These results indicate that dehydration, as expected, depends on the number of beads, the type of container and the sucrose pretreatment. Ideally, larger containers should be used to handle more beads; i.e., glass petri dishes with a larger numbers of beads. However, the cost of a glass petri dish is 250 times higher than a baby food jar. A routine was established to clean the growth room. It perrnits rnite control by combining chernical treatments specific for adults and eggs. Every 15 days we also apply rnint or eucalyptus oil scents on the floor as a preventive measure to keep rnites away .. Future activities • Recover clones lost by mi te attack • Continue freeze-testing using core collection • Analysis of clone behavior and group definitions after freezing o 5 10 15 20 25 30 • c20 --ec20 .. c40 ----- ec40 • p40 -- pe40 1 Figure l. Behavior of beads treated in 0.75 M silica gel for 3 days across different dehydration times. Two types of container were tested: Baby food jars (e) and petri dishes (p) References Escobar, R.H.; Mafla, G.; Roca, W.M. 1997. A methodology for recovering cassava plants from shoot tips maintained in liquid nitrogen. Plant Cell Rep 16:474-478 Escobar, R.H.; Manrique, N.C.; Debouck, D.G.; Tohme, J.; Roca, W.M .. 2000. Cryopreservation of cassava shoot tips using encapsulation-dehydration technique. In: Annual Report, Project SB-01: Assesing and utilizing agrobiodiversity through biotechnology. CIAT (Centro Internacional de Agricultura Tropical), Cali, CO. p. 178-181. Escobar, R.H.; Maruique, N.C.; Debouck, D.G.; Tohme, J.; Roca, W.M .. 2001. Cryopreservation of shoot tipsd: The encapuslation-dehydration technique. /n: Annual Report, Project SB-01: Assesing and utilizing agrobiodiversity through biotechnology. CIAT (Centro Intemacio.nal de Agricultura Tropical), Cali, CO. p. 243-247. Roca W.M.l984. Cassava: /n: Sharp, W.R.; Evans, D.A.; Amrnirato, P.; Yamada, Y. (eds). Handbook of plant cell culture. V.2. Crops species. MacMillan Publishing Co. New York. P. 269-301 322 2.2.6 In vitro systems (conventional tissue culture and RITA~ to support CIA T research agenda R.H. Escobar, L. Muñoz, J. Tohme SB-2 Project Introduction Last year the SB-2 project supported others projects using its tissue culture facilities to provide them sorne interesting materials (Escobar et al., 2002). The massive cassava propagation system using RITA® was developed by CIAT. This system has been validated with a wide range of cassava materials (Escobar et al., 2000), as well as with other crops (Escobar et al., 2002). Using this system malees it possible to increase propagation rates 6- 11 times vs 3-4 times under solid conditions. A summary of this activity is presented here. Material and methods Conventional in vitro propagation using 4E medium was implemented; rooting was doing in a 17 N medium (Roca, 1984). Users were responsible for hardening and greenhouse management, following Roca et al. (1984). The quality and quantity of new explants recovered by RITA depend on the type and amount of cytokinins used in the media. For cassava best results were observed after treatment with 0.5 mg/1 IDZ; it release shoot tips from apical dominance and allow recover more than one shoot per explant. Combining RITA® -Solid-RIT A® gave the best combinations and enabled recovery of plants without hyper hidrycity (Escobar et al., 2002). Technical support was given to Corn Products lnc. (Malambo) with respect to the management of sorne in vitre problems associated with water quality in their laboratory. Results Six different plots were established (Table 1). As only a few plants were requested each time, the solid conditions could be used. For the Cauca farmers this project could be considered a unique option for obtaining clean materials. As discussed in the article "Cassava propagation using low-cost in vitre propagation techniques and conservation of native varieties from southwestern Colombia" presented elsewhere in this report, material planted in San Rafael, Cauca, last year was certified by ICA functionaries as FSD-free and could be used as a seed source for the rapid propagatíon system. Table l. Material involved in smaU propagation set and their potential users. Characteristic of Interest No. Clones Commercial material for 7 regional testing Wild material for Bemisia 5 tabaci testing Xanthomonas testing* 22 FSD testing FSD testing 1 Material requested from Cauca 6 farmers Propagation and embryogenesis testin * For DNA extraction using in vitro tissues. U ser or Partner IP-3 PE-1 GDF-61 IP-3/ IP-1 ASOPROSA FIDAR SB-2 Material Involved CM 523-7, M Col 1468, M Bra 383, CM 1438-2, M Col 3306-4, CM 2772-3, M Tail 6 No.of Plants 484 M. aesculifolia jlavelifolia 444- 67 002, M. peruviana 417-003, M. peruviana 417-005, M Nga 2, M Nga 11 M Bra 685, M Bol 3, M Bra 12, M Bra 110, M Bra 383, M Bra 881, M Bra 900, M Col 1468, M Col 22, M Col 1522, M Col 2215, M Cub 51, M Mal 2, M Mal 48, M Mex 59, M Pan 51, M Tai 1, M Ven 25, M Col 1505, CM 2177-2, CM 3306-9, M Nga 1 MCol2063 MCol2063 HMC-1 , M Bra 383. CM 523-7, CM 6740-7, M Per 183, M Col 1522 MTai 16 762 90 96 600 1500 Whiteflies and FSD are the most limiting problems for cassava in the Cauca regions. In 2002, CORPOICA, MADR and CIAT released a clone with high resistance to the whitefly (Aleurotrachelus socialis Bondar) named Nataima-31 (CIAT code CG489-31) (Bonilla et al. , 2002). This is a promising clone that could be used under Cauca conditions. Gustavo Jaramillo (pers. com.) has provided us with a list of sorne materials that he has observed with FSD tolerance. For that reason we included sorne clones with whitefly and FSD resistance in propagation schemes that could be u sed for future testing plot under farmers' conditions. At present, we are running sorne materials with RITA® in · order to cover the needs for whitefly and FSD materials (CG 489-31 , SM 7951-5, M Per 183, HMC-1), commercial clones (CM 523-7, CM 6740-7) and sorne for Cauca farmers (M Col 1522, M Bra 383). For CM 6740 the propagation rate was 1:12. Conclusions In vitro facilities could play a critica! role in producing clean material to make comparisons with conventional cuttings, releases of new breeding material, renewing materials, clone dissemination, pest and insect testing among others uses. 324 If CIA T projects are thinking of encouraging the use of these materials they must plan their agenda jointly, making propagation schemes, agreeing upon objectives and supply requirements. Coordination among projects is critica! to meet everyone's research agenda. The in vitro system could be used initially for whitefly and FSD management in Cauca by preliminary testing of this material on farm conditions. This scheme could also be implemented in the biofortification program for distributing, testing and promoting high carotene-content materials. Punctual requirements could be covered with RITA® because this system has the highest propagation rate in a short period time. The GRU usually provides 5 plants per clones after signing the Material Transfer Agreement. References Bonilla, H.L.; Rey, L.; Arias, B.; Bellotti, A. 2002. Nataima-31. Variedad de yuca (Manihot esculenta Crantz) resistente a mosca blanca (Aleurotrachelus socialis Bondar) para el Valle Cálido del Alto Magdalena. CORPOICA (Corporación Colombiana de Investigación Agropecuaria), MADR (Ministerio de Agricultura y Desarrollo Rural), CIAT (Centro Internacional de Agriculture Tropical). (Plegable Divulgativo) Escobar, R.H.; Muñoz, L.; Montoya, J.E.; Chavarriaga, P.; Tohme, J.; Roca, W.M. 2002. Propagating commercial clones and transgenic cassava plants with RITA~. In: CIA T (Centro Internacional de Agricultura Tropical). Assessing and utilizing agrobiodiversity through biotechnology. Annual report, Project SB-02. Cali, CO. p 281-285. Escobar, R.H.; Muñoz, L.; Tohme, J. 2002. Tissue culture to support CIAT research agenda. /n: CIAT (Centro Internacional de Agricultura Tropical). Assessing and utilizing agrobiodiversity through biotechnology. Annual Report, Project SB-02. Cali, CO. p 288-289. Escobar, R.H.; Muñoz, L.; Tohme, J.; Roca, W.M. 2000. Cassava micropropagation for rapid "seed" production using temporary immersion bioreactors. In: CIAT (Centro Internacional de Agricultura Tropical) Assessing and utilizing agrobiodiversity through biotechnology. Annual Report, Project SB-02. Cali, CO. p 185-187. Roca, W.M. 1984. Cassava. In: Sharp, W.R.; Evans, D.A.; Amrnirato, P; Yamada, Y. (eds). Handbook of plant cell culture. v. 2 . Crop species. MacMillan Publishing Co., New York. p. 269-301. Roca W.M.; Rodríguez, J.A.; Mafia, G; Roa, J. 1984. Procedures for recovering cassava clones distributed in vitro. Centro Internacional de Agricultura Tropical, CIAT. 8 p. 2.2.7 Embryo Rescue of Sexual Seeds from Breeding Populations for Molecular Assisted Selection (MAS) of CMD Resistance Luis Guillermo Santos, Danilo Moreta, Samy Judy Moreno, Adriana Alzate, Nelson Morante, Heman Ceballos, Martín Fregene CIAT Introduction CMD breeding at CIA T aims to develop Latín America cassava gene pools adapted to the disease should in case it ma.kes an accidental entry into the region. A second important objective is to facilitate gennplasm shípment of CIA T' s elite cassava gennplasm to regions, such as India and Sub Saharan Africa where CMD is endemic, via the introgression of CMD resistance into CIAT's elite gennplasm. To permit marker-assisted selection (MAS) of CMD resistance at CIA T for Latin America and at the same time fulfill plant quarantine conditions for the shipment of the CMD resistant CIA T germplasm to India and Africa, it is necessary to germinate and maintain in vitro breeding populations. This season more than 3000 controlled crosses were made between CMD resistant parents introduced from llT A and elite cassava parents or backcross derivatives of M. esculenta sub spp flabellifolia for resistance to the green mite. A total of 2315 seeds were harvested as mature seeds and have been germinated in vitro from embryo axes. A summary of results obtained from the germination of these seeds is presented below. Once germinated, the plantlets were multiplied, molecular-assisted selection (MAS) performed using the marker NS158 and another SCAR marker RME1, and CMD resistant genotypes will be transferred to the screen house for further hardening and evaluation as well as for shippment to collaborators in India and Africa. Methodology The culture of embryo axes from mature or immature seed has become routine in the establishment of cassava CMD breeding populations in vitro. The method being currently used is that developed at CIA T and modified by Okogbenin (2003), in summary, seeds are selected by a floatation test in water, seeds that float (indicating of vain or little developed seeds) are discarded, incubation in 97% sulfuric acid to 97% for 50 minutes, to allow for easy scarification, and washing with water to eliminate the acid. Following, the seeds are disinfected with 70% alcohol to for two minutes then with a solution of sodium hypoclorite (0.5%) and a drop of 20 tween for 12 minutes, finally they are rinsed three times with sterile deionized water. The embryo axes is then removed together with the cotyledon and placed on 17N media and left in the dark for 5 days at temperatures between 28 and 31°C. Next, the embryos are incubated at the same temperature under a photoperiod of 12h light and 12h darkness until they grow into full plants which are then multiplied in 4E media. 326 Table l. Sexual seeds of cassava that were embryo rescued and multiplied In vitro in the year 2003. 1 1 1 1 !lb. á !lb. of seeds 1 10tal l'b.of l l'b.of P lantsl % reOOYery Mother 1 F ather Pu r ¡:ose 1 seeds 1 recei\UI discaroed ~eds r eco-ered ofplants CR-i OW 18.'M 1 C-!27 A_LW~CMD j 28 _ 0 0 _ 1 :B 6 5 5__ 83 18 - CR-2 -- --C"-4- AM at4-B1 ACMD ! 6 rn-a C-4 1 CGt89-3i ACMD-M$ 1 56 1 $ 21 38 CR-4 C-4 CW6&f!J ACMD-ACR 1 11 O ll 3 'Zl CR-5 C-4 !¡ CWfi'r'iJ ACMD-ACR 12 2 JO 1 1 10 CR~ C-4 CW07-42 ACMD-ACR 18 2 16 9 56 CR~A C:-4 , MOOL1734 ACMD-'IRT ! 132 8 124 55 44 CR.:rB MOOL 1734 C:-4 ACMD-'IRT 46 3 43 15 35 CR-eA C:-4 ! MOOL?aXi ACMD-'IRT 1 168 7 161 31 1 19 C:R-88 MOOL 2<03 1 C:-4 ACMD-'IRT l 24 6 18 1 7 1 39 CR-9A C-4 1 MTA18 ACMDl 1 :J:'iJ O 300 , 1$ L-._!4 C:R-98 M-TAI8 ! C-4 ! ACMDl-,-38 6 32 al 1 63- ~~ 1 M~~~734 -~~~i~-~-~~~~:~ ~--~--·----~- t-~-- -~~-~ ·t--· fs CR-ilA C-i'Zl - M OOL 22:.Xi Í ACMD-vRT 1 47 1 t6 29 l 63 CR-llB MOOLZOO C-127 í ACMD-vRT 23 4 19 J 15 79 CR-12 C-l'Zl MTAl 8 1 ACMD-1:01 1 50 3 47 1 23 49 CR-13 MOOL 1734 C:-33 YRT-ACMD ! 1 29 3 23 9 ! 35 CR-l4A CM2H C:-4 ACMD2 11 O ll 8 1 73 CR-i4B C-4 CM5234 ACMD2 : fb 1 49 ! 36 +-73-_ CR-l!iA. CM2H • C-33 ' ACMDL..._ 5 2 3 2 l 07 CR-lSB C-33 CM5z57-t- ACMD2 1 1 O 1 1 ;-- m- CR-!6 CM2H C-B9 1 ACMD2 J 13 1 1 12 2 r--1-7 -- CR-!7 CM2H C.OO : ACMD2 17 5__j__ 12 í 3 : 25 CR-lB 1 CM:m>-4 ! C-4 ! ACMD1 t-"32 3 l ::9 1 15 1 52 CR-19 1 CM:m>-4 ¡ (:¡8 ACMD1 ' 3 ?.__L 1 F-0--l - --0 - C:R-:n\ . __ g-1~_-4 __ ! -· C:-33 ¡ AQviD1 1 12 1 ... j 11 ~ 6 j 55 ~-~d~~~++~-+1~----6-+---~-+-1H---~-- CR-22A CM075H~ C:-33 ! ACMD! ¡ 2 1 l 1 1 1 ID CR-228 C-33 ! CM6íS4-a ! ACMD1 1 3 1 2 ! O 1 O CR-ID CM7951-5 ! C-4 i ACMD4 32 4 :B ! ~ 93 CR~ CM7951-5 í C:-18 ' ACMD4 12 1 ll 10 ! 91 CR-25 CM7951-5 1 C:-33 1 ACMD4 7 O 7 1 6 ! 86 CR-ffi CM7951-5 ' C-B9 1 ACMD4 5 3 2 1 50 CR-27 CM7951-5 C.OO 1 ACMD4 i 35 4 31 25 81 C:R-2&\ SM!m-25 C-4 j 1 ACMD2 j 07 17 ff) 29 58 CR-iBB C:-4 SM~ ACMD2 i 58 4 54 34 63 CR~ SM!m-25 3 O C-33 ACMD2 CR-33 CR-a4A SM16ffi-2 SM 1741-l SM1741-l C-lB - SM~ C-B9 C-413 C:-33 C:-4 C:-18 SM1741-l ACMD2 ACMD2 ACMÓ2 ACMD2 ACMD4 ACMD4 ACMD4 54 130 24 ! 1 3 7 8 o 3 2 07 4 1 25 13 9 69 o o o 51 1 13 25 123 77 63 16 14 88 1 o o - ~~-f----~~1.!741-l .. -----~~-~~-º-'!.. __ 1.~----L- ____ 1?.._ __ !_9_·-·---~ CR-ffi SM 1741-l 1 C-B9 1 ACMD4 15 2 13 8 1 62 Table 1 (contd) e recovery 1 reoei'led discarded of seeds recovered 1 of plants ~CR~~ ~~e~~ -4--CM~~47~-+~~A~CM~D2~~~--~1~5--,'--~0-- lli ~ 1 00 ! ClHl8 e~ OW28J-l ACMD-P1N \ 31 1 2 -~-- _ _E_ ___ j ro j CR-69 e~ SM ~~ ACMD2 l 6 l O 6 4 1 01 1 I--:CR:=-::~==-+----::e::-:-l~8·-+-;CM~;-;:~4-;;~:-+-'7A~CM~D2~-+: ---· 3 1 1 . - ¿- ·· --- 0-----, - 0----j CR~1 e-l8 MCOL :rró ACMD-mT i 12 1 O ~ ¡ ~ 1 1CO ' ~CR::::,:-.;~ ~--,:e:-:l==-8---+-~M:-:::C0-:3:::':--L :::izrs:::::::- t-A-:-;:;;CM:-:D~-mT=::::--i!--::32:;:;-----"-- 5 'Zl 15 56 CR~ e.:l3 CM4574~ ACMD2 ; 28 4 24 15 1 63 CR~ e-as CM~ ACMD1 15 1 14 11 79 i CR~ e-as CM457H ACMD2 25 3 22 : 15 68 1 ~CR:::::-;~7-~-,:e:-:ag~-4--::SM~~~-+-7A~CM:-;D~2~~~--~5~----~0-- 5 4 1 00 ¡ CR~1 e-243 OW28J-l ACMD-PT j 34 6 28 21 í 75 CR-52A e-243 SM~~ ACMD2 i 94 1 10 84 1 43 ! 51 l CR~ e-243 MTA,.:.:;!,.::.-8 f---':A-'-::CM:::-:-::D..;:-1 L., ---:-9:-:.--- O 9 j 7 ; 78 t---::e=-R~-=-=,.-+---:-M-:::1:=-c'A,-.,.1 -=-8--4--:::C ~ ACMD 1 110 - :fl 9J ! 70 78 1 ¡.-..:::CR:==:~;:--.J---,M~BRA;::::..;;lA:-l--;:;e;:;;-l8~:-+~A.;:;;CM:-=;D;¿:1'RT~:;:--'~'--- -'=:7;;:;- ~ -0 - 7 5 71 1 CR~ MCOL :rró e-l'Zl ACMD-mT ! 7 1 6 2 1 33 1 CR-57 MCOL zr6 e-l8 ACMD-mT ! ~ 5 7 7 1 1CO ! CR-58 MMAL f:.6 e -i8 ACMD-mT ' 11 7 4 4 j 1CO \ CR~ MTAI 2 e-l8 ACMD-mT 13 1 ~ 10 1 83 CR~ MTAI 8 C-l8 ACMD1 7 2 5 5 1 1CO CR-ú1 MTAI 8 e.:l3 ACMDl 1 4 1 2 2 O J Qo i CR-62 MTAI 8 e-as ACMDl 1 18 6 ~ 10 ! 83 ! 1DTAL t 2315 1 214 2101 1128 ¡ 54 1 Results A total of 2315 mature seeds of cassava originating from 62 F1 families corresponding to breed.ing populations for resistance to CMD, denominated "CR" were received from the breed.ing section. Of this 214 seeds were rejected by the floatation test and 2101 seeds were germinated. Of this number a total of 1128 fully formed plants were recovered and multiplied, a little more 54% recovery. The low % recovery of plants, down from the normal 80% was dueto a very severe attack of green mites in the growth room this year. Plants that are observed to be contaminated with green mites are immed.iately removed and destroyed. A concerted effort of weekly application of acaricide in the growth room, elimination of infected plants, and cleaning of individual tubes with 96% ethanol has been initiated to control the mi te infestation. A number of other cassava tissue culture facilities at CIA T also had mi te problems this year. The source of the mi te problem is not known but it is thought they may ha ve gained entrance via infested in vitro plants transferred from one facility to another. 328 One mature plant of each of the 1128 genotypes was sent to the cassava molecular marker lab for DNA extraction and marker analysis for MAS. Genotypes that have the Cl\ID2 resistance gene after marker analysis are sent to the green house for hardening and further field evaluation and to partners in Africa and India. At the moment 633 genotypes resistant to ACl\ID are being transferred to greenhouse, a total off 125 have been established to far, and also being multiplied, at least 10 copies per shipment, to make a shipment to Nigeria, Tanzania and India. Also in the green house are 96 genotypes of 3 S2 families developed under the S2 recurrent selection for tolerance to inbreeding and 47 genotypes of inter-specific crosses established last year from immature sexual seeds and propagated in vitro this year. These immature seeds are from plants that had to be removed from the field due to the "zero-cassava" rule of no plants for one month on the experimental station at Palmira. Other activities in the tissue culture section this year include Embryo rescue of 173 and 123 seeds respectively from 2 St families (AM313 and AM320) of cassava for molecular mapping of cyanogenic potential in collaboration with the Swedish Agricultura! University (SLU), Uppsala, Conclusions The establishment CMD resistance breeding populations at CIAT for molecular márker- assisted breeding of resistance has become routine. Although a severe attack of mites reduced considerable the% recovery of plants, more than 1000 plants ha ve been processed this year, compared to about less than 500 Iast year. Future perspectives are to tackle the problems discovered this year with mite infestation in the growth rooms, as well as the current size of the growth room and raise this number to at least 5000 plants every year. References Okogbenin E. (2003). QTL mapping of early root yield, morphology and root quality in cassava. Ph.D dissertation submitted to the University of Ibadan, Ibadan, Nigeria. 2.2.8 Dissemination of lmproved Cassava Varieties as Tissue Culture plantlets Adriana Alzate, Luis Guillermo Santos, Bernardo Os pina, Heman Ceballos, Martin Fregene CIA T Introduction CIA T cassava project has responsibility to malee available improved germplasm to partners in the NARs in Asia, Latín America and Africa. The cassava tissue culture facility was therefore used to propagate clean materials of improved varieties for shipment to NARs in Latin America and Africa on request from partners. The facility has also been used to clean-up and transfer into in vitro field plants from the field that have were requested for by partners. An example is the shipment to Vietnam of 57 genotypes selected from a diallel experiment for further evaluation in Vietnam. We present a report of shipments of plants in vitro of cassava made to several countries from February to September of the 2003. Methodology In vitro plants of the improved varieties were received from the genetic resources unit (GRU) and multiplied using 4E media according to standard protocols established at CIAT (Roca and Mroginski 1991). The method utilized for clean-up and transfer of field plants to tissue culture is that routinely used at CIAT with sorne modifications, in sumrnary: stems with apices and nodes are obtained from plants in the field or green house, the leaves are removed leaving approximately 5mm of petiole. The stem cuttings are washed with sufficient tap water and cut into nodes and apices, with one node per piece. In the flow hood, the fragments are placed into a 250ml sterile flask and washed with 70% ethanol for 30 seconds with vigorous shaking. The ethanol is discarded and the washed again with 10% sodium hypochlorite (0.5% v/v sodium hypochlorite) and one drop of liquid soap for 5 minutes, with vigorous shaking. The hypochlorite solution is discarded and the cuttings washed 3X with sterile double distilled water. The cuttings are then transferred to a sterile petri dish and, using a sharp sterile scapel, the extremes that have been in contact with the solutions are eliminated. The stem cuttings are then planted in 4E medium and placed in a growth room at 28oC with photo-period of 12h light and 12h hours darkness to develop into full plants. 330 Table l. Summary of shipments of in vit.ro plants of cassava (Manihot esculenta Crantz) made to several countries from February to September 2003. ICOUNTRY/ DATE OF SHIPMENT GENOTYPE HAITI JPERU CUBA NIGERIA NICARAGUA j_INDIA Re p. DOMINICANA Miy6/03 11Vey7/03 1;~ Jun 11/03 Jul18/03 J§_~ 2/03 ~eQ 10103 -1 - ----·--'-:#"""'P=lants lf Plants : # Plants - # Plants # Plants !# Plants # P:..lan:..::t:.;s :=- -- 1 ·BRA383 30 '30 ~ 5 1- 30 110 120 2 CM1335-4 - • - - a - .10 - 3 CM2'n2-3 !15 5 15 a 30 --+-'-!1..:::..0 ___ -4=20.:: ______ 1 4 CM3306-4 30 30 .5 - 30 10 20 ~ lg~~~~ ~ - ·--~-~i- --t --~-. J+~ - io-------·-- 7 ICM4843-1 •30 30 5 a ~---b_Q -~ ------ a CM489-1 '- ·- - - 1- :10 •-fo-!~~~i--~--- ..... t~ ··---~}- t~ --· ---~-----· .. ·-··f~~ - ---- ~---- ------ 11 CM523-7 130 30 ,5 ! - 30 10 20 L2 ¡ cM5306-~. J 30 ¡30 _ 15 . ¡a ... -. ~ ¡ 1_0 20 H-lg~::~4 i;6 -- !~--- ---~-~-- ---.. ¡~-------~-- ····-- --~-~-ó- --- -·~ -- -----··-- ~~ ;g~~~:~ .~ ~ ~ · ~ ¡~ ~~---~:-~-------1 17 ,CM6921-3 30 ~ _ ;~. la 130 :10 _ L20 _ ____ _ 1a CM7033-3 130 1a 5 ia -- ¡30 '10 120 19 CM7on."i~- - +3o ---is ia -- ---~ -- 1o-- ----· ~'2o ______ _ 20 CM7514-a 30 30 '5 a 30 10 20 21 CM7951-5 30 30 i5 ¡a 30 10 20 ~ ~~2?:.~-~---1~º-.. ---~~ ....... : ... t~ ----·--· ªº---------· }§---····· ~----·--------- 24 ¡tvCOL 1468 1- 1 - i - ' 6 - i10 - 25 ¡rvcoL 1684 1- _J:_ i - ,_ 1- 10 - tvCOL 20631 : 1 , 1 26 tSecundina) - - .. ! - ' - ' 5 27 tvCOL 1734 - ·• !. • - 10 2a tvCOL 2215 • - 1- a - 10 ~ ~~~;:~\-~ ~ - 11 5~ ~ - ·- ~ ~~ --~-- -----------~ 31 SM 1433-4 30 30 a 30 10 20 32 SM1460-1 30 !30 5 -~ -1: _ ---~----~ 20 * ~~~~:~5 ~ - -]30 --- ~ .1_aa - ¡30- - r1100 - ~ 35 SM 1821-7 30 --t;o 5 1 ~ 20 36 SM2019-9 30 5 1 - !- 37_l!SMa05-15 130_ j 30 ~ 5 ~ SM 9<}9-2~ _ _Jl' 30 - 130 .. ·15 39 TAl-a 30 Í30 15 1 - T~ TOTAL 1795 ¡716 130 10 110 ]10 1370 20 20 20 1525 Table 2. List of plants rescued from tbe greenhouse and placed in vitro for shipment to Vietnam (Piants for Ms Cach) Genotype IGenotype IGenotype Genotype 1 Genotype 1 Genotype CM 9106-18 CM9923-30 CM9954-18 GM237-22 GM255-2 GM281-24 CM9106-7 CM9926-17 CM9954-23 GM238-29 GM258-2 GM281-28 CM9148-2 CM9945-22 CM9957-1 GM246-25 GM258-3 GM287-10 CM9148-3 CM9945-27 CM 9957-21 GM246-3 GM259-3 GM287-13 CM9703-17 CM 9946-11 CM9958-1 GM247-27 GM266-24 GM289-9 CM9703-24 CM9946-12 CM9958-6 GM248-26 GM272-1 GM 291-11 CM9907-1 CM 9949-1 CM9966-27 GM250-24 GM272-4 GM291-7 CM9907-3 CM9949-25 GM236-26 GM250-29 GM273-13 CM9921-25 CM 9952-1 IGM236-7 GM251-12 GM280-15 CM 9923-1 CM 9952-19 IGM 237-13 GM251-9 GM280-24 Results A total of 3506 plants from a list of 39 improved genotypes were shipped to collaborators in Haití, Peru, Cuba, Nigeria, Nicaragua, India and the Dominican Republic this year (Table 1). Aside from this, 5 plants of 44 genotypes were also shipped to Crop Research Institute (CRI), Kumasi Ghana, for the Ph.D. study of Ms Elizabeth Okai who is looking for heterotic pattems between germplasm from Africa and Latín American based on SSR marker clustering. Ten plants each of a selection of 57 genotypes from a diallel study was also shipped to Vietnam for the Ph.D. study of Ms Cach who was at CIAT in 2002 for 6 months analyzing the same diallel experiment (Table 2). Stakes from the 57 genotypes in the field were planted in the sreen house, disinfected and nodal cutting established in vitro in 4E media. Conclusions The cassava tissue culture facility is being used to share valuable germplasm with collaborators all over the world within a reasonable period of time, germplasm is shipped within 2 months from when the request is received, smaller requests of 5-10 plants per genotype take a month while larger ones take up to 2 months. Future perspectives include an enlargement of thee growth room to allow for the simultaneous processing of several shipment requests. References Roca WM; Mroginski LA. 1991. Cultivo de tejidos en la agricultura: Fundamentos y aplicaciones. [Tissue culture in agriculture: principies and applications.] Centro Internacional de Agricultura Tropical (CIAT), Cali, Colombia. 970 p. 332 2.2.9 A Simple Method for the Rapid Multiplication of Clean Cassava Planting Material Dr James George1 ; Elizabeth Okai2, Emmanuel Okogbenin2 ; Chris Okeke3 Dr Acheampong4 Bernardo Ospinas Martín Fregene6 1CfCRI, Trivandrum, India; 2CRI, Ghana, 3NSM, Nigeria; 4University of Legon, Ghana, sCLA YUCA, 6CIA T Introduction Good quality and healthy planting material is crucial for high yields in cassava. Cuttings obtained from diseased and/or pest infested plants can reduce yield by 30% to 80% (Guritno1985). Another problem with unhealthy planting material is sprouting. The ability of stakes to sprout is closely related to their starch content at planting and growth during the first 20 days after plating is exclusively at the expense of the nutritive reserve previously accumulated in the stakes (Molina et. al. 1995). Starch content of stakes is reduced with poor soil fertility and disease/pest attack. A 20% drop in sprouting and a yield loss of 10% due to the use of poor quality planting material in a 2000ha farm translates to a loss of 8,400 metric ton of fresh cassava roots, at the national average yield of 14tlhectare. Ata conservative estímate of US$50/ton this is a loss of US$0.42 million. Such a loss quickly wipes out the profit margin and puts the entire venture at risk. As part of its assistance package to the Nigerian commercial cassava production company, the Nigeria Starch Mills (N S M), CIA T and CLA YUCA agreed to pro vide technical know- how on rapid multiplication of clean planting material. In Nigeria, commercially available cuttings are often taken from plants that are diseased, damaged by insect and pests and inadequately fertilized with a consequent reduction in yield (Okeke 1994, Ospina and Fregene, Personal observation 2003). Tissue culture of cassava meristematic tissue has been successfully applied to obtain in vitro cassava plants that are disease- and pest-free (Roca et al. 1991). The technology also permits the mass production of in vitro plants compared to traditional multiplication methods. A multiplication ratio of 100 is possible using new efficient multiplication systems, such as the automated temporary immersion systems (ATIS) or "bioreactors". Apart from the production of healthy planting materials , rapid tissue culture multiplication can be used to mass-produce and deploy a new promising variety over a short period. However, the initial high start-up costs of the automated ternporary immersion system and support tissue culture facilities increases the costs of establishing a commercial farming operation and diverts funds from other much need activities, making it less attractive compared to other multiplication schemes. To reduce costs, a cheaper tissue-culture based rapid multiplication scheme that combines an initial step of tissue culture muitiplication, hardening, 4 months of field growth, then a 2-node multiplication scheme in a special nursery, as practiced at CfCRI, Trivandrum, India, followed by another cycle of field growth and multiplication in the special nursery was proposed to NSM. The rapid multiplication scheme is expected to produce 2-3 million plants by July 2004, with enough cuttings to plant a 200ha "seed bank" to be used as a source for good quality healthy planting material. The scheme can also become a commercial source of planting material for sale to other large scale multiplication projects, for example the pre-emptive management of the cassava mosaic disease (CMD) projects that is being initiated in Nigeria by the Government of Nigeria, NDDC, sorne oil companies, and USAID in collaboration with liT A, Ibadan .. Specific objectives were: Tissue culture rapid multiplication of 44 improved liT A varieties to produce at least 200 plantlets per genotype Hardening of the plants in the green house and transfer to the field Harvest of 2-node cuttings at 4 months after planting from the field and transfer to the rapid multiplication nursery; transfer of plants from the nursery to the field after one month. Harvest of 2 node cuttings from the new plants at 4 months after planting and transfer to nursery; transfer of plants from the nursery to the field after one month Methodology Five stakes from 44 liTA improved cassava genotypes were obtained from Alfred Dixon, liT A, Ibadan, for tissue culture multiplication. A list of the improved lines is shown in Table l. A request to multiply the materials at the tissue culture facility of the University of Legon, Acera was granted and multiplication began first week of July. The five stakes were planted in plastic bags and after 2 weeks, meristematic or nodal cuttings was harvested, cleaned with sodium hypochlorite and cultured in 112 MS media supplemented with BAP and GA (4E media). After 4, 8, and 12 weeks after planting the plantlets were subcloned to obtain the target of 200-250 plants per genotype. Hardening of the plants will be in the NSM screen house (65% shade), currently under construction, using a soil mixture of 3 parts sand and 1 part top-soil in black plastic bags. Tissue culture plants received will be gently removed from the glass tubes and placed in the plastic bags with soil, a fungicide will be applied to control fungus growth and plants fertilized with a commercial fertilizer rich in phosphorus. A stryofoam cup with holes will be used to provide the plantlets with high humidity. After one month in the screen house, the plants will be transferred to the field and watered regularly. About 500kg of NPK 15:15:15 fertilizer will be added in split applications at 1 and 2 month after transplanting to the field in bands. 334 After 4 months of growth, 2-node cuttings will be obtained from each plant using a sharp knife and transferred to the special covered nursery. The plants will be watered regularly and at 1 month after planting they will be transferred to the field. The special nursery is simply a canopy of a mesh that allows in only 65% of light while excluding the rest over a nursery bed. The shade of the canopy, combined with constant watering provides an area of high humidity adequate for the fast growth of cassava. A picture of the special nursery is shown in Figure l. During this period, special care is taken to rouge out diseased plants and to keep the area free of pests by spraying appropriate pesticides. After the plants multiplied in the special nursery have grown, another round of multiplication is carried out as described above, this is the final round of multiplication. The plants are then allowed to grow until maturity in the field and used as a seed bank of planting material for storage root production Results Ten to 20 plantlets of 44 improved varieties from llT A were established from nodal cuttings obtained from 2 week old potted plants in tissue culture (4E media) at the University of Legon. After 4 weeks, the plantlets were subcloned to obtain a 3 to 5 time multiplication of the original plantlets. Two further rounds of multiplication were conducted at 8 and twelve weeks to give between 200 and 250 plants per genotype. The plants are ready for green house hardening and are awaiting the completion of the NSM green house facility under construction at Ihiala, Nigeria. In the mean time an import permit to bring the plants from Ghana into Nigeria has been applied for via the Nigerian Plant Quarantine office and a phyto-sanitary certificate has been requested from the Ghanian authorities to ship the plants to Nigeria. Table l . List of 44 improved new cassava varieties that ha ve been multiplied by tissue culture Cassava mosaic Bacteria! Root CNP(mgHCN/ Rootdry Fresh root flesh Genotype disease blight mealiness IOOg fresh root Matter(%) yield (liba) color l. 92JOOS1 Resislant Resislant Mea! y Medium 30 25-30 White 2. 928100068 Resistant Resislant Mea! y Medium 32 30-3S White 3. 9210326 Resislant Resislant Mea! y Low 30 30-3S White 4. 93/0098 Resistant Resistant Mealy Medium 30 32-3S White S. 9210325 Resislant Resistant Mea! y Low 3S 20-25 White 6. 97/0162 ResisWit Resistant Mea! y Low 30 30-3S White 7. 97/4769 Resislant Resistant Mea! y Low 30 30-3S White 8. M9810028 Resistant Resistant Mea! y Medium 30 30-3S White 9. 98/0SOS Resistant Resistant Mealy Medium 3S 30-40 White 10. 98/0S!O Resistant Resistant Mod. mealy Medium 3S 40-4S White 11. 991!S90 Resistant Resistant Mod. mealy Low 3S 25-32 White 12. 99/6()12 Resistan! Resistant Mod. mealy Low 3S 3S-40 White 13. M9810040 Resislant Resislant Mod. mealy Low 32 40-4S White 14. 91/02324 Resistant Resistant Non-mea! y Medium 3S 3S4S White !S. 9210067 Resislant Resislant Non-mea! y Medium 30 25-30 White 16. 928100061 Resistant Resistant Non-mea! y Medium 30 30-3S White 17. 9410561 Resistant Resistant Non-mea! y Medium 30 30-3S Yellow 18. 9410026 Resistant Resistant Non-mealy Medium 32 30-3S White 19. 9410039 Resistant Resistant Non-mealy Medium 32 30-40 White 20. 9S/0!66 Resistant Resistant Non-mealy Medium 30 3S-40 White 21. 9SI0379 Resistant Resistant Non-mea! y Medium 30 30-3S Yellow 22. 9SI0289 Resistant Resistant Non-mealy Medium 32 30-3S White 23. 96/!S6S Resistant Resistant Non-rnealy Medium 30 3S-40 White 24. 96/1089A Resistant Resistant Non-mealy Medium 32 30-3S White 25. 9ÓI0603 Resistant Resistant Non-mealy Medium 30 30-3S White 26. 96/1642 Resistant Resistant Non-mealy Medium 30 30-3S White 27. 9713200 Resistan! Resistant Non-mealy Medium 32 3S-40 White 28. 91n2os Resistant Resistant Non-mealy Medium 30 30-3S White 29. 97/4763 Resistant Resistant Non-mealy Medium 32 3S-40 White 30. 9812226 Resistan! Resistan! Non-mea! y Medium 30 30-3S White 31. 9810002 Resistan! Resistant Non-mea! y Low 3S 40-4S White 32. 97/0211 Resistant Resistant Non-mea! y Medium 30 30-3S White 33. 961!S69 Resistant Resistant Non-mealy Medium 30 30-3S White 34. 96/1632 Resistant Resistant Non-mealy Medium 40 3S4S White 3S. 97/4779 Resistan! Resistant Non-mealy Medium 30 30-3S White 36. Z9710207 Resistant Resistant Non-mea! y Low 30 3S-40 Whitc 37. 98/0581 Resistant Resistant Non-mealy Medium 30 30-32 Whitc 38. 9812101 Resistant Resistant Non-rnealy Medium 30 30-3S Whitc 39. M98/0068 Resistant Resistant Non-mealy Medium 3S 40-4S Whitc 40. 99/1903 Resistant Resistant Non-mealy Medium 39 40-4S Whitc 41. 99n123 Resistant Resistant Non-mealy Low 3S 30-32 White 42. 9913073 Resistant Resistant Non-mealy Medium 30 30-3S Whitc 43. 9610523 Resistant ResisWit Non-mealy Medium 30 3S-40 White 44. 96/1317 Resistant Resistant Non-mealy Medium 32 30-3S Yellow Conclusions A rapid multiplication of clean improved planting material has been embarked upon for the Nigeria Starch Mili (NSM), if successful it could be a very rapid method for the production of clean planting material for both the medium and large scale cassava production sector in Nigeria as well as small scale rural farming. Although the scheme is still in progress, expected outputs is a 400X multiplication within a single growing season compared to the 336 tradüional lOX multíplicatíon or lOOX vía tíssue culture alone or 200X vía 200X vía 2- node cutting alone. References Cock J.H. 1985. Cassava: New Potential for a Neglected Crop. Westview Boulder, CO Guritno B. Influence of planting material on plant performance in cassava. Ph.D. thesis, University of Brawijaya, Malang, Indonesia, 158pp. Molina J.L. and El-Sharkawy M.A. 1995. Incresing crop productivity of cassava by fertilizing production of planting material. Field Crops Research 44:151-157 Okeke J.E. Productivity and yield stability in cassava (Manihot esculenta Crantz) as affected by stake weight. J. Agríe. Sci., Cambridge, 122:61 -66 Roca WM; Mroginski LA. 1991. Cultivo de tejidos en la agricultura: Fundamentos y aplicaciones. [Tissue culture in agriculture: principies and applications.] Center Internacional de Agricultura Tropical (CIAT), Cali, Colombia. 970 p. 2.2.10 Temporary Immersion System (RITA) for Anther Culture ofRice M. Quintero2, E. Tabares 1, R .Escobar 1, G. Delgado 2, Z. Lentini I.Z 1SB2 2IP4 ' Introduction Plant in vitro culture usíng temporary immersíon (RITA) offers all the advantages of a liquid medium system (automation, Iarge scale production, easy changes of medium, filter sterilization, easy cleaning) without any of its drawbacks (reduce gas exchange, vitrification). lmmersion time, i.e. duration or frequency, is the most decisive parameter for system efficiency. The optimization of the nutrient medium volume and the container volume also substantially improves efficacy, especially for shoot proliferation. Severa! reports confirmed large gains in efficacy from temporary immersion when using liquid medium for micro propagation. The main parameters involved reducing production costs are, frrstly the drastic reduction of work labor, followed by a reduction in shelving area requirement and in the number of containers used. Scaling up the use of temporary immersions for embryogenesis and shoot proliferation procedures are currently taking place in order to commercialize this process (Berthouly & Etienne, 2002). This system has preved its efficacy for somatic embryogenesis of banana (Alvard et al, 1993; Escalant et al, 1994), coffee (Berthouly et al, 1995; Etienne et al, 1997), citrus (Cabasson et al, 1997), oil palm and rubber plant (Etienne et al, 1997), and at CIA T for cassava (Escobar and Roca, 1999). High efficiency has also been demonstrated for clona] propagation through micro- cuttings of coffee, and sugar cane (Lorenzo et al, 1998); for proliferation of meristems of banana, and pineapple, and for micro-tuberization of potato (Teisson and Al varad, 1998). We ha ve previously reported preliminary results using RITA for the induction of embryogenic callus derived rice from mature zygotic embryos (Tabares et al., CIAT SB2 Report 2000) and from anther culture (Tabares et al., CIAT SB2 Report 2001). This year we reporta comparative analysis including various indica and japonica rice genotypes. Materials and methods Anther culture of the indica rice Cica 8, PN1, Cimarron, Fedearroz 2000, and CT 11275, and of the japonica breeding line CT 6241-17-1-5-1 were used. Donor plants were grown in the field, panicles harvested, and anthers cultured according to Lentini et al. (1995). Tissues were either culture in liquid medium contained in RITA vessels or in liquid medium in baby food jars (control). lnduced callus from each culture system was then transfer onto solid plant regeneration medium according to Lentini et al. (1995). The effect of different culture media was evaluated. The medium NL commonly used in the rice anther culture laboratory (Lentini et al, 1995) was used as control. This medium was supplemented with 2,4-D 2mg/L; picloram 0.07mg/L; k.inetin 1mg/L; maltose 8%; with or without sil ver nitrate10mg/L. A modified medium was evaluated consisting of NL basal medium but replacing picloram with phenyl acetic acid (PAA) 10 mg/L, with or without silver nitrate (medium M 1). The rest of the culture procedure, including plant regeneration, was according to Lentini et al. (1995). The optimal immersion frequency was determined by evaluating callus induction and embryogenesis at 3 different immersion frequencies. Treatments were conducted using immersions of 1 min every 4, 6, or 8 hr for a total of 4 weeks of culture. Three RITA vessels were used per genotype with 1,000 anthers per 200 ml medium; and 4 baby food jars (permanent immersion system, PIS) per genotype each with 250 anthers per 10 ml culture medium (control). Cultures were incubated at 24 a 26°C. Three different treatments were tested to increase plant regeneration. A water stress treatment was induced on callus prior culture by incubation on 1% agarose-containing medium in the dark at 27°C to dehydrate callus. After two weeks of culture, stressed callus from 1% agarose-containing medium were transferred to 0.4% agarose solid medium for regeneration and incubated in light. Another set of callus was not treated with water stress, and was cultured directly on the medium semi-solidified with 0.4% agarose. The effect of osmotic stress was evaluated by sub-cultured on medium containing 3% sorbitol for 24 hr, after this partial desiccation treatment the callus were transferred on regular plant regeneration medium. Control consisted of callus transferred from callus induction medium without treatment to regular MS regeneration medium. A factorial experimental completely randomized design was used. At least 10 replicates of 10 callus each was evaluated per treatment. 338 Results and Discussion A significant higher callus induction was obtained when immersion was conducted every 6 hr than every 4 hr and 8 hr independently of the genotype (Figure lA). The japonica line cr 6241 showed about 7 fold more callus respect to the indica varieties (Figure lB). The maximum callus induction was noted at 40 days after culture on PIS and at 50 days after culture on RITA. But it seems the slower process of callus induction in RITA allows an optimal induction of embryogenesis. A significant higher number of embryogenic callus (95%) were obtained for both indica and japonica genotypes respect to the permanent immersion system (PIS, 45%) (Figure 2). There was an interaction between the callus induction medium and the culture vessel used. No significant differences were seen between the different media on RITA, although there was a tendency of higher induction when using media M 1 with or without silver nitrate. However with PIS, the indica genotypes with intermediate to low response such as Cfll275, Cimarron and Cica 8 showed significant higher induction on M1 medium with or without silver nitrate, but Fedearroz 2000 optimal induction was noted on NL medium with sil ver nitrate. It has been reported that PAA mode of action is similar to that of IAA, although higher levels of PAA are needed and it is more stable in culture inducing a larger number of organized structures. It seems PAA effect is related to the inhibition of ethylene production from the cultured tissues (Ziauddin et al., 1992). Silver nitrate has also been reported to inhibit the action of ethylene of tissues culture in vitro (Lentini et al., 1995). The beneficia! effects of replacing sucrose by maltose increasing the callus induction from rice anther of recalcitrant genotypes has also been associated with a reduction of ethylene effects. These results j ointly with the effects noted when using PAA and RITA in this work suggest that ethylene might be a critica! factor detennining the induction of androgenesis from microspores in rice. Plant regeneration from embryogenic callus was not affected by the composition of the callus induction medium nor the callus induction culture vessel, indicating that the regeneration capacity depends on the leve! of embryogenesis. Once embryogenesis is obtained the capacity for plant differentiation is similar. Thus, the optimal callus induction medium and culture vessel should be selected based on the larger number of embryogenic callus produced per anther cultured. Significant higher green plant regeneration was obtained when water stress treatment was applied using agarose 1% for 1 week followed by agarose 0.4% for the rest of the culture period (Figure 3). Twice as many green plants were obtained with this treatment respect to the control. Osmotic stress with sorbitol inhibited plant regeneration (Figure 2). Independently of the callus induction or plant regeneration treatments, about 50-60% of the green plants were doubled haploids, which is in the range previously reported (Lentini et al., 1995). (A) (a) f_ 1 - Different letter denote statistical si¡nificanc:e (P < 0.0.5). Duncall's Mulliple Range Test Figure l.Callus induction in RITA system using 4 hr, 6 hr, and 8 hr temporary immersion frequencies. (A) Mean values of recalcitrant indica varieties. (B) Comparison between different genotypes. Values refer to the mean number of callus induced per 1,000 cultured anthers. ltt M l!l 3 .. u li o~ 'C .. .&J ~ M •c-.~ •- DfW a"" Differcnt letter denole swistical significaoce (P < 0.0.5). Ryan- Einoi-Gabriei-Welscb Multiple Ran¡e Test Figure 2. Percentage of callus embryogenesis on permanent immersion system (PIS) and RITA at different immersion frequencies (8 hr, 6 hr, and 4 hr, respectively). 340 2 3 4 Different lettcr denote statistical significance (P < 0.05). Ryan- Einot- Gabriel-Welsch Multiple Range Test Figure 3. Percentage of green plant regeneration on medium containing (1) agarose 1% for 1 week and then agarose 0.4%; (2) agarose 0.4%; (3) sorbitol; and ( 4) control Future activities • To study systernatically different factors affecting the emission and action of ethylene on in vitro culture • To test different PAA concentrations and its interaction with maltose level for optimizing androgenesis in rice • To evaluate modifications of culture vessel allowing aeration andlor temporary immersion for reducing current cost for the implementation of the RITA system References Alvarado, D.; Cote, F.; Teisson, C. (1993). Comparison of methods of liquid mediurn culture for banana micropropagation. Plant Cell Tiss. Org. Cult. 32: 55-60 Berthouly, M.; Dufour, M. ; Alvarad, D.; Carasco, C.; Alemanno, L.; Teisson, C. (1995). Coffee micropropagation in liquid medium using temporary imrnersion technique. ASIC, Kyoto, Vol JI: 514-519C. Cabasson, D.; Alvarad, D.; Dambier, D.; Ollitrault, P.; Teisson, C. 1997. Improvement of Citrus somatic embryo development by temporary imrnersion. Plant Cell Tiss. Org. Cult. 1-5 Escalant, J.V.; Teisson, C.; Cote, F. (1994). Amplified somatic embryogenesis from male flowers triploid banana and plantain (Musa sp.). In Vitro Cell. Dev. Biol. 30: 181-186 Escobar, R. H.; Roca, W. M. (1999). Cassava micropropagation for rapid "seed" production using temporary immersion bioreactors. CIAT Annual Report Project SB-02 p 84-86 Etiene, H.; Lartaud, M.; Michaux-Ferriére, N.; Carron, M.P.; Berthouly, M.; Teisson, C. (1997). Improvement of somatic embryogenesis in Hevea brasiliensis (Müll. Arg) using the temporary immersion technique. In Vitro Cell Dev. Biol. 33:81-87 Khanna, H. y Raina S. (1998). Genotype x culture media interaction effects on regeneration response of the indica rice cultivars. Plant Cell Tissue and Organ Culture 52: 143-145 Lentini, Z.; Reyes, P.; Martíriez, C.; y Roca, W. (1995). ~drogenesis of highly recalcitrant rice genotypes with maltose and silver nitrate. Plant Science 110: 127-138 Lorenzo, J.C.; González, B.; Escalona, M.; Teisson, C. ; Espinosa, P .; Borroto, C. (1998). Sugarcane shoots formation in an improved temporary immersion system. Plant Cell Tiss. Org. Cult. 54: 197-200 Murashige T.; Y Skoog F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol. Plant. 15: 473 - 497 Teisson, C. & Alvarad, D. (1998). In vitro propagation of potato micro tubers in liquid medium using temporary immersion. Conf. Patato Seed Production by Tissue Culture, Brussels, Cost 822, European Commission. Tabares, E.; Delgado, G.; Escobar, R.; Lentiní, z. (2001). Temporary immersion system (RITA) for anther culture of rice. Annual Report, 200 l. CIA T, Cali-Colombia. pp 19 - 20. Tabares, E.; Delgado, G.; Escobar, R.; Lentini, Z. (2002). Osmotic stress as a tool to enhance plant regeneration in rice. Annual Report, 2002. CIAT, Cali-Colombia, pp. 55- 57 Tsukahara and Hirosawa (1992). Retention and reviva! of regenerating ability by osmotic adjustment in Iong-term culture of four varieties of rice. Plant Cell Reports. Vol 11, pp. 550-553 Ziauddin, A.; Marsolais, A.; Simion, E.; and Kasha, K.J. (1992). Improved plant regeneration from wheat anther and barley microspore culture using phenyl acetic acid (PAA). Plant Cell Report Vol. 11, pp.489-498 342 Activity 2.3 ldentification of points of genetic intervention and mechanism of plant stress interation 2.3.1 Exploring the genetic potential to improve micronutrient content of cassava C. Cardenas3, A. L. Chávez\ T. Sánchez2, N. Morante2, J. Tohrne1, M. Ishitani1 and H. Ceballos2 1SB-2 Project; 2IP-3 Project, 3Universidad del Valle Introduction The overall objective of this project is to improve the nutritional status of people living in marginal environments of the tropics, by selecting and promoting cassava genotypes with high and good bio-availability of micronutrients and vitarnins. Related traits are the need for a better understanding of the biochemical and genetic basis of post-harvest physiological deterioration (PPD). Early work was conducted to demonstrate the existence of genetic variability in cassava for carotene content in roots and foliage and to measure it. Carotene contents in the roots ranged from 0.102 to 1.040 mg/100 g fresh tissue (Ff), whereas in the foliage the figures were much larger: 12.05 to 96.42 mg/100 g Fr (CIAT 1999; 2000; 2001). PPD is apparently reduced or delayed in roots with high carotene content. Correlations between carotene content in the roots and severa! agronomic and nutritional traits were estimated, with not relevant information except for the effect on PPD. The current research focus shifted towards better understanding the nature of the carotenes accumulated in cassava roots and its stability after .common processing and storage. This report summarizes relevant information with the introduction of a new methodology (HPLC) for measuring carotene content in cassava roots more precise! y. 2.3.2 Stability of carotene content after alternative storage methodologies for overcoming the problem of Post-Harvest Physiological Deterioration (PPD) Introduction The short post-harvest storage life of cassava is a characteristic that limits the marketability of the root and requires either consumption or processing shortly after harvesting. Post- harvest physiological deterioration (PPD) of cassava roots begins within 24 hours of harvest, and results in crop and product quality Iosses, high marketing margins and risks, and restricted management flexibility for farmers, traders and processors. The reduction of PPD has been identified as a priority target for strategic research. It would also benefit processors in Brazil, who face price discounts for samples processed even 48 hours after harvest, dueto grayer and lower quality starch than that derived from fresher roots. In many respects, PPD resembles wound responses found in other better studied plant systems but cassava appears to lack the wound healing capacity which is normally associated with the inhibition of wounding responses. Normally, such defensive wound responses are inhibited by wound repair. However, this repair process does not occur in the harvested cassava storage root, leading to the hypothesis that unrestrained cascarles of wound responses ultimately result in deterioration (Oirschot, Q., 2000). An important component of these wound responses are the oxidati ve processes. Ascorbic acid and carotene a¡e known to have antioxidant properties and seem to delay the onset of PPD (CIAT, 1999; 2000; 2001). Because of the rapid deterioration of cassava roots, they cannot be stored until they can be anal yzed for caro ten e con ten t. Therefore, onl y a few samples ( < 40) can be anal yzed in one day, with the available facilities at CIAT. A study was conducted to analyze if carotene contents change after different storage conditions for varying periods of time. Two different storage conditions were evaluated: storage at -80 oc and at- 20 °C. Results Storage at -80 °C.. Stability of carotene content was measured on root samples from the clon MCOL 2508 by the colorimetric and HPLC methods on fresh roots. This information was used as a check for data obtained after different periods of storage at - 80 °C. Samples of roots were stored for 22, 73, 84, 96, 126, 210 and 230 days and then, carotene contents evaluated using the same measuring methods. As it was usually the case, total carotene content measured by HPLC was slightly higher than when measured by the colorimetric method (1.07 vs. 0 .88 mg 1 100 g fresh tissue). Samples of roots analyzed by the colorimetric method averaged 0.88 mg across the different dates (the same value as the original samples) with a standard deviation of 0.038. Values ranged from 0.81 to 0.92 mg 1 100 g fresh tissue. Measurements based on HPLC showed larger variation. The average content was 1.01 mg 1 100 g fresh tissue, with a standard deviation of 0.156. Individual measurements ranged from 0.81 to 1.26 mg. The larger error associated with HPLC coincides with the findings reported in section 3.1. There was no clear trend suggesting that carotenes measurements will decrease or increase as a consequence of storage. Therefore, it is valid to conclude that root samples can be stored for as long as seven months until carotenes are finally measured. The logistic advantages of this conclusion are very important for different projects involving cassava and carotenoid compounds. UOy---------------------------""""'1 .. :::J .. .. 1.20 ;¡ o ~ 0.10 Q .§. .. o.eo e S e 8 0 . .00 Cl e Cl o 0.20 .. 11 (.) Measurement by HPLC Measurement by colorimetry - .O - Total.cdorimelry ----Totai·HPLC ...:neal (TOial-colorimally) --ou·neal (Totai·HPLC) o.oo~--------------------------1 o 100 1!10 Days under storage Figure S. Stability of carotene content measured on root samples from tbe clon MCOL 2508. Original Jevels of carotenes were measured by the colorimetric (A) and HPLC (B) methods on fresh roots. Samples of roots stored for 22 to 230 days at - 80 oc were taken and carotene contents evaluated using the same measuring methods. Storage at -20 °C. In a separate experiment roots from 20 different clones were analyzed by colorimetry soon after harvest and then stored for six months at -20 °C. The root samples were then recovered and analyzed again for carotene content. Figure 6 illustrates the results from this experiment. During the storage period the roots were maintained at - 20 oc in open petri dishes. There was a differential dehydration process of root samples from different clones, depending on their position in the storage container. This may explain the obvious lack of correlation between non-stored and stored roots samples of sorne clones (MCOL2435, MCOL2508, MCOL 2266, AM 273-7 and AM 273-23). Similar results were obtained when carotenes were evaluated by the HPLC method (Figure 7). In this case, the lack of association between data from non-stored and stored root samples was found in clones MCOL 2580, MCOL 2410, MCOL 1468, and MCOL 2596. Because of the errors associated after storing roots at - 20 oc using both the colorimetric and the HPLC methods the potential of maintaining root samples at that temperature needs further confinnation under closed containers. The next evaluations should consider the potential of liophylization as well. Until then, roots samples should be maintained at -80 oc. 345 4,--------------------------------------------------------------------------, 3.5 Non-stored roots Cl o ~ 2 0 -. '()· . .§. = 1.5 e: ~ 1 111 o 0.5 o l!l ... "' ~ ~ ~ ~ ~ ... "' "' "' d ~ i ~ ~ ~ § :¡ :8 :! ~ a: ~ d w 0.. ~ ~ . Q' ' Roots stored at -20 C ... i ~ co co ~ ~ ..... .., o ..... "' do ;¡ ..; "' l ... "' do .;. ,;, ..; "' "' M ¡;; ..... ~ d ¡;; M ¡;; ¡;; "' ..... ¡;; ~ ~ ~ "' ~ ~ ~ ~ ~ ~ ~ Cl Cl Cl Cl Cl Cl Cl Clon Figure 6. Comparison of carotene content (measured by the colorimetric method) on roots from 20 clones at harvest time and then after 6 months of storage at -20 oc 5,-----------------------------------------------------------------, ~ 4.5 111 111 ; 4 ~ 'ti 3.5 Cl o ~ 3 -Cl s 2.5 -e ~ e 8 1.5 Cl e ~ 1 2 ~ 0.5 O. Roots stored at • 20 C .. . Noon-stored roots ~ o 5i § ~ ~ :; "' i a: ~ ~ ~ ~ ~ ::1 ::1 ... .... ~ ~ u ~ ' •ó ó ~ : ~ : ~ ¡:¡ .... "' l "' ;;; ~ "' ;;; ;; ;; ,;, ~ ;; ;; ;; ;; "' ;; ;; ~ ::1 ~ ~ ~ ~ ~ ~ ~ ~ ~ " Clon .., ;¡ ¡;; ~ Cl Figure 7. Comparison of carotene content (measured by the HPLC method) on roots from 20 clones at harvest time and then after 6 months of storage at -20 oc 2.3.3 Comparison of colorimetric and HPLC methods for measuring carotene content in cassava roots from 100 clones Introduction Cassava roots were harvested and analyzed for carotene content using the colorimetric and the HPLC methods as described above. The purpose of this study was to continue the screening for clones w~th high-carotene in the roots, but more important, to determine the correlation between results from both methods and to estímate the proportion of different carotenoids (alfa and beta-carotene and lutein) by HPLC. The later information is relevant because the production of vitamin A is twice as effective in ~-carotene compared with a- carotene (Combs, 1992). Methodology Sampling. Root samples of 100 clones from the germplasm or the cassava breeding project at CIAT were used for this study. Harvest took place at 9-10 months after planting (normal harvesting time for cassava at CIA T) and commercial size, disease-free roots were taken to represent each clone. Colorimetric method. The extraction procedure outlined by Safo-Katanga et al. (1984) was adjusted by extracting root parenchyma with petroleum ether .. A sample of 5 g was taken out of the root or leaves, taken at random 10 to 11 months after planting. The quantification of total carotenes was done by ultraviolet spectrophotometry using a Shimadzu UV-VIS 160A recording spectrophotometer. UV detection was done at A = 455nm for root extracts andA= 490 nm for leave extracts. HPLC method. Starting from the method used for the spectrophotometric quiantification of total carotenes, aliquots (20 ml) of petroleum extract were completely dried by rotaevapotration. Then the dry extract was dissolved in 1 ml of HPLC mobile phase (methanol:methyl-t-butylether :water, 81:15:4 v/v), centrifugated at 14000 rpm and 10 J.d were injected in the HPLC system using a YMC-C30 column (250 mm, ID:4.6rnm, Waters). Separation was done by a linear gradient elution from methanol:methyl-t- butylether:water, 81 :15:4 v/v to methanol:methyl-t-butylether :water, 20:76:4 v/v during 90 minutes at 1 ml min"1 and 23°C. ¡3-carotene was detected by monitoring absortion at 450 nm. Identificaction and quantificaction was done by comparing retention times and uv- vis spectra with a standard of 13-carotene (Sigma C-0126). Results Based on a sample of roots from 100 clones, (results peviously reported in annual report 2002) the correlation between the two methods of measurements was excellent. If 347 correlations are measured using fresh tissue data the coefficient was 0.922 (Figure 1). Correlation coefficient was slightly higher (0.935) when taken based on dry tissue data (Figure 2). Results illustrated in Figures 1 and 2 help to highlight that most cassava clones have low levels of carotene in the roots. As the total carotene concentration increases, so does the disagreement between the two measuring methodologies. 1.80,-------------------------------, u _J ... o a.. 1.20 J.: ~ 'i 1.00 :; lQ o.eo Q) E g¡ o.eo e: Q) e Cd 0.40 ü 0.20 Correlation coefficient = 0.922 • • • • • • • • Y = 1. 345x 1.2979 R2 = 0.8794 o.oo..__--------~----------------~ 0.00 0.20 0.40 o.eo o.eo 100 120 Carotenes quantfied by colorimetry Figure l. Relatiooship between caroteoes (mg/100 g FT) measured by the colorimetric and HPLC methods (samples from 100 clones, measured in three repetitioos). 7.00r-------------------------------. :: a.. J: a.oo 5.00 >- 4.00 .Ll (1) CD e: ~ 3.00 !U () 2.00 1.00 Correlation coefficient = 0.935 • • • •• • • • • • y= 0.9708x1·2338 R2 =0.8924 • o.oo.¡...._--=--~--......---~--------------------J 0 .0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Carotenes (rrg 1 100 g dry tissue) by colorirnetry Figure 2. Carot.enes (mg/100 g dry tissue) measured by the colorimetric and HPLC methods (based on sample from 100 clones evaluated in three repetitions. Data in Table 1 present the descriptive statistics to the two measuring methods. The advantage of HPLC is that it allows for the partitioning of total carotenes into its components (a- and P-carotene and lutein). Total carotene was, on average, 13% higher using the HPLC method compared with the colorimetric method. It was very interesting to observe that most of carotene extracted from cassava roots was P- carotene. On average, about 97% of carotene measured was P-carotene, ranging from 91 to 100% (Table 1). This is advantageous because of the recognized bio-efficacy of this type of carotene to be metabolized into retino! (the active form of vitaminA). The proportion of P-carotene in cassava roots is higher than that found in carrots which was found to ha ve the following proportions (Rodriguez-Amaya, 2001): a.-carotene (28.0%); p-carotene (56.6%); y-carotene(13.2%) and Lutein (2.1 %). Coefficient of variation for colorimetry averaged 7.1% but was 20.9% for measurements based on HPLC. The higher variation for HPLC measurements may be a result of the higher values generally observed for HPLC, but also, and more likely, due to the adjustments of a new methodology in the laboratory. Based on these results sorne p procedures ha ve been modified to reduce experimental error in HPLC measurements. 349 Table l. Statistical description of the measurements of carotene content in root samples from 100 cassava clones using tbe colorimetric and HPLC metbods. Colorimetry HPLC Total a-carotene ¡3-carotene Lutein Total Data based on fresh tissue (mg carotene 1 100 g of fresh tissue) Mean 0.421 0.016 0.455 0.005 0.476 St.Dev. 0.233 0.018 0.331 0.007 0.350 Maximum 1.038 0.077 1.381 0.044 1.459 Mínimum 0.141 0.000 0.083 0.000 0.083 Data based on d.ry tissue (mg carotene 1 100 g of dry tissue) Mean 1.359 0.052 1.468 0.016 1.536 St.Dev. 0.908 0.064 1.196 0.025 1.267 Maximum 4.255 0.249 5.672 0.165 5.950 Minimum 0.366 0.000 0.213 0.000 0.213 2.3.4 Stability of carotene accumulation across different locations Introduction Improving the efficiency with wich cassava acquires micronutrients and accumulates them in the roots and leaves, can have an enormous potential not only in terms of human nutrition, but also in terms of crop production. The purpose of this short study was to evaluate the relative stability of carotene content depending on the locations where the cassava clones have been grown. Results are preliminary given the small number of genotypes grown. Methodology Colorimetric method (already described above) HPLC method ( already described above) Sampling The study was conducted with three clones grown in four locations with three replications/location. Locations are localized in Llanos Orientales: Location 1 =La Libertad; Location 2=Santa Cruz; Location 3=Cumaral; Location 4= Cabuyaro. Results Because the clones had to be harvested at about the same age after planting, and taking into consideration that the trials at the different locations were planted at different dates, depending on the availability of rains, the sampling and carotene measurements were performed on roots that had been harvested at different times. Therefore, there was a confounding effect between batch for carotene measurement and location. Figure 3 shows the mean carotene content of each clone at each location. The amount of carotenes obtained for clone SM 1859-26 in the third location was unexpectedly high. Because of this only outlying data point, the genotype x environment interaction could reach statistical significance at the 5% probability level. In general, however, the performance at the other 11 data points suggests that genotype x environment interaction is not prevalent for carotene content in cassava roots .. 1 .2~-------------------------------------------------------, 1 ~--- ... _.______ ~ ........--------. - MBRA502 f go.a r o ..- o, E 0.6 -Q) - SM1363-11 . .. . ..... ~ 0.4 e 11-------11- -- .. ---11 ... as () 0.2 0+-------------~------------~-------------r------------~ Loc1 Loc2 Loc3 Loc4 Location Figure 3. Stability of carotene content (mg/100 g fresh tissue) on roots from tbree different cassava clones grown in four different Iocations in Colombia. Table 2 presents the result of the analysis of variance for this experiment. It is important to highlight that, regardless the significance of the genotype by environment interaction observed, it was not strong enough to induce cross-overs in the performances of the clones evaluated. These analyses were carried out by the HPLC methodology and results are based on the addition of a- carotene, ~- carotene and lutein. Table 2. Analysis of variance for carotene content on roots from three clones evaluated in four different Iocations with three replications per location. So urce of V ariation Degrees Freedom Mean S~uares Location 3 0.097644""' Rep (Locatiol!) 8 0.064507 Clones 2 1.325868** Clones • Location 6 0.024867* Error 16 0.007845 Total 35 0.106727 . . . . NS= Statistically non-significant. •• S1gnificant at 1% probability level. • S1gnificant at S% probability level • 351 In conclusion, no significant differences m carotenoids conteñt were found between cassava roots sampled in different areas. 2.3.5 Effect of home processing on carotenes content of cassava roots. Introduction Most of the data available on the carotenes content of foods refer to the raw materials. It is evident, however, that data relating to the form by the foods are consumed by the population are needed and the influence of processing on carotenes levels has to be determined. In addition to the quality of carotenes present in cassava their bio-efficacy will depend on the amount lost upon processing the roots. The purpose of this study is to determine losses of the vitarnin A potency during processing of cassava roots. Methodology Carotene contents were measured by HP~C on fresh roots, and after boiling, sun-drying and oven-drying the roots from the same three clones (MBRA 502, SM 1859-26 and SM 1363-11) grown and harvest at four different locations in Llanos Orientales: Location 1=La Libertad; Location 2=Santa Cruz; Location 3=Cumaral; Location 4= Cabuyaro. Processing preparations. At the laboratory, cassava roots from 2 different plants of the same genotype are peeled and prepared. 3-5 roots are divided into 5 subsamples that are processed in various ways. Boiling: The subsample (500 g) was boiled in an aluminium pot with 1 liter tapwater for 20 min. Sun dry: The subsample (500g) was dryed under sun during a period of 76 hours. Oven dry: The subsample (500g) was dryed in oven at 60 oc during 24 hours. Lyophilization : The subsample (500g) was dry by Iiophylization at -20 °C. Unprocessed: 5 g of fresh cassava root chopped in small pieces was analized (results in previous report (Stability of carotene accumulation across different locations). HPLC method ( already described above) Results Figure 4 illustrates the results (averaged across the four locations). Results demonstrate that relatively small amount of carotenes are lost upon boiling the roots (up to 55% of the originallevels recovered), whereas drying the roots by the common sun-drying process or in the oven, resulted in a recovery of 13 and 8%, respective! y. These results contrast with previous ones (previously reported, CIAT, annual report 1999) based on the colorimetric method of quantifying carotene contents (with recoveries of 60, 40 and 63% after boiling, sun- or oven-drying, respectively). It is not clear if the differences between these contrasting results are because of the changing in the method for measuring carotenes (colorimetric versus HPLC methods), or else sorne error in the process. Losses of the vitamin A potency during processing can be caused by oxidation by wich the total alpha and beta carotene content of the roots is reduced, and by isomerisation, in such case the total carotene content, quantified by colorimetric method, is not altered, but the vitamin A potency (quantified by HPLC as beta carotene content) is reduced because of the transformation of the naturally occurring all-trans isomers into the biologically less active cis-isomers. Further analyses will be conducted to corroborate or correct results so far obtained. 3~-------------------------------~------------¡ Gl ¡ 2.5 ~ ~ , 2 DI o o ... -g 1.5 111 Gl e S e as u s o 0.5 1- o Unprocessed O MBRA 502- S1 SM 1859-26 S1 1111 SM 1363-11 S1 •Average 8% Boiled Sun-dried Oven-dried Proceslng method Figure 4. Total carotenes (mg/100 g of dry tissue) adjusted for dry matter content in fresh roots and after different processing methods. Measurements were made using the HPLC methodology. 353 2.3.6 Inheritance of carotene content based on Sllines. Inheritance of carotene content in cassava has been described as relatively simple (Iglesias, et al. Euphytica 94:367-73), but not definite mode of inheritance has been clearly specified. 16 S1 lines obtained from the elite clone MTAI 8 (Rayong 60, derived from the cross between MCOL 1684 and Rayong 1) were evaluated. MTAI 8 (= Rayong 60) has roots that are almost white in color. The S1 progenies, surprisingly, exposed a large range of variation for carotene content. Measurements with HPLC were higher than those with the colorimetric method. Also, the coefficient of variation was higher in HPLC than in colorimetry. Moreover, as illustrated in Figure 8, the overestimation of HPLC (or underestimation by the colorimetric method) seems to be much higher at high values of carotene content. It is possible that the increased difference between the colorimetric and HPLC methods at high levels of carotene is due to lack of adequate fit of the standardization curves for one or both methods. This explanation makes sense because the curves were originally devised for a range of carotene content lower than 1 mg 1 100 g tissue. It is not clear yet if it is a matter of overestimation by HPLC or underestimation by colorimetry. Results presented in Table 3 clearly suggest that carotene content can indeed be increased over the natural ranges so far obtained. It is important to emphasize that the highest concentration of carotenes based on fresh tissue so far found had been 1.06 mg 1 100 g fresh tissue. Four of the sixteen S 1 clona! lines evaluated showed much higher levels of carotene concentration (around 1.70 mg 1 100 g fresh tissue). Also promising is the fact that these high values were obtained from a clon (MT Al 8 or Rayong 60) whose actual concentration of carotene was rather low (0.31 mg 1 100 g fresh tissue). Future activities • To standarize the new HPLC methodology for carotenoid compound analysis, adjusting the parameters that decrease reproducibility in the results. • To measure sampling variation for ~-carotene in cassava roots from differets plants, between roots of the same plants and among different sectios of the same root, to obtain more uniform, comparable and reliable data. • To determine the effect of different preparation or processing methods on B - carotene and total carotenes content of cassava roots. Table 3. Dry matter and carotene content in roots from 16 S11ines derived from MTAI 8 (Rayong 60) measured by the colorimetric and HPLC metbods on fresh and dry tissue. Colorimetric method HPLCmethod Dry matter Carotene content (mg/100 g) Carotene content (mg/100 g) (%) Fresh tissue Dry tissue Fresh tissue Dry tissue AM 320-140 37.47 1.01 2.71 1.76 4.71 AM 320-136 33.62 0.83 2.48 1.66 4.93 AM 320-135 34.47 0.87 2.54 1.78 5.17 AM 320-133 32.61 0.81 2.49 1.49 4.58 AM 320-123 37.36 0.44 1.19 0.66 1.78 AM 320-143 36.89 0.36 0.98 0.25 0.68 AM 320-147 42.69 0.37 0.86 0.40 0.93 AM 320-127 37.20 0.41 1.11 0.49 1.32 AM 320-124 38.86 0.31 0.80 0.40 1.04 AM 320-146 40.74 0 .32 0.78 0.42 1.04 AM 320-139 33.78 0.26 0.77 0.21 0.63 AM 320-144 33.91 0.26 0.77 0.26 0.77 AM 320-142 36.41 0.15 0.41 0.10 0.27 AM 320-138 37.63 0.14 0.37 0.10 0.27 AM 320-130 31.53 0.14 0.44 0.11 0.34 AM 320-121 40.82 0.14 0.35 0.10 0.24 MTAI8 39.98 0.27 0.69 0.31 0.78 Average 36.63 0.43 1.19 0.64 1.79 St.Dev 3.17 0.29 0.85 0.64 1.87 M in. 31.53 0.14 0.35 0.10 0.24 Max. 42.69 1.01 2.71 1.78 5.17 &m r--------------------------------------------, S1 clonal llne Figure 8. Carotene content in roots from 16 S11ines derived from the elite clone MTAI 8 (or Rayong 60) measured by tbe colorimetric and HPLC method. Data based on dry tissue. 355 References Rodriguez-Arnaya. D.B. 2001. A guide to carotenoid analysis in foods. ll..SI Human Nutrition Institute. One Thomas Circle, NW, Washington, OC 20005-5802, USA. Combs, G.F. Jr. (1992) The Vitarnins. Fundamental aspects in nutrition and health (2d edition). Academic Press. San Diego, New York, Boston, London, Sydney, Tokyo, Toronto. CIAT, 1999. Intemational Center ofTropical Agriculture. Annual Report for Year 1999. Cali, Colombia. CIAT, 2000. Intemational Center of Tropical Agriculture. Annual Report for Year 2000. Cali, Colombia. CIAT, 2001. Intemational Center ofTropical Agriculture. Annual Report for Year 2001. Cali, Colombia. Safo-Katanga, 0 ., Aboagye, P.,Arnartey, S.a., and Olaham, J.H. 1984. Studies on the content of yellow- pigmed cassava. In: Terry, E.R. et al. (eds). Tropical Oirschot, Q., O'Brien Gerard and Dufour, D. 2000. The effect of pre-harvest prunning of cassava uppon root deterioration and quality characteristics. Joumal of the science of Food and agriculture, 80: 1866-1873. 2.3.7 Evaluation of S1 Families for Waxy Mutants Allison Smith1, Chris Hylton1, Nelson Morante2, Teresa Sanchez2, Heman Ceballos2 , Martin Fregene2 1JIC, Norwich, 2 CIAT Introduction There is a growing interest from both the prívate and public sectors to develop a waxy starch phenotype in cassava. Three approaches have been embarked upon, via genetic transformation, anti-sense, and sense silencing, irradiation of cassava seeds, and screening the germplasm bank for waxy starch phenotypes. The heterozygous nature of cassava however makes the second and third option difficult, given the low chances that a natural mutant of the waxy gene will be found in the homozygous state. A decision was made to screen many S1 families developed under a population development effort to for tolerance to inbreeding. About 14 S1 families were available for evaluation. To further increase the precision of the chemical assay for percent amylase in the above evaluations, Prof Allison Smith's group at the nc was contacted for assistance in preparing pure cassava amylose and amylopectin for making an amylose determination standard curve. Current assays use commercially available pure potato amylose and amylopectin in preparing the standard curve, but the chain lengths of amylose in cassava and potato differ and this may introduce errors in the percent amylose measured. Pure amylose and amylopectin was isolated from root starch of the cassava variety MCol 2216 and used in developing an amylose determination standard curve. Methodology 14 S1 fanúlies, developed under an S2 recurrent selection program to develop populations tolerant to inbreeding were the plant materials for the above experimene. They were planted last year at the CIAT station in Santander the Quilichao and harvested July this year. At harvest starch samples was collected from 3 roots of all progenies, a total of 514 individuals, and taken to the laboratory for analysis. Pure cassava amylose and amylopection preparations were made using a sepharose separation column at the John Innes center, Norwich following methods described by Zeeman et. al. (2003). Mixtures of amylose and amylopectin with varying proportions of amylose from 0% to 100% were prepared, the samples were then dispersed in ethanol, hydrolyzed by acid, iodine added to a final concentration o 2% (v/v) and the absorbance measured at wavelengths of 700nm and 525nm in a spectrophotometer. Measuring the sample at this two wavelengths and using a ratio of 700:525 rather than the traditional method of measurement of a single wavelength , 620nm, have been found to be more accurate in determining amylose content (Zeeman et al. 2003). The absorbance and of the samples with varying proportions off amylose was used in generating an amylose determination curve. Results Pure cassava amylose and amylopection were obtained from fractions of a sepharose column at maximum absorption at wavelength of 595nm (Fig 1). The preparations were used in making mixtures of different proportions of amylose and the absorbance measured as describd above. Figure 2 and table 1 shows results of the development of an amylose content determination curve. The curve will be used for determining amylose content in the starch samples obtained from the S1 fanúlies. Analysis of the starch samples are ongoing and should be completed by November. Any samples that shows less than 5% amylose will be sent for further analysis, including a quantitative purification of the amylose and amylopectin, and analysis of the molecular structure, to ensure that the low amylose content is not due to extremely long amylose molecules whose absorbance in solution closely mirrors amylopectin molecules 357 0.20 0.18 0.16 0.14 J 0.12 o 0.10 ~ :11 ~ 0.08 0.08 0.04 0.02 0.00 .... ... 5 6 7 8 Figure l. ; ... • !\ ~ : ; : c.-va arch: S.pharoa Cl2B (20108103) IIOul F + 140ul IIO%Lugol .. .... .... . • ~- .. .... .... . ........... .. ·. ... · ..... . .. ............... . 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 26 29 30 31 32 33 34 35 36 37 38 39 40 Fraetlon Absorbance of a sepharose column fractions in the preparation of pure cassava amylose and amylopectin from the cassava variety MCol2216 Mlytose CXX'IIent vs ratio ol /'i!OO ;nd P62S ol glucao-lodine complex (27108103 20.119 '030822 Casuva • t.W1in Freo-.xls") 1 .0r---------------------------------------------------------------------~~~--. o.a - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0.2 y • 0.499ex . 0.1056 R'·0.1199 0.0+-----~~--~------------~----------~------------~----~----~------~--~ 0.0 0.2 0.4 o. a 0.8 1.0 1.2 1.8 1.8 2.0 2.2 ,. .. 0 (A700:A525) Figure 2. An amylose determination standard curve prepared using pure cassava amylose and amylopectin 2.4 Table l. Data for tbe preparation of tbe amylose content determination curve Amylose AP Proportlon A525 A700 Ratio .mylo•• A700".A525 ul ug ul 1 ug 1 1 11 Mean 1 1 11 Mean o 0.000 103.1 17.816 0.00 0.2249 1 0.2391 0.2320 0.0506 0.0497 0.0502 0.2162 3 0.891 98.0 16.926 0.05 0.2427 ¡ 0.2414 ! 0.2421 0.0748 0.0747 0.0748 0.3088 6 1.782 92.8 16.035 0.10 0.2403 i 0.2369 1 0.2386 0.0993 0.0983 0.0988 0.4141 9 12.672 87.7 15.144 0.15 0.2422 ! 0.2353 0.2388 0.1210 0.1253 0.1232 0.5158 - - 1-----+---1--·-- 12 3.563 82.5 14.253 0.20 0.2515 ! 0.2533 0.2524 0.1584 0.1593 0.1589 0.6294 18 5.345 72.2 12.471 0.30 1 0.2537 i 0.2524 0.2531 0.2046 0.2048 0.2047 0.8089 24 7.127 61 .9 10.690 0.40 0.2520 ¡ 0.2562 0.2541 0.2510 0.2673 0.2592 1.0199 30 8.908 51.6 8.908 0.50 0.2483 1 0.2512 0.2498 0.2971 0.2974 0.2973 1.1902 36 10.690 41 .3 7.127 0.60 0.2489 1 0.2652 0.2571 0.3526 0.3781 0.3654 1.4213 42 !1 2.471 30.9 5.345 0.70 0.2557 r 0.2671 0.2614 0.4016 0.4207 0.4112 1.5729 ¡-·-- t·-- >---- --- ·· 0.4804 1 ,....- - 1-- - 48 14.253 20.6 3.563 0.80 0.2645 : 0.2558 0.2602 0.4519 0.4662 1.7919 1 ' 54 16.035 10.3 1.782 0.90 0.2666 ¡ 0.2663 0.2665 0.5313 0.5336 0.5325 1.9983 60 17.816 0.0 0.000 1.00 1 0.2617 0.2561 0.5577 0.6009 0.5793 2.2624 0.2504 1 1 Conclusions A more accurate method for the determination of amylose and amylopection proportions in cassava has been developed in collaboration with the John Innes center. It consists of using an amylose deterrnination standard curve that was prepared using cassava amylose and amylopectin rather than those from patato. The method is currently being used to evaluate starch samples from St families References Zeeman S ., T iessen A., Pilling E., Kato L., Donald A., and Smith A. (2003)Starch synthesis in Arabidopsis thaliana: Granule synthesis, composition and structure. Plant physiology 129:516-529. 359 2.3.8 Irradiation of Sexual Seeds for the Production of W axy Cassava and other Mutants Chikelu Mba1 ; Ms Sarah Tafur; Heman Ceballos3, Nelson Morante3 Martin Fregene3 1 IAEA, Vienna, Austria, 2 National University, Palrnira, 3CIAT Introduction The globalization of econornies has meant a search for local crops that are competitive and will preserve local agriculture. Economic surveys in the past ten years in cassava growing regions of Africa, Asia and Latín America have revealed that cassava is an important factor in the improved livelihoods of the rural population of these regions (Nweke et al 2001; Kawano 2001; Ceballos 2002). A change in starch quality, for example the elimination of amylose (waxy starch), via a knockout of the GBSS 1 gene, implies access to new markets for cassava growers. For most rural communities, a better standard of living depends on increasing income from their crop harvest. An important disadvantage of cassava is the short shelf life of its roots, which have to be consumed or processed within a few days after harvest. This trait, post-harvest physiological deterioration (PPD), results in losses and higher marketing costs and its elimination will lead to higher profit margins for the small producer. PPD is thought to be a wound response cascade that goes out of control, and the knockout of certain key genes in the cascade should lead toa reduction in PPD. Cassava also possesses a wide range of cyanogenic glucoside content that is often cited as a health risk and a stigma for its acceptance as raw material in certain food and feed industry. Cyanogenic glucosides are produced by a biosynthetic pathway that has a cytochrome P450 as catalyst in the rate-limiting step (Andersen et. al. 2000). Removal of the gene that expresses the cytochrome P450 gene should lead to low cyanogenic potential (CNP) cassava plants. Increased productivity or increased value of cassava roots such as novel starch types and improved marketing through elimination of losses from post-harvest physiological deterioration and the removal of the stigma of cyanogenic glucosides, stands to improve livelihoods of cassava farmers. Mutagenesis has been applied extensively in the production of novel phenotypes in crop species (Van Harten 1998). The project will take advantage of simultaneous research, currently under way, that will facilitate to routine production of inbred materials, for the first time in a cassava breeding project. Once mutants have been identified, molecular genetic analysis will be used to track down genes responsible for the novel traits. The use of these genes as markers or in genetic transformation will permitan increase in the efficiency of transferring these traits to other cassava gene pools through conventional breeding. Specific Objectives were: Irradiate, using gamma rays (a Cobalt-60 source) and fast neutrons, of sexual seeds from elite cassava genotypes under improvement for tolerance to inbreeding depression Establish plants from the irradiated seeds and non-mutated parental genotypes and evaluate them for useful traits such as delayed post harvest physiological deterioration (PPD), low cyanogenic potential (CNP), high dry matter content (DMC), and novel starch types. Develop selfing of the mutated lines to obtain So progenies and their evaluation for the above root quality traits and any other potentially useful trait Carry out DNA analysis for genes known to be involved in biosynthetic pathways of the above traits to identify mutants. Methodology Cassava has seldom been inbred, and the large "genetic load" hidden in its heterozygous background will likely hinder the production of viable homozygous plants, a phenomenon known as "inbreeding depression". Selection for tolerance to inbreeding depression has therefore been initiated to make cassava populations amenable to the production of inbred lines. Sexual seeds from cassava genotypes tolerant to inbreeding were the source of genetic material for mutagenesis. About 2000 sexual seeds were shipped to IAEA for irradiation, using gamma rays (a Cobalt-60 source), 1000 seeds, and fast neutrons, 1000 seeds. The leve! of irradiation with gamma rays was 200Gy. The irradiated seeds were sent back to CIA T for germination and establishment of the plants in the field. The heterozygous nature of cassava implies that mutations in a recessive gene will not be observed in the Mo phenotype. There is therefore a need to self the Mo plants to permit ídentification of the recessive mutants. However, the task of selfing thousands of plants is beyond capacity at CIA T and a selection of mutants for genes of interest will first be carried out to identify mutants. DNA analysis that can identify single nucleotide polymorphisms (SNPs) or insertions/deletions (INDELs) in genes of interest will be employed. At 1 O months after planting, the plants from the irradiated seeds selected above will be evaluated for ability to produce flowers. Plants that flower will be cloned and planted the following year in a clonal observation trial fashion of 10 plants per genotype. Plants in the clona! observation tria! above will be selfed to obtain the M1 (So) generation. The seeds from the MI (So generation) will again be established in the field at CIA T and other key target environments and thoroughly evaluated for the traits mentioned in the previous section. Progeny identified with useful root traits will be introduced into the cassavabreeding program. 361 Results A total of 2000 full-sib seeds from full- and half- families of MCol1505, HMC-1, C4 (CMD resistant parent) were sent to IAEA for irradiation with gamma rays and fast neutrons. About 1000 seeds were irradiated with adose of 200Gy of gamma rays. They were moisture equilibrated over a 69% glycerol solution in a dessicator prior to radiation. These seeds ha ve been sent back to CIA T where they have been planted in a seedling nursery. Seeds irradiated with fast neutron experiment, aboutlOOO seeds, are still being expected back. Conclusions This project seeks to use novel methods of mutagenesis, conventional plant breeding and molecular genetic analysis to identify cassava genotypes with value added traits. lt will take advantage of the recently initiated research to produce inbred cassava germplasm. The project will also use tools of genomics to track down genes responsible for the above traits, markers associated to these genes can be used to efficiently move these genes around the different cassava gene pools defined by agro-ecologies. The new methods will not only accelerate the production of improved germplasm but also be a model for the development of other traits of interest to the market and farmer. References Andersen MD, Busk PK, Svendsen 1 and M!llller BL 2000. Cytochromes P-450 from bcassava (Manihot esculenta Crantz) catalyzing the first steps in the Biosynthesis of the cyanogenic glucosides linamarin and lotaustralin. Cloning, functional expression in Pichia pastoris, and substrate specificity of the isolated recombinant enzymes. J Biol. Chem. 275-1966-1975. Ceballos, H. 2002. La yuca en Colombia y el mundo: nuevas perspectivas para un cultivo milenario. In: Ospina B. and H.Ceballos (Eds.) La Yuca en el Tercer Milenio: sistemas modernos de producción, procesamiento, utilización y comercialización. CIAT Publication Number 327. Cali, Colombia. CIAT (lnternational Center for Tropical Agriculture), 2001. Annual Report Project IP3: Improved cassava for the developing world. Cali, Colombia. FAO (2000). Food Outlook December 2000. FAO, vía delle tenne di Caracalla, 00100 Rorne, Ita! y. Kawano, K. (2001). The role of improved cassava cu1tivars in generating income for better farm management. ln:Howeler R.H., and Tan S.L. (eds). Cassava's Potential in the 21st Century: Present Situation and Future Research and Development Needs. Proceedings of the Sixth Regional Workshop, held in Ho Chi Minh city, Vietnam. Feb 21-25, 2000, CIAT, Bangkok, Thailand, pp 5- 15 Kawano, K., K. Narintaraporn, P. Narintaraporn, S. Sarakarn, A. Lirnsila, J. Lirnsila, D. Suparhan, and W. Watananonta. 1998. Yield improvement in a multistage breeding program for cassava. Crop Sci. 38:325-332. Nweke, F., Spencer, D .• Lynam, J. (2001). The Cassava Transforrnation: Africa's Best-Kept Secret. Michigan State University Press, East Lansing, USA. 273 pp. Salehuzzarnan SNIM, Jacobsen E y Visser RG F. (1993) Isolation and characterization of a cDNA- encoding granule-bound starch synthase in cassava (Manihot esculenta Crantz) and its antisense expression in patato. Plant Mol. Biol. 23:947-962. Van Harten A. M. (1998). Mutation breeding: theory and practica! applications. Cambridge University Press, Cambridge. 2.3.9 Development of Populations Tolerant to lnbreeding Depression in Cassava N. Morante, H. Ceballos, M. Fregene. CIAT Introduction A S2 recurrent selection program was set-up for the development of cassava populations tolerant to inbreeding. The principal reason for the program is the development of populations tolerant to inbreeding and for the development of pure cassava lines via the doubled haploid technology. A second reason is the identification of genetic stocks for gene mapping studies, for example excellent segregation of beta-carotene content and cyanogenic poten ti al (CNP) was observed in 2 of the S 1 farnilies from M Col 72 (beta- carotene) and MTAIS (beta-carotene and CNP). A preliminary selfing of an initial 14 genotypes produced 300 S1 lines that were evaluated in a clonal observation triallast year (CIAT2002). More than 30 of those S1 lines that flowered profusely were selfed to produce S2 farnilies. We describe their production and establishment in the field. Furthermore additionallarger sized St farnilies have been developed from 20 varieties and have been established in a seedling tria! this year. 363 Table l. Summary of ~ families developed from embryo rescue of selfed S1 individuals !Date of transplant Grand Parent Parent Code No.of Plants ~un-03 MColl505 AM244-10 AM 244-10-5 2 ~un-03 MColl505 AM244-16 AM 244-16-1 ~un-03 MCol1505 AM244-16 2 AM 244-16 ~un-03 MCol1505 AM244-31 AM 244-31 5 ~un-03 MCol1505 AM244-31 1 AM 244-31 5 ~un-03 MColl505 AM244-31 2 AM 244-31 5 ~un-03 MCol1505 AM244-35 AM 244-35 2 ~un-03 MCol1505 AM244-35 1 AM 244-35 4 ~un-03 MCol1505 AM244-35 2 AM 244-35 7 ~un-03 MCol1505 AM244-35 3 AM 244-35 3 ~un-03 MCol1505 AM244-38 AM 244-38 1 ~un-03 MColl505 AM244-40 1 AM244-40 2 ~un-03 MColl505 AM244-101 1 AM 244-101 3 pun-03 MCol1505 AM244-109 1 AM 244-109 2 ~un-03 MColl505 AM244-109 4AM 244-109 5 ~un-03 MCol1505 AM244-l09 5 AM 244-109 5 ~un-03 MColl505 AM244-109 6 AM 244-109 2 ~un-03 MCol1505 AM244-l09 7 AM 244-109 5 ~un-03 MCol1505 AM244-l09 8 AM 244-109 2 pun-03 MCol1505 AM244-l09 10 AM 244-109 1 ~un-03 MColl505 AM244-109 12 AM 244-109 4 ~un-03 MColl505 AM24-135 1 AM 244-135 1 ~un-03 MColl505 AM24-135 6 AM 244135 2 ~un-03 MCol1505 AM24-135 7 AM 244-135 3 ~uJ-03 HMC l AM266-2l 4AM 266-21 2 ~ul-03 HMC 1 AM266-41 1 AM 266-41 ~ul-03 HMC 1 AM266-41 5 AM 266-41 2 ~ul-03 HMC 1 AM266-41 6 AM 266-41 ~uJ-03 HMC 1 AM266-45 2 AM 266-45 7 ~ul-03 HMC 1 AM266-50 4AM 266-50 2 ~ul-03 HMCl AM266-76 1 AM 266-76 Pul-03 HMC l AM266-76 6AM 266-76 ~ul-03 CM849-1 AM312-42 l AM 312-42 5 tfotal 95 Methodology About 30 S¡ progenies were selfed and seeds harvested at 40-60 days after pollination, the earl y harvest of the seeds was due to the mandatory remo val of all cassava plants in CIA T before the beginning of the "zero cassava" month at CIAT, one of the measures adopted to control the incidence of white flies at CIA T. The S2 families were established from embryo rescue as described in above in the section on the development of mapping populations in cassava. After one round of multiplication in vitro they were transferred to the screen house for hardening and to the field. At 10 months after planting, the progenies will be evaluated for fresh root yield, dry matter content, foliage weight, harvest index, culinary quality, starch content/quality, and frog skin disease according to standard CIA T procedures. Results An initial 14 cassava genotypes were chosen for the development of populations tolerant to inbreeding. The genotypes were chosen due to their good general combining ability performance for yield, dry matter yield or root quality. They include the following lines: MCOI22, CM523-7, MCOL1684, MBRA12, MCOL2060, MVEN77, MCOL1522, MTAI1, l\1PAN51 , MECU169, MCOL1468, MCOL72, CM849-1, HMCI. More than 300 pollinations were made per genotype and between 30-150 seeds were obtained per genotype. During a clona! observation of the abo ve families, selfings were carried out and 15 small sized S2 families, of 2-10 progenies were developed (Table 1). These seeds were germinated from embryo axes. Between 5 and 10 plants per S2 progeny was hardened in the green house and transferred to the field in July this year. Conclusions The development of populations tolerant to inbreeding has continued with the development of 15 small sized S2 families. These families will be evaluated at 10 months after planting. References CIAT (International Center for Tropical Agriculture), 2001. Annual Report Project IP3: Improved cassava for the developing world. Cali, Colombia. 365 OUTPUT 3 Collaboration with public and private partners enhanced Activity 3.1 New Collaborative Arrangements and Networks 3.1.1 Update on HarvestPius (Biofortification Challenge Program) Project team members continued to played a major role in the formulation, organization and fund raising of the Biofortification Challenge program convened by CIAT and IFPRI. The Biofortification Challenge Program (now called HarvestPlus) was approved in October, 2002. World Bank funding of $3 million for 2003 was made available in December, 2002. A first Project Advisory Committee (PAC) meeting was held in March to initiate/approve the selection of the Program Director and to set up/approve initial operating procedures and initial project activities for 2003. A second PAC meeting will be held in November 2003 to discuss/approve the workplans developed for 2004, coming out of the various planning meetings. Renewed contacts with the Gates Foundation were initiated by CIA T and IFPRI in January, 2003. A formal proposal for 25 millions over four years was submitted on behalf of the CGIAR lead consortium to the Gates Foundations including the submission of the extemal reviews cornrnissioned by the interim Science Council (iSC) and iSC. While the proposal was approved in Aug 1, 2003 a formal announcement and the launch of Harvest Plus took place in Washington on October 14, 2003. A press release was issued announcing the grant, and a press conference was held in Washington, DC. Discussions continue with a number of additional donors in the hope of meeting the proposal target budget of $12.5 million per year over the first four years. Full start-up of the project is envisioned in January, 2004. Severa! planning meetings were organized during the year. The most relevant ones were: I- Planning meeting of core collaborators and selected stakeholders, June 2-6, CIAT headquarters: 75 persons attended the five-day meeting. The planning cornrnittee of 15 persons met for the first time as a group for one full day on May 30. The following objectives were achieved: To introduce and discuss the different broad components of the project to those who will be participating in the project and other stakeholders; there was an initial unevenness among the collaborators/participants in their familiarity with the project. To flesh out and revise the conceptual framework laid out in the proposal to ensure that the different components are designed to complement each other technically -- thus, it was essential to engage in interdisciplinary learning and exchange at the meeting; To develop an operational plan to implement coordinated activities over the next 12 months; a workplan and budget allocation for 2003 was agreed upon. To open and set up lines of communication across the components -- collaborators got to know each other much better, and crop leaders became better familiarized with what would be in volved in building and managing their crop teams. To hear, at a very broad, global level, what might be stakeholder or collaborator concems on the concept of the project as a whole and to get their bu y-in . TI- Private sector meeting, IFPRI headquarters, July 20: Representatives from DuPont/Pioneer and Monsanto corporations, ILSI, and USAID met with Howarth Bouis and Joe Tohme of the BCP to discuss collaboration. The meeting was organized by Bill Neibur of the COlAR Prívate Sector Committee. There was strong interest expressed on both sides in cooperating on development of micronutrient-dense, transgenic varieties. Private sector representatives were encouraged to attend the upcoming technical meetings and crop meetings during September-October. . Further meetings are planned for early 2004. m- Nutrional breeding objectives and Crops meetings: Team members attended the Nutritional Breeding Objectives, and the maize, sweet potato and rice crop meeting and organized the bean and cassava meetings. 3.1.2 Biofortification : Linking The Strategic Breeding Work To Downstream lmpact: Reaching And Engaging End-Users L. Sperling CIAT, Rome- ltaly Considerable progress was made this year in identifying strategies to ensure that the biofortified products deve1oped reach users-and meet their needs. Analysis showed that priority client groups are diverse (rural producers and consumers; urban producers and consumers; intermediary processors; micro-nutrient deficient population; women and children) Further, in three separate sets of meetings (Copenhagen Oct/2002, Cali, June 2003; and subsequent crop-specific workshops; September-October 2004), participants identified the varied activities needed address this diversity, and to encourage widespread utilization of products. As summarized in Table 1, sorne of these more downstream activities might be programmed several years from prototype development (eg seed multiplication) while others need to start relatively quickly (eg basic community-level nutritional R&D). Table: l. Broad overview of activities implied in the thrust in :Reaching and engaging end-users Activity Set Description Variety Development and Adaptation Identification of range of end-users needslpreferences to guide initial selection criteria Elh Partners (community level and Adaptive testing schemes which reach clients (decentralized and/or participatory}-and which give beyond). good feedback V arieta! testing models which can work with the poor and build oo fanner skills Prefcnbly testing modtls which can be replicated over many sites Adaptivt agronomic research which follows use of biofortified materials under realistic management conditions Community-level Nutritional Research Range of activities for understanding the possible constraints/opportuuities for use of biofortified and Development rnaterials ínter and intra-household, community and region Determination of poteotial nutritional beoefits for different population groups , given existing dietary pattems Determination of possibilities for expanding the use of biofortified products, e.g. by incorporating the crop into culturally acceptable recipes, developing new recipes, etc. Identification and use of lin.ks to global initiatives to improve infant and young child feed, being spearheaded by UNICEF and WHO Communicatioo Outteach Strategies -- Developmeot of strategies to reach di verse major audieocelpartoer groups: at regional leve! (iocludes 'Social Decision-makerslpolicy makers marketing' to guide behavioral change) Potential producers and coosumers, traders, processors Micro-nutrient deficient population Tailoring of: • content of messages to specific audience • forrnat of messages (policy brief, radio, drama, extension bulletins) • work in multiple lan¡zua¡zes Production and delivery of Production and delivery of seed /food plus substantial inforrnatioo on use and nutritional seedlproducts (along with its associated consequenceslbenefits of biofortified materials knowledge base) Acti vities effected through normal seed channels, nutritional channels and uncon ventional ones ( eg 'coca cola ttucks) Program monitoring and feedback at the Eocouraging of co-leaming and feedback among varied stakeholders community/regional leve! Confirrnation /adaptation CP program directions Identification novel mechanisms for enhanced irnpact Development of specific tools for program implementers to track progress on a regular basis- yearly ata mínimum. Most closely look at the uptake by producers and the uptake by consumers Creation!Fostering of Enabliog Seositization/feedback from decisioo-makers as to magnitude of micronutrient problem Policy/Stakeholder environment Stimulation of interest in 'paradigm' shifts to link agricultura! research sector to health research sector (new institutional environments) Establishment of flCSt 'buy in' from key collaborators-early Shaping environment for favorable policies oo: variety release, seed certification Getting systernatic feedback from higher-level producer and consumer groups on: biofortified solution and use of GMOs as ooe ¡¡ossible thrust One of the opportunities presented by working across crops in this biofortification program is that it allows research and development workers to address the different dietary needs of the poor, and to work in integrated farming systems, rather than focus on one crop or another. For instance, the four major staples given focus in the Africa program--: maize, beans, sweet potato and cassava-- are often grown in single homesteads, and sometimes even on the same plot. Because of the intercrop complementarity within this Challenge Program, activities of this 'reaching and Engaging User'' group should be programmed across crops, when possible. Table 2 suggests broad strategies for across crop coordination. Such intercrop work will demand considerable facilitation, but ultimately could lead to more comprehensive impacts at the field level-as well as increased CP program efficiency. 368 Table 2: Coordination of thrusts for reaching and engaging users with the Biofortification Challenge Program Within Crops * Variety Development and Adaptation with users Across Crops * Cornmunity-level Nutritional R&D * Cornmunication Outreach * Program Monitoring and Feedback * Creation/Fostering of Enabling Policy/Stakeholder Environment Within and Across Crops * Seed production and Multiplication * Marketing of Products 3.1.3 Models of Food Safety Assessment of Transgenic Crops, Workshop funded by USAID and the Rockefeller Foundation, Washington DC, May 6-8, 2003 Joe Tohme1 and Rector Quemada2 1SB-2 Project; 2Crop Technology, USA The safety assessment of transgenic plants is a major hurdle to their deployment in agriculture. Researchers and regulatory authorities in deveioping countries find such hurdle especially difficult to overcome. On the one hand, researchers lack the guidelines and experience in the types of tests that need to be conducted to assess safety. On the other hand, regulatory authorities are not experienced in assessing the safety of crops in terms of their general release. These difficulties will be felt more acutely when crops that have been developed within these countries reach the stage of general reiease. In these cases, developing country researchers and reguiatory authorities will not be abie to depend upon the previous decisions of other reguiatory bodies, as is presently the case with transgenic crops released in all countries except the United States, Canada, China. Instead, they will have to detennine the appropriate questions to ask, and appropriate data to be gathered, in order to arrive at ajudgment of safety for these indigenously developed transgenic crops. For many regulatory authorities faced with the problem just described, there is Iikely to be the inclination to rnirnic the requirements imposed by regulatory authorities in countries such as the United S tates, where commercialization of transgenic crops has taken place and is relatively routine. However, there is little understanding of the reasoning behind the requirements that have been imposed upon specific transgenic crops in countries such as the U.S., where regulatory data requirements have arisen out of regulatory frameworks and philosophies that rnight be quite different from those of other countries. Furthermore, there is little understanding of the cost and types of expertise required to complete the safety assessments that have already been done. Consequently, the imposition by regulatory authorities of requirements for safety testing that are beyond the means of their countries, either in terms of money or human capacity, will effectively prevent the deployment of the results of biotechnology research in those countries. In order to assist developing country and the CGIAR researchers in gaining an understanding of the types of food safety assessments needed for the deployment of transgenic crops, and to determine which requirements are appropriate for developing countries a workshop was held in May,2003, with funding from the USAID and the Rockefeller Foundation. The meeting was organized by loe Tohme and Rector Quemada from Crop Technology. The goal of the workshop were to provide researchers and regulators with the means to examine 1) the protocols used in the safety assessments of current commercialized transgenic crops, and 2) the rationale for the requirements imposed (or are likely to be imposed) on crops representative of different transgenic technologies. The workshop focused only on food safety assessment. The participants were from NARS representatives from Brazil, Colombia, Mexico, Egypt, India, Kenya, South Africa, Philippines, Thailand and Uganda; from CG centers: CIAT, CIP, CIMMYT.ICRISAT, !CARDA and liTA; from intemational lnstitutions and donors: AATF, AgBios, USAID, Rockefeller Foundation, ll...SI and USDA and from the prívate sector. Three representative case studies were commissioned to leading expert and were presented at the workshop: 1) Bt patato, 2) virus resistant papaya, 3) mustard oil with enhanced levels of vitaminA. The outputs of the workshop were the following Examination of the protocols used in the food safety assessments of current commercialized transgenic crops Understanding of the rationale for the requirements imposed (or likely to be imposed) on crops representative of different transgenic technologies Awareness of the regulatory steps to enable researchers to integrate regulatory issues and experiments related to regulatory approval in their overall research and development strategy. Definition of a roadmap for food safety assessment 370 3.1.4 Updating CassavaDB, a ACEDB-type data base for Results of Genome Mapping C. Buitrago; F. Rojas; J.Tohme, M. Fregene (CIAT) Introduction The cassavaDB is an AceDB-type genome database designed specifically for handling bioinformatic data flexibly. It includes tools designed to manipulate genornic data that is graphic, flexible and portable. It can be operated on various Unix workstations (SUN, DEC, NEXT, SGI .. . ), The first principie of the program is that any piece of data stored in AceDB can very easily be exported in flat ascii files to be used by other programs. The second principie of the program is that write acces to the database is organised in macro transactions that is called sessions. All the information is stored in objects, which fall into a number of classes. The classes are standard units such as genes, alleles, strains, c1ones, papers, authors, journals, etc. , and the names are in most cases the standard names. The CassavaDB was initially hosted at the plant databases at the National Agricultura! Library at http://probe.nal.usda.gov. This was later moved to the ARS Genome Database Resource (GDR) server formerly at http://ars-genome.cornell.edu which is now permanently off-line. At the moment it is hosted at http://ukcrop.net/, the website of the UK Crop Plant Bioinformatics Network (UK CropNet) established in 1996 as part of the BBSRC's Plant and Animal Genome Analysis special initiative, the focus is the development, management, and distribution of information relating to comparative mapping and genome research in crop plants. Since CassavaDB was developed in 1998, hundreds of additional markers, predorninantly SSR markers have been developed and mapped in cassava. This information is currently not in the database. Also not in the database are additional genetic maps of cassava made since the first map was published in 1997. Data from several QTL mapping studies have also since become avai]able since then. There is also a need to have the CassavaDB situated on a local server at CIA T to enable more frequent updates and also make it more visible to the international cassava community. Activities to update CassavaDB. A compilation of all cassava SSR marker mapping data and map location data from work done by Mba (et al. 2001), Zarate (2002) Garcia (2002) and unpublished or undocumented work was initiated in August 2003. This information was formatted according to requirements for AceDB. Sirnilarly, graphics of parental surveys, progeny data and maps were developed from data from the three mapping studies and prepared for CassavDB. All the above information will be up-loaded into CassavaDB in collaboration wi th the CIA T bioinformatics unit and maintained locall y at CIA T. A link to the above will be made on the cassava web page of the CIA T website. References Mba, RE.C., Stephenson, P .• Edwards, K., Melzer, S., Mkumbira, J., Gullberg, U .• Apel, K .• GaJe, M., Tohme, J. and Fregene, M. (2001) Simple Sequence Repeat (SSR) Markers Survey ofthe cassava (Manihot esculenta Crantz) Geno me: Towards an SSSR-Bassed Molecular Genetic Map of Cassava. Theoretical and Applied Genetics. 102: 21-31. Garcia, T .. 2002. Mapeo genético de una población F1 de yuca cultivada ( Manihot esculenta Crantz) utilizando Microsatélites que proviene de cONA. Tesis de pregrado. Universidad del Valle, Cali. Colombia. Zarate, L.A. 2002. Mapeo genético de una población Fl de yuca cultivada ( Manihot esculenta Crantz) utilizando Microsatélites. Tesis de pregrado. Universidad del Tolima. !bagué. Colombia. 3.1.5 Report of the Second Tri-Annual Workshop of the Molecular Genetic Diversity Network of Cassava (MOLCAS) H. Rosling1 ; M. Akerbolm2 ; E. Kizito3, L. Chiwona-Karltun3, U. Gullberg3 ; J. Jiggins4 G. Muhlen5; E. Okai6 ; M.Fregene7 1Karolinska lnstitute, Sweden, 2IPICs, Sweden 3SLU, Sweden,4W AU, Netherlands; SU. de Rondonia, Brazil; 6CRI, Ghana 7CIAT lntroduction Understanding the extstmg genetic diversity and its distribution within and among individuals, populations, species and gene pools is crucial for an efficient management and use of germplasm collections. The large amounts of cassava genetic resources held by farmers have been demonstrated to represent a critical resource for the future productivity and stability of production of cassava. A highly successful breeding program at the International Center for Tropical Agriculture (CIAT) in the early 1970s began with a great initial germplasm variation, sorne 2,218 local clones collected from Colombia, Brazil, Venezuela, and Peru (KA W ANO 2003). That program achieved more than 100% increase in fresh root yield and more than 20% in dry matter content by the early 1980s. The improved varieties from Latin America combined with local varieties in South East Asia formed the basis of a very successful breeding program for the South East Asia Sub- continent, particularly Thailand. The new Thai cultivars are now planted on more than one million hectares in Thailand alone and the economic benefits from the increased productivity is in the order of one billion US dollars and the rural communities in sorne of the poorest parts of Asia captured a large proportion of these economic benefits (Kawano 2003) .. The molecular diversity network of cassava comprises of scientists drawn from institutes in Africa, Latin America, Europe and the USA. In its four years of existence, MOLCAS 372 has studied diversity of local varieties in Tanzania, Nigeria, Ghana, Uganda, Guatemala, , Peru, Sierra Leone anda subset of the World germplasm bank held at CIAT (MOLCAS 2003). Studies are ongoing of diversity in Cuba and Brazil. The ultimate aim of these studies is to identify and exploit useful variability for increased crop productivity and value addition. Between September 3 and 5 this year, members of the workshop from Brazil, Ghana, Uganda, Netherlands, and Colombia got together in Uppsala with network members based in Sweden and the director of the IPICs donor for the 2nd MOLCAS workshop to discuss progress and draw priorities for the next application to IPICs for the period 2004- 2006. The meeting was held at the Plant biology department of SLU, and was well attended by members of the department including the head of department, Prof Per Bergmann. Report of the 2nd MOLCAS meeting. The first day of the meeting saw presentations on SSR assessment of local cassava varieties from Ghana, Brazil, Nigeria, Sierra Leone, Guatemala, and Uganda. During second day, members of the plant biology department at SLU working on cassava presented a summary of their findings. They include cloning and transformation of starch biosynthetic genes, molecular biology of the cassava mosaic virus, and breeding cassava for small holders. A discussion was held in the aftemoon of the second day on how an application for the third phase of BIOEARN, the Sida funded East African training network could be streamlined with activities being carried out in Africa. Of particular interest was how the wide gap between upstream biotechnology could be applied to secure the cassava crop as a food security crop and also provide improved livelihood through value addition. It was decided that BIOEARN members draw up a draft of their application and circulate it to MOLCAS members for comments and their inputs. The third day, Prof Rosling presented results of an epidemiology of the Konzo disease in Mozambique conducted in the 1970s as an example of a proper approach to sampling genetic diversity. Two presentations were also made by Prof Janice Jiggins of WAU, Netherlands and Dr L. Chiwona Karltun on a farmer participatory research project taking place at the moment in Malawi. The aftemoon of the third day was spent in a priority setting exercise for projects to be conducted in the next phase of MOLCAS. From a total of 10 project, the following were prioritized for the next phase beginning 2004 until: Evaluation of highly differentiated gene pools in cassava for heterosis ( Ghana) Highly differentiated cassava gene pools may represent heterotic pools. The study seeks to evaluate these accessions for heterosis or hybrid vigor by making genetic crosses between and within representative members of the clusters. The study will be conducted in CRI, Kumasi , Ghana. Latin American germplasms were shipped to Ghana as tissue culture plants, hardened, and established in a crossing block along with African genotypes. Tracing the lineages of local African Cassava varieties: towards a better understanding of sub-structures in African cassava gene pools (Uganda and Sweden) To better understand the diversity sub structures in African local varieties, records and germplasm from breeding programs that existed in East Africa in the early part of the 19th century will be examined and genotypes compared to modero day local varietíes. This study will be carried out by SLU in collaboratíon with researchers from NARO, Uganda. A search f or useful variability in local cassava land rae es based upon the structure of SSR marker diversity analysis (Nigeria) Further evaluation of sorne genotypes that have been observed during the SSR marker study to have novel characteristícs, for example novel starch quality, will be carried out to confirm the earlier observations. Should they be confirmed, genetic crosses will be made to attempt a transfer into improved and other local varietíes and also for inheritance studies A comparison of clustering of cassava germplasm from Brazil and Malawi based upon bitter or sweet taste ofthe roots. (Brazil, Malawi, Sweden) Studies of genetic diversity in Brazil revealed a clustering along the lines of taste. A similar result was also obtained for cassava germplasm from Malawi. The objectíve of this study is to combine both data sets and how they cluster. Dr Gilda Muhlen of University of Rondonia, Brazil will be responsible for the study in collaboration with SLU and Malawian researchers. The meeting was attended by Prof H. Rosling of the Karolinska lnstitute, Stockholm; Dr M. Akerbolm, Ms L. Sjobolm of IPICs, Uppsala; Dr JaniceMs E. Kizito, Dr L. Chiwona- Karltun, Prof U. Gullberg, Prof P. Bergmann (head of department), Prof Chris Jassen, Dr Roger Anderson, Mr Yona Baguma, of the plant biology department, SLU, Uppsala; Dr G. Muhlen of the Universidad de Rondonia, Brazil; Ms E. Okai of CRI, Ghana, and M.Fregene of CIA T. References Kawano. 2003. Thirty Years of Cassava Breeding: Biological, and Social Factors for Success. Crop Science 43:1325-1335. 3.1.6 Progress in the Development of a Web Accessible Data Base for SSR Marker Assessment of Diversity of Local Cassava Varieties Akerbolm1 C. Buitrago2, F. Rojas2, M. Fregene2 1IPICs, Sweden; 2CIA T Introduction The web accessible database of the molecular diversity network of cassava (MOLCAS) was set up to make the results of its SSR marker assessment of local cassava diversity in selected countries of Africa and Latín America widely available. The database is updated as completed country studies become available. A total of 8 country studies, 5 in Africa and 3 in Latín America, have now been completed by the network. The database has been 374 updated with 2 additional country studies, bringing the total on the website now to 5, namely Tanzania, Peru, Nigeria, Guatemala and Ghana. A very rewarding outcome of the development of the MOLCAS web is the high number of visits the web has recorded since its inception last year. We describe here progress made this year. Results The MOLCAS web-based data base (http://www.ciat.cgiar.org/molcas and Fig 1) was conceived as a mechanism to document and make freely available to the cassava community information being generated by the network. It is hosted at CIA T. Results of the SSR analysis of local cassava land races from Ghana and Guatemala, including names and characteristics of local cassava varieties, allele size data by SSR marker locus, genetic diversity analysis, principal component of genetic distance etc, are now available for viewing on the MOLCAS web data base. Results from 3 other country studies namely Uganda, Sierra Leone and Brazil are being compiled and will be available on the web before the end of the year. The MOLCAS web-based data base has since its launch last year received a total of 34,477 visits with an average of 2700 visits per month. Table 1 shows the summary of all visits and visits in the last 3 months .... ----- .. ~:~..t !J' G--. rw-- O= k ., •. r,~ -=- -)IU--.-.. .... - ..... ----.IIIIOU: .. - ....... ~ .... ---.. ..., . ..,. __ .. ____._~..... . .-,.-............. ~._.....--..._.- --........ -,_.-. ------ --"'~ Figure 1. MOLCAS borne page on the web showing links to the data base of country studies Table l. A record of visits to the MOLCAS web-based data base on the country studies of SSR assessment of local cassava varieties Request Total Hits SeE-03 Aug 2003 Ju12003 /molcas/imagen.jsp 8,003 130 1,303 496 /molcas/locus.jsp 7,605 247 1,228 517 /molcas/alelosp.jsp 5,515 374 1,002 252 /molcas/estudios.jsp 3,250 125 266 370 /roo leas/ 2,162 88 138 130 /molcas/markers-det.jsp 1,892 37 101 187 /molcas/imagenbioquim.jsp 1,401 37 72 156 /molcas/appendix 1.jsp 1,381 35 69 179 /molcas/intrap _data2 .j sp 1,377 37 86 149 /molcas/appendix2.jsp 858 16 40 145 /molcas/studies.jsp 482 11 16 38 /molcas/pcr_cond.jsp 448 10 48 41 /webapps/molcas/ 103 2 5 4 Total 34,477 1,149 4,374 2,664 Conclusions and Perspectives The MOLCAS web accessible database has increased the completed country studies available for viewing to 5 and 3 more will be added before the end of the year. Two other country studies, Cuba and Mozambique are ongoing and will also be added once completed. Future perspectives for the database is to add a number of useful links, for example, sites where genetic diversity analysis software can be downloaded freely. Other perspectives is to link the MOLCAS data base to and genotyping activities on cassava in the genetic resources challenge program which is expected to begin next year. 3.2 Cassava Biotechnology Network's activities for 2003 Alfredo A. C. Alves SB-2 Project Introduction The Cassava Biotechnology Network for Latín America and the Caribbean (CBN-LAC) is a network of cassava researchers and end-users united by the goal of mobilizing the development and application of biotechnological tools for the enhancement of the value of cassava for food security and econornic development in the poorest rural areas of the LAC. The network is jointly funded by Directorate General for Intemational Cooperation (DGIS) of the Netherlands' government and the Canadian Intemational Development Research 376 Center (IDRC) under the project "The Cassava Biotechnology Network in Latín America: Strategies for lntegrating Small-Scale End-Users in Research Agenda-Setting, Testing and Evaluation ", which initiated activities in 2001. The IDRC had committed funds for supporting the network for 2 years, 2001 and 2002 while the DGIS is providing funds for four years, 2001 to 2004. The CBN-LAC's objectives are to: a) integrate the needs of small-scale cassava farmers, processors and consumers into biotechnology research planning; b) stimulate research on high-priority topics; and e) stimulate the exchange of information, techniques and research materials. Its outputs will enable small-scale cassava farmers, processors and consumers to benefit from advances in cassava biotechnology The third year of CBN-LAC (2003) was devoted mainly to the following activities: 1) Monitoring and guidance for on-going projects at pilot sites in Colombia, Brazil, and Ecuador. 2) lmplementing new projects under the CBN-LAC Small Grants Scheme in Colombia, Brazil, Cuba and Ecuador. 3) Scholarships for postgraduate studies in biodiversity under the Ginés-Mera Memorial Fellowship Fund. 4) Organizing the Sixth Intemational CBN meeting (CBN-VI) Report on Activities at Pilot Si tes COLOMBIA 3.2.1 Application of low-cost in vitro propagation techniques to conserve native varieties and produce quality cassava seed in southwestern Colombia José Restrepo (Coordinator), FIDAR ; Gloria Ospina, FIDAR; Roosevelt Escobar, CIAT Joe Tohme, CIAT; Carlos Hemández, Farmer Introduction This project has been conducted since 2001 aiming to use biotechnology tools to conserve native cassava varieties while enhancing the food security of small farmers in Colombia's Andean region. The 2003 activities focused on fine-tuning the phases of hardening in vitro-produced material and transplanting it to the field. Severa! native cassava varieties were also collected and identified using morphological descriptors, and will be subsequently analyzed by AFLP technique. Since this project has been concluded this year, this analysis and other follow-up activities is being continued by a new CBN project under Small Grant Scheme, which was approved and initiated this year (see 'Projects under CBN-LAC Small Grants Scheme' section) Achievements and constraints • A methodology adjusted for the adaptation of in vitro material to the greenhouse phase and its subsequent planting in the field. Plants were placed for 8 days in water and then transferred as such to the definitive planting site, where they were planted in plastic bags containing sterilized soil. This practice eliminates the additional expenses occurred when transporting the material between the laboratory and the farms, while also reducing the possibility of lodging during transportation. • Continued incorporation of new equipment and less expensive reagents into the process. A laminar flow chamber was built with local materials, costing 10 times less than an imported chamber. • Advances were also made in the adjustment of the MS basal media. Cassava seedlings showed best growth when the commercial product Ferrovital-NF was used. • Use of the rapid propagation system based on two-budded cuttings by the group of farmers to complement the in vitro system. With this methodology, farmers multiplied nearly 6000 plants of the variety MCOL 1468 from in vitro material certified by the Colombian Agriculture and Livestock Production lnstitute (ICA, its Spanish acronym). • Establishment of a bank in farmers fields with in vitro material from 6 clones of interest to farmers (MCOL 1522, HMC 1, CMC 523-7, CM 6740-7, :MBRA 383, and MPER 183) to provide FSD-free material (FSD = frog skin disease). When these clones are harvested in 2003, they will be multiplied by the rapid propagation system, using two-budded cuttings, and distributed to local farmers. • Collection of 14 native cassava varieties in the municipalities of Caldono, Piendamó, Morales, Santander de Quilichao, and Caloto (Department of Cauca) and their identification using the morphological descriptors applied by CIAT's germplasm bank. A sample of each variety was planted on two farms in the municipalities of Caldono and Piendamó, as well as in the greenhouse at CIA T -Palmira. These varieties will be evaluated in CIA T s Biotechnology Unit Lab, using AFLP to determine the degree of diversity among materials. • Systematization of the traditional knowledge of local men and women regarding native varieties, especially their tolerance to different stress factors and their resistance to the 378 most common pests and diseases affecting cassava in the region, uses, starch quality, and use of plant parts (other than roots). • Definition of a participatory work scheme involving men and women to strengthen the fonnal cassava seed production system using in vitro technology and rapid propagation in the rural communities of Bajo Santa Ana, Alto Santa Ana, and Quinamayó. • Easy adoption by the group of farmers of the rapid propagation technique using two- budded cuttings. This method was implemented by the project as complementary to the in vitro method and fanners described it as easy and simple, and said it produced immediate results. Outstanding results • An in vitro cassava seed production scheme to be managed by small fanners was defined. The scheme consists of 6 phases: (1) receipt of certified material ; (2) in vitro multiplication in the rural laboratory; (3) adaptation in the greenhouse of the material multiplied through tissue culture; (4) planting of material in farmer fields with adequate management of irrigation, pest control, and fertilization; (5) multiplication of cuttings obtained in the field by rapid propagation (two-budded cuttings); and (6) production of good quality cuttings through the conventional system. • During both phases of this project, rural laboratories were built with their corresponding work areas, and the use of locally available, low-cost materials in their construction was evaluated. In addition, most of the components of the culture medium, such as salts, honnones, and sugars, were replaced with products obtained in the local market. The multiplication rate (1 :3-4) obtained when local inputs were used to prepare the culture media was similar to that obtained when imported raw materials were used. • Severa! of the native materials collected in central northem Cauca, for example Algodona Amarilla and Algodona Grande, present good starch quality and high percentage of starch. Therefore, an agreement was reached with several farmers and starch producers to increase the area planted to these two materials using in vitro propagation. FIDAR assumed production costs and CIA T provided the certified materials. These two cassava clones will hopefully be able to compete better with the varieties imported by starch factories in Ecuador and other regions of Colombia. • The women belonging to the group were able to reach a consensus about their needs and ways to solve them. Priority will be given to the search for income-generating altematives. The in vitro production of cassava seed is one of these altematives, but because of the time involved, they decided to participate in complementary production activities such as the planting of cassava, pineapple and bean, individually and in association, and the multiplication of commercial cassava varieties using rapid propagation. • The results of group participation reported in this phase are characterized by the empowerment and commitment of fanners, who have showed interest in undertaking other initiatives that benefit the community, for example a program to improve the quality and coverage of basic secondary education and allowing the access of young people and adults, who because of economic problems cannot attend nearby schools. An improved educational level has allowed the group to better understand and analyze the processes of in vitro technology. Difficulties with the work plan • The cost of an in vitro cassava plant produced in the rural laboratory was estimated at US$0.29 (see final project report 2001). This value is still quite high and not sufficiently competitive for small farmers to purchase these plants. The project continued working on this aspect during the current reporting phase and was able to integrate in vitro production technology with the rapid propagation system using two- budded cuttings. The data needed to determine whether the integration of these two systems reduced costs should be available at the end of 2003. • New systems are being studied to reduce the cost of handling and transporting cassava seed in the greenhouse, its definite adaptation in farmers' fields, and the diversification of use of rurallaboratories for tissue culture of banana and fruit trees. • The sharing of a common language and the leve! of confidence demonstrated by the group of farmers in the facilitating farmer ensured that project participants completed the six phases of the in vitro cassava production process in the rural laboratory and the multiplication in the field. However, the lack of academic training prevented several farmers from converting measurements of volume and weight and from making decisions to sol ve problems of contamination. • Macro and microeconomic factors affected community participation in the project, for example: - High fluctuations in the exchange rate of the Colombia peso versus the US do llar - Low price of cassava roots in Colombia over the last 10 months - Legal and illegal importation of sour starch by large companies in the region - Increase in the number of agronomic and plant health problems (whitefly and diseases such as frog skin) over the last two years However, the adoption of several complementary measures (production projects, training in cassava cultivation) has allowed the group to continue. The importance of acting in an 380 organized fashion to achieve project/group objectives in benefit of the community has been recognized. Communication and dissemination of information Project results have been disseminated at the local and national levels through the participation in different forums and workshops on topics related to fanner application of cassava tissue technology. • Seminars and Workshops Researchers and technicians working wüh the project participated, m 2002, in the following seminars and workshop: First Regional Workshop on Rapid Propagation (In Vitro) and Genetic Transformation of Cassava. CIAT, 25 February - 2 March 2002. Biotechnology in the Development of Colombia. First Colombian Congress on Biotechnology. Universidad Nacional de Colombia. Bogotá, June 2002. Intensive training course in modern cassava production and processing systems. CIA T, June 2002. Rapid propagation as a technology to support the multiplication of in vitro cassava materials by farrners. Santa Ana (Cauca), September 2002. Cassava seed production workshops held for the farrner associations of El Agrado, the Toez Indigenous Councíl, and La Arrobleda. August, September, and October 2002. Publications During 2003, two articles sumrnarizing the project's experience wiii be subrnitted to the journals Illeia and Scientific American Latinoamérica Tissue culture for farrners: participatory adaptation of low-input cassava propagation by a resource-poor rural cornrnunity in southern Colombia. R.H. Escobar, C.M. Hemández, G. Ospina, J. Restrepo, L. Muñoz, J. Tohme, and W.M. Roca. 2003. Applicación de la tecnología de propagación in vitro para producir semilla de yuca (Manihot esculenta Crantz) por pequeños agricultores: descripción y análisis de la experiencia. J.M. Restrepo, G.l. Ospina, C.M. Hemández, R.H. Escobar, J. Tohrne and W .M. Roca. 2003 Lessons learned and future work By implementing and assessing this in vitro cassava seed multiplication technology, farmers were able to maintain informal cassava seed production systems that yielded propagation rates of 1:3-4 every 45-60 days. The system was capable of producing 3250 plants per initial explant (plant), reaching an efficiency of 400% compared with the conventional vegetative seed propagation system currently used by regional farmers. Outstanding results were also obtained in the identification of technical parameters to build a low-cost rural laboratory, that could be easily operated by farmers, as well as of equipment and inputs to prepare the culture media, achieving an efficiency similar to that of specialized laboratories. In addition, the farmer-farmer training methodology implemented by project researchers and technicians proved to be correct because it ensured that participants understood the concepts and acquired the skills needed to operate the rural laboratory. However, more time was needed than that initially planned for the group to understand and self-manage the different processes (prepare the culture media, plant the tissues, hardening the plants, establish plantlets in the field). Although the simplification of the tested technology significantly reduced the costs of the infrastructure, inputs, and culture media used, the labor costs implied by the different processes continue to be quite high and make it impossible for farmers to assume in vitro seed multiplication. It is therefore important to continue evaluating new systems and diversify the use of the laboratory with other crops to ensure its long-term sustainability. Mechanisms must also be sought to attract the participation and support of different local institutions. The use of in vitro technology by farmers is an altemative that solves the problem of availability of good quality seed, especially in the case of new varieties or when seed is scarce because of climatic and plant health problems. The rural laboratory can also be used to multiply native cassava varieties of northem Cauca. These are currently being identified and cleaned at CIAT's Biotechnology Unit Laboratory for subsequent redistribution to the communities for their in situ conservation and multiplication of seed of those varieties enjoying greatest acceptance by farmers and starch producers. 382 BRAZIL 3.2.2 Farmer participatory in vitro cleaning and multiplication of local and improved cassava varieties Institution: Embrapa Cassava and Fruits (Embrapa/CNPMF) Address: Caixa Postal 007,44380-000, Cruz das Almas, Babia, Brazil Ernail: cfukuda@cnpmf.embrapa.br Collaborating institution Empresa Baiana de Desarrollo Agricola (EBDA) Farming communities of Caetité (Southeast of Babia, Brazil) Staff directly in volved Chigeru Fukuda (Coordinator), Embrapa Wania Fukuda, Embrapa Osvaldo Pereira da Paz, Embrapa Osório Vasconcelos, EBDA Josué Cerqueira, EBDA Introduction The project took off effectively in March 2002 with the general objective of introducing and implementing participatory biotechnology methodologies with small-scale cassava farmers in the cleaning and multiplication of cassava varieties. This is to be done using the low-cost rapid in vitro multiplication techniques developed at CIA T. The project site is the Mania~u region, Caetité municipality, Bahia, Brazil. Achievements and constraints • Training for Osvaldo Pereira da Paz (from Embrapa) on low cost cassava in vitro rapid multiplication techniques in the Biotechnology Research Unit of CIA T. The training was performed in two farmer communities: a) Santa Ana, in Santander de Quilichao city, where is located a low-cost in vitro laboratory, and b) San Rafael, in Jamundi city, where was observed a local low-cost facilities for hardening in vitro-produced material to produce clean seeds for rapid propagation and distribution of clean plants to the farmers. • Visit of the technician responsible for low-cost biotechnology (Osvaldo Pereira da Paz) and social scientist (Maria das Gra9as Sena) to seven Caetité's communities (Tanquinho, Contendas, Junquinho, Lagoa do Barro, Passagem de Areia, Ingazeira and Lagoa de Fora) in order to select the pilot si te for the work of low-cost mu1tiplication. In this oportunity, a great interest of all the communities was observed in participating actively of the work. There is a large necessity of clean and improved seeds in the region. The farmer' s time availabily for training and water quality and availability for the work were analyzed. Considering these aspects was chosen the community 'Lagoa de Fora' . This community has available area for clone multiplication, water of good quality and farmers able to receive training and transfer techniques of low-cost cassava propagation. This community housing 60 families, with approx. 300 people, which are highly dependent of the cassava crop. Two women farmers and two EBDA's technician were selected to attend a training on low-cost cassava propagation at the pilot laboratory built in the CNPMF/Embrapa. • Construction of the pilot laboratory for low-cost cassava multiplication at CNPMF/Embrapa. This lab will be the base for training of farmers and technicians, starting with the four persons already selected. • Cleaning and in vitro multiplication of the following clones: 003, 005, 1318, 1389, 1393, Aipim Cachorro (Local), and Laza (Local). In CNPMF, there are 1000 plants of these clones to be transfer to the low-cost lab at 'Lagoa de Fora'. • Seeds (cuttings) of the improved and local selected varieties were supplied to CNPMF's Biotechnology Unit for meristem multiplication and cleaning. Difficulties with the work plan The initial work plan was delayed on account of personnel changes at CNPMF/Embrapa leading to the late take off of the project, effectively March 2002. Since this date, however, the pace of work has moved really fast. There is ever indication that the project will expand to other regions from Caetite. This prognosis is based on the numerous requests for the sighting of the pilot si te in severa! other places and in the general enthusiasm observed amongst the farmers, extension agents and community leaders from other regions. Communication and dissemination of information The project has received a lot of publicity through the radio e.g. the local radio station, Radio Educadora de Caetité, meeting with farming communities, leaders of rural associations, Caetité Municipality, women associations, and the exchange of information between the farmers 384 ECUADOR 3.2.3 Diagnosis of the use and production of cassava in Manabí province Introduction In order to update existing infonnation on the status of cassava in Ecuador, a diagnostíc study on the production and utilization of cassava in the Manabi Province of Ecuador was carried out as a pilot study. Manabi is the major cassava producing area of Ecuador. The main objective for this study is to determine the components of the major cassava systems in Manabi and by so doing relate cassava production and use with social and environmental dynamics in the communities. In all, 650 surveys were conducted in Manabi and the data from this already inputted. The analysis is on going. At the end of 2001 and beginning of 2002 were carried out the surveys, concluding the part corresponding to data acquisition. Due to lack of personal for data analysis and transition of CBN coordination staff, the evaluation was delayed and started at the beginning of 2003. Visits to Ecuador were perfonned in order to get the data and to update sorne infonnations related to cassava cultivation with communities and farrner's associations. These informacions will allow CBN to undertake actions to complete the research and to provide sorne support to the sector of small production of cassava. This study will serve as means for evaluating the status of cassava projects that have already been executed in Ecuador, constitute a guide for the execution of development in the cassava cultivating community of Manabi and also serve as the base line data for planning other development projects. CBN has in close collaboration with PROA identified sorne pararneters that could be used to elicit certain urgent information from the data collected so that further work in Ecuador could be commissioned and these are summarized thus: a number of questions that would give a good indication of preferences that could be analyzed by gender (detectable by the respondent's narne), age, wealth (as measured by landholdings and tenure status), principal occupation and agroecological zone. • Questions that relate to the importance of cassava overall compared to other productive activities or community priorities include: - preference for cassava vs. other crops, - importance of cassava for family subsistence, - importance of cassava to household well-being, and - indication of what are considered the most important problems faced by the community. • Questions that indicate preferences, problems and priorities for cassava include: - input requirements and constraints, - disease problems, - varieties known, used, preferred and reasons, - yield/production, - preferred characteristics for cassava plants and reasons - responses may overlap with previous, - demand for mix of varieties, - might give a rough idea of the importance of cassava for marketing vs. subsistence, - importance of being able to delay harvest, - asks about seed knowledge and supply within the community. May give an indication of the importance of cassava, as well as the degree of innovation and cohesion present in the community, - importance of cassava by-products, - varietal preferences of buyers of cassava or cassava products, - difficulties encountered in marketing fresh cassava, cassava starch and cassava flour, - difficulties encountered in processing (spec grating), and - productivity and profitability of cassava over time. • Questions that give an indication of for whom cassava is important and the degree of voice/asset control that person has are: - who decides variety selection, - division of labor for cassava and other household production activities, - control over income from sales of cassava and cassava products, and -control over household income and expenses) 3.2.4 Projects under CBN-LAC Small Grant Scheme Introduction The CBN-LAC Small Grants Scheme aims at achieving the following: • Foster cassava biotechnology projects with developing country - developed country or advanced laboratory linkages; and • Permit and encourage developing country laboratories to work on cassava biotechnology research topics relevant to end-users' , national and overall CBN objectives. This small grants scheme primarily supports the planning of specific proposals for intemational collaborative research on priority cassava biotechnology research needs as defined by cassava end-users and farmers. Sorne grants are also awarded in the form of grants-in-aid for developing country biotechnology research operational expenses; for 386 emergency biotechnology research bridging funds; or for short-term training in specific biotechnological methods. Due to unforeseen tragic occurrences, none of these grants could be awarded earlier than early 2003 implying a reserve of funds for this line item. CBN Coordination decided to make 2 rounds of awards for 2003. The first round of the awards advertised late 2002 has been approved. In all, a total of 11 awards amounting to US$116,900 were made to cassava research groups in the 4 countries where the network maintains pilot sites. The distribution of these wards by country and by research theme is presented in Tables 1 and 2 In Table 3 is showed the projects and their status. The Small Grants Scheme was meant to be competitive but on account of the peculiarity of the region, especially with regards to relative inexperience with competing for research funds, CBN Coordination had to extend deadlines and ask for the revision and resubrnission of a majority of the grants. Table 1- Distribution ofCBN-LAC SmaU Grants Awards for 2003 by country Country Brazil Colombia Cuba Ecuador Total No. of awardees 4 1 3 3 11 Table 2- Distribution of CBN-LAC Small Grants Awards for 2003 by research theme Research Area Genetic Transformation Geno mies Germplasm Conservation and Characterization Tissue Culture, Rapid Multiplication and Seed Cleaning Total No. of awardees 2 3 4 2 11 Table 1 - Projects or the CBN Small Grants Awards 2003 Proiect Tille Country lnsti tution Award (US$) Period Genetic manipulation of proline biosynthesis in cassava aiming to increased tolerance to Brazil Embrapa Cassava and Fruits 10,000 May/03 to water stress Apr/04 Development of protocols for genetic transformation of cassava genotypes from Northeast Brazil Federal University of Ceará (UFC) 10,000 May/03 to Brazil Apr/04 lsolation of genes in volved in the sugary phenotype in the storage root of cassava (Manilwt Brazil Embrapa Genetic Resources and 10,000 May/03 to esculenta Crantz) Biotechnology_ Apr/04 Maintainance and characterization of cassava living collection at the University of Brasflia Brazil University of Brasilia (UnB) 7,000 Dec/03 to Nov/04 Use of in vitro technology by small farmers to clean and preserve native cassava varieties Colombia Fundación para la Investigación y 13,000 Mar/03 to in Southem Colombia's Andean Region Desarrollo Agrícola (FIDAR) May/04 Rescue and production of high quality seed of local cassava and cocoyam varieties by Cuba Instituto Nacional de Ciencias 9,900 Jun/03 to biotechnology tools adapted to Cuban rural conditions Agrícolas (INCA) May/04 Application of new techniques to control sorne viral diseases in cassava (Manillot Cuba Instituto de Investigaciones de 10,000 Dec/03 to esculenta Crantz) in combination with massive propagation tec.hnique. Viandas Tropicales (INIVI1) Nov/04 Application of molecular tec.hniques for genotypes differentialion of Manihot spp. Cuba Instituto de Investigaciones de 10,000 Jun/03 to Viandas Tropicales (INIVI1) Mar/04 Establishing the contribution of Manihot leptophylla to the genetic constitution of cassava Ecuador Pontificia Universidad Católica del 13,000 Jun/03 to and the differentiation of sweet and bitter types Ecuador (PUCE) May/04 In situ conservation of cassava varieties cultivated by Kichwas from • Alto Napo' Ecuador Federación de Organizadores de la 12,000 Dec/03 to Nacionalidad Kichwa de Napo Nov/04 (FONAKIN) Cleaning of released varieties and recovery of cassava genetic resources in production area Ecuador Instituto Nacional Autonomo de 12,000 Dec/03 to ofEcuador Investigaciones Agropecuarias Nov/04 --- (INIAP) 388 3.2.5 Genetic manipulation of proline biosynthesis in cassava aiming to increased tolerance to water stress Institution: Embrapa Cassava and Fruits (Embrapa/CNPMF) Address: Caixa Postal 007,44380-000, Cruz das Almas, Bahia, BRAZIL Email: adilson@cnpmf.embrapa.br Collaborating institution Instituto Agronómico do Paraná (IAP AR), Brazil Staff directly in volved Adilson Kenji Kobayashi (Coordinator), Embrapa/CNPMF Alfredo A. C. Aives, Embrapa/CNPMF Femanda Vidigal Duarte Souza, Embrapa/CNPMF Luiz Gonzaga Este ves Vieira, IAP AR Introduction The generation of transgenic cassava plants has a high potential as a complement to traditional plant breeding program in improving agriculturally valuable traits. However, plant genetic transformation efficacy of cassava, like many other crops, requires the establishment of efficient protocols for plant regeneration. Cassava plant regeneration efficiency is known to be genotype dependent, varying between 5 and 70% (Zhang et al., 2001) which may constrain the use of genetic transformation technology in cultivars with Jow regeneration frequency. Therefore, the first step of our research project was aimed to the establishment of efficient plant regeneration protocols of the selected genotypes. Simultaneously, elaboration of the gene constructs and introduction in the Agrobacterium was also performed. Methodology Plant material: Cuttings of cassava genotypes MCol 22, Aramaris, Aipim-Brasil and three genotypes selected for their high ~-carotene content (BGM1153, BGM1692 e BGM1728) were grown in pots under greenhouse conditions. Establishment of in vitro culture of explant donor plants: One month-old newly formed shoots of all six genotypes were surface decontaminated with 10% (v/v) of a commercial solution of sodium hypochlorite for 15 minutes and rinsed five times with sterile distílled water. Shoot tips were excised and transferred to culture flasks containing E4 medíum (Table 1) for establishment of vitro conditions. After three weeks of culture on E4 medíum, the explants were transferred to 17N medium (Table 1) for multiplication. Multiplied shoots were then transferred to MS (Murashige & Skoog, 1962) medium without growth regulators for root formation. All cultures were maintained at 27±1 °C under 16 hours photoperiod. Table l. Macronutrient, micronutrient, vitamins and growth regulators composition of the culture media used in the establishment of in vitro explant donor plants. Macronutrient (mgll) 4E 17N NI4NÜJ 1650 577.5 KN03 1900 665 CaCh.2H20 450 154 MgS0 •. 7H20 370 129.5 KH2PO• 170 59.5 FeS04.7H20 27.8 9.73 Na2EDT A2H20 37.3 13.055 Micronutrient (mgll) KI 0.83 0.291 H~03 6.2 2.17 MnS0 • .4H20 22.3 7.805 znso •. 1H20 8.6 3.01 Na2Mo04.2H20 0.25 0.088 cuso •. 5H20 0.025 0.009 CoCh.6H20 0.025 0.009 Thiamine - HC1 1.0 1.0 Inositol 100 100 ANA 0.02 0.01 BAP 0.04 AG 0.05 0.01 Sucrose 20 gil 20 gil Agar 7gll 7gll Phytagel 1.8 gil 1.8 gil Gene construct and transformation vector preparation: The binary plasmid pBI121 containing the P5CS gene from Vigna aconitifolia under the control of the constitutive promoter CaMV35S, with the genes nptll (neomycin phosphotransferase, for resistance to the antibiotic kanamycin) and uidA CP-glucuronidase or GUS, reporter gene) was amplified using the strain DH5a of Escherichia coli. Bacteria were grown ovemight on 25 mlliquid LB medium. Plasmids were harvested using the mid-prep alkali lysis protocol (Sambrook et al .,1989). The binary plasmid was transferred to the disarmed hypervirulent Agrobacterium tumefaciens strain EHA105 were transformed by electroporation. Bacteria! cultures were grown overnight at 28°C, in liquid YMB medium (0.5 g r K2HP04, 0.1 g 1"1 MgS04.?H20, 0.1 g 1"1 NaCl, 5 g r 1 glucose, 10 g r 1 de mannitol and 0.4 g r 1 yeast extract) supplemented by 20 mg r1 rifampicin and 30 mg r1 kanamycin. Glycerol at 20% final concentration was added to the culture. The bacteria! culture was separated in aliquots in 1.5 mi Ependorf tubes and stored at -80°C. 390 Results In vitro culture of explant donor plants was established for the six genotypes. Plants are currently on rooting medium (Figure 1). This step is essential in order to obtain plant material for the following experiments on plant regeneration and genetic transformation using protocolos based on Roca et al., (1984) and Mathews et al. (1993) as we had proposed in the research project. Transformation vector in the Agrobacterium strain EHA105 was concluded and, in addition, a gene construct containing the gene bar (resistance to the herbicide ammonium glufosinate) instead of nptil is currently under way. This gene has also an agronomic interest and has been successfully used in other cassava genotypes (Sarria et al., 2000). Figure l. Explant donor plants of the six genotypes on rooting medium Observations The present research project is according to the proposed time table presented in the grant application. Three genotypes of cassava with high ~-carotene content were added in the experiments. Dr. Femanda V. D. Souza (CNPMF) was included in the research group and Dr. Luiz F. P. Pereira (fonner researcher at IAPAR), unfortunately, had to leave the research group due to job exchange. References Mathews, H. ; SchOpke, C.; Carcamo, R.; Chavarriaga, P.; Fauquet, C.; Beachy, R.N. (1993). Improvement of somatic embryogenesis and plant recovery in cassava. Plant Cell Rep. 15, 328-333. Roca, W.M.; Rodrigues, J.A.; Mafia, G.; Roa, J. (1984). Procedures for recovering cassava clones distributed in vitro. CIAT, Colombia. Sambrock, J.; Fritsch, E .F.; Maniatis, T. (1989). Molecular cloning: a Iaboratory manual. 2nd edition. Cold Spring Harbar Laboratory Press, Cold Spring Harbar. Sarria, R.; Torres, E.; Angel, F.; Chavarriaga, P .; Roca, W.M. (2000). Transgenic plants of cassava (Manilzot esculenta) with resistance to Basta obtained by Agrobacterium-mediated transformation. Plant Cell Reports 19: 339-344. Zhang, P.; Phansiri, S.; Puonti-Kaerlas, J. (2001). Improvement of cassava shoot organogenesis by the use of silver nitrate in vitro. Plant Cell Tissue and Organ Culture 67: 47-54. 3.2.6 Development of protocols for genetic transformation of cassava genotypes from Northeast Brazil Institution: Federal University of Ceará (UFC) Address: Caixa Postal6039, 60451-970, Fortaleza, Ceará, BRAZIL Ema.il: bioplant@ufc.br Collaborating institution Centro Intemational de Agricultura Tropical (CIA T), Colombia Embrapa Tropical Agroindustry (Embrapa/CNP AT), Brazil Staff directly involved Francisco A. P. Campos (Coordinator), UFC Terezinha Feitosa Machado, DSc Student, Embrapa/CNP AT Paul Chavarriaga, CIAT Introduction This is a training program in cassava transformation technology at the Biotechnology Unit of CIA T aiming to develop protocols for the gene tic transformation of major cassava varieties from Northeast Brazil. The graduate student ·who is being trained is Terezinha Feitosa Machado who is a research scientist from Embrapa/CNPAT and PhD student at UFC. The main steps for this project are: a) developing friable embryogenic calli (FEC) for cassava varieties adapted to NE Brazil; b) developing transformation protocols using Agrobacterium carrying marker genes; and e) monitoring progress on modified cassava plants 392 Methodology The cassava varieties that are being evaluated are listed in Table 2. The plant material prepared in Brazil and brought to CIA T were soma tic embryos obtained from stem apex induced on medium MS2-50Pi (salts and vitamins MS, sucrose 2%, CuS04 2~, Pidoran 50~, Agar 0.6%) Table 2 - Cassava varieties from NE Brazil that are being evaluated for genetic transformation CNPMF BRA Collected in Code Code Vulgar name (City-State) Institution BGM 0365 000086 ÁguaMoma Pacatuba-CE EPACE BGM0394 008087 Rosinha Mage-RJ IPEACS BGM0260 007277 Rosa N.S. das Dores-SE IPEAL BGM0549 012611 Amansa Burro ltapirema-PE IPA BGM0123 004502 Aparecida Cruz das Almas-BA EAUBA BGM0004 005533 Milagrosa Cruz das Almas-BA !PEAL BGM 1063 073245 Tapicina Berberibe-CE CENARGEM BGM 1467 102482 Bujá Preta na na na = not available The embryos are being subcultivated in the mediums MS2-50Pi, MS2-BAP 2~ (salts, vitamins MS, sucrose 2%, 0.4 mg/L of BAP and agar gelrite 3:1); and medium GD2-50Pi (salts and vitamins GD, sucrose 2%, 50J..LM (12mg!L) of picloram, agar-gelrite 3:1). The genetic transformation is being performed in green cotiledons using the technology of Agrobacterium tumefaciens with lineage Agl-1 1305-2 pCambia. The transformation process is going on. 3.2.7 Isolation of genes involved in the sugary phenotype in the storage root of cassava (Manihot esculenta Crantz) Institution: Embrapa Genetic Resources and Biotechnology (Embrapa/CENARGEN) Address: Parque Estacrao Biológica -Final W3 Norte, Brasilia, DF, BRAZIL Email: carvalho@cenargen.ernbrapa. br Collaborating institution Universidade Estadual de Santa Cruz (UESC), Brazil Staff directly in volved Luiz J. C. B. Carvalho (Coordinator), Embrapa/CENARGEN Claudia Regioa B. de Souza, Postdoctorate Student Julio Cezar Cascardo, UESC Introduction This project provide a financia! suport to a post-doc student to perform a research that addresses the issue of food security in Brazil by improving food quality through exploitation of novel Amazon cassava genotypes with increased diversity of storage root quality traits. The research focuses on the biochemistry and molecular genetic analysis of recently discovered genotypes with unusual properties of accumulating sugar, and starch in the storage root. Modern genome biotechnology will be used to study the biodiversity and biology of the storage roots of cassava. The variability of free sugar and novel starch accumulated in the storage root of identified wild clones of cassava will be explored by a) biochemical characterization of starch synthesis and degradation; b) isolation of mutated candidate genes with homologous and heterologous genes sequence coding for enzymes of the starch synthesis and degradation pathway, and gene expression (RNA) analysis as well as gene sequence variability; ande) isolation of genes differentially expressed in these contrasting phenotypes by sequencing several subtractive cDNA libraries. The main outputs from this project are: • Better understanding of the regulatory biochemical mechanism on the diversity of starch naturally occurring phenotypes. ldentification of the molecular structure of the isolated novel starch, and amylopectin structure phenotype isolated. ldentification of the molecular structure of the candidate gene coding for the enzyme responsible for the diversified amylopectin structure isolated. Obtain biochemical tools to be used in the identification and isolation of natural diversity of starch structure in cassava gennplasm. Obtain molecular biology tools to be used in the identification and isolation of new clones of cassava. Offer new alternative of use of the storage root of cassava to the farmer and the industry. · 394 • Clone and sequence the cDNA of known genes coding for starch synthesis and degradation enzyme in cassava. Gain knowledge about the presence or absence of specific gene coding for starch synthesis and degradation enzyme by gene expression analysis at the leve! of mRNA. Obtain molecular biology tools to be used in further advanced technology to alter starch structure in local cassava varieties and/or other crops. Gain knowledge on the structure and regulation regions of the genes of starch synthesis enzyme, needed to establish new technological strategies. • Gain knowledge of the sequence of the genes differentially expressed, that would help to explain the storage root tissue phenotype in sugary clones. Generate a data base of EST from the diversity of the storage root of cassava. Obtain genomic tools to be used in data mine to dissect the formation of the storage root of cassava. Training scientist on the new technology as well as promote the distribution of the biotechnology to the poor. Methodology Plant material Field grown plants in Embrapa Genebank collection, composed of 26 high sugar clones. Experimental strategies Multiple experimental strategies are being used, depending on the starch phenotype. The high sugar class of storage root allowed to identify three types starch phenotypes, being amylose free starch, phytoglycogen starch type and differentiated branched amylopectin when compared with starch from the commercial cassava. Therefore each type of phenotype is worked under appropriated framework of mutation already identified in other plant system. In the amylose free starch clone, the strategy uses the waxy mutation framework of rice and com, and uses the candidate gene of the granule bound starch synthase as target gene. In the phytoglycogen accumulating clone the strategy use the sugary mutation framework of pea and com, and the candidate gene coding for the debranching enzyme is the target gene. In the highly branched arnylopectin it uses the soluble starch isoforms mutation frarnework of com, and the candidate gene coding for the soluble starch synthase I is the target gene. The differential expressed gene of the high sugar storage root in relation to the commercial cassavas is used to build an EST database with the sequence of a subtractive cDNA library to generate the needed molecular tools. Biochemical characterization For the biochemical characterization will be performed the following main steps: • Storage root tissue sarnpling and preparation for carbohydrate analyses; • Reducing sugars quantification; • Quantification of glucose, sucrose, total starch, and apparent amylose; • Storage root tissue sampling and preparation for enzyme activities and protein quantification; • Soluble protein quantification; • Estimation of Proteins bound to the starch granule; • Activity of the enzymes involved in the starch synthesis: ADP-Glucose Pyrophosphorylase (AGP), Soluble Starch Synthase (SSS), Granule-Bound Starch Synthase (GBSS), Starch Branching Enzyme (SBE), and Starch Debranching Enzyme (SDE); • Storage root tissue sampling and preparation for starch granules characterization; • Microscopy of the starch granules observation; and • Estimation of size and specific area of starch granules. Results Functional genomics applied to the diversity of storage root of cassava Six subtractive cDNA libraries were assembled and are being sequenced. The results so far, can be summarized as follow: 1) A collection of more than 22000 cDNA clones has been assembled. 2) A pilot sequence initiative generated more than 2500 high quality sequence (PHRED>20). 3) Conventional RNA blots probed with cDNA probes belonging to several classes of genes confirmed the efficiency of the subtraction techniques used to building the differential expressed cDNA Iibrary. 4) Several classes of new genes have been revealed in our preliminary analysis of the sequence of our EST bank. These new tools resources will be used in this project to isolate the genes related to sugary phenotype. 396 3.2.8 Maintainance and characterization of cassava living collection at the University of Brasilia Institution: University of Brasilia (UnB) Address: Campus Universitario- Asa Norte, CEP 70919-000, Brasília, BRAZll... Email: nagnassa@rudah.com.br Staff directly involved Nagib M. A. Nassar (Coordinator), UnB Introduction The target of this project is a cassava living collection, housed in University of Brasilia, which consists of rare wild Manihot species, that have been extincted in their natural habitats, and sorne interspecific hybrids. In order to preserve this collection, which has 16 wild species and 13 hybrids of cassava with wild species, this project aims for the following objectives: • To propagate old and degenerated accessions of both wild and interspecific cassava hybrids • To characterize these introductions morphologically and cytogenetically • To improved the field facilities for maintaining the living collection • To create an adequate database system for data characterization • To create a seed genebank Outputs The main outcome of the project will be the maintainance of wild species of cassava that already disappeared in their natural habitats from complete extinction, saving the rare examples of interspecific hybrids from degeneration and extinction to enable use them in the future for cassava improvement, using their invaluable gene pools. 3.2.9 Use of in vitro technology by small farmers to clean and preserve native cassava varieties in Southern Colombia's Andean Region Institution: Fundación para la Investigación y Desarrollo Agrícola (FIDAR) Address: Calle 6A n.61-109, AA 25074, Cali, COLOMBIA Email: fidar@colombianet.net Collaborating institution Centro lntemational de Agricultura Tropical (CIA T), Colombia Staff directly involved José Restrepo (Coordinator), FIDAR Gloria Ospina, FIDAR Rooseve lt Escobar, CIA T Joe Tohme, CIAT Carlos Hernández, Farmer Results Collection of varieties Thirteen new native materials of yuca were collected in the municipalities of Buenos Aires, Caldono, Morales and Santander of Quilichao. These materials with the 14 gathered at the end of the second semester of the 2002 were sowed in three nurseries in the fanners' parcels in the municipalities of Caldono (1.500 m), Piendamó (1.700 m) and Santander de Quilichao (1.000 m). For each one of these materials fields in the nurseries were evaluated the following morphological describers: color of apical leaf, pubescens, shape of central lobe, plant height, height of first rarnification, degree of branching, length of the filament, branching habit, habit of stem growth, color of the stem epidermis, color of the root cortex, color of the terminal rarnification, flowering, plant type, peduncle in root, cortex color, root pulp color, texture and constrictions of the root. For each one of these collected materials a collection record was elaborated, with the purpose of comparing them with the database of the Genetic Resources Unit of CIA T. Cleaning and identi.fication of the varieties In different nurseries, the technicians of CIA T ha ve taken samples of each varieties with the purpose of carrying out meristems cultivation starting from mature stakes of each genotype. With the in vitre materials will begin a termoterapy process and of the materials coming from the selected stakes young leaves were harvested and total DNA was extracted by protocol of Dellaporta (Figure 1). The DNA was quantified by fluorometer. 398 Figure 1 - A) Gel showing the total DNA extracted from young leaves of mature stakes of the local varieties collected in Cauca. B) Plants of six varieties of commercial interest ready to be propagated and distributed to the farmers. Farmers' verbal registration about the collected cassava varieties With the purpose of systematizing the local farmers knowledge about the native and introduced varieties of cassava collected in the North of the Cauca, meetings were carried out with sorne groups using open type questions and obtaining the following answers: Question 1: What is the opinion that yo u ha ve about the native cassava varieties m relation to those introduced or improved? Opinions are divided, 60% of the farmers has preference for the native materials, arguing that they have seed permanently and most of the varieties ha ve double purpose (starch and fresh consumption). However, they consider that sorne native varieties have diminished their growth and development and many of them are disappearing. In relation to the improved or introduced varieties they consider that they are more precocious, they have higher production and they get less disease and pests, but sorne of them after being cultivated during 3 or 4 cycles show low yield. Another difficulty is the lack of planting material. Question 2: What are the main characteristics ofthe collected varieties? In Table 1 a summary is presented on the farmers' opinions about the collected varieties. Question 3: What are the main factors that the fanners keep in mind to cultivare a native variety? • Farmers from medium and high zones ( 1400 - 1700 m) - The seed should be obtained in the region. - The variety should be for double purpose (fresh consumption and starch). - Good content and quality of starch. - Good yield and resistance to pests and diseases. - Do not need so much fertilizers. - The varieties of more acceptance in order of importance for the farmers from high zones are the following: "algodona grande" (verdadera algodona), "bajuna", "chirosa" y "parroquiana". • Farmers of low or warm area (900-1300 m) - The variety should be for double purpose. - Good yield and resistant to pests and disease, especially to white fly and frog skin. - Varieties with more acceptance in this area are: "Rojita", "Raya 7", "Falsa chirosa". Question 4: What other parts of the cassava, different to the roots, are used by the communities? - The root peel is used as organic fertilizer. - The leaf is used for feeding of cattle and pigs, the same as the by-products of starch processing. - The hearts are used as greenness for the sausages or stuffs. - The water from starch processing is used to control stomach aches and diarrheas. 400 Table 1 - Main characteristics of the collected varieties according to the farmers VARIETY(Iocol name) COUECTION'S PLACE PRODUCTION'S POTENTIAL ADVANTAGES OF THE VARIETY LIMITATIONS 1- PARROQUIANA Vereda La Unión (Piendamó). 1700 m. Thick roots and of good size when it is lt is known in the region since 30 years. It lt is very late, needs more than 11 It has been lcnown for 30 years in the harvested after 16 months. Yield is is to1erant to the excess of humidity and months to obtain good yields. region. good when ferti lizer is applied. has little disease incidence. lt is used for fresh consumption. 2- VARITA Vereda La Unión (Piendamó). 1700 m. Regular production and it produces Big and small roots stuck to the stem, lts yield in the last years is not ver It has been known for 1 O years in the little planting material. favorable to the harvest. good. re~ion. 3- FALSA CHIROSA Vereda Pescador (Caldono). 1500 m. lt Good yield and formation of roots. lt is well accepted in the market due to lt is susceptible to Phytopthora. has been known for 12 years in the cortex pink color. 1t has good market for region. fresh consumption. 4- ALGODONA GRANDE Vereda El Mango (Piendamó). 1700 m. The yield is good when the weeds are lt is a variety of high acceptance for those lt is late, it needs from 16 to 18 mooth lt has been lcnown for 40 years. controlled and is fertlized. lt resists the that produce starch, for its quantity and to produce good starch. drought. Good ramification and seed quality. The true 'algodona'. production. 5- ROJITA Vereda San Miguel (Buenos Aires) Using stake of good size its yield is lt is used for fresh consumption and lt produces little planting material. 1100 m. lt has been known for 14 better. starch. In the last years the processors of years. starch prefer it for starch quality and precocitv. 6- ALGODONA GIGANTE Vereda El Mango (Piendamó). 1700 m. lt presents good yield when fertilized lt has high acceptance by the processors lt is susceptible to bacteriosis. lt need 1t has been known for 12 years. and weeds controlled. lt adapts better of starch for its quality and quantity. between 16 and 17 months to produc above 1500 m. well. 7- VENENOSA O Vereda Mary López (Buenos Aires - lt is a rustic variety that supports lt is a variety for industrial use lts production is low when the see MATASUEGRA Cauca). lt has been known for 20 drought. (production of starch). doesn't have good quality. lt is no years. good for fresh consumption. 8- ALGODONA RAPIDA Vereda La Independencia (Piendamó). Variety of good yield when fertilized Good quality of starch. lt is susceptible to Phytopthora. 1500m. and when doesn't receive a lot of water. 9- SATA Vereda La Marfa (Piendamó). 1500 m. lt presents good yield, especially when lt presents early branching avoiding the Thin stems and it is a late variety (1' lt has been known for 15 years. using thick seed of good size. proliferation of weeds. lt is tolerant to months to be harvested). bacteriosis. 10- AMARILLA Loma de los reyes (Piendamó) 1700 m. Plant of good size, stems of good Good acceptance for fresh consumption lt takes between 16 and 18 months t' lt has been known for 25 years. diameter and short intemodes. Good due to its thick roots with conical shape. obtain good yield. yields especially when fertilized with chicken manure. 11- VALLUNA Vereda El Mango (Piendamó). 1600 lt has good yield when fertilized and Variety of double purpose (starch and lt presents not well formed roots, but i m. lt has been known for 15 years. lt the weeds controlled. fresh consumption). is very accepted for fresh consumpti01 was brought from coffee area of the by the families . 'Valle'. 12- TOTOQUENA Vereda el Mango (Piendamó). 1600 m. Variety of good yield, big and medium lt is good for fresh consumption and to lt presents so me difficulties fo lt has been known for 12 years. roots. process starch. lt is a precocious variety, harvesting (pull out the roots) can be harvested at 12 months. 13- CHIROSA El Porvenir (Caldono) 1500 m. lt has Variety of good yield. lt produces big, 1t is very good for the fresh consumption The plant material found in the regio' been known for 12 years. lt was medí u m and small roots. market. lt gets better prices than other of 'Caldono' and 'Piendamó' is ver introduced by severa} farmers from varieties. deteriorated and its productioo is low. · ~· VARIE7Y(Ioca/ nan!i!) COUECfiON'S PLACE PRODUCf/ON'S POTENTIAL ADVANTAGES OF THE VARJE7Y UMITATIONS Armenia (Quindfo). 14· VERDEO VERDECITA Vereda Independencia (Pieodamó). hs yield in this area is not the best, it It is a very good variety for fresh It doeso't present a lot of starch. It is 1500 m. It was introduced from behaves well between 1000 and 1300 coosumption. very susceptible to caterpillars. Buenos Aires to Caldono. m. 15· PATA PAVA La María (Piendamó). 1600 m. h has It presents good yield with big and It is very accepted for the fresh Late variety, needs between 16 and I8 been lcnown for 1 O years. medium roots. consumption in the farm. months to obtain good production. 16- BAJUNA Piendam6. 1400 m. It has been lcnown Good yield when ferti lizcd and the h is good for fresh consumption and Late variety, needs from 15 to 18 for 40 years. It is one of the oldest weeds is controlled in the first months. starch production. months. The plant material is varieties. deteriorated .. 11· CORREITA Pescador (Caldono) 1500 m. h has Its yield is not very good. The planting Variety recommended for fresh Late variety, needs between 16 and 18 been lcnown for 10 years. material is deteriorated. consumption in the farm. No very good months. It doesn't produce so much acceptance by pests. planting material. 18· MEJORADA Vereda Independencia (Piendamó), It presents good yield and thick roots of Variety of double purposc (fresh lt needs good fertilizers. The cortex INDEPENDENCIA 1500 m. Variety introduced by Fídar good size. consumption and starch production). color is not very attractive for the and CIA T since 4 years. market. 19- CHlROSA ROJA La María (Piendam6) 1700 m. h has lt has an acceptable yield when using h presents very good acceptance for the The plaot material is oot of good orBATATA been lcnown for 20 years. planting material of good quality. fresh market. quality. Befare it produced more. 20- BLANQUITA El Turco (Santander de Quilichao) Good yield when using planting I has very good acceptance for processing This variety is disappearing for the loss I 300 m. It has been lcnown for more material of good quaJity. duc to quality and quantity of starch. of the quality of thc plant material. than 1 O years. 21- YUCA PAPA Mondomo (Cauca) 1300 m. lt has been Good yie1d whcn the weeds is It is a variety of doublc purposc (fresh lt is not vcry attractive for market due known for 15 years. controlled in the first months. consumption and starch). It is used for to shape and color of the roots. fami ly consumption. 22· REGIONAL MORADA Piendamó (S. Jost) 1700 m. It has Its yield has been good (25 ton/ha) but h is used for fresh consumption and has Susceptible to bac teriosis. been lcnown for 10 _y_ears. in the last y_ears ha ve diminished. ve'l'Aood acceptance for market. 23· PANAMENA Cajibfo (Cauca) 1800 m. Not very well- h presents good production poten tia l. h is used for fresh consumption and Susceptible to bacteriosis. known varicty, it was introduced by starch. CIAT. 24- SM 850-1 Cajibfo (Cauca) 1800 m. Variety It has good yield especially when It is a variety of double purpose (fresh h doesn' t produce so much plant ' iotroduced by CIA T. fertilizcd. COOSUm_lltiOil and starch). material. 25· CHlROSA (2) La María (Piendam6). 1650 m. Jt is It presents good yield. Good quality for fresh consumption in the The color of the cortex is not pink and known as 'Clúrosa' but is not the true farm. therefore it is not very good for market onc. 26- YUCA BLANCA Vereda El Carmen (Piendamó). 1700 It presents a good yield ( 18 ton/ha). It is a variety of double purpose. Good h doesn't produce so much plant m. It has beco known for more than 1 O The plant material is being deteriorated. cootent of starch. material. Many branclúng. years. 27- SAUCE Villa Clemencia (Piendam6). 1700 m. Low yield. Can be used for frcsh consumption and Susceptible to bacteriosis. i h has been known for more than 15 starch. years. -- -- 402 Evaluation and maintenance of cassava varieties multiplied in vitro During the last 8 months the evaluation and monitoring of the materials multiplied in vitro by farmers, in the first phase of the project, were continued. The varieties or materials that are available to be harvested in September and then multiplied by fast progatation technique through rooting induction of 2 nodes stakes, are the following: HMC-1, MBRA 383, MPER 183, MCOL 1522, CM 6740-7 and CM 523-7. These materials multiplied by fast propagation will be planted in the farmers' plots in San Rafael, Santa Ana and Caldono with the purpose of evaluating their yield and sanitary quality of the planting material by technicians and farmers. Communication and socialization of the information The diffusion of the project's results has been carried out at locallevel in forums with the schools of Santander of Quilichao, where the project's advances were presented toa group of 12 teachers interested to know how to use the different tools used for cassava in vitro propagation. During the second semester a group of 6 teachers from 'Femández Guerra' Scholl, in Santander of Quilichao will be trained in in vitro seed propagation of cassava. With the participation of the farmers 'group of Santa Ana and Caldo no the training on cassava seed quality and the need for conserving local varieties, have continued. Emphasis is also been focused in aspects related to better nutritional content of cassava and beans varieties, to include them in the diet and to strengthen the food security and the nutritio of the families. AH these activities are guided to strengthen the organization and the labor participation of men and women in equal form in the different home and cornmunity acti vities. Also, sorne ideas have been discussed on low cost cassava in vitro propagationt, with University's professors (National University and Nariño University, from Colombia) and with Dr. James Gorges from CTCRI, India. Difficulties in the work plan The main difficulties during this phase of the project were related to the following aspects: - Sorne collected materials were confounded in their identification due to different vulgar names in the same region. This situation was corrected identifying them in the field using morphological descriptors by CIAT's expert. - The materials sent to CIAT and multiplied in vitro have presented a slow growth and it has been necessary to change the cultivation medium to stimulate their growth. These two difficulties have retarded the proposed chronogram of activities. - In relation to the participation of the farmers' groups (Santa Ana and Pescador) sorne farmers' demotivation has been reported dueto low prices of the cassava roots and starch, situation that has diminished the planted area. Future activities - To begin AFLP's evaluation. - To continue the thermoterapy treatments and in vitro cultivation until finding negative plants for frog skin. - To certify for frog skin by graft with 'Secundina' variety. - Once certificated, the multiplication activities should start to give the materials to the groups of Santa Ana, Caldono and Piendamó. - To begin the activities of fast propagation and to plan a field day to delivery materials to the farmers involved in the project. - To facilitate the delivery to CIAT of the native materials for low areas. - To elaborate a new proposal to present to a donor in order to keep in mind the advances achieved by the project. - To strengthen the participation and the farmers' organization linked to the project keeping In mind the activities that men and women carry out in the communities. - To consolidate and to strengthen the training of the farmers for handling m vitro technology. 404 3.2.10 Rescue and production of high quality seed of local cassava and cocoyam varieties by biotechnology tools adapted to Cuban rural conditions Institution: Instituto Nacional de Ciencias Agrícolas (INCA) Address: San José de las Lajas, La Habana, CUBA Email: rnhedez@inca.edu.cu Collaborating institution Instituto de Investigaciones de Viandas Tropicales (INIVIT), Cuba Staff directly in volved Maria Margarita Hemandez Espinosa (Coordinator), INCA Humberto Ríos Labrada, INCA Lorenzo Suárez, INCA Miruldys Valcárcel, INCA Jorge López, INNIT Aymé Rayas, INIVIT Background The main objective of this project is the rescue and multiplication of high quality planting material of cassava and cocoyam varieties selected according to farmers' requirements and preferences, through biotech tools adapted to the conditions of the selected rural communities. The application of in vitro techniques for micropropagation of cassava and cocoyam varieties will facilitate the increment of the yields in these crops due to high quality of the seed produced and better tolerance to main pests and diseases, since the planting material will be free of contaminations. This effort is a high priority for food security in Cuba and also will open new possibilities for incorporation of the experience from rural women and technicians in the rural communities of Cuba. Methodology and results Diagnostic This phase had two activities: 1) Survey'application to know how the flow of cassava and cocoyam seeds is performed within communities, its entrance and exchange points, and leadership in these crops. With the results of this survey will be evaluated the relationship among farmers' groups and their roles within the processes of acquisition, distribution and use of planting material ; and 2) Realization of the first workshop with participation of the farmers from severa! regions involved in the project, as well as researchers, farmers' association representatives, local decision-makers politicians, and other stakeholders; the objective of this workshop is to divulgate the project, discuss the general problems related to cassava and cocoyam cultivation, and establish the priorities of future actions. These two activities have been accomplished and the data analysis is in process. Farmers' training andfarmer participatory research on in vitro plant adaptation During the first workshop many farmers have visited facilities utilized for micropropagation, perceived more details about this technique, and observed the genetic diversity of cassava and cocoyam when visiting INIVIT's collections. In this phase was accomplished the second project's workshop aiming to establish the characteristics of in vitro plants and to arrange local conditions for in vitro plants of cocoyam, which currently are being multiplied and will be distributed to the actors in November 2003. Also, an adjusted timetable of activities and agreements were established. Clones of cassava and cocoyam, selected by farmers, have been rnicropropagated. In vitro cocoyam plants and cassava stakes from micropropagated plants from INIVIT's elite clones will be cultivated in farmers' experimentation fields. Two actors from INCA have attended a training at INIVIT about cassava and cocoyam rnicropropagation and farmers from La Palma, after training, will assemble the facilities for in vitro plants climatization. These activities have been performed in the municipality of Santo Domingo, province of Villa Clara (homogeneous environment), and La Palma, community of San Andrés, province of Pinar de Río (heterogeneous environment). 406 3.2.11 Application of new techniques to control sorne viral diseases in cassava (Manihot esculenta Crantz) in combination with massive propagation technique Institution: Instituto de Investigaciones de Viandas Tropicales (INIVIT) Address: Apartado 6, Santo Domingo, 53 000, Villa Clara, CUBA Email: inivit@enet.cu Collaborating institution Centro Internacional de Agricultura Tropical (CIA T), Colombia Staff directly involved Ricardo Hemández, INIVIT Graciela Mafia, CIA T Norma Cristina Flor, CIA T Benjamín Pineda, CIA T Introduction The systemic pathogenic organisms, as virus, spread with the planting material generation after generation; these agents not only affect the yield of the local cassava cultivars but al so limit the maintenance of the germplasm bank, the regional and intemational exchange of clones and the supply of certified seed for extensive cassava production. Such pathogens are the cassava vein mosaic virus (CVMV), cassava common mosaic virus (esCMV), cassava virus X (esVX) and the agent of the cassava frog skin disease (CFSD). This project will become available an altemative technology for virus elimination, with less time and less infrastructure requirements, which will decrease the cost of basic seed production of cassava. This outcome will facilitate the adjsutment of biotech tools for massive multiplication of certified seeds by cassava farmers, mainly those involved with cassava production for industrial starch. Outputs • Altemative technology for cassava virus elimination. The procedure (electrotherapy) would replace the current thermotherapy method applied in cassava (40° e at day-light and 35° e at night, during 3 to 4 weeks) to obtain, in combination with apical tissue culture, plants free of virus and other systemic pathogens. This technique will speed the growth of in vitro plants and get more regenerated plants. • Personnel trained on application of electrotherapy at eiAT and other institutions interested in production of cassava planting material free of virus. • Reduced cost of production of free disease cassava plants. • Cornmunications in speciaJized forums. 3.2.12 Application of molecular techniques for genotypes differentiation of Manihot spp. Institution: Instituto de Investigaciones de Viandas Tropicales (INIVIT) Address: Apartado 6, Santo Domingo, 53 000, Villa Clara, CUBA Email: inivit@enet.cu Collaborating institution Centro Internacional de Agricultura Tropical (CIAT), Colombia Staff directly involved Y oel Beovides García (Coordinator), INIVIT Martin Fregene, CIA T Introduction With the development of molecular biology, several molecular markers have been used for identification of genetic variability in cassava cultivars from different regions and countries (Fregene et al., 1997; Fregene et al., 2003). The rnicrosatellite technique or SSR (Simple Sequence Repeats) is attractive to study dueto its abundance in the plant genome, its high polimorphism and adaptability to the automation. It has been appropriate for germplasm cassava characterization (Dixon et al. , 2002; Azudia et al., 2002), and can contribute to evaluate the genetic diversity within cassava Cuban collection and to establish the relationships between the accessions and their wild or cultivated relatives. The objective of this study is to evaluate the cassava genetic diversity and its phylogenetic relationships with culti vated relatives of Africa, South and Central America, in order to assist a sustainable management of the Cuban genetic resources. Methodology A total of 94 cultivars was collected in May 2003 in the Cuban cassava collection, according to its economic and/or genetic importance. Also 54 clones from Africa and America were incorporated (12 from Nigeria, 10 from Tanzania, 12 from Guatemala and 20 of South America), and other 13 genotypes of genetic interest. Total DNA was extracted following Dellaporta et al. (1983); DNA concentration and quality were evaluated by fluorometry and electroforesis in gel, respectively. The samples of DNA were diluted to a concentration of 10 ng/ml for the subsequent analyses of PCR and to evaluation of 36 SSR markers that represent a wide covering of cassava genome. The PCR amplification for diversity studies with SSRs, the electroforesis in poliacrilamida gel and the dye with sil ver were developed according to adjusted approaches from Fregene et al. (2002). 408 PreJJminar results A total of 18 SSR has been evaluated, which have showed a high polimorphism, high amplification level and a great number of bands per primer (Figures 1 and 2), except the SSRY-132 primer that was monomorphic not allowing differentiation among the individuals. Figure 1 - PCR amplification of cassava cultivares from Cuba, Nigeria, Tanzania, Guatemala and South America with primer SSRYSl. Figure 2 - PCR amplification of cassava cultivares from Cuba, Nigeria, Tanzania, Guatemala and South America with primer SSRYlSl. Other SSR Y primers will be evaluated up to complete 36. The reading and determination of band's weight will be carried out by the program 'Quantity one' and the statistic analysis will be performed using the following softwares: "Microsat","Gensurvey" and "FSTAT". A comparison matrix among countries will be designed with Fst values and the relationship will be analyzed by a cluster analysis, using the method UPGMA of NTSYS- PC. References Azudia C, L. Monte, D. Debouck y M. Fregene. 2002. Simple Sequence Repeat (SSR) Marker Assesment of Genetic Diversity of Cassava Land Races from Guatemala. Annual Report-2002, Project IP3: lmproved cassava for the developing world. CIAT-2002, Output 8-17. Dixon A, A. Raji, J. Mario , M. Fregene. 2002. Simple Sequence Repeat (SSR) Marker Assesment of Genetic Diversity of Cassava Land Races from Nigeria. Annual Report-2002, Project IP3: lmproved cassava for the developing world. CIAT-2002, Output 8-12. Fregene M, Angel F, Gomez R, Rodríguez F, Chavarriaga P, Roca W, Tohrne J, Bonierbale M (1997) A molecular genetic map of cassava. Theor Appl Genet 95: 431-441. Fregene M, M. Suárez, J. Mkumbira, H. Kulembeka, E. Ndedya, A. Kulaya, S. Mitchel, U. Gullberg, A. G. O. Dixon, R. Dean y s. Kresovich. 2003. Simple sequence repeats marker diversity in cassava landraces: genetic diversity and differenciation in an asexually propagated crop. Theoretical and Applied Genetics, accepted. Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg, U., Apel, K., Gale, M., Tobme, J. and Fregene, M. (2001) Simple Sequence Repeat (SSR) Markers Survey of the cassava (Manihot esculenta Crantz) Genome: Towards an SSSR-Bassed Molecular Genetic Map of Cassava. Theoretical and Applied Genetics. 102: 21-31. 3.2.13 Establishing the contribution of Manihot Ieptophylla to the genetic constitution of cassava and the differentiation of sweet and bitter types Institution: Pontificia Universidad Católica del Ecuador (PUCE) Address: Av. 12 de Octubre y Roca, Quito, Ecuador Email: anarvaez@puce.edu.ec Collaborating institution Institut Francais pour la Recherche et le Developpement (IRD), France Staff directly involved Alexandra Narváez-Trujillo (Coordinator), PUCE Carolina Portero, PUCE Juan Lizarzaburu, PUCE Gérard Second, IRD Introduction Allem (2002) proposes that M. leptophylla, a species whose distribution range includes southern Colombia, coastal Ecuador, Perú and reaches to Belém in Brazil, could be part of the synonymous complex of sub-species which includes M. esculenta ssp. esculenta, M. esculenta ssp. flabellifolia, and M. esculenta ssp. peruviana. The M. leptophylla botanical type is reported from the coastal province of Manabí in Ecuador. Research dedicated to elucidate the position of M. leptophylla, and whether it may have contributed to the genetic constitution of cassava on the western side of The Andes was based on the following initial observations: i) extensive review of herbariurn specimens indicated that only specimens from the western coast of Ecuador and the southwestern part 410 of Colombia are morphologically similar to the M. leptophylla type reported from the Manabí Province of Ecuador and ii) specimens of the species available from the Amazon Basin are morphologically distinct from those from the West leading us to believe that there may be a misidentification of specimens, given the plasticity of Manihot species. In this case, specimens fonn the eastern side of the Andes more probably correspond to M. jlabellifolia than to M. leptophylla, which would be restricted to the lowlands on the westem side of the Andes in Ecuador and Colombia. Molecular marker studies using AFLPs (Narváez & Second, 2002) clearly indicated that M. leptophylla is genetically very djstant from cassava and from M. jlabellifolia. Eventhough AFLPs give a total evaluation of molecular diversity given that they are distributed in the total genome and are considered reliable markers for diversity studies, their potential in an outbreeding plant such as cassava may be limited. Furthennore, as one of the objectives of this study is to evaluate possible hybridization ancl/or introgression microsatellite markers were chosen for their reproducibility, enabling compilation of data from different data sets and for their high allelic diversity as co-dominant markers which could generate markers that indicate specific introgressed si tes on the genome of the species under evaluation. The objectives of this project include: a) To deterrrune the relation of Manihot leptophylla to the other species of the genus and in particular to cassava (Manihot esculenta ssp. esculenta) and b) To provide proof that M. leptophylla has contributed to the genetic constitution of cassava through hybridization and introgression, events that may have also led to the genetic differentiation between sweet and bitter cassava types. Methodology Plant samples Samples from wild putative M. leptophylla populations at different geographical sites in Ecuador were collected in silica gel throughout 2002-2003. Peral plants encountered at these sites were also collected. Collection number and site are listed in Table l . M. esculenta samples were collected in traditional fields in the westem provinces of Manabí and Esmeraldas, collection number and si te are listed in Table 2. A selection of samples for SSR analysis was made considering geographical location and ecological characteristics. The final data sets (Data 1 and Data 2) are listed in Tables 3 and 4. Primers An initial set of 15 mapped primers (Mba, et al 2001) were selected based on previous studies on cassava (Morillo, 2002) for preliminary evaluation of the data sets. These primers are: SSRY3 (CA) 17, SSRY9 (GT) 15, SSRY30 (CT) 22, SSRY31 (GA) 21, SSRY38 (CA) 17, SSRY40 (GA) 16, SSRY55 (GA) 16, SSRY68 (CT) 12 CC( CT) 17, SSRY80 (GA) 25, SSRY103 (GA) 22, SSRY108 (CT) CCT, SSRYlOO (CT) 17 TI( CT) 7, SSRY135 (CT) 16, SSRY169 GA19A3GAA2, SSRY179 (GA) 28. PCR conditions were based on Mba et al (2001); silver staining was perfonned according to Promega®. Data analysis included principal coordinate analysis using NTSYS v. 2.1. Results Results were obtained for each of the data sets. Data set 1 included samples from different geographical origin within Ecuador for M. leptophylla in order to establish genetic relations to M. esculenta, which included samples of bitter and sweet types, the first represented by samples from French Guiana and the latter from westem and eastem localities in Ecuador. Samples of M. flabellifolia included ones from French Guiana and from the CIA T world collection. Samples for Central american species were al so obtained from CIAT. As can be seen in Figure 1, the first axis clearly shows a differentiation of M. leptophylla, asan extreme in the PCoA analysis in regard to M. esculenta (bitter and sweet types). M. jlabellifolia samples, from French Guiana, is found at a center position. M. leptophylla could actually be genetically more similar to M. aesculifolia, although these two species differentiate on the second axis of variation. Regarding M. leptophylla, there seems to be a geographical differentiation. The samples Antl , Ant2 and Ant7, the most extreme in the PCoA analysis (Figure 1), are from the geographical site from were the botanical type is reported, Camarones, Manabí Province, Ecuador; therefore these can be considered "true" M. leptophylla. The samples from other localities, especially those from Guayas and Los Ríos Provinces warrant further analysis. Analysis of allele diversity shows that M. jlabellifolia has the largest number of alleles (Table 5), followed by M. esculenta and M. leptophylla. Of the species analyzed, M. jlabellifolia shares a large number of alleles with M. esculenta, while M. leptophylla and M. esculenta have few common alleles (data not shown). The second sample set (Data set 2), includes samples of M. leptophylla and M. esculenta, especially traditional varieties grown in the coastal provinces of Manabí and Esmeraldas, as well as putative hybrids and feral plants. In this case, there is also a clear differentiation of M. leptophylla from M. esculenta. Within M. leptophylla there is not a high genetic diversity (Figure 2) while within M. esculenta there is a large diversity (revealed on the second axis of variation); especially noteworthy are the feral plants that may represent hybrid plants and, within M. leptophylla, sample Ant17 which has unique alleles and also an interesting position in the PCoA analysis (Figure 2). These results give way to considering a new sample set, combined from the two described here. For this purpose, further sampling of Ríos and Manglares-Churute (Guayas) sites must be carried out. 412 Table 1- List ofwifd (Manütat leptophylla) or feraJ cassava SampleNo. Reference code Species Locality AntO 1-04-0 1 M. leptophylla Tabuga, Manabí Ant02-04-0l M. leptophylla Tabuga, Manabí Ant03-04-0 1 M. leptophylla Camarones, Manabí Ant04-04-0 1 M. leptophylla Camarones, Manabí AntOS-04-0 1 M. leptophylla Tabuga. Manabí Ant06-04-01 M. leptophylla Camarones. Manabí Ant07 -04-01 M. leptophylla Tabuga, Manabí Ant08-04-0 1 Ecu 1- Gerard M. leptophylla San Mateo, Esmerladas Ant09-04-0 1 Ecu 2- Gerard M. esculenta Esmeraldas Antl0-04-01 Ecu 3- Gerard M. esculenta Caimito, Esmeraldas Ant 11-04-0 1 Ecu 4- Gerard M. leptophylla Quingue, Esmeraldas Ant12-04-01 Ecu 5- Gerard M. leptophylla San Mateo, Esmeraldas Ant13-04-0 1 Ecu 6-Gerard M. esculenta Unión del Toachi, Pichincha Ant 14-04-0 1 M. brachyloba Jatun Sacha, Napo Ant15-04-01 CB-Guayas M. leptophyl/a BP Cerro Blanco, Guayas Ant17-02-02 Mgll- Manglares 1 M. leptophylla Cerro Pancho Diablo, Guayas Ant18-02-02 PPl- Patricia Pilar 1 M. leptophylla Patricia Pilar, Los Ríos Ant19-02-02 PP2- Patricia Pilar 2 M. leptophylla Patricia Pilar, Los Ríos Ant20-02-02 Mgl2- Manglares 1 M. leptophyl/a Cerro Pancho Diablo, Guayas Ant21-02-02 Mgl3- Manglares 2 M. leptophylla Cerro Pancho Diablo, Guayas Ant22-20-12-01 ETl-Cerro Blanco M. leptophylla BP Cerro Blanco, Guayas Ant23-20-12-01 ET2-Cerro Blanco M. leptophylla BP Cerro Blanco, Guayas Ant24-20-12-0 1 ET3-Cerro Blanco M. leptophylla BP Cerro Blanco, Guayas Ant25-20-12-01 ET4-Cerro Blanco M. leptophylla BP Cerro Blanco, Guayas Ant26-12-09-02 #9 colección de la Tola M.esculenta La Tola del Indio, Guayas Ant27-12-09-02 #8 colección de la Tola M.esculenta La Tola del Indio, Esmeraldas Ant28a Ecu 7 .1-Gerard M.esculenta germinations Ant28b Ecu 7 .2-Gerard M.esculenta Ant28c Ecu 7 .3-Gerard M.esculenta Ant28d Ecu 7 .4-Gerard M.esculenta Ant28e Ecu 7 .5-Gerard M. esculenta Table 2 • List of local cassava varieties from Manabí and Esmeraldas Provinces SamEleNo. Local name Species Localidad Province Coordenates MOOl-03-2002 Mcol2215 M. esculenta Bijahual Manabí N 0002 29.1 W77 19 36.9 M002-03-2002 Cascaruda "2" M. esculenta San Vicente Manabí S 01 04 32.3 w 80 19 32.2 M003-03-2002 Cascaruda M. esculenta San Vicente Manabí S 01 04 32.4 w 8019 32.3 M004-03-2002 Blanca M. esculenta San Vicente Manabí S 01 04 32.5 w 80 19 32.4 M005-03-2002 3 Meses M. esculenta Sapaniyal Manabí S 01 05 13.2 w 80 20 17.5 M006-03-2002 Peruana M. esculenta Quebrada Guillén Manabí S 01 04 38.7 w 8018 28.5 M007-03-2002 Negra M. esculenta Quebrada Guillén Manabí S 01 04 38.8 w 8018 28.6 M008-03-2002 Tres meses M. esculenta Charapotó Manabí S 00 49 51.6 w 8029 34.9 M009-03-2002 Yuca de año M. esculenta Charapotó Manabí S 00 49 51.7 w 8029 34.1 M010-03-2002 Tres meses M. esculenta San Clemente Manabí S 0045 59.7 w 80 30 03.5 M011 -03-2002 Colorada Colombiana M. esculenta San Clemente Manabí S 00 45 59.8 w 8030 03.6 M012-03-2002 Yema de huevo M. esculenta Tablones, Junín M a nabí S 00 57 13.5 w 80 13 56.0 M013-03-2002 Tres meses M. esculenta Tablones, Junín Manabí S 00 57 13.6 w 80 13 56.1 M014-03-2002 Blanca M. esculenta Junín Manabí S 00 5624.6 w 8014 10.6 M015-03-2002 Colorada M. esculenta Junín Manabí S 00 56 24.7 w 8014 10.7 M016-03-2002 Tres meses M. esculenta Junín Manabí S 00 56 24.8 w 8014 10.8 M018-03-2002 Huevo cambiado M. esculenta Potrerillo, Calceta Manabí S 0052 24.2 w 8010 58.7 M019-03-2002 Huevo cambiado M. esculenta Potrerillo, Calceta Manabí S 00 52 24.3 w 801058.8 M020-03-2002 La babita M. esculenta Potrerillo, Calceta Manabí S 00 52 24.4 w 8010 58.9 M021-03-2002 Seis meses M. esculenta Salida de Calceta Manabí S 00 50 40.1 w 8010 22.1 M022-03-2002 Yema de huevo M. esculenta Pay Pay Manabí S 0048 40.9 w 80 13 33.1 M023-03-2002 Espada M. esculenta Pitahaya Manabí S 00 48 17.9 w 8015 00.6 M024-03-2002 Mulata M. esculenta Pitahaya Manabí S 00 48 l7.l0 w 8015 00.7 M025-03-2002 Yuca de año M. esculenta Jama Manabí S 00 12 16.5 w 8015 35.4 M026-03-2002 Tres meses M. esculenta Jama Manabí S 00 12 16.6 w 8015 35.5 M027-03-2002 Blanca M. esculenta Jama Manabí S 00 12 16.7 w 8015 35.6 M028-03-2002 Pata de Paloma M. esculenta Pedernales Manabí N000406.3 w 80 03 20.5 M029-03-2002 Espada M. esculenta Pedernales M a nabí N000406.4 w 80 03 20.6 M030-03-2002 Cimarrona 1 M. esculenta Bilsa Esmeraldas N 00 17 54.7 W79 57 04.3 M03l-03-2002 Cimarrona 2 M. esculenta Bilsa Esmeraldas N 0017 54.8 w 79 57 04.4 M032-03-2002 Cimarrona 3 M. esculenta Bilsa Esmeraldas N 00 17 54.9 W7957 04.5 M033-03-2002 Cimarrona 4 M. esculenta Bilsa Esmeraldas N 00 17 54.10 W 79 57 04.6 M034-03-2002 Amarilla M. esculenta Estero Ancho Esmeraldas N 00 42 07.3 W79 54 53.8 M035-03-2002 Blanca M. esculenta Estero Ancho Esmeraldas NOO 4207.4 w 79 54 53.9 M036-03-2002 Negra M. esculenta Agua Clara Esmeraldas N 00 46 16.4 W7955 49.9 M037-03-2002 Añera M. esculenta Albergues, Tonsupa Esmeraldas NOO 52 43.5 w 79 48 57.3 M038-03-2002 Blanca M. esculenta Albergues, Tonsupa Esmeraldas N 00 52 43.6 W7948 57.4 M039-03-2002 Negra M. esculenta Albergues, Tonsupa Esmeraldas N 00 52 43.7 W7948 57.5 M040-03-2002 Yema de huevo M. esculenta Albergues, Tonsupa Esmeraldas N 0052 43.8 w 79 48 57.6 M17-03-2002 Serrana M. esculenta LaMijarra Manabí S 0053 46.0 w 8010 55.7 414 Table 3 • Data set 1 · wild Manihot species aod M. esculenta Sample No. Species Observation Origin VRH3 M.esculenta sweet Napo, Ecuador VRH12 M.esculenta sweet Napo, Ecuador M20 M.esculenta sweet Manabf, Ecuador M37 M.esculenta sweet Manabf, Ecuador GI M.esculenta bitter French Guiana G20 M.esculenta bitter French Guiana G2l M.esculenta bitter French Guiana G28 M.esculenta bitter French Guiana G31 M.esculenta bitter French Guiana G34 M.esculenta bitter French Guiana G92 M.esculenta bitter French Guiana GI12 M.esculenta bitter French Guiana GI25 M.esculenta bitter French Guiana GUYI5 M. baccata French Guiana GUY17 M. baccata French Guiana GUY22 M. baccata French Guiana GUY23 M. baccata French Guiana GUY25 M. baccata French Guiana GUY 32-1 M. flabellifolia French Guiana GUYI M. flabellifolia French Guiana GUY36 M. flabellifolia French Guiana GUY41-2 M. flabellifolia French Guiana Aa318 M. flabellifolia CIAT world collection Per4-12 M. peruviana CIAT world collection Ant 1 M.leptophylla Manabf, Ecuador Ant 2 M.leptophylla Manabf, Ecuador Ant6 M.leptophylla Manabí, Ecuador Ant7 M.leptophylla Manabf, Ecuador Ant12 M.leptophylla Manabf, Ecuador Antl9 M.leptophylla Manabf, Ecuador Ant 21 M.leptophylla Manabf, Ecuador N°l M.leptophylla Manabf, Ecuador NOJ M.leptophylla CIA T world collection Chl2 M. chlorostica CIA T world collection ChiS M. chlorostica CIA T world collection Chl9 M. chlorostica CIA T world collection Maes l M. aesculifolia CIA T world collection Maes3 M. aesculifolia CIA T world collection Maes7 M. aesculifolia CIA T world collection Rub4 M. rubricaulis CIA T world collection Rub29 M. rubricaulis CIA T world collection Aal9 M. flabellifolia CIA T world collection Aa6l M. flabellifolia CIA T world collection Aa68 M. flabellifolia CIA T world collection Ecu75 M. esculenta CIA T world collection Table 4 - Data set 2 - M. esculenta local varieties and wild M. leptophylla Sample No. Species Origin Blo 001 M.brachyloba CIAT world collection Ecu 82 M.esculenta CIAT world collection Ecull6 M.esculenta CIAT world collection M31 M.esculenta Esmeraldas, Ecuador M32 M.esculenta Esmeraldas, Ecuador M33 M.esculenta Esmeraldas, Ecuador M lO M.esculenta Manabí, Ecuador M24 M.esculenta Manabí, Ecuador M25 M.esculenta Manabf, Ecuador M26 M.esculenta IManabf, Ecuador Antl4 M. brachyloba Napo, Ecuador M02 M.esculenta IManabf, Ecuador M04 M.esculenta Manabí, Ecuador M05 M.esculenta Manabí, Ecuador M09 M.esculenta Manabí, Ecuador M15 M.esculenta Manabí, Ecuador M17 M.esculenta Manabí, Ecuador Ml8 M.esculenta [Manabf, Ecuador Ml9 M.esculenta IManabf, Ecuador Ant 1 M. leptophylla Manabf, Ecuador Ant2 M. leptophylla Manabf, Ecuador Ant3 M. leptophylla Manabí, Ecuador Ant5 M. leptophylla Manabí, Ecuador Ant 8 M. leptophylla Esmeraldas, Ecuador Ant9 M. Ieptophylla Esmeraldas, Ecuador Ant 10 M. Ieptophylla Esmeraldas, Ecuador Antll M. leptophylla Esmeraldas, Ecuador Ant 13 M. leptophylla Pichincha, Ecuador Ant 15 M. leptophylla Guayas, Ecuador Ant17 M. Ieptophylla Guayas, Ecuador Ant 18 M. leptophylla Los Ríos, Ecuador Ant20 M. leptophylla Guayas, Ecuador Ant22 M. Ieptophylla Gua y as, Ecuador Ant25 M. Ieptophylla Guayas, Ecuador Ant28a M. esculenta gerrninations Ant28b M. esculenta gerrninations Ant28c M. esculenta gerrninations Ant28d M. esculenta germinations M28 M. esculenta Manabí, Ecuador M36 M. esculenta Manabí, Ecuador 416 -0.34 ··· ;~··:·· ... o c) ANT7·M.¡· .•• 0ANT2-M-I \ N"1 ¡ o : ·••· .• Q.ANT12 :.~···/ ········· ,,--, 1 Maes1 ' o \ M. flabellifolla GUY32·1 (Fla) GUY41-2(~ GUY1 ANT19 ° Aa318 O o Loa Aloa o GUY36(Aa) Fla 19 Ecu75 ANT~M-c o o o Aa68 o PER412 o Ch30~a61 ~ o Rub4 UY o o :19 M. escu/enfJI {bl tter & sweet) ••••• •••••• VRH3 . • .• o :" G34 G31 \ !G92 00 ~12 ~ 0 Gi? oo ~ Ó O• M37·· o Gto ".. Gl Biner : ·. .· ...... ~~3 ,' Rl~---'~8 Tree manloc , , , M. aescullfolla __ ., -0.50 -0.24 0.03 X=15.2% 0.29 0.55 Figure 1 - PCoA analysis showing relations among seven Manihot species (M. leptophylla, M. esculenta, M. flabellifolia, M. baccata, M. aesculifolia, M. chlorostica and M. rubricaulis), based on analysis with 15 SSR primers (see text for details). Table 5 - Total number of alleles eer seecies corresponding to Data set 1 SSR Primer (Mba et al, 2001) Total Number Species 9 100 31 3 169 68 55 108 103 38 88 135 40 179 30 of Alleles M esculenta 6 5 5 3 3 6 8 4 6 3 5 5 3 6 4 72 M. baccata 2 4 2 1 1 3 2 2 2 1 3 3 2 30 M. jlabellifo/ia 4 7 7 5 6 6 7 5 6 5 3 4 5 6 4 80 M. peruviana 1 1 1 1 1 1 2 1 2 1 1 1 16 M. /eptophylla 3 6 6 5 6 4 5 3 4 3 2 5 2 4 3 61 M. ch/orostica 2 2 1 1 2 2 2 2 2 2 2 3 1 26 M. aesculifo/ia 2 2 2 2 2 1 2 2 1 2 2 2 2 26 M. rubricaulis 2 1 1 1 3 2 2 1 1 2 21 0.58 ~11 M. leptophyll• -o.3 '-419 M05 1.126'" Cultlvated M • ..cu/ent• J-428 ,'-424 '-436• r------------------- 1 1 1 J-417 Cultiva lid M. NC~Hnl6 J-418 :_~------------ ---- _1 f.nt28d FeraJ M. escu~M~m f.nt9 ¡..nl28a'~o . 7 La T ola-Esmenlldas -o.sa+---.--.---r---.--.---r---r-...-....---,--...---.--,--...-~~=;:=::;::::~~ .0.74 .0.45 -o.15 0.15 0.44 X=24% Figure 2 - PCoA analysis showing structuring of M. leptophylla and M. esculenta, based on analysis with 15 SSR primers (see text for details). Conclusions SSR data confirm AFLP data that indicated that M. leptophylla is genetically distinct from M. flabellifolia . Additionally, there seems to be a closer relationship to M. aesculifolia than to M. esculenta or M. flabellifolia. Within species diversity of M. leptophylla could be considered small, although there seems to be a geographical differentiation, being the samples found in Camarones, Manabí Province the most extreme on the first axis of variation, which is in agreement with the botanical data that point to this locality as the botanical type for the species. Samples from other geographical sites, specially from Guayas Province on the westem coast and from Los Ríos on the Andes foothills may be candidates for introgression. 418 Up to the moment there is no clear evidence for intogression of M. leptophylla and M. esculenta or for differentiation of sweetlbitter types based on genetic contribution of M. leptophylla. Future Plans Based on current results, further sampling from interesting sites is needed in order establish a more comprehensive data base that will enable a better evaluation of introgression. Additional primers to be included in the analysis are those used by Dr. Martín Fregene in cassava diversity studies (CIAT Anual Report 2002). Contact was established with Dr. Fregene in June of this year in which was requested an aliquot of these rutine primers in order to lower costs, Dr. Fregene also recommended using interna! controls, that would be supplied by himself, in order to make our data compatible with other data sets. These last points are the present constraints in order to continue work; the number of SSR loci analyzed up to date is low given that we had a failure with our power pack and therefore had to send it for repair. We have overcome this and have full capacity to advance quickly. Additional data analysis is necessary, especially related to genetic diversity. This will be carried out on the final data. Contact was established with the Instituto Nacional de Investigaciones Agropecuarias (INIAP), the Ecuadorian NAR, in order to transfer a copy of the material we have collected. It seems that there is an interest to make a new collection of cassava for which our collection data was made available to them. References Allem, A. 2002. The origins and taxonomy of cassava. In: L.J. Hillocks, J.M. Thresh and A.C. Belloti. eds. CAB Intemational 2002 Cassava: Biology, Production and Utilization. Mba, REC, P. Stephenson, K. Edwards, S. Melzer, J. Nkumbira, U. Gullberg, K. Apel, M. GaJe, J. Tohme, M. Fregene. 2001 Simple sequence repeat (SSR) marlers survey of the cassava (Manihot escu/enta CRANTZ) genome: towards an SSR-based molecular genetic rnap of cassava. TAG 102. 21-31. Morillo, E., F. Fuenrnayor, P. Cesar de Carvalho and G. Second. 2001 AFLP and SSR Polymorphism: Evidence of Significant Levels of lntrogression from Manihot glaziovii and M. carthaginensis into Traditional Varieties of Cassava in their Area of Origin. In: Fauquet, CM y Taylor. NJ, eds. Cassava: An ancient crop for modero times. [Compact disc] Proceedings 5th Intemational Meeting of the Cassava Biotechnology Network. 2001 November 4-9. St. Louis Mo. USA. Narváez-Trujillo, A. & G. Second. 2002. Manihot Ieptophylla (Euphorbiaceae), una especie restringida a la costa pacífica de Colombia y Ecuador y su relación con otras especies del género. En: Rangel-Ch., J.O., J Aguirre-C & M.G. Andrade-C (eds.) Libro de Resúmenes VIII Congreso Latinoamericano de Ciencias y Segundo Colombiano de Botánica. Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá. 3.2.14 In situ conservation of cassava varieties cultivated by Kichwas from 'Alto Napo' Institution: Federación de Organizaciones de la Nacionalidad Kichwa del Napo (FONAKIN) Address: Calle Augusto Rueda #242, Casilla Postal 217, Tena, Napo, ECUADOR Email: fointena @uio.satnet.net Collaborating institution Centro Internacional de Agricultura Tropical (CIA T), Colombia Staff directly involved Rita Mamallacta, FONAKIN Sergio Yurnbo, FONAKIN Cínthya Peñaherrera, FONAKIN Fabricio Guarnán, FONAKIN Elizabeth Caicedo, CBN-CIAT Roosevelt Escobar, CIAT Introduction Although cassava is the main crop for the nationality Kichwa of the Amazonía, it has not been considered in research and development projects in Ecuador. In this region the cassava is perceived as part of the life: it has always been and it will be present in the life of the Kichwas especially of the women. This lack of cassava studies is due to the wrong idea that cassava is an abundant and unperishable crop. However a reduction of the cultivated varieties is observed. This reduction is due to the devaluation of the culture (the boys and girls are educated more and more to live a "modern" style in detriment of traditional life style). Considering these circumstances, FONAKIN has proposed this project aiming the importance of cassava crop for the family subsistence and for the culture, and the narrow relationship among the chakra (cassava agroforesty system) and the woman. This project correspond to the first phase of the project "Documentación del conocimiento Kichwa sobre la Chakra y Fortalecimiento al Grupo de Mujeres de Base de la FONAKIN", designed with participation of the FONAKIN's Women Group and has the following main objectives: • To document the knowledge of Kichwa's Nationality from 'Alto Napo about production systems in chakra. • To promote women participation in planing, execution and evaluation of FONAKIN's projects. 420 Outputs • Diagnostic socioeconornic of the chakra, including the systernization of cassava management. • Characterization of the plant diversity within representative chakras. • In situ conservation of cassava varieties of high basin of the river Napo. • Genetic and morphological characterization of cassava varieties. • Invigoration of the administration and negotiation capacity of FONAKIN regarding to property rights on genetic resources. • Women kichwas with capacity of fonnulating projects airning gender analysis, including a training for 25 women on severa} tapies, such as molecular biology and support to women's events. 3.2.15 Cleaning of released varieties and recovery of cassava genetic resources in production area of Ecuador Institution: Instituto Nacional Autónomo de Investigaciones Agropecuarias (INIAP) Address: Estación Experimental Portoviejo, Km 12 via Portoviejo-Santa Ana, Portoviejo, ECUADOR Email: iniap@iniap-ecuador.gov.ec Collaborating institution Centro Internacional de Agricultura Tropical (CIAT), Colombia Staff directly involved Francisco Hinostroza García, INIAP Gloria Cobeña Ruiz, INIAP Alma Mendoza de Arroya ve, INIAP Flor María Cárdenas Guillén, INIAP César Tapia Bastidas, INIAP Elizabeth Caicedo, CBN-CIAT Roosevelt Escobar, CIA T Introduction The INIAP, in Ecuador, has released two cassava varieties since 1992. According to a survey, perfonned in the main production areas of Manabí, those varieties have decreased their good characteristics. The importance of cassava for human food, animal feeding and agroindustry malees necessary to clean the released and local varieties to recovery their best performance, which will allow to have these valuable genotypes as source of diversity to obtain cassava products and by-products required by national and intemational markets. Objectives • To clean the released cassava varieties 'INIAP-Portoviejo 650' and 'INIAP-Portoviejo 651' . • To adjust a methodology for conservation and management of the cleaned materials under farm facilities belonging to five 'Asociaciones de Productores y Procesadores de Yuca' (APPY) by training processes, participatory research and gender analysis. • To recover genetic materials of local cassava varieties through collection and ex situ conservati on .. The direct beneficiaries will be the partners of the APPYs from: Tablones (Junín), Bijahual y San Vicente (Portoviejo), Jaboncillo y San Miguel (24 de Mayo), of Manabí province. The indirect beneficiaries will be the non partners farmers and their families . 3.2.16 Other CBN Activities Recruitment of CBN coordination staff A Regional Coordinator, Dr. Alfredo A. C. Alves has been appointed and have resumed duties at the Cali Headquarters of CIAT in February 2003. On his resumption of duty, the recruitment of the Social Scientist have been concluded with the recruitment of Elizabeth Caicedo in May 2003. The Ginés-Mera memorial fellowship fund for postgraduate studies in biodiversity The Canadian Intemational Development Research Center (IDRC) has generously provided funding in the total sum of US$750,000 spread over 5 years for "The Ginés-Mera memorial fellowship fund for postgraduate studies in biodiversity". This fund is aimed memorializing Dr. Maria Jesús (Chusa) Ginés and Ms. Verónica Mera, CBN Coordinator and Social Scientist, respectively, who lost their lives in a tragic airplane accident (Jan 2002) while on an official trip from their base in Quito, Ecuador, to the headquarters of the Intemational Center for Tropical Agriculture (CIAT), Cali, Colombia. Specifically, the fund aims at achieving the following over the 5-year span: To provide opportunities and support to female and male master's students from the developing countries of the world to undertake thesis research addressing key elements of the sustainable use and conservation of agricultura} biodiversíty, in particular: a) intellectual property rights and access to agricultura! genetic resources b) molecular characterization of agrobiodiversity e) community-based conservation of genetic agrobiodiversity. 422 To promote the bridging of the research/ development divide, by encouraging researchers and their home universities to develop linkages with research for development projects, and to undertake applied research which informs development processes. To explore opportunities for further expansion of this initiative in order to involve other stakeholders. To encourage and support the exchange of information, knowledge and technology between the stakeholders in agricultura! biodiversity conservation in these countries. A total of 22 proposals were evaluated and 7 fellowships were approved and started in October/2003 (Table 1) References Thro, AM and C. Spillane (2003). Biotechnology-assisted Participatory Plant Breeding: Complement or Contradiction? Working Document No. 4. CIAT, CaJi, Colombia. Fauquet, CM and Taylor, NJ, eds (2003). Cassava: An Ancient Crop for Modem Times. [4 Compact Discs) Proceedings of the 5th Intemational Meeting of the Cassava Biotechnology Network. 200 l November 4- 9, St. Louis MO, USA. Web based information dissemination A website, http://www.ciat.cgiar.org/biotechnology/cbn/index.htm, dedicated to highlighting the activities of CBN-LAC has been launched. The site is also aimed at dissemination information to network members and in due course would also serve as an interactive forum for the exchange of ideas. Capacity building of NARS partners In order to ensure more efficient performance by NARS partners involved in projects at pilot sites the following have received CBN support to perform training at CIAT: Yoel Beovides García, Researcher (INIVIT), Cuba Terezinha Feitosa Machado, Researcher (Embrapa/CNP AT) and Doctorate Student (Federal University of Ceará), Brazil Maryluz Folgueras, Researcher (INIVIT), Cuba Table 1 • Projeds supported by the "Ginés-Mera Memorial Fellowship Fund for Postgraduate Studies in Biodiversity". 2003 Student Project Title University Award (US$) Adriana Mercedes Alzate Identificación por medio de marcadores moleculares de la diversidad genética de yuca de pequeños Universidad Nacional de 12000 agricullures y evaluación del nivel de adopción e impacto de varieddaes mejoradas por CIAT en la Costa Colombia Atlántica Colombiana Aslrid Johanna A rango Uso de la diversidad del totumo (Crescentia cujete): un árbol multipropósito para pequeños agricultures Georg-August-Universitat- 8000 Ulloa en regiones secas Goettingen, Germany Constanza Maria Aplicación de una metodología para caracterizar germoplasma de frijol común (Phaseolus vulgaris L) Universidad Nacional de 12000 Quintero Valencia mediante polimorfismo de un solo nucleótido (SPN) Colombia Javier Uacsa Tacuri Factores de distribución y fluctuación de la variabilidad de papas natives en las familias de las Universidad Nacional San 12000 comunidades de Amaru, Chahuyatire y Viacha del Distrito de Pisac-Calca, Peru Antonio Abad del Cu~Peru Juliana Chacón Pinilla Patrones filogenéticos inter-e-intra específicos de la yuca (Manihot esculenta spp. Esculenta, Universidad de los Andes, 14000 Euphgorbiaceae): biogeografia y ecología compararada de las especies de la amazonía y la region andina Colombia Nelson Arturo Ro yero Caracterización molecular de la variabilidad genética de guanábanos (annona muricata L.) y especies de Universidad Nacional de 8000 Moya anonáceas ralacionadas de importancia hortfcola Colombia Roosevell Escobar Pérez Establecimiento de técnicas biotecnológicas a bajo costo para la limpieza y conservación de variedades Universidad Nacional de 8000 locales de yuca y uso sostenido en programas de seguridad alimentaria Colombia 1 - - -- ~-- -- 424 Familiarization with Projects at Pilot Si tes and Monitoring of on-going Activities The CBN coordination staff has traveled to visit on-going projects at Cauca, Colombia; Bahía, Brazil; and Ecuador. Plans are underway for initiating visits to the on-going projects under small grants scheme. Participation in Scientific Meetings The network's activities have been highlighted at the following meetings: Needs Assessment Study on Cassava, Ghana, 12-14 May 2003. Worshop on Needs Assessment Study on Cassava, Cali, CIAT, 19-23 May 2003. Planning Workshop on the Development of a Protocol for the Generation of Cassava Doubled-Haploids and their Use in Breeding, Cali, CIAT, 11-12 June 2003. Organization of the Sixth International CBN meeting (CBN-VI) The CBN is organizing its Sixth International Scientific Meeting (CBN-VI), which will take place at CIAT, Cali, Colombia on 8-14 March 2004. The theme of the meeting is: Adding Value to a Small-Farmer Crop. The aim is to discuss the better uses of biotechnology tools to add value to this small-farmer crop. Presentations at the meeting will focus how biotechnology can assist cassava farmers by developing, for example, more suitable varieties, disease-free planting materials, and better ways to conserve and process cassava after harvesting. The announcements and other informations about this meeting can be found at CBN website: http:/ /www .ciat.cgiar.org/bi otechnology/cbn/index .htm Activity 3.3 Database and Libraries 3.3.1 Data bases about distribution of wild relatives of crops D.G. Debouck SB-1 Project We have continued with the establishment of databases about the distribution of wild relatives of beans (Phaseolus), cassava (Manihot) and rice (Oryza) in the Neotropics. This year we have done an inventory at the following herbaria: COL, EBUM, INB, NY, QCA, and US. The data basically include: taxonomic identification, location, date, phenology and notes. For easy consultation, types are in red, collectors in blue, and state/ province in green (examples below). To date, the following herbaria have been visited: ARIZ, BAA, BM, BR (part), BRIT, COL, CR, CUZ, DES, ENCB, F (part), G, HAO, HUf, K, Lll., LPB, MA, MEXU, MICH, MO, MSC, NY, QCA, SGO, SI, US, USCG, USJ, and USM. This work has been developed bearing in mind the following perspectives: Correct identification of materials collected and kept in ex situ conservation facilities (namely CIAT genebank, and from there collaborating institutions). Geographic distribution of wild relatives of direct interest in breeding activities (namely acquisition of germplasm useful to the breeders). Distribution of wild relatives genetically compatible with the crop, in view of introduction and management of transgenical crops. Monitoring of modificationl destruction of natural habitats and disappearance of populations. Along perspective # 1, a major outcome of that work has been a revision of Phaseolus (Freytag & Debouck 2002). Example 1 Flora de Costa Rica, Fabaceae/ Pap. Vigna umbellata (Thumb.) Ohwi & Ohashi, det. N Zamora, enero 1996. San José, cantón de Aserrí. Z.P. Cerros de Escasú. Cerros de Escasú -La Carpintera. Cedral. Bosque primario y secundario en la falda norte del Cerro Pico Alto. Cuenca del Río Poás. 09°50'57"N 84°08'25"W. 1600-2300 m. Planta rastrera, flores fucsia. JF Morales 194. 13 December 1991. Instituto Nacional de Biodiversidad, en colaboración con el Missouri Botanical Garden (MO). 111 Instituto Nacional de Biodiversidad. Fabaceae/ Pap. JF Morales 194. Phaseolus coccineus L. ssp. darwinianus Hem-Xol & Miranda, identifica N Zamora, octubre 1996. /// DGD: costaricensis, 1 raceme, début floraison, exemplaire pauvre non typique, peu vigoureux. [INB; 21-VIII- 2003]. 426 This example shows a population of a new species described for Costa Rica (Freytag & Debouck 1996), that has been shown to belong to the phylum of the comrnon bean (Schmit et al. 1993), thus widening possibilities for wide crossing. Example 2 EBUM7394. Fam, Euphorbiaceae. Manihot intermedia Weatherby. n.v. "teyapu". Loe. Rancho Galeana. Edo. Michoacán. Mpio. Apatzingán. Hab. Terreno plano, rocoso basáltico, suelo arcilloso. 310 msnm. Selva baja caducifolia. Col. X Madrigal Sánchez no. 3167. Obs. Latex muy irritante. Det. X Madrigal Sánchez. Fecha Die 6/ 1978. /// DGD: tomatophylla, végétatif, fe palmatilobées, a 5 lobes arrondis, pétioles 12-16 cm long. [EBUM; 11-II-2003]. This example shows an additional record to the few known for this wild species of cassava, the distribution of which seems restricted to Michoacan, Mexico. It does not exist in genebanks. Example 3 Plants of Louisiana. Herbarium of Southern Methodist University. Oryza sativa L.. Allen Parish, 1.7 miles northeast of Oberlin. Shallow water, roadside ditch, and silty clay; bordering rice field. Severa! plants. Lloyd H Shinners 22092. 8 October 1955. /// Herbarium of Northern Kentucky University (KNK), det. John W Thieret 1984. /// DGD: sativa, la panicule a perdu pratiquement tous ses épillets, certains épillets ont les glumes brunatres, type 'red rice' ! [BRIT; 10-VI-2002]. This example opens the possibility of the presence of 'red rice' on borders of rice fields in the southern USA, as early as 1955. Example 4 Herb. Le Jolis [s.n.], Phaseolus, Mexique Occidental, Acapulco, [Guerrero] , presqu'lle Griffon, Legurnin. Oct. 1866. /// Herbier Barbey-Boissier. /// Durand 1913 (acquisition ou cession?!). /// DGD: mcvaughii, go vertes, 3 racemes, dimorphisme dans les gousses! [G; 15-ll-2002]. This case shows how modification of habitats affects wild population, because this site has now been converted into a tourist resort. lt has been shown elsewhere (Bayuelo et al. 2002) that this species is promising for salinity tolerance. References Bayuelo, JS, DG Debouck & JP Lynch. 2002. Salinity tolerance in Phaseolus species during early vegetative growth. Crop Sci. 42: 2184-2192. Freytag GF & DG Debouck. 1996. Phaseolus costaricensis, a new wild bean species (Phaseolinae, Leguminosae) from Costa Rica and Panama, Central America. Novon 6: 157-163. Freytag GF & DG Debouck. 2002. Taxonomy, distribution, and ecology of the genus Phaseolus (Leguminosae-Papilionoideae) in North America, Mexico and Central America. SIDA Bot. Mise. 23: 1-300. Schmit, V, P du Jardín, JP Baudoin & DG Debouck. 1993. Use of chloroplast DNA polymorphisms for the phylogenetic study of seven Phaseolus taxa including P. vulgaris and P. coccineus. Theor. Appl. Genet. 87: 506-516. 3.3.2 Updating the data base and maintaining the ceparium of the Biotechnology Research Unit J. L. Bocanegra 1, A. F. Salcedo2, F. Rojas2, P. Chavarriaga2 andA. Mejia2. Universidad del Tolima, !bagué, Tolima, Colombia SB2 Project, CIAT, Cali, Colombia Introduction The ceparium was established in 1996 to maintain a collection of Agrobacterium andE. coli strains. It consisted of 97 accessions of A. tumefaciens, 17 of A. rizhogenes and 156 of E. coli. The strains are kept frozen at -80 °C, in four boxes, whit 3 copies per box, two of which are the stocks and one is the working collection. The last renovation of the entire ceparium was done in 1998. Since then new accessions have been added, in sorne cases without a back up copy, or lacking crucial information. Due to the overwhelming increase in information on strains, vectors and genes, it was necessary to create a database in year 2000, using Oracle developer 5, Edition Release 8.1.7.4.0 (Figure 1). The data base was not completely filled due mainly to the lack of a full time person devoted to this activity. We hired a visiting scientist anda "pasante" from the Universidad del Tolima to update the ceparium, renew part of the collection andintroduce new strains. Objectives Update the ceparium and database Renew and purify existing strains in critica! shape Enter new accessions into the ceparium and database. Review and collect references relevant to the ceparium Digitalization of information (pictures, maps, MT A, etc) available for strains, vectors and genes. 428 Materials and Methods All boxes, stocks and working collections, were checked and all damaged accessions of A. tumefaciens, A. rizhogenes, andE. coli, were renewed by growing them on LB media with the proper antibiotics. Bacteria! cultures were mixed with conservation media (Glycerol 30%, MgS04 7-H20 100 mM, TrisHCl pH 7.5 10 mM) and stored in cryogenic vials. All available drawings, maps, pictures, restriction pattems of vectors and MT As (Material Transfer Agreements) were scanned and included in the database. Using an Oracle tool called TOAD (Figure 1), the available information on strain, vector and genes was also included in the ceparium database. To eliminate contaminants in Agrobacterium strains, a genus-specific Ketolactose test (Bemaerts & DeLey , 1963) was used to purify and renew strains. - - -:::::- - -- -· - ··- - ·- _...,,_ _......,., .... _.. ... _....,,_ - ·- - ·- - ·-u- _ , _ - ·-:::!.:!.-... , = ::: = -- -- u u ........ u-· u-· • w.,. , .. _.., ........ ......... ... ..,, u.,. ... _... , .. _... ......... --· ......... ........ _, ..... , ·-··!:':a' ......... ......... ....... , u-· .. _. u_. Figure l. TOAD. a too! from Oracle that facilitates entering the data in the ceparium database. On the right is the actual application of the ceparium, as it shows on the screen, used also to enter and search for data. Results The ceparium was partially renovated by replacing damaged accessions, or reincorporating lacking ones in all boxes. Currently the collection is maintained by triplicate (Stock 1, Stock 2 and Working Collection) at -80°C. Each strain is in triplicate within each box. There 67 strains of A. tumefaciens, 7 of A. rhizogenes and 36 strains of E. coli. A ketolactose test for all renovated Agrobacterium accessions confirmed that , in fact, they belong to this gen u. All the information available about strains, vectors and genes was digitized and entered into the ceparium database application. Ten New articles related to strains, vectors, were filed. The plasmids pBSK 7, pUC9 psl, pTo 76, pUC9#44, pUC9 GA, pT3T7, pBI22i were introduced in E. coli DH5 O by electroporation, creating se ven new accessions. Filling a data sheet established for such purpose will monitor each new strain that enters the ceparium (Figure 2). Future Activities Entering to the ceparium collection and database new accessions; these included the plasmid vectors of JIRCAS (Dreb) and pRip. Incorporated into ceparium collection and database the accessions belong to rice transformation collection. Training transformation personal in ceparium database application manage. References Bernaerts, M. J. and DeLey, J. (1963) A biochemical test for crown gall bacteria Nature (London) 197,406- 407 430 ~lodfversity-lfld. Blotec1molo&Y' ~ Project·-CIA~ • Data· sheet fOl' lntrodutiDe-straiDr fp]asmids· and-IJl~ 1n1o-SB2·collectim1 1 ' 1 1! ~~~----------~~ f , __ f 1 1 bdtnutio,... f 1 ,,1 .. . 1 ~-· 1 1 .•. ""!"'" •• _ ......... 1 • ll r r r ~ ~ r ~ - Ycd • h1~,-~~.-. . -----------------------· -l&b~=,a,1~----------------·L ! • . . • • . ~ • ~ ¡ -···-·-·-------·-· w ·--- '=~- -- .... -~ UTA. .. ...,..· Ytf4 - / ·No-1 · ··· /R.Rrlalcapcom- ···' · - · ·/-s..,-oo ~=- ..... , .y., .. - - -- / ·Nol • • • . ., "1" 1. J" j " J" ~~~~,~~~~ ~--------------- l. . l [~ -- - 1 1 1 ~ _.... Yud ¡. 1 · o 1 1. D&rof1~ ~ • 1 ~ ... , . .._. 1" 1" l " Figure 2. Data sheet to fill in when entering new strains in the ceparium. Activity 3.4 Training and Workshops 3.4.1 National and international Collaboration Joe Tohme and Howdy Bouis visited the Gates Foundation January 2003 to discuss the biofortification proposal. Joe Tohme visit to the Danforth Center, Kansas State University and Yale University, January 2003 to discuss collaborative project on Cassava and rice Joe Tohme. January, 2003 Corpoica, Ministerio de Agricultura. Bogotá -Colombia Daniel Debouck, Costa Rica, January 2003. Observations about gene flow in wild-weed- crop complexes in the Central Valley of Costa Rica. Field work for the BMZ supported project on gene flow analysis. University of Costa Rica. Steve Beebe. January, 2003. Guatemala. SDC follow up workshop on the future of the Central American networks Daniel Debouck, Mexico, February 2003. Annual meeting of the SGRP; upgrading plan for the genebanks of the CGIAR. University of Morelia: Conference: "Evolución temprana del frijol en el Occidente de México". Invited seminar at the University of Guadalajara: "Entre Maguey y Cempoalxochitl. Observaciones sobre los recursos fitogenéticos del Occidente de México". Mathew Blair. Alexandria, Egypt, February 2003, to attend Genetic Resource Challenge Program stakeholders meeting Steve Beebe. February, 2003. Uganda. Training workshop on Marker Assisted Selection on project funded by Rockefeller Foundation. Steve Beebe. February, 2003. Rwanda. Field visits. Steve Beebe. February, 2003. Kenya. Planning workshop for a bioefficacy trial with the University of Nairobi. Mathew Blair. Kisumu and Nairobi, Kenya, February, 2003 to attend Bean Biofortification project meeting. 432 Mathias Lorieux. March 2003. m Genoplante Congress. Presentation of one PostersPoitiers, France. Mathew Blair. Medellin, Colombia, March, 2003.To visit field experiments and plan for collaborative activities with CORPOICA. Mathew Blair. Santa Cruz, Bolivia, March - April, 2003, to visit field experiments and plan for collaborative activities with PRONOLAG Martín Fregene. March, 2003 Trip to CRI, Kumasi, Ghana, on the collaborative project, " Analysis of genetic diversity of local land races from Africa and Latín America and the search for heterosis", being carried out as a Ph.D. study by Ms Elizabeth Okai. Steve Beebe. March, 2003. Haiti. Field visit and training of local CIA T staff. Joe Tohme. March, 2003. Meeting CIO on Biofortification. Biosafety Meeting. Washington. Organizer and participation. Steve Beebe. April, 2003. Honduras. To attend annual meeting of Honduran CIALs (Local Agricultura} Research Committees) Steve Beebe. April, 2003. Honduras. To attend annual meeting of PCCMCA. Martín Fregene. April 2003 Trip to NARO Namulonge, Uganda, on the collaborative project " Genetic mapping of cyanogenic potential (CNP) in cassava, being carried out as a Ph.D. study by Ms Elizabeth Kizito, under the BIOEARN program (SLU with Sida funding) Manabu Ishitani. April, 2003. Objective: to build research networks with research institutions in Japan and to discuss specific research projects to seek for future collaborations Institution: April. RIKEN Plant Science Center ; Mitsui Global Strategic Studies Institutes; JIRCAS ; RIKEN; Okayama University; TOYOTA R & D Center; Nagoya University ; National Institute for Basic Biology Mathew Blair. Cusco, Peru, April, 2003 to visit field experiments and plan for collaborative activities with INIA Daniel Debouck, Bogotá, April, 2003 (with Dr J Tohme of CIA T). Meeting with Genetic Resources and Biotechnology Program of CORPOICA: planning and future activities. Mathias Lorieux. May 2003. Planning Meeting. CIO- CIAT. Montpellier, France. Martín Fregene. May 2003 Trip to ARI, Mikocheni, Tanzania, on discussions of the project "A Molecular marker-assisted, farmer-partipatory breeding project to improve local cassava varieties in Tanzania with resistance to pest and diseases" Martín Fregene. May 2003 Trip to ll..RI, Nairobi, to visit the bioscience facility particularly the molecular marker capacity and ongoing cassava work (by Dr Morag Ferguson, TITA). Martín Fregene. May 2003 Trip to RF-Nairobi, in company of Dr Ferguson to present to Dr Joe DeVries, RF Assistant Director of Food Security, outcome of the meeting in Tanzania. Manabu lshitani. May, 2003. To meet staff and exchange information. Japanese Embassy. flCA Colombia Office Mathew Blair. Chicago, USA, May, 2003 attend technical meeting of the Bean/Cowpea CRSP. Daniel Debouck, New York, USA, June 2003, invited serninar at the New York Botanical Garden: "Bees and beans: diversity of sweets, tongues, mutualisms, and consequences"; review of NY herbarium. Martín Fregene. June 2003. Visit to Nigeria, in company of Bernardo Ospina and Luis Fernando Cadavid to assess the farrning and seed multiplication operations of the Nigerian Starch Mills (NSM). Joe Tohme. June, 2003. Visiting Instituto von Humboldt. Ministerio de Agricultura Joe Tohme. June, 2003. Coffee Genornics. Meeting Coordinator. Cenicafé. Manizales- Colombia. Joe Tohme. July, 2003. GEF Project Meeting. Bogotá- Colombia Daniel Debouck, Lima Peru. July 2003, invited presentation (with Dr W Roca of CIP) at the Regional Workshop on Access to Genetic resources in the Andean Region: "Mecanismos de intercambio de germoplasma de los Centros del COlAR"; review of QCA herbarium. Daniel Debouck, Costa Rica, August 2003, meeting with INBio Staff in view of future collaborative activities; review of INB herbarium. Daniel Debouck, Washington, USA, August 2003, invited serninar at the Srnithsonian Institution, National Museum of Natural History: "Phaseolus beans in the Pliocenic transit 434 lounge of Panamá: last findings of markers and cpDNA analyses"; review of US herbarium; meetings with Smithsonian Staff in view of future collaborative activities. Manabu Ishitani. August, 2003. To attend technical planning workshop for Genetic Resources Challenge Program. Wageningen Science Center Joe Tohme. August, 2003. PBA Meeting. Bogotá- Colombia Martín Fregene. August 2003. Visit to the John Innes Center (JIC), Norwich, to attend a one da y conference in honour of Prof Mike Gale on his retirement from JIC; and to discuss a collaborative project with Prof Allison Smith. Mathias Lorieux. August, 2003. Planning Meeting of the Challenge Program. "Unlocking Genetic Diversity of Crops for the Resource Poor". Wageningen, The Netherlands, Daniel Debouck, Bogota, Colombia, September 2003, invited presentation (with Dr E Torres of lvH) at the Regional Workshop on Access to Genetic resources and protection of indigenous knowledge: "Propuesta de mecanismo de seguimiento y control, luego de un contrato de acceso a recursos genéticos". Mathew Blair. Davis, USA, September, 2003 to organize Phaseomics meeting. Mathew Blair. Nairobi, Kenya, September - October, 2003 to attend Harvest Plus Challenge Program - bean project meeting. Martín Fregene. September 2003. Visit to Swedish Agricultura! University (SLU), Uppsala, to attend the 2nd MOLCAS meeting and discuss with Dr Eva Ohlsson of Sida a proposed visit in October by Hernan and Joachim. Martín Fregene. September 2003. Visit to Mukono, Uganda on the work planning meeting of the cassava biofortification challenge program (BCP). Martín Fregene. September 2003. Return visit to CRI, Kumasi, to review progress made with establishing the crossing block in the project " Analysis of genetic diversity of local land races from Africa and Latin America and the search for heterosis" . Cesar Martínez. September, 2003. Attend Plant and Animal Genome XII Conference. Miami -USA. César Martínez. October, 2003. IRRI - Philippines. Attend the Rice Crop Meeting of the Biofortification Challenge Program. Mathias Lorieux. October, 2003. m International Conference on Rice Breeding. Universidad Central de Venezuela. Maracay, Venezuela. Zaida Lentini. October 2003. Intemational Biosafety Workshop. Agrobio. Canada. Steve Beebe. October, 2003. Biofortification Challenge Program planning meeting for common bean, in Nairobi, Kenya Manabu Ishitani. September, 2003. To discuss possibility of research collaboration and to ha ve scientific discussion with Drs. Leon Kochian and Mark E. Sorrells Comell University. Joe Tohme. November, 2002. Workshop on Biosafety. Ministerio del Medio Ambiente. Bogotá -Colombia. Steve Beebe. November, 2002. Guatemala. SDC workshop on the future of the Central American networks. Joe Tohme. September, 2003.- Biofortification Sweet Potato and Cassava crop meeting, Uganda. Visiting II.RI Nairobi Joe Tohme. October, 2003. Meeting Biofortification. Gate Announcement. Visiting World Bank. Washington- USA. Joe Tohme, October 2003. Meeting PBA. Santa Marta- Colombia Joe Tohme. November 2003. Agronomic Meeting. Attend PAC. Washington- USA. Joe Tohme. November 2002. Meeting on Biofortification. Rio de Janeiro, Brazil. Myriam C. Duque, Apoyo a Instituciones de Agronatura. IPGRI. Análisis de colecciones de Capsicum y de Phaseolus vulgaris en Cuba y Guatemala: Dra. Carmen de Vicente, Sr. Félix A. Guzmán. Myriam C. Duque. Apoyo a Instituciones de Agronatura. Biotec: Análisis de ensayos en guanábana: Sres. Nelson Royero y Juan Jairo Ruíz. Myriam C. Duque. Análisis de redundancia /CANOCO. Ofrecido por el Dr. Chris Blackwood en curso coordinado por el Dr. Edmundo Barrios. 436 3.4.2 Training, Workshops, International and National Conferences for CIA T Personnel SB-2 Staff. September 2002 . XXXVIII ACCB Congress. Pasto -Colombia SB-2 Staff. October,2002. Second International Meeting on Agricultura! Biotechnology. Cartagena- Colombia. Palmira, Colombia, presentation during the Open House at the Genetic Resources Unit, 7 December 2002: "A race, a revolution, a treaty". Fort Collins, USA, 31 January 2003, invited seminar at the Genetic Resources Conservation Center of USDA: "Searching for new bean germplasm in the Americas". Matthew Blair - Kampala, Uganda, February 5-8, 2003. to Jecture at Rockefeller Foundation sponsored marker workshop at CIA T -Africa SB-2 Staff .. March 7, 2003. Meeting CEGA- PBM. Bogotá- Colombia SB-2 Staff. March 31- Abril4, 2003. Taller Gestion y conservación de recursos genéticos . Cartagena- Colombia SB-2 Staff. March 24, 2003. Cita con representante agregado agropecuario de USDA en la embajada de Jos USA en Bogotá. Bogotá Colombia Matthew Blair- Bogotá, Colombia, April 30-May 1, to attend the meeting of the National Congress of Food Scientists - ACTA "Asociacion Colombiana de Ciencia y Tecnología de Alimentos", and presenta keynote address on biofortification. ACTA. Asociación Colombiana de Ciencia y Tecnología de Alimentos. April 30-May 1, 2003. Bogotá- Colombia Biosafety Workshop for Journalists: J. Tohme, P. Chavarriaga, Z.Lentini,. Coordination of Workshop on Agriculture Biosafety for Joumalists. April, 2003. CIAT, Cali, Colombia. A total of 15 participants. Matthew Blair- Dakar, Senegal, May 15-21, to attend the Forum on Agricultura! Research in Africa and present a talk on biofortification for a USAID sponsored work~hop. SB-2 Staff. May 18-22, 2003. Asistencia y presentación de los trabajos en el Molecular Breecling of Forage and Turf, Tirad Symposium. Dalas. USA. SB-2 Staff. June 18, 2003. Semi Quantitative RT -PCR. Training. Perpignan - France SB-2 Staff . June 9, 2003. Course CS!ll... and training on Microarrays. lllinois University. USA. SB-2 Staff. June 26, 2003. Course on Bioinformatics and Genomics. Madrid- España. SB-2 Staff. June 10, 2003. Visiting Bath University. Myriam C. Duque. Conferencia Magistral en el XXIV Congreso Nacional de Fitopatología y Ciencias Afines. Armenia, Colombia. June 27, 2003 Mauricio Soto Suárez was a speaker in the "XXIV Congreso Nacional de Fitopatología y Ciencias Afines". June. 25-27. 2003. Carmenza Muñoz - RNA extraction for cornmon bean nodular cortex and physiological studies on nitrogen fixation potential of tepary beans under phosphorous deficiency. INRA-ENSA, Montpellier, France- June- July, 2003. Congreso Sociedad Colombiana de Fitomejoramiento y Producción de Cultivos. July 2-5, 2003. Bogotá- Colombia Paul Chavarriaga, XXX Congreso de SOCOLEN en calidad de conferencista. July 17-19, 2003. SB-2 Staff. July 1, 2003. Monitoring and guidance for the Project "Limpieza y Multiplicación in vitro de variedades de yuca tradicionales y mejoradas con la participación de agricultores. Salvador, Brazil. Matthew Blair - Bogotá, Colombia, July 3-5, to attend Plant Breeding Association "Congreso Nacional de Fitomejoramiento" meetings and give a keynote address on plant biotechnology in the one-day genomics symposium. Soto-Suárez, M, Restrepo, S, Mosquera, G., Lopez, C., Tohme, J., and Verclier, V. 2003.Gene expression profiling of cassava responses to Xanthomonas axonopodis pv. manihotis infection. 11-th Intemational Congress on Molecular Plant-Microbe Interactions - July 18-27 2003, St.-Petersburg, Russia. 438 SB-2. Staff. August 1, 2003. Agroexpo. Bogota- Colombia Matthew Blair- Wageningen, Netherlands, August 24-30, to attend the Genetic Resource Challenge Program technical meeting. Matthew Blair - Clemson, USA, Sept 3-7, to organize genornic resources and conduct hybridization experiments SB-2 Staff. Oracle Linux Da y. AS V OC. Bogotá - Colombia SB-2 Staff. Oct 27, 2003. Universidad del Tolima. SB-2 Staff. Oct, 23, 2003 Ministerio de Agricultura!. Conference. Bogotá- Colombia J. Salcedo. Universidad del Valle, Colombia. BSc thesis. October 2002-September 2003. R. González. Universidad Nacional de Colombia. MSc. thesis. October 2002-September 2003. Silvia Restrepo . October, 2002. Workshop on the application of Marker Assited Selection to bean breeding, at CIA T headquarters. Attended by 10 national program breeders. Myriam C. Duque support "Fundamento y Explicación de los procesos estadísticos de análisis". Master' s Degree Programme m Plant Genetic Resources, Universidad Nacional de Colombia, October 2002. Biosafety Workshop for the Ministry of Environment: Z.Lentini, P. Chavarriaga, J. Tohme. Coordination of Workshop on Agriculture Biosafety for the Colombian Ministry of Environment. November, 19-20, 2002. CIAT, Cali, Colombia. A total of 25 participants. Carlos Cesar Caula- training in rnicrosatellite mapping, marker assisted selection and gene tagging, at CIAT. Gloria Santana - CORPOICA - Rionegro, Antioquia, Colombia (Aug - Sept 2003) - training in molecular marker techniques and indirect selection for BCMV resistance, at CIAT. Emeterio Payro - Departamento de Biologia, Centro de Investigación Fitogenética de Yucatán, Mexico (August 2003). training in molecular marker techniques for diversity assessment of wild beans, at CIAT. Carlos Andrés Hidalgo, Adrian Alejandro Carrera, César Augusto Vera - Escuela Politécnica del Ejercito del Ecuador, Quito, Ecuador (September 1 - 26, 2003). training in molecular marker techniques and indirect selection for BCMV resistance at CIAT. Ms Elizabeth Okai (Ghana) Ph.D. student, University of the Free State, Bloemfontein, South Africa (6 months, left January 2003) Ms Elizabth Kizito (Uganda) Ph.D. student, University of the Free State, Bloemfontein, South Africa (June-December 2003) Henry Ojulong (Uganda) Ph.D. student University of the Free State, Bloemfontein, South Africa (6 months, left January 2003) Martha Isabel Moreno (Colombia) M.Sc. student, Universidad de Valle, Cali (September 2002- August 2003) Wilson Castel Blando (Colombia) B.Sc. student, Universidad Nacional de Colombia, Sede Bogota (Jan-Dec 2003) Liliana Cano (Colombia) B.Sc. student, Universidad de Santander, Bucaramanga (Jan-Dec 2003) Ana-Maria Correa Colombia) B.Sc. student, Universidad de Valle (June 2003 - May 2004). Catalina Oviedo. Universidad Eafit Medellín. Process Engineering. Six Months: August 2003 - J anuary 2004. Developing of a process for sterilizing Phaseolus sedes with Ck vapors Joel Beovides, INIVIT, Cuba. CBN small grant recipient (June- December 2003) Dr Chiedozie Egesi, NRCRI, Nigeria. One week on breeding methods at CIA T after the doubled haploid workshop (June 2003). 440 Catalina Oviedo. Universidad Eafit Medellín. Process Engineering. Six months: Augst 2003- January 2004 . Developing of a process for sterilizing Phaseolus seeds with Cl2 vapors Juan Jairo Ruiz M.Sc. Thesis; 3 years. National University of Palmira, Colombia; - Corporación Biotec. Field evaluation of in vitro propagated trees of soursop Silvio Cadena; M.Sc. Thesis; 3 years. National University of Palmira, Colombia - Corporación Biotec. Greenhouse management of in vitro propagated plants of soursop Nelson Royero M.Sc. student. National University of Palmira, Colombia - Corporación Biotec Molecular characterization of accessions of soursop and related annonaceus species. Jhon Alex Cambindo. Instituto Técnico Agropecuario (ITA). Buga - Valle del Cauca. Specialization as technician in agronomy. Six months. Sandra Lorena Acosta Lopez. System Plus. Palmira. Valle del Cauca. Database management and large data set organization. Two months. Intemational Biosafety Workshop: Organized by Agrobio, Canada. October 2003. Cartagena, Colombia. Z.Lentini Lecturer. A total of 50 participants Adriano Alejandro Carrera. B.Sc. Military Polytechnic School. Quito, Ecuador. Training on tissue culture, anther culture, and genetic transformation. Carlos Hidalgo B.Sc. Military Polytechnic School. Quito, Ecuador. Training on issue culture, anther culture, and genetic transformation. Cesar A. Vera. B.Sc. Military Polytechnic School. Quito, Ecuador. Training on tissue culture, anther culture, and genetic transformation. Ángela Mina. B.Sc. El Valle University. Training on microsatellites analysis of rice and their use for gene flow analysis into wild/ weedy relatives Paola Olaya. B.Sc. El Valle University. Training on microsatellites analysis of rice and their use for gene flow analysis into wild/ weedy relatives Margarita Pineda. B.Sc. National University of Colombia, Palmira. Training on tissue culture. Ines de Alba Aragon Nieto, BIOMOL. Training course of Real Time PCR technology from September 8 to September 12. February, 2003. Uganda. Training workshop on Marker Assisted Selection. Attended by about 15 national program breeders. Myriam C. Duque, Curso Métodos estadísticos aplicados en Biología Molecular. Fundación DANAC- Venezuela. January 13-17, 2003 Myriam C.Duque, Curso "Métodos estadísticos aplicados en ensayos de campo. Fedearroz, V alledupar- Colombia. February 5-7, 2003 Myriam C. Duque, Curso "Análisis de variables categóricas por tablas de frecuencia"Fondo para el desarrollo del Recurso Humano. CIAT. May 15- 16, 2003 Myriam C. Duque. Capacitación en Biometría para el Ing. Edwin lquize, funcionario del Instituto Boliviano de Tecnología Agropecuaria-Proyecto. ffiTA-Chapare. Actvity 3.5 Publications 3.5.1 Refereed J ournals, Books Aluko,G.K., CP Martinez, Carolina Castaño, J. Tohme, J.H.Oard.2003. QTL, non- parametric, and population structure analysis of agronomic traits in doubled haploid lines derived from the intespecific cross O.sativa (L)/O.glaberrima(Steud). Poster. PAG meeting. San Diego ,California. January,2003. Anderson J., Delseny M., Fregene M., Jorge V., Mba C., Lopez C., Restrepo C., Piegu B., Verdier V., Cooke R., Tohme J., Horvath D. 2003. An EST Resource for Cassava and Other Species of Euphorbiaceae. Plant Molecular Biology (submitted)2. Bayuelo, J. JS, DG Debouck & JP Lynch. 2002. Growth, gas exchange, water relations, and ion composition of Phaseolus species grown under saline conditions. Field Crops. Research 80: 207-222. Bayuelo, J. JS.; D.G. Debouck, Lynch, J.P. 2002. Salinity tolerance in Phaseolus species during early vegetative growth. Crop Science 42(6):2184-2192. Beebe, Steve. Improvement of common bean for mineral nutritive content at CIAT. Chapter for an encyclopedia. 442 Blair MW, Pedraza F, Buendia HF, Gaitán-Solís E, Beebe SE, Gepts P, Tohme J (2003) Development of a genome-wide anchored microsatellite map for common bean (Phaseolus vulgaris L.) Theor Appl Genet (accepted) Blair MW, Garris AJ, Iyer AS, Chapman B, Kresovich S, McCouch SR. (2003) High resolution genetic mapping and candidate gene identification at the xa5 locus for bacteria} blight resistan ce in rice ( Oryza sativa L.). Theor Appl Gene t. 107: 62-73 Blair MW, Beebe S, Tohme J (2003) CIAT Bean Genomics. In Legume Genomics. Special issue, ed. G. Stacey, K Vendenbosch. Plant Physiology Broughton WJ, Hemandez G, Blair MW, Beebe SE, Gepts P, Vanderleyden J (2003) Beans (Phaseolus spp.)- Model Food Legumes. Plant and Soil 252: 55-128. K Borrero,J. Martínez,CP.,Almeida,A., Duque,MC.,Correa,F., Carabalí,SJ., Delgado D., Tohme,J., Silva,J. 2003.Incorporando genes valiosos de parientes silvestres al arroz cultivado.Paper presented at Vlli Congreso Soc. Col. Fitomejoramiento y Producción Cultivos. Bogotá, Julio 2-5,2003. Buendia, HF, Beebe SE, Blair MW, Terán H, Pedraza F (2003) Identificación de marcadores moleculares asociados a genes de rendimiento en una población RC2F3.7 de frijol común Phaseolus vulgaris L. (DOR390 x 019892). Fitotecnia Colombiana 3: 57-64. Coulibaly, I; J . Louarn, M. Lorieux , A. Charrier, S. Hamon, M. Noirot. 2003. Pollen viability restoration in a Coffea canephora P. and C. heterocalyx Stoffelen backcross. QTL identification for marker-assisted selection. Theoretical and Applied Genetics 106:311-316 Cortés, D.F.; Reilly, K.; Okogbening, E.; Beeching, J .R. ; Iglesias, C.; Tohme, J. 2002. Mapping wound-response genes involved in post-harvest physiological deterioration (PPD) of cassava (Manihot esculenta Crantz). Euphytica. -128(1);47- 53 Chavarriaga P, Pachico D and Tohme J (2003) Transgenic cassava as a tool to control the stemborer Chilomima clarkei in Colombia. In: "Forging Links: Southem Perspectives on Biotechnology and Trade", Chapter X. Intemational Centre for Trade and Sustainable Development ICTSD, Switzerrland (In press). Coulibaly, I. B. Revol, M. Noirot, V. Poncet, M. Lorieux, C. Carasco-Lacombe, J. Minier, M. Dufourand P. Hamon. 2003. AFLP and SSR polymorphism in a Coffea interspecific backcross progeny [( C. heterocalyx x C. canephora ) x C. canephora ]. Theoretical and Applied Genetics 107:1148-1155 Dom, B.; Mattiacci, L. Bellotti, A.C.; Dom, S. 2003. Effects of a mixed species infestation on the cassava mealybug and its encyrtid parasitoids. Biological Control 27(1): 1-10. May Elsevier Science. Echeverry M, Mancilla U , Cortes DF, Chavarriaga P y Tohme J (2003) Evaluación preliminar de la expresión del gen bar en plantas transgénicas de yuca (Manihot esculenta) mantenidas en reproducción vegetativa por cerca de diez años. Revista de la Asociación Colombiana de Ciencias Biológicas 15:43-52. Freytag GF & DG Debouck. 2002. Taxonomy, distribution, and ecology of the genus Phaseolus (Leguminosae-Papilionoideae) in North America, Mexico and Central America. SIDA Bot. Mise. 23: 1-300. Fregene M., Mba C., Buitrago C., Zarate A., Garcia T., Tohme J. 2003 A Predominantly Simple Sequence Repeat (SSR) Marker Map of Cassava (Manihot esculenta Crantz). Plant Molecular Biology (submitted) Fregene M., Suarez M., Mkumbira J. , Kulembeka H., Ndedya E., Kulaya A. , Mitchel S. Gullberg U. , Rosling H., Dixon A., Kresovich S. (2003) Simple Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta Crantz) Landraces: Genetic Structure in a Predominatly Asexually Propagated Crop Theor Appl Genetics 107:1083-1093 Fregene M., Matsumura H., Akano A., Dixon A., Terauchi R. 2003. Serial Analysis of Gene Expression (SAGE) of Host Plant Resistance to the Cassava Mosaic Disease Resistance (CMD) Plant Molecular Biology (submitted) Gaitan, S.E.; Duque, MC.; Edwards, K.J.; Tohme, M.J. 2002. Microsatellite repeats in common bean (Phaseolus vulgaris L.): Isolation, characterization, and cross- species amplification in Phaseolus ssp. Crop Science 42(6):2128-2136 Gómez OJ, Blair MW; Frankow-Lindberg BE, Gullberg U (2003) Molecular and phenotypic diversity of common bean (Phaseolus vulgaris L.) landraces from Nicaragua. Genetic Resources and Crop Evolution (accepted) González-Torres, R.l., E. Gaitán, M.C. Duque, O. Toro, J. Tohme & DG Debouck. 2003. Monitoring gene flow between wild relatives and landraces of common bean in Costa Rica. Annu. Rept. Bean Improvement Coop. (USA) 46:1-2. House, W.A.; Welch, R.M.; Beebe, S. Cheng, Z. 2002. Potential for increasing the amounts of bioavailable zinc in dry beans (Phaseolus vulgaris L) through plant breeding. Joumal of the Science of Food and Agriculture. 82(13):1452-1457. 444 Langar, K. ; M. Lorieux, E. Desmarais, Y. Griveau, L. Gentzbittel, A. Bervillé. 2003. Combined mapping of DALP and AFLP markers in cultivated sunflower using F9 recombinant inbred lines. Theoretical and Applied Genetics 106:1068 - 1074 Lentini Z., Lozano 1, Tabares E., Fory L., Domínguez J ., Cuervo M., Calvert L. 2002. Expression and inheritance of hypersensitive resistance to rice hoja blanca virus mediated by the viral nucleocapsid protein gene in transgenic rice. Theoretical and Applied Genetics 106: 1018-1026. Published online: 14 December 2002. Lopez, C.E.; Acosta, I.F.; Jara, C.; Pedraza, F.; Gaitan-Solis, E.; Gallego, G.; Beebe, S.; Tohme, J. 2003. ldentifying resistance gene analogs associated with resistances to different pathogens in common bean. Phytopathology. 93(1): 88-95. Lorieux M.; G. Reversat, S. X. Garcia Diaz, C. Denance, N. Jouvenet, Y. Orieux, N. Bourger, A. Pando-Bahuon, A. Ghesquiere. 2003. Linkage mapping of Hsa-1°8, a resistance gene of African rice to the cyst nematode, Heterodera sacchari. Theoretical and Applied Genetics 107:691-696 Mahuku, G.S. , C. Jara, C.E.; Cajiao, V.; Beebe, S. 2002. Sources of resistance to angular leaf spot (Phaeoisariopsis griseola) in common bean core collection, wild Phaseolus vulgaris and secondary gene pool. Euphytica. 130(3):303-313. Mahuku, G.S. , C. Jara, C.E.; Cajiao, V. ; Beebe, S. 2002. Sources of resistance to Colletotrichum lindemuthianum in the secondary gene pool of Phaseolus vulgaris and in crosses of primary and secondary gene pools. Plant Disease. 86(12):1383- 1387. The American Phytopathological Society. Martinez, C.P., Borrero, J ., Almeida, A., Duque, M. C., Correa, F., Carabalí, S. J., Delgado. D., Tohme, J., Silva, J.2003.Análisis de la adaptación regional de líneas interespecíficas. Paper presented VITI Congreso Sociedad Col. Fitomejoramiento Y Producción Cultivos. Bogota, Julio2-5,2003. Martinez, A.K.; Gaitán-Solis, E.; Duque, MC.; Bemal, R.; Tohme, J. 2002. Microsatellite loci in Bactris gasipaes (Arecaceae): their isolation and characterization. Molecular Ecology Notes. 2(4):408-410. Okogbenin E. and Fregene M (2002). Genetic Análisis and QTL mapping of early root bulking in an Fl population of non-inbred parents in cassava (Manihot esculenta Crantz). Theoretical and Applied Genetics. 106(1):58-66 Sallaud, C. ; M. Lorieux, E. Roumen, D. Tharreau, R. Berruyer, P. Svestasrani, O. Garsmeur, A. Ghesquiere, J.-L. Nottéghem. 2003. Identification of five new blast resistance genes in the highly blast-resistant rice variety IR64 using a QTL mapping strategy. Theoretical and Applied Genetics 106:794-803 Sperling, L. 2002. Emergency seed aid in Kenya: Sorne case study insights on lessons leamed during the 1990s. Disasters. 26(4):329-342 Sperling, L.; Longley, C. 2002. Beyond seeds and tools: Effective support to farmers in emergencies. Disasters. 26(4):283-287 Tomkins J., Fregene M., Main D., Kim H., Wing R., and Tohme J. 2003. Bacteria! Artificial Chromosome (BAC) Library Resources for Positional Cloning of Pest and Disease Resistance Genes in Cassava (Manihot esculenta Crantz). Plant Molecular Biology (submitted) Thornson M.J., Tai T.H., McClung A.C., Hinga M.H., Lobos K.B., Xu Y., Martínez C., McCouch S.R. 2003. Mapping quantitative trait loci for yield components, and morphological traits in an advanced backcross population between Oryza rufipogon and the Oryza sativa Jefferson. Theoretical and Applied Genetics 107:479-493. Wenzl, P.; Chavez, A.L.; Patiño, G.M.; Mayer, J.E.; Rao, I.M. 2002. Aluminium stress stimulates the accumulation of orgnic acids in root apices of Brachiaria species. Joumal of Plant Nutrition and Soil Science. 165(5):582-588. 3.5.2 Proceed.ings, Abstract and Others Blair MW, Giraldo MC, Duran L, Beaver J, Nín JC (2003) Phaseolin characterization of Caribbean common bean germplasm. Annual Report of the Bean Improvement Cooperati ve. Blair MW, Iriarte G, Beebe SE (2003) QTL analysis of an Andean advanced backcross population for yield traits derived from wild P. vulgaris. Annual Report of the Bean Improvement Cooperative Blair MW, Pantoja W, Muñoz LC, Hincapie A (2003) Genetic analysis of crosses between cultivated tepary bean and wild Phaseolus acutifolius and P. parvifolius. Annual Report of the Bean Improvement Cooperative Blair, MW (2003) "Mejoramiento nutricional mediante biotecnologia" ACTA; Asociacion Colombiana de Ciencia y Tecnología de Alimentos, April 30-May 1, Bogotá, Colombia (Abstract). Blair, MW (2003) "Introducción a la Genómica Vegetal" VIII Congreso Sociedad Colombiana de Fitomejoramiento y Produccion de Cultivos. July 2-5, 2003. (Abstract). Checa O, Blair MW (2003) Trait correlations in climbing beans. Annual Report of the Bean Improvement Cooperative. 446 Checa O, Blair MW (2003) "Evaluacion de caracteres asociados con la capacidad trepadora y rendimiento en frijol voluble (Phaseolus vulgaris)" VIII Congreso Sociedad Colombiana de Fitomejoramiento y Produccion de Cultivos. July 2-5, 2003. (Abstract). Florez, C.P.; Emmerling, M.; Fory, L.F.; Spangenberg, G. and Lentini, Z. 2003. Isolation and characterization of a caffek acid 0-methyl transferase from signal grass. Molecular Breeding of Forage and Turf, Third International Syrnposium, May 18-22,2003. Dalias-Texas, USA. Poster# 136, p. 131. Florez, C.P.; Arroyave, J.A. ; Duque, M.C. ; and Lentini, Z. 2003. Irnproving Somatic embryogenesis and plant regeneration in Brachiaria decumbens. Molecular Breeding ofForage and Turf, Third lntemational Symposium, May 18-22, 2003. Dalias-Texas, USA. Poster # 129, p. 124. Iriarte, GA, Blair MW, Beebe SE (2003) "ldentificacion de qtls en una retrocruza avanzada entre una accesion silvestre y una variedad cultivada de frijol comun" VIII Congreso Sociedad Colombiana de Fitomejoramiento y Produccion de Cultivos. July 2-5, 2003. (Abstract). Lentini, Z. 2003. Genetic transformation as a tool for germplasm development at CIA T. Models of Food Safety Assessment of Transgenic Crops. May 6-8, 2003. IFPRI, Washington DC. USA Lentini, Z. 2003. The use of doubled-haploids in breeding of a self-pollinated crop: The case of rice. Planning Workshop on Development of a Protocol. for the Generation of Cassava Doubled-Haploids and their Use in Breeding. June 11-12, 2003. Cali, Colombia. Lentini, Z. 2003. Developing cassava haploid technology. Planning Workshop on Development of a Protocol for the Generation of Cassava Doubled-Haploids and their Use in Breeding. June 11-12, 2003. Cali, Colombia. Lentini, Z. 2003. Unique Challenges and Opportunities for Environmental Assessment of GMOs in the Tropics. In : Craig R. Roseland. (Ed.). LMOS and the Environment. Proceedings of an Intemational Conference (2002). November 27-30, 2001. Raleigh, North Carolina, USA. OECD, Paris. Muñoz C, Blair MW, Debouck D "Diversidad Genetica en Frijol Tepari (Phaseolus acutifolius)" Vlli Congreso Latinoamericano de Botánica, Cartagena, Colombia. October 13-18, 2002 (Abstract). Santana G, Blair M, Morales F, Mahuku G, Jara C, Castaño M (2003) "Uso de tecnicas clasicas y avanzadas para identificar genotipos de frijol resistentes a antracnosis y mosaico común" VIll Congreso Sociedad Colombiana de Fitomejoramiento y Produccion de Cultivos. July 2-5, 2003. (Abstract) . 3.5.3 Thesis Andrea Frei - (January - July 2003, thesis submitted Sept 2003) ETH, Switzerland - studying the quantitative trait loci involved in resistance to the leaf-feeding insect, Thrips palmi in common bean (collaboration C. Cardona, S. Dorn, H. Gu) Andrés Salcedo. 2003. BSc. Clonaje y caracterización parcial de genes implicados en la biosíntesis de carotenos en raíces de yuca. Universidad del Valle. Catalina Romero. 2003. Aproximación genómica al fenómeno de resistencia de Brachiaria al salivazo (Aeneolania varia):Correlación con homólogos de genes de resistencia (HGR) y aislamiento de genes expresador diferencialmente en la respuesta de defensa. Universidad Nacional de Colombia, Sede Bogotá. (Laureada). Claudia Patricia Flores. Ph.D. Thesis. Desarrollo y uso de la transferencia genetica de Brachiaria decumbens stapf en fitomejorarniento.Universidad Nacional de Colombia. Sede Palmira. (Meritoria). Elizabeth Okai (Ghana) Ph.D. student, University of the Free State, Bloemfontein, South Africa (6 months, left January 2003) Elizabth Kizito (Uganda) Ph.D. student, University of the Free State, Bloemfontein, South Africa (June-December 2003) Hernando Ramírez. Ph.D. Thesis. Universidad Nacional de Colombia. Sede Palmira (graduated June 2003). Incorporación de resistencia al cogollero "tuta absoluta" en la variedad de tomate UNAP AL-Arreboles, por transformación genética, vi a Agrobacterium. Universidad Nacional de Colombia, sede Palmira - Colombia. 193p. Henry Ojulong (Uganda) Ph.D. student University of the Free State, Bloemfontein, South Africa (6 months, left January 2003) Ivan Ochoa- (Jan 2003 - Feb 2003) Pennsylvania State University, USA- conducting laboratory and field studies to understand the inheritance and mechanisms of low 448 phosphorous tolerance in common bean and the role of adventitious rooting in adaptation to low phosphorous stress (collaboration with J. Lynch). Juan Jairo Ruiz M.Sc. en Fitomejoramiento. 2003. Evaluación del comportamiento agronómico de clones de guanábano "Annona muricata L." propagados por microinjertación in vitro. Universidad Nacional de Colombia, Sede Palmira. 146p. Martha Isabel Moreno (Colombia) M.Sc. student, Universidad de Valle, Cali (September 2002- August 2003) Manuel Quintero, Under-Graduate Thesis. Universidad Nacional de Colombia. Sede Palmira graduated September 2003). Ajuste del sistema RITA para la inducción de callos embriogénicos y regeneración de plantas a partir del cultivo de anteras de arroz. (Meritoria). Maria Eugenia Buitrago. 2003. Evaluación de la tolerancia al aluminio, de una población segregante de Brachiaria decumbens x Brachiaria ruziziensis. Universidad del Valle. Mauricio Soto. 2003. Herramientas de genornica funcional: Análisis de ESTs, Librerías sustractivas y microarreglos de ADN para avanzar en el conocimiento de las respuestas de defensa de la yuca a la infección x Xanthomonas axonopodis pv manihotis. Universidad Nacional de Colombia, Sede Palmira. (Meritoria). Osear Checa- (complete year, from 2001 onward) Universidad Nacional de Colombia - Palmira, Colombia - studying the inheritance of climbing ability in common bean and the importance of genotype x environment interaction in this trait. Paola Ruíz. 2002. Caracaterización fenotípica y genética del arroz rojo ( Oryza sativa f. Spontanea L.) de Saldaña (Departamento del Tolima) y Huila. Universidad Javeriana, Bogotá. Rosana Pineda, M.Sc. Thesis. Universidad Nacional de Colombia. Sede Medellín. Evaluación de Flujo de genes desde arroz transgénico (Oryza sativa L) y no transgénico (V. Purpura) hacia la maleza arroz rojo (Oryza sativa F. Spontanea) bajo condiciones experimentales de campo. Universidad Nacional de Colombia, Sede Medellín. 2003. Silvio Cadena; M.Sc. Thesis; 3 years. National University of Palmira, Colombia - Corporación Biotec. Greenhouse management of in vitro propagated plants of soursop Sil vio James Carabalí. Master Thesis: Genetic gains in rice grain quality obtained through severa! cycles of recurrent selection. Universidad Nacional de Colombia Secciona! Palmira. Febrero 2000-Diciembre 2002. Vanessa Segovia. M.Sc. Thesis. Universidad Real de Madrid, Spain. Optimización de la regeneración de lulo (Solanum quitoense) orientada a la transformación genética de plantas. Universidad Internacional de Andalucía, Sede Rabida-España. Maestría en Biotecnología de Plantas. April 2003. Yina J. Puentes Páramo. Ensayos preliminaries para la transformación genética con orientación antisentido del gen Waxy en yucca (Manihot esculenta Crantz) para la obtención de un almidón 100% amilopectina. Universidad Nacional de Colombi, Sede Palmira. (Meritoria) 3.6 Projects 3.6.1 Project approved or on going High through-put genetic diversity characterization of germplasm with a DNA chip. Donor: IPGRI. Expanding the range of uses of cassava starch: A source of income generation. Donor: USAID Development of strategies for better targeting of seed relief and linking relief and rehabilitation. Donor: FAO Models of food safety assessment of transgenic crops - Workshop. Donors; USAID, Rockefeller Foundation Nutritional Genomics. Donor: USAID Maize - Vit A Biofortification. Donor: USAID Model of Food Safety Assessment of Transgenic Crops. Donor: USDA Biofortified Crops for Improved Human Nutrition. Donor: WBIIFPRI Bean Genomics for Improved drought Tolerance in Africa and Latín America. Donor: GTZ A molecular marker-assited, farmer-participatory breeding project to improbé local cassava varieties in Tanzania with resístanse to pest and disease (Rockefeller Foundation). 450 Gene tic Mapping of the Linamarin biosynthetic genes CYPD 1 and D2 and the development of markers for CNP in Cassava, in collaboration with Prof. Birger Moller, Royal Agriculture and Veterinary University, Copenhagen (DANIDA). Genoplante. Project for phenotypic and characterization of a series of T -DNA mutants. 3.6.2 Project Submitted, in preparation and concept notes Combating hidden hunger in Latin America: Biofortified crops with improved Vitamin A, essential minerals and quality protein A collaborative project submitted to CIDAon behalf of A Partnership of Intemational Agricultura) Research Centers, and National Agricultura) Research Systems in Latin America Using biotechnology tools and GIS to conserve biodiversity in Colombia Development of micro-satellite markers to facilitate use of the cassava Molecular genetic maps by African collaborators working on gene mapping resístanse to CMD Towards the development of industrial cassava varieties: genetic and molecular análisis of early bulking in cassava gennplasm collection Marcadores moleculares asociados a resistencia a pudrición radical por phytophthora drechsleri, Phytophthora nicotianae y Phytophthora cryptogea en una población segregante de yuca. Sustainable oil palm production as a source of employment and mcome for rural communities and small-scale farmers in tropical Latin America. Molecular Characterization of Genetic Diversity and the Definition of Heterotic Groups in Cassava. Applications of Spatial Statistics and GIS to Cassava Bacteria! Blight Management. Developing and exploiting expressed séquense Tags for cassava Starch and Bacteria! Blight Resístanse. Expanding the range of uses of cassava starch: A source of income generation. Seed aid and gennplasm restoration in disaster situations: síntesis of lessons leamed and promotion of more effective practices. Identificación de marcadores moleculares para la resistencia a la enfermedad de la hoja blanca del arroz en programas de mejoramiento. Development of an in vitro protocol for the production of cassava doubled-haploids and its use in breeding. Combating hidden hunger in Latín America: Biofortified crops with improved Vitamin A, essential minerals and quality protein. "Comparative genomics and genetics in legumes" a collaborative research Project between CIAT and University of Aarhus, concept note prepared for DANIDA. "Phaseomics" wRUIG-GIAN. (Submitted by Univ. Of Geneva with CIAT collaboration) "Utilización de hierro y zinc en modelo animal y respuesta clínica al consumo habitual de fríjol de alta densidad mineral en mujeres y niños" Submitted by Universidad del Valle (with CIA T) to Colciencias. The molecular diversity network of Cassava (MOLCAS). Mutagenesis of Cassava (Manihot esculenta Crantz) for the generation, identification and Molecular Análisis of Novel traits. Research Contract submitted to the Intemational Atomic Energy Agency (IAEA), Viena, Austria. W orkplan in sub-programs of the Gene tic Resources Project for 2004. Fregene Challenge Program "Unlocking Genetic Resources in Crops fort the Resource-Poor". Genoplante. Project for phenotypic and characterization of a new series of T-DNA mutants. Development and use of inbred lines in cassava breeding. Submitted to the Rockefeller Foundation. New York . Development of an In vitro Protocol for the Production of Cassava Doubled-Haploids and its Use in Breeding. Submitted to ZIL, Switzerland. 3.6.3 Projects funded and their Donors (Oct 2002- Sept 2003) Cana da Intemational Development Research Centre. (IDRC) 452 Strategies for integrating small-scale end-users in cassava biotechnology research (Latín America) Colombia Fundación para la Investigación y el Desarrollo Agrícola. (FIDAR) Rice Functional Genomics Consortium Ministry of Agriculture and Rural Development. (MADR) Regeneration capacity and genetic transforrnation potential of cornmercial cassava varieties in Colombia Propagation and certification of FSD-free cassava Biotech Fruits Corporación BIOTEC Molecular and agromorphological characterization of native genetic variability of soursop and related Annonaceae species Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnología. (COLCIENCIAS) Characterization of cassava resistance to vascular bacteriosis and its use in breeding Instituto de Investigaciones de Recursos Biológicos Alexander von Humboldt. Use of morphological and molecular techniques to study the diversity and conservation of endangered Colombian palrn trees Investigación sobre etiología, epidemiología y control de la Mancha Anular de la Palma de Aceite de la Zona Occidental de Colombia productora de Palma de Aceite Belgium Belgian Administration for Development Cooperation. (AGCD/BADC) Genetic hnprovement of comrnon beans using exotic gerrnplasm and biotechnology France Advanced Research Platform. (AGROPOLIS) Genoplante (IRD) Developing and exploiting expressed sequence tags for cassava starch and bacteria! bligh resistance Genoplante- evaluation and multiplication of 5000 lines of TDNA-mutants Gennany German Agency for Technical Cooperation. (GTZ) An integrated approach to genetic improvement of aluminum resistance in crops on low- fertility acid soils Gene flow analysis for assessing the safety of bio-engineered crops in the tropics New Zeland Govemment of New Zeland (NZ) Ro me Food and Agriculture Organization. (F AO) Meeting to measure baselines of genetic diversity Intemational Plant Genetic Resources Institute (IPGRI) Hihg through-put genetic diversity characterization of germplasm with a DNA chip. The Netherlands Ministry ofForeign Affairs and Trade. (MFA) Directorate General Intemational Cooperation. (DGIS) Cassava Biotechnology Network m- CBN United Kingdom Wallace Genetic Foundation (WGF) Department for International Development (Dfl.D) Knowledge and tools for the modulation of post-harvest physiological deterioration in cassava. 454 USA Rockefeller Foundation. (RF) Legume genomics meeting between US and CGIAR Research development of a molecular maps of cassava (Manihot esculenta) Delivery of transgenic rice cultivars to seed producers and farmers in tropical A.merica, following a multi-step approach involving biosafety assessment, nutritional testing, and negotiations on intellectual property rights Molecular marker-aided analysis of traits of agronomic importance in cassa va Rice biotechnology Research Agency for International Development. (USAID) Crop Biofortification Initiative Yale University Rice Functional Genomics Consortium Development of molecular markets for the breeding of sustainable pest resistance in common beans - a novel strategy Department for International Development. (DFID) Reviving the agricultura! base of a region: Use of genetic transformation and interactive testing to restare predominant locally adapted cassava varieties. Knowledge & tools for the modulation of post-harvest physiological deterioration in cassava. IFPRI Biofortified Crops for Improved Human Nutrition Venezuela Centro Tecnológico Polar Ensuring stable and durable resistance of rice to pathogens and pests: rice Hoja Blanca Virus, Rhizoctonia solani, and Sogata Activity 3.7 Project SB-2 Project 3.7.1 Current SB-2 Investigators: Discipline, position and time fraction N ame Discipline Al ves Alfredo CBN Regional Coordinator Beebe Steve Bean Breeding Bellotti Anthony Cassava Entomology Blair Mathew Bean Genetics and breeding Ceballos Hernan Cassava Breeding Chavarriaga, Paul Transgenesis, Cassava Debouck Daniel Botan y Fregene Martín Cassava Genetics and breeding Ishitani Manabu Molecular Biologist Lentini Zaida Biology/Genetics Lorieux Mathias Rice Genetics and Biotechnology Martínez César Breeding Mejía Alvaro Cell Biology Sperling Louise Seed Systems TohmeJoe Genornics, Project manager Tissue Culture/Cryopreservation/Plant Transformation Escobar, Roosevelt - Biochemists/Education Galindo, Leonardo F.- Agron. Engineering González, Eliana- Biologist Ladino, Janeth Julieta- Agron. Engineering Manrique, Norma - Agron. Engineering Muñoz, Liliana - Biochemists/Education Segovia, V anesa - Agron. Engineering Tabares, Eddie- Biologist López, Danilo - Agron. Engineering Echeverry, Morgan- Biologist Juan Jairo Ruiz Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Time dedication% 100 30 20 70 40 100 20 60 100 80 50 49 100 20 lOO Fory, Luisa- Biologist Rios, Auradela - Biochemistry Tech. Bolaños, Eugenio Rice Biotechnology Research Coordínator Technician Technician Dorado, Carlos Technician Herrera, Pablo Technician Ríos, Alexander Technician Tigreros, Humberto Technician Muñoz, Carmenza - Visiting Researcher 456 Genome Diversity Gallego,Gerardo.- Biologist Gaitán, Eliana. - Biologist Barrera, Edgar. - Biologist Gutiérrez, Janeth P. - Biologist Bohorquez, Adriana.- Biologist Vargas, Jaime. - Biologist Quintero, Constanza. -Biologist Muñoz, Monica. -Biologist Cortés Diego F. - Biologist Giraldo,Olga X.- Biologist Galindo, Lonardo M.- Biologist Reyes, Nidia Londoño, Claudia Plant-Stress interactions Chaves Alba L.- Chemistry Soto Mauricio. - Microbiologist Leonardo Miguel Galindo Andrés Felipe Salcedo Maria Eugenia Recio Administra ti ve Cruz, Oiga L. Zuñiga, Claudia S. Duque, Myriam C. Institute A. von Humboldt Palacio, Juan D. - Agron. Engineer Carolina Villafañe Tania García Corporación BIOTEC Royero, Nelson. - Biologist Cadena, Sil vio.- Agron. Engineer García Víctor Rugo- Biologist CORPOICA Sanchez, I., PhD. Genome Research Coordinator Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Research Assistant Technician Technician Research Associate Research Assistant Research Assistant Research Assistant Technician Bilingual Secretary Bilingual Secretary Statistical Consultant Visiting Researcher Visiting Researcher Visiting Researcher Visiting Researcher Visiting Researcher Visiting Research Visiting Researcher 3.7.2 Current Graduate Students Andrea Freí- PhD. ETH, Switzerland- studying the quantitative trait loci involved in resistance to the leaf-feeding insect, Thrips palmi in common bean (collaboration C. Cardona, S. Dom, H. Gu) Constanza Quintero. Recursos Fitogenéticos Neotropicales. MSc. Plant Genetic Resources. Universidad Nacional de Colombia Sede Palmira Eliana Gaitán; Molecular markers and diversity of palm trees. PhD. Plant Breeding Program, Universidad Nacional de Colombia, Palmira, Colombia. Eliana González. BSc. 2001. Univalle. Diversidad genética de 3 poblaciones de colombo balanus excelsa (fagacia) especie endémica de los Andes Colombianos Edgar Barrera; Molecular markers for ACMD resistance- MSc Plant Breeding, Universidad Nacional de Colombia, Palrnira, Colombia. Eyvar Andrés Bolaños Vidal. BSc. Caracterización de la diversidad genética en cuanto a contenidos de caroteno de raíces y hojas de 682 genotipos de yuca. Fabio Escobar; Molecular markers to certify seeds of rice - MSc Program, Agronornic Sciences, Universidad Nacional de Colombia, Palrnira, Colombia. Gerardo Gallego. Gene cloning of rice disease resistance genes. PhD. Plant Breeding Program, Universidad Nacional de Colombia, Palrnira, Colombia. Iván Ochoa. PhD. Pennsylvania State University. USA. Conducting laboratory and field studies to understand the inheritance and mechanism of low phosphorous tolerance in common bean and the role of adventitious rooting in adaptation to low phosphorus stress (collaboration with J. Lynch). Jaime Vargas. MSc. Plant Breeding. Identificación de marcadores moleculares rnicrosatélite asociados con el gen de resistencia a mosca blanca en yuca. Universidad Nacional de Colombia, Sede Palrnira. Leonardo Fabio Galindo. Master Bussiness Administration. Universidad del Valle. Martha Isabel Moreno, Post Graduate (M.Sc). Universidad del Valle. Gene Cloning of CMD2) Meike Anderson. PhD. 2002. Genetic diversity and core collection approaches in the multipurpose shrub legumes Flemingia macrophylla and Cratylia argentea. University of Hohenheim Nelson Rayero; Molecular markers and diversity of Anonna spp - MSc Plant Breeding, Universidad Nacional de Colombia, Palmira, Colombia. 458 Osear Checa- PhD. Plant Breeding. Universidad Nacional de Colombia, Sede Palmira. Studying the inheritance of climbing ability in common bean and the importance of genotype x environment interaction in this trait. Paola Fory. 2002. MSc. Plant Genetic Resources. Improving the breeding of lulo (Solanum quitoense LAM) through an understanding of the species genetic diversity . Universidad Nacional de Colombia, Sede Palmira Roosevelt Escobar. Genotypic stability of cryopreserved cassava plants- MSc. Plant Genetic Resources Program, Universidad Nacional de Colombia, Palmira, Colombia. Rosana Paola Pineda. 2002. MSc. Medición de flujo de genes con microsatelites en arroz. Universidad Nacional de Colombia, Sede Medellín. Vanessa Segovia. 2001 M.Sc. Thesis .. Optimización de la regeneración de lulo (Solanum quitoense), orientada a la transformación genética. Universidad Internacional de Andalucía, Spain Yamileth Cortés. MSc Plant Breeding. Analyzing the genetic diversity of the Colombian plantain collection (Musacea collection) using microsatellites. Universidad Nacional de Colombia, sede Palmira An Argentine PhD candidate was hosted for two weeks to prepare data analysis for his thesis work, involving a statistical analysis of multi-locational trials carried out over a 15 year period in the north-west of Argentina. A thesis plan, including field work, was established with a Cuban MsC student. The student will carry out a physiological analysis of lines derived from the cross of DOR 364 x BAT 477, the latter of which has expressed resistance to multiple abiotic stresses. The study will reveal the physiological relationship between resistances to low P, nitrogen and drought stress. 3.7.3 Undergraduate students (current) Andrés Bolaños, Universidad Nacional de Colombia, Sede Palmira Angela Zarate, Universidad de Tolima, lbague (Conversion of RFLP to SSCP markers) Carolina Castaño, Universidad de los Andes, Bogotá Carolina Astudillo, Universidad del Valle Carolina Ramirez Rodríguez, Universidad del Tolima Catalina Romero, Universidad Nacional de Colombia, Sede Bogotá Gloria Iriarte, Universidad de Tolima Rector Fabio Buendía, Universidad de Tolima Jaime Marin, Universidad de Tolima, lbague. (QTL mapping of early bulking) Juan Felipe Calderon, Universidad Nacional de Colombia, Sede Palmira (Humboldt) Leonardo Bocanegra, Universidad del Tolima (Micras, Rice) Manuel Quintero, Universidad Nacional de Colombia, Sede Palmira Maria Eugenia Buitrago, Universidad del Valle, Cali. (Plant Stress) Paola Ruíz. Universidad Javeriana, Bogotá Paula Andres. Universidad Javeriana, Bogota (Gene tagging of CBB resistance) Paola Cardenas, Universidad Javeriana Bogotá (Humboldt) Sergio Prieto- Universidad Nacional de Colombia Wilfredo Pantoja, Universidad del Valle 3.8. List of Acronyms and Abbreviations U sed in the Text Acronyms ADB AHI Bean!Cowpea CRSP BoT CA CARDER CARE CATIE CBN CENIPALMA ClALs CIFOR CIMMYT CIP CIPASLA CIRAD CLODEST CNPMF CODESU COLCIENCIAS CONDESAN CORPOICA CSIRO CURLA DAN IDA DFID OOIS DICTA DNP EAP·Zamorano EC ECABREN E CLAC EMBRAPA EPMR ETII FAO FCRI FLAR FONAIAP GRU GWG lBS RAM Asian Development Bank African Highland lnitiative Bean!Cowpea Collaborative Research Support Program (ofthe University of Georgia. USA) Board ofTrustees (o/CIA1) ~partement des Cultures Annuelles (o/CIRAD) Corporación Autónoma Regional de Risaralda, Colombia Cooperative for American Relief Everywhere Centro Agrónomico Tropical de Investigación y Enseñanza, Costa Rica Cassava Biotechnology Network Centro de Investigación en Palma de Aceite, Colombia Comit6 de Investigación Agrícola Local, Colombia Centre for Intemational Forestry Research, Indonesia Centro Internacional para Mejoramiento de Maíz y Trigo, Mexico Centro Internacional de la Papa, Peru Consorcio Interinstitucional para una Agricultura Sostenible en Laderas, Colombia Centre de Coopération lntcmationale en Recherche Agronomique pour le ~veloppement, France Comitl! Local para el Desarrollo Sostenible de la Cuenca del Río Tascalapa. Honduras Centro Nacional de Pesquisa de Mandioca e Fruticultura Tropical (o/EMBRAPA) Corporacióo para el Desarrollo Sostenible de Ucayali, Peru Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnología "Francisco José de Caldas", Colombia Consorcio para el Desarrollo Sostenible de la Ecorregión Andina, Peru Corporación Colombiana de Investigación Agropecuaria Commonwealth Scientific and Industrial Research Organisation, Australia Centro Universitario Regional del Litoral Atlántico, Honduras Danish Intemational Development Agency, Denmark Department for lntemational Development. UK Directorate-General for lntemational Co-operation, lhe Netherlands Dirección de Ciencia y Tecnología Agropecuaria. Honduras Departamento Nacional de Planeación, Colombia Escuela Agrícola Panamericana aJ Zamorano, Honduras Economic Commission (ofthe European Union) Eastem and Central Africa Bean Research Network Economic Commission for Latín America and the Caribbean Empresa Brasileira de Pesquisa Agropecuária. Brazil Extemal Program and Management Review (ofCIA 1) EidgenOSSische Technische Hochschule, Switzerland Food and Agriculture Organization of the United Nations Field Crop Research lnstirute, Thailand Fondo latinoamericano y del Caribe para Arroz de Riego, based at CIA T Fondo Nacional de Investigaciones Agropecuarias, Venezuela Genetic Resources Unit (o/CIA 1) Gender Worlcing Group (ofthe CGIAR Systemwide Programme on Participatory Research and Gender Analysis for ... ) lntemational Board for Soil Research and Management, Thailand 460 !CA !CARDA ICER ICIPE ICRAF ICRISAT IDEAM IDIAP IDRC IFDC IFPRI IGAC IGDN IGER IIA IIASA IICA liLA liTA ILRI INBIO INIA IN !AA INIAP INIFAP Mexico INNIT INTA lPGRI IPRA IRRI NITA IWMI JIRCAS LSU MT NARO NRI NRMG OA ORSTOM L' PABRA PASOLAC PBG PROCITROPICOS PRO DAR PROFRUOL PROFRIZA RNM SABRN SDC SINCHI SINGER SP-IPM SP-PRGA Instituto Colombiano Agropecuario, Colombia l.otematiooal Center for Agricultura! Researcb in the Dry Ateas, Syria I.otemaUy Commissiooed Externa! Review (ofCIA 1) l.otematiooal Centre of lnsect Physiology and Ecology, Kenya l.otematiooal Centre for Research in Agroforc:stty, Kenya l.otematiooal Crops Rc:searcb lnstitute for the Semi-Arid Tropics, l.o