ANNUAL REPORT 2001 PROJECT SB-02 ~·;.~ _,.~·((·~=~:-·¡.-:;~'-~? ' ¡ .. ..... . . . ... "·' t;:. l. :... · r. ·. 1 ·: · •. :; ¡•]N y ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH BIOTECNOLOGY CIAT For Internal Use only October, 2001 CONTENT ~;:. ~~'. ~¡('1 ,,....--) ""'fo11J ' ,.... ! 1 o ...., 1 • 1 • . .... ··í : l .. . . ·. - - ··-- .... 1 . . . - t' ' ·1 (1 -!'i:A C/Cri Y '- .. L ~u~ 1 ,.,CI(JN PROJECT SB-2 : ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH BIOTECHNOLOGY ....................................................................................................................... 1 PROJECT OVERVIEW ........................... ........... .................................... .................. ............ .......................... 1 WORK BREAKDOWN STRUCTURE ........................... .................. .......... .. .............. ........................... .................. 11 LOG FRAME- WORK PLAN FOR SB-2, 2002 .. .......................................................... ....................................... 111 PROJECT SB-2 HIGHLIGHTS 2001 ............................................................................................ ...................... VI OUTPUT l. GENO MES OF WILD AND CULTIVA TED SPECIES OF MANDATED AND NON MANDA TED CROPS, AND ASSOCIA TED ORGANISMS CHARACTERIZED ................. 1 ACTIVITY 1.1 CHARACTER1ZATION OF GENETIC DIVERSITY .............. ............................................................... 1 1.1.1 1.1 .2 1.1 .3 1.1.4 1.1.5 1.1.6 1.1.7 1.1 .8 1.1.9 1.1.10 1.1.11 1.1.12 1.1.13 1.1.14 Evaluation ofthe P. coccineus and P. po/yanthus core collections with molecular markers ............ 2 Genetic diversity ofthe CIAT tepary bean (Phaseolus acutifolius) collection measured with AFLP and microsatellite markers ..................................................................................................... 8 Genetic diversity of microsatellites in common bean ..................................................................... 13 ldentification of common bean genotypes using " Fingerprints" of metabolism enzymes and seed proteins: the case of the En ola variety ................................................ .................................... 16 Simple Sequence Repeat (SSR) marker diversity in cassava landraces: genetic diversity and differentiation in a predominantly asexually propagated crop .. ...................................................... 18 Assessment of genetic diversity among African cassava accessions resistant to the Cassava Mosaic Disease using SSR markers ............................................ .................................................... 25 Analysis of genetic diversity in Cassava landraces from the coastal, andean and forest region ofPeru ................................................ ......... ....................................... .......... .............................. ..... 31 Root quality and pest resistance genes from wild relatives of Cassava for broadening the crop genetic base ................... ....... ......... ............................................ ........ ... ........................................... 35 Assessing the genetic variability of Xanthomonas axonopodis pv. manihotis in Togo by using RFLP .......................................................... ..................................................................................... 40 Microsatellites to study genetic diversity in Indica and Japonica rice ............................................ 43 Gene flow analysis for assessing the safety oftransgenic rice in the Tropics ................................ 47 Genetic diversity and core collection approaches in the multipurpose shrub legumes Flemingia macrophy//a and Craty/ia argentea ............ .. .......... .......................................... ............. 50 Use ofmolecular techniques for the studies ofthe genetic diversity and conservation studies of endangered palms in Colombia .............................................. .......... .......................................... 59 Molecular and agro-morphological characterization of the genetic variability of Soursop (Annona muricata L.) accessions and related Annonaceus species ................................................ 62 ACTIVITY 1.2 IDENTIFICATION ANO MAPPING OF USEFUL GENES ANO GENE COMBINATIONS ......................... 64 /.2.1 Wild QTL pursued in population ofDOR 390 and Gl0022 ...... ............... ...................................... 65 1.2.2 QTL analysis of an Andean advanced backcross population for yield traits derived from wild P. vulgaris .............................................................................. ......................................................... 66 1.2.3 QTL mapping ofmicronutrient content in two populations ofcommon bean ................................ 68 1.2.4 Analysis of ferritin and other candidate genes for micronutrient accumulation in common bean se e d ............................................................ ............................ ................................................. 71 1.2.5 Tagging genes for resistan ce to Apion goodmani in common bean ............................................... 73 1.2.6 Identitication ofrelevant traits and development ofsegregating progenies in Cassava for further molecular marker analysis ............. .. ....... ............................................................................. 75 1.2. 7 Identitication of relevant traits and development of segregating progenies for further molecular marker analysis: leafretention during plant growth in the absence ofwater stress ....... 80 1.2.8 1.2.9 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 1.2.19 1.2.20 1.2.21 Identification of relevant traits and development of segregating progenies for further molecular marker analysis: "root bullking capacity" ...................................................................... 84 Improving the experimental error in the measurements of post harvest physiological deterioration ........ ................. ...................................................... ....... .............................................. 85 Progress Towards a PCR-Marker Based Map ofCassava and its in Cassava Breeding ................. 86 Marker-Assisted Breeding ofResistance to CMD in Latín American Cassava Gene Pools ........... 87 The novel CMD resistance gene (CMD2) confers high Jevels of resistance to the aggressive Ugandan strain (UgV) .......................................................................... ........................................... 90 QTL mapping in an F1 population from non-inbred parents in cassava: yield, yield related and root quality traits ............................................................................ ........................................ .. ....... 93 QTL mapping in an F 1 population from non-inbred parents in cassava: morphological traits ..... 1 O 1 Marker Fidelity Study of QTLs identified for early bulking in Cassava ...................................... 106 QTL mapping of cyanogenic potential (CNP) in cassava ............................................................. 108 Identification ofmarker linked genes conferring resistance to white fly in cassava ..................... 109 Mapping and pursuit of QTLs for yield and yield components in rice populations derived from backcrosses between wild species and cultivated rice ......................................................... 116 Evaluation and selection of inter-specific populations vía conventional breeding methods ......... 123 Influence ofwild rice species on the grain quality, nutritional va1ue, the eating and cooking quality of inter-specific progenies ........................................ .. ....................... ................................ 125 M.A.S. for BGMV resistance in the common bean ........ .............. .................. .............................. 128 M.A.S. in Rice .............................................................................................................................. 129 ACTIVITY 1.3 OEVELOPMENT OF MOLECULAR TECHNIQUES FOR ASSESSING GENETIC DIVERSITY ANO 1.3.1 1.3.2 1.3.3 1.3.4 1.3.5 1.3.6 1.3.7 1.3.8 1.3.9 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 MAPPING USEFUL GENES ........................... ................................................................... .......... 131 Oissection and sequen ce analysis of a cluster of resistance gene analogs associated with resistance to angular Ieaf spot in common bean ............................................................................ 132 Screening of a common bean cONA library to isolate full-Iength Resistance Gene Analogs (RGAs) ........................................................................................................................................ .. 134 Microsatellite repeats in common bean (Phaseolus vulgaris): Iso1ation, characterization and cross-species amplification in Phaseolus ...................................................................................... 136 Microsatellites isolated from common bean cONA and small-insert genomic libraries ............... 139 Legume microsatellites tested in common bean ..................... .. ........... ......................................... 143 Enhanced microsatellite map developed for common bean .......................................................... 147 Isolation and characterization ofTyl-Copia group retrotransposon L TR sequences in Phaseolus vulgaris ........................................................................................................................ 148 Analysis of Ty 1-copia retrotransposon L TR Sequences and Their U se for genome organization studies ...................................................................................................................... 152 Application ofthe Oiversity Array Technology (DarD in beans for mapping and germplasm characterization .................................................................................. ................ ........................... 15 5 Application of the Oiversity Array Technology (DarD in genepoo1 characterization and marker assisted selection for Cassava irnprovement.. ................................................................... 158 Genome location of SSR markers from a cassava cONA library ................................................. 160 Annotation of SAGE tags (Transcripts) differentially expressed in CMD resistant and susceptible genotypes .............................................................. ..................................................... 163 A comparison of marker assisted selection (MAS) and conventional selection for the rapid deployment ofthe novel CMD resistance gene (CMD2) in cassava gene pools .......................... 166 cDNA-AFLP analysis of differential gene expression in the Cassava - Xanthomonas axonopodis pv. manihotis interaction .................................................... ........................................ 168 Study of gene expression during pathogenesis of Xanthomonas axonopodis pv. manihotis using an AFLP-based microarray ................................................................................................. 1 72 Gene expression in cassava stems in response to infection by Xanthomonas axonopodis pv. manihotis ........................................................................................ .............................................. 174 Toward Fine-Mapping of Major Genes for Blast Resistance in Rice ........................................... 177 Constructing cONA Express ion Libraries for Oryzica Llanos 5 and Irat 13 ........................ ......... 179 11 1.3.19 1.3.20 1.3.21 1.3.22 Oetection ofOifferentially Expressed Genes Related to Apomixis using cONA Subtraction Coupled to Microarray Hybridization ...... .. ........................ ........................................................... 181 Isolation ofResistance Gene Analogs (RGAs) from Brachiaria .................................................. 183 Construction of a molecular genetic map of Brachiaria and QTL analysis of spittlebug resistance ...................................................................................................................................... 186 Isolation and characterization of microsatellite loci in Bactris gasipaes ...................................... 192 OUTPUT 2 GENES AND GENES COMBINATIONS MADE A V AILABLE FOR BROADENJNG THE BASE OF MANDATED ANO NON MANDATED CROPS ......................................... 195 ACTIVITY 2.1 TRANSFER OF GENE AND GENE COMBINA TIONS USING CELLULAR ANO 2.1.1 2.1.2 2.1.3 2.1.4 2.1.5 2.1.6 2.1.7 2.1 .8 2.1.9 2.1.10 2.1.11 MOLECULAR TECHNIQUES .................................... ....................................................... 195 Genetic transformation oftepary beans and breeding of common bean genotypes amenable to Agrobacterium tansformation... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Transformation of cassava cultivars with a crylAb gene for insect resistance ............................. 202 Transformation and Regeneration of sorne CIAT Elite Cultivars: Towards Testing Candidate CMD Resistance Genes ................................................................................................................ 205 Control ofRHBV (Rice Hoja Blanca Virus) through nucleoprotein mediated cross protection in the greenhouse and in the field .............................. ........................ ....... .................................... 207 Characterization ofTransgenic Rice Containing the RHBV Non-Structural4 {NS4) Gene from the RNA 4 .. .................................................. ........................................................................ 211 Foreign genes as novel sources of resistance for fungal resistance .............................................. 215 Oevelopment of genetic transformation of Brachiaria mediated by Agrobacterium tumesfaciens ......... ........................................................... ........ ..................................................... 218 Isolation of lignin biosynthetic genes from Brachiaria decumbens .............................................. 220 Genetic transformation oftomato variety UNAPAL Arreboles for resistance to udworm (Tuta absoluta) ............... .............. ........................................ ........................ ........................ ......... ......... 222 Resistance to sugar cane yellow leaf virus (Se YL V): Genetic transformation an altemative aiding breeding of sugar can e ....................................................................................................... 224 Expression ofrecombinant CRYI(A)b Protein in E. coli .............................................................. 226 ACTIYITY 2.2 OEYELOPMENT OF CELLULAR ANO MOLECULAR TECHNJQUES FOR THE TRANSFER OF 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6 2.2.7 2.2.8 2.2.9 2.2.10 2.2.11 2.2.12 2.2.13 2.2.14 GENES FOR BROADENING CROP GENETIC BASE ............... ................................................. ...... 229 Regeneration of cassava plants from friable embryogenic callus (FEC) by combining conventional so lid media and temporary immersion using RITA~ ......... ..................................... 230 Induction of Friable Embryogenic Callus (FEC) in commercial cassava cultivars ............. .. ....... 233 Oevelopment of a novel backcross methodology for producing fertile cornmon x tepary beans hybrids from otherwise incompatible genotypes .......................................................................... 235 Farmer's cassava seed production using in vitro low cost system .......................... ....................... 237 Estimation ofthe costs involved in a rapid multiplication scheme based on the use ofmicro- stakes ............................................................................................ ................................................ 239 Rapid propagation of planting material by temporary immersion bioreactors .............................. 241 Cryopreservation of cassava shoot tips using the encapsulation -dehyd.ration technique ............ 243 Costing all expenses about the different methods to conserve cassava germplasm ...................... 248 Preliminary studies on the cryopreservation of meristems and seeds of wild Manihot species .... 248 Temporary Immersion System (RITA) for Anther Culture of Rice ............................................. 252 Development of selection systems for the generation of transgenic rice according to current food biosafety requirements ......................................................... ............... .................................. 254 A fU"St initiative to support cassava seed production for the industry .............. .. ........................... 257 Development of methodologies for in vitro multiplication, plant regeneration, and genetic transformation ofnaranjilla (lulo) ............................................................................... .................. 259 Cryopreservation of Friable Embryogenic Callus (FEC) lines ..................................................... 262 111 2.2.15 In vitro propagation through micrografting of selected clones of soursop (Annona muricata L.): Optimization ofthe technique and field evaluation ofthe agronomic performance of propagated trees ............................................................................................................................ 264 2.2. 16 Developing cryopreservation alternatives for tropical fruits useful in National Programs: The case ofTree Tomato (Cyphomandra betacea) .......................... .................. .. ................................ 269 ACTIVITY 2.3 IOENTIFICATION OF PO!NTS OF GENETIC INTERVENT!ON ANO MECHANISMS OF PLANT- STRESS INTERACTION ............................................................................... ............................. 271 2.3.1 Characterizing genes ofthe carotene pathway in cassava ................ ........ ....... ..... ................. ........ 272 2.3 .2 Genetic variation in total carotenes and minerals of cassava genotypes ....................................... 274 2.3 .3 Evaluation of genetic diversity for total carotene content cassava le aves and roots ..................... 277 2.3.4 Enab1ing Genomics Tools for Understanding and Exploiting the Genetic Potential ofCassava Starch ......................................................... .. .......................................... .............. ....... .................. 281 2.3.5 Functional Genomics Tools ofPost-Harvest Physiologica1 Deterioration in Cassava .... .. ...... ..... 282 2.3.6 Identifying target points for the control of post-harvest physiological deterioration in cassava ... 284 OUTPUT 3. COLLABORATION WITH PUBLIC AND PRIVA TE PARTNERS ENHANCED ............. 288 ACTIVITY 3.1 NEW COLLABORATIVE ARRANGEMENTS ANO ORGANIZATION OF WORKSHOPS ANO TRAINING COURSES .............................. .......................... ....................... ....... ......................... 288 3.1.1 Highlights From The Cassava Biotechnology Network's Activities For 2001.. ........................... 289 3.1.2 Collaboration with Public and private sectors ......................... ..... .. .............................................. 290 3 .1.3 International Scientific Meetings .................. ................................. ............................................... 292 3.1.4 Workshops, training and Conferences .......................................................................................... 295 3 .1.5 Visiting Research ... ....................... ......................... .................................................... ............... .... 298 ACTIVITY 3.2 ASSEMBLING OATABASES, GENETIC STOCKS, MAPS PROBES ANO RELATEO !NFORMATION ... 299 3.2.1 Molecular genetics database constructed for a microsatellite parental survey of common bean germplasm ..................... ......................................... .................. ............. ...................................... .. 299 3.2.2 Flora Map: Release 1.01 with newclimate grids .......................................................................... 302 3.2.3 Genetic constructs, stocks and databases ........................... ........................................................... 304 ACTIVITY 3.3 PROJECT PROPOSALS ANO PUBUCATIONS ..................... .................. ....................................... 305 3.3.1 Projects approved or on going ...................................................................................................... 305 3.3.2 Projects submitted, in preparation and concept notes ................................................................... 306 3.3.3 Pub1ications ................................................................................................................................... 306 ACTIVITY 3.4 DONORS CONTRJBUT!NG TO PROJECT SB-2 IN THE PERlO O OCT, 2000 SEPT. 200 ! ............... 310 ACTIVITY 3.5 PROJECT SB-02 STAFF (2001) ..................................................................................... 312 3.5 .1 SB2 Investigators: Discipline, position and time fraction ............................................................. 312 IV PROJECT SB-2: ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH BIOTECHNOLOGY PROJECT OVERVIEW The Challeoge: The Project's approach to the challenges involved in increasing agricultura! productivity, with probably less water and soil, and agricultura! competitiveness in tropical developing countries, is based in the application of modem biotechnology to enhance our ability to develop improved strategies for the characterization and sustainable utilization of genetic diversity in crop improvement and conservation. Common bean, cassava and rice are vital to food security and human welfare and along with tropical forages, are grown in developing countries. Our research on the mandated crops expands to other crops of current or potential economic importance in Latín America Objective: To employ modem biotechnology to identify and use genetic diversity for broadening the genetic base and increasing the productivity of mandated and selected nonrnandated crops. Outputs: lmproved characterization of the gene tic diversity of wild and cultivated species and associated organisms. Genes and gene combinations used to broaden the genetic base. Collaboration with public-and private-sector partners enhanced. Milestooes: Cassava cryopreservation implemented. Screening with microarray technology initiated. Gene transfer used to broaden the genetic base and enhance germplasm of rice, cassava, and the forage grass Brachiaria. Marker-assisted selection implemented with cassava and beans. Marker-assisted selection implemented for rice, beans, cassava, and Brachiaria. ESTs generated for cassava starch and CBB. Efficient transformation system devolved for beans. Transgenic cassava tested for resistance to stemborer. Bioreactor technology implemented for cassava and rice. Collaboration with public and private partners strengthened. lntegration of genotype x environrnent GIS system with molecular characterization. High throughput screening of germplasm bank and breeding materials implemented, using microarray technology. Marker-assisted selection for ACMV and whitefly resistance initiated. Transgenic rice resistant to a spectrum of fungal diseases. Users: CIAT and NARS partners (public and private) involved in crop genetic improvement and agrobiodiversity conservation; AROs from DCs and LDCs, using CIA T technologies. Collaborators: lARCs (IPGRI through the Systemwide Genetic Resources Program, CIP, and liTA through root and tuber crop research); NARS (CORPOICA, ICA, EMBRAPA, INIAs); AROs of DCs and LDCs; biodiversity institutions (A. von Humboldt, INBIO, SINCHI, Smithsonian); and corporations and private organizations. CGIAR system linkages: Saving Biodiversity (30%); Enhancement & Breeding (60%); Training (1 0%). CIAT project linkages: Inputs 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: Genetic and molecular techniques for the gene bank, crop productivity, and Soils (microbial) projects. Identified genes and gene combinations for crop productivity and protection projects. Methods and techniques for propagation and conservation for gene bank and productivity projects. Interspecific hybrids and transgenic stocks for crop productivity and IPM projects. WORK BREAKDOWN STRUCTURE PROJECT SB-02: ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH BIOTECHNOLOGY PROJECT GOAL To contribute to the sustainable increase of productivity and quality of mandated and other priority crops, and the conservation of agrobiodiversity in tropical countries. PROJECT PURPOSE To ensure that characterized agrobiodiversity, improved crop genetic stocks, and modem molecular and cellular methods and tools, are used by CIA T and NARS scientists for impn;wing using and conserving crop genetic resources. 1 OUTPUT l. Genomes characterized: Genomes of wild and cultivated species of mandated and non-mandated crops, and associated organisms characterized. - Molecular characterization of genetic diversity - Identification and mapping of useful genes and gene combinations - Development of molecular- genetic techniques for assessing genetic diversity. -r OUTPUT 2. Genes modified: Genes and gene combinations utilized for broadening the genetic bases of mandated and non-mandated crops. - Transfer of novel genes and gene combinations by means of cellular and molecular gene transfer techniques. - Identification of points for genetic intervention in plant/stress interactions. - Development of cellular and molecular techniques for genome modification. 11 1 OUTPUT3. Collaboration with public and prívate sector partners enhanced. - Organization of conferences, networks, workshops and training courses. - Assembling of data bases, genetic stocks, maps and probes, and related information. - Publications, project proposal development and contribution to IPR and biosafety management. Are a: Log Frame- Work Plan for SB-2, 2002 Genetic Resources Research M . ., JoeToh NarratJve SllJDDlUY Measurable lodlc:aton Meaos ofVcrifw:atloo Goal To contribute to the sustainable increase of • CIA T scientists and partners using CIA T and NARS publications. productivity and quality of rnandated and other biocechnology infonnation and tools in crop Statistics on agriculture and biodiversity. priority crops, and the conservation of research. agrobiodiversity in tropical countries. • Genetic stocks available to key CIA T partners . Purpoee To ensure that characterized agrobiodiversity, • Inforrnation on diversity of wild and cultivated Publications, reports, project proposals. improved crop genetic stocks, and modem molecular species. and ccllular methods and tools are used by CIA T and . Mapped economic genes and gene complexes . NARS scientists for improving, using. and conserving • Improved genetic stocks, lines, populations . crop genetic resourccs. Output 1 Genomes characterizcd: Characterization of genomes • Molecular information on diversity of mandated Publications, reports, project proposals. of wild and cultivated species of mandated and and norunandated crops species, and pathogenic Gerrnplasm. nonmandated crops and of associated organisrns. and beneficial organisrns. • Bioinforrnatic techniques . Output 2 Genomes modified: genes and gene cornbinations • Transgenic lines of rice and advances in Publications, reports, project proposals. used to broaden the genetic base of mandated and cassava, beans, BrachiDrla, and other crops. Gerrnplasm. nonmandated crops. • Cloned genes and preparalion of gene constructs. • lnformation on new transfonnation and tissue culture tedmiques. Output 3 Collaboration with public- and private-sector partners • CIA T partners in LDCs using informalion and Publicalions. enhanced. genetic stocks. Training courses and workshops. • New partnerships with private sector . Project proposals. lll Assumptioos Pro-active participation of CIA T and NARS agricultura! scientists and biologists. Availability ofup-to-date genomics equipment and operational fWlding. IPR management to access genes and gene promoters. Biosafety regulations in place. Govemment and industry suppon national biotech initialives. · - Narrativt Summary Mcasurable lodicators Mea os of Verilication Importan! Assumptioo OUTPUT 1: Genomes characteriud Activity I. I.Molecular characterization of • Characterization of core collections Report, articles, databases of molecular • Availability of structure collections genetic diversity • ldentification of sources of resistan ce to fingerprinting • Material supplied by GRU disease • Genetic structure ofwild and cultivated beans and cassava available Phylogenic 1 trees based on ITS sequences. • Characterization of genetic diversity of endangered palms and soursop in Colombia Activity 1.2 ldentilication and rnapping, of • Marker assisted scheme established for Draft articles, Annual Rcport, pub! ications • Availability ofmapping populations and useful gene and genes combinations bean rice, and Brachiaria phcnotypic characterization • Linkage detected betwcen markers and importan! agronomical traits • QTL analysis for quality traits,diseasc rcsistance, and ag¡onomic performance in bean cassava and rice. Activity 1.3 Dcvelopment of molecular • Sean, Cassava and Brachiaria Sequences available • Access to facilities in advanced labs 1 techniques for assessing genetic diversity and microsatellites developed Report, draft articles mapping useful genes • New technologies- SAGE, cONA AFLP , and microarray implemented. • Resistance genes analogues identified, characteriz.cd and rnapped in bean and Brachiaria. • Mapping resistance genes in Brachiaria rice. OUTPUT 2. Genomes modified • Activity 2.1 Transfer of novel genes an gene • Express ion of insecticida! protein Transgenic plants in biosafety field • Biosafety regulation approved combinations by cellular/molecular techniques • lsolation of lignin biosynthetic genes Report, draft articles • Biosafety greenhouse space available from Brachiaria • Collaboration with NARS • Generation oftransgenic Brachiaria, cassava. rice, tomato, sugarcane. • Field test of transgenic rice with virus resistance • Backcross conversion from transgenic rice. Activity 2.2 Dcvelopment of cellular and • Rapid propagation rates of cassava Farmer reports, leve! of adoplion of technology • Access to farmers associaJion molecular techniques for the transfer of genes cullivars improved by bioreactors. • Access to RITA system for broadening crop genetic base • Low cost cassava in vitro propagation method transferred to farmers association. • Cost analysis ofpropagation and conscrvation of cassava germplasm by different methods. • Use of bioreactors for rice anther culture lV • Adaptation and use of selection system for genetic transformation non dependent on antibiotic resistance • Development of propagation, plant regeneration and transformation of naranjilla • Cryopreservation of cassava and trce tomato • In vi/ro propagation of soursop improved Activity 2.3 ldentification of points of genetic • Genomics tools uscd to understand and • intervention and mechanism of plant stress exploit diversity for cassava starch and post-harvest deterioration • Characterization of genetic diversity and key pathway genes for carotene and mineral contents in cassava OUTPUT J. CoUaboratlon tnhanttd • Activity 3.1 Organization ofNetworks, • The Cassava Biotechnology Network was Reports • Funding available Workshops, training courses in biotechnology re-established • Contribution to training courses • Organization of a legume genomics meeting between CG and US universities. • Organization ofthe CG planning workshop on biofortification. • Al least 70 people received training • Participation ofteam members to intemational regional conferences Activity 3.2 Data Base and Genetic Stocks • Database for bean microsatellites Number of register users, report, access to • Continued core support established databases, publications • New version of Flora Map distributed • Database for gene constructs, plasmid and vectors established Activity 3.3 Project proposals and Publications • Five new projects approved and 11 Number of projects approved Continued core support proposals submitted Activity 3.4 Oonors contributing • Twenty five donors contributed Number of current donors Continued core support Activity 3.5 Project SB-2 staff • 8.9 Senior Staffperson/time Total number of staff Continued core support • 41 Support Staff • 3 Administrative Support Staff • 18 Gra4uate Students • 14 Undergraduate Students ---- V PROJECT SB-2 HIGHLIGHTS 2001 Output 1 • Diversity of P. vulgaris, Phaseolus coccineus, P. polyanthus, P. acutifolius, P. parvifolius and P. tenuifo/ius with AFLP and microsatelies provides insights on gene pool structure and taxonomic relationships. • Genetic diversity of common bean for a range of important agronomic traits was assessed by microsatellite and incorporated into the Bean Gene Bank database. • Assessment of genetic diversity of cassava land races from South America, Central America, and Africa using microsatellites identified substantial variation for CMD resistance, yield, and other agronomic traits. • Microsatellite markers were identified to assess gene flow from transgenic rice into rice varieties and wild/weedy relatives. • No correlation between geographical origin of Xanthomonas strains and DNA polimorphism was found in various populations within and between ecological zones in Togo. • Molecular characterization of Flemingia macrophylla and Cratylia argentea was initiated to integrate with morphologic~l and agronomical data for conservation and management purposes. • Characterization of Colombia endangered palm species using microsatellite was initiated in collaboration with the von Humboldt Institute. • High iron trait was transferred from wild bean into cultivated background. • QTLs for iron and zinc were identified on chromosomes from the Andean and Mesoamerican bean populations used. • MAS for CMD hs been initiated with liT A after identifying at CIA T molecular markers linked to a qualitative and high leve! of resistan ce gene so urce. • Candidate genes for CMD has been identified. • The use of Annotation of SAGE tags differentially expressed in CMD resistant and susceptible genotypes has allowed the identification of genes involved in SAR response to disease in plants. • QTL from O rufipogon associated with improved agronomic traits were identified in inter- specific hybrids with cultivated rice. • The dissection and sequence analysis of a cluster of RGA associated with resistance to angular leaf spot was completed. • The isolation of full-length RGAs in beans was initiated. VI • The number of sequences available in the common bean Genebank public database was increased at minimum a 600% by adding over 3,000 microsatellite sequences. • Microsatellite markers developed for soybean and cowpeas were screened to adapt microsatellite available for other Phaseolus legume crops to bean. • Bean T y 1-copia group retrotransposon L TR sequen ces and 24 d ifferent sequen ces corresponding to RNAse-polypurine tract-long terminal repeat were isolated and characterized. • The implementation of novel microarray based technology Oiversity Arra y Technology (DarT) was initiated on bean and cassava. • New set of 85 microsatellites were located on the cassava genome map. The integration analysis of another new set of 157 SSR on the current cassava map is in progress. • Analysis using cONA-AFLP technique allowed the identification of putative molecular markers linked to CBB resistance in cassava. • The construction of a molecular Brachiaria map using grasses RFLP, RAPO, SCAR, AFLP and SSR was achieved. • Two major QTLs for spittlebug resistance in Brachiaria were identified. OUTPUT2 • Scaling up genetic transformation of cassava was accomplished by broadening from one to four the cultivars used. • Transgenic cassava lines containing Crylab insect resistance genes were established in the biosafety greenhouse. • Two fie1d trials with transgenic rice resistance to RHBV were evaluated. Transgenic rice with the highest level of resistan ce to RHBV were identified. • A selection system for generating transgenic rice based on the use of mannose isomerase selection gene was developed. The system will aid the generation of transgenic plants non- dependant on antibiotic resistance selection marker genes. • A cONA library to clone key genes involved in linging biosynthesis in Brachiaria was accomplished. • Lines of Friable Emrbyogenic Callus (FEC) from various cassava commercial cultivars were established. • Novel backcross methodology for producing fertile common x tepary beans hybrids from otherwise incompatible genotypes was developed. Vll • Excellent agronomy and phenology perfonnance in the field of 3-year-old soursop clones propagated in vitro, validated the in vitro micrografting as a altemative for multiplying pathogen free materials. • A large scale test ofthe cryo-preservation cassava protocol was implemented. About 43% of the core collection was tested, and 82% ofthe accessions showed > 30% recovery rates after freezing. • The cryo-preservation protocol has been extended to cassava wild relatives including Manihot esculenta subsp flabellifo/ia, subsp. Peruviana and subsp. Carthaginensis. • FEC from two cassava cultivars were recovered after freezing allowing long tenn storage of suitable material for genetic transfonnation. • Plants from the tropical fruit tree tomato were recovered after freezing. • Propagation methods for cassava commercial clones and doubled haploid generation from rice anther culture were optimized using RITA bioreactors. • In vitro cassava plants propagated by small farmers are under field testing and new clones selected by the farmers were included in a new cycle of propagation. • A methodology for the reproducible plant regeneration ofnaranjilla fruit (i.e. lulo) was developed. • Characterization of key genes involved in carotene biosynthesis pathway was initiated. • Genotypes of cassava core collection with higher levels ofFe, Zn, and pro-vitamin. A were identified for improving cassava quality through breeding. • The genetic variation of carotene content in lea ves and roots of 682 cassava accesions was detennined. Carotenes concentrate much more on leaves suggesting a higher nutritive value of cassava leaves. • Two classess of secondary metabolites, hydroxyconmarins and flavan 3-ols, were -identified as antioxidants and antimicrobials for the control of post-harvest deterioration in cassava. Output 3 • CIA T obtained approval of a project from BMZ, to integrate approach for genetic improvement of aluminum resistance of crops on low-fertility acid soils. • During period Oct 2000-2001 more than 70 people (researchers, joumalists, visitors and others) received training with SB-2 Project Staff. • In collaboration with two scientists from the Biotechnology Centers ofRutgers University, a training course was held at CIA T to upgrade knowledge of SB-2 staff on molecular approaches for disease resistance, and modulating gene expression in transgenic plants. Vlll • A course on the use of microarray and Diversity Array Technology (DarT) was conducted at CIA T by CAMBIA (Australia) for 30 assistants from SB-2. • A second workshop on Biotechnology and GMOs biosafety was given by CIA T to Colombian journalist. • An updated version ofFloramap was released in 2001. Sorne 200 registered users from several countries obtained a copy. • Data base for bean microsatellites was established • The cassava Biotechnology Network was re-established for Latin America with funding from DGIS and IDRC. • The first Planning Workshop to develop a biofortification proposal at the CG level was organized and conducted at CIA T. • In the period Oct 2000-Sept 200 l, SB-2 members published 40 Scientific Papers in refereed journals and books, abstracts and posters in proceedings. • In the same period, 5 new projects were approved and 11 proposal were submitted. • A total of24 donors contributed funding projects in SB-2 ix Otttput l. Genomes ofwild and cultivated species of mandated and non mandated crops, and associated organisms characterized Activity 1.1 Characterization of genetic diversity Main Achievements • A core collection of Phaseolus coccineus and Phaseolus polyanthus was evaluated with AFLP markers, demonstrating that very little gene pool structure exists in these two species, although Mexican and Guatemalan accessions of P. coccineus separate slightly, and an ecotype of P. polyanthus exists in South America. • Diversity in the CIA T collection of tepary beans (Phaseolus acutifolius) was analyzed with AFLP and microsatellites to distinguish taxonomic relationships with P. parvifolius as well as within the species between P. a. var. acutifolius and P. a. var. tenuifolius. • Genetic diversity of microsatellite alleles was determined for two parental surveys of comrnon bean that provide the basis for" mapping and genetic tagging for a range of important traits; including biotic and abiotic stress resistance 1 tolerance, micronutrient accumulation. This information was incorporated into a new molecular genetics database constructed for microsatellite parental surveys and the Bean genes database. • The study of cassava land races using microsatellites was extended to assessing genetic diversity and differentiation of cassava land races from 5 countries in South America, 2 in Central America, and 2 in Africa, and to African cassava genotypes resistant to the Cassava Mosaic Disease (CMD). The analysis showed a substantial amount of genetic diversity in CMD resistance germplasm appropriate for the genetic improvement of CMD resistance as well as of other traits, particuJarly yield. • A set of microsatellite markers were identified detecting polymorphism between transgenic rice, other rice varieties, wild Oryza species and red rice. This tool is being used to monitor gene flow from transgenic rice into rice varieties and into wild/weedy relatives. With this project we initiated studies on environmental biosafety jointly with the generation of transgenic plants to set guidelines for their safe use in agriculture. • The assessment of the origin and the genetic diversity of the Xam was extended to cassava populations within and between ecological zones in Togo. The cluster analysis revealed the existence of 7 groups at 70% similarity. No correlation between the geographical origin of the strains and DNA polymorphism was observed. • The Genetic diversity of the multipurpose shrub legumes Flemingia macrophylla and Cratylia argentea was initiated using AFLPs. The molecular data will be integrated with morphological and agronomical data to improve the use and management, including conservation, of the collections. • As a collaboration with the Humboldt Institute, the analysis of the genetic structure of Colombian endangered palm species was carried using microsatellite isolated at CIA T. - 1.1.1 Evaluation of the P. coccineus and P. polyanthus core collections with molecular markers S. Beebe1, G. Gallego1, E. Tovar1, C. Cajiao3, M.C. Duque 1•2, J. Tohme1 1SB-2 Project; 2IP-4 Project, 3IP-1 Project Introduction In a project financed by the Belgian government to explore the potential of P. coccineus (PC) and P. polyanthus (PP), a core collection was formed as reported in past years. No previous study ofthe genetic structure of these two species had been performed, to determine if they display gene pools comparable to other species of the genus such as P. vulgaris. Therefore the core was studied with molecular markers to determine if gene pools exist within these species. Materials and metbods A total of 178 accessions of P. coccineus and P. polyanthus were studied, including the PC-PP core collection and others that had been designated as promising for specific traits, such as resistance to BGYMV or bean fly. The PC accessions included 66 cultivated types, 40 wild accessions, 1 intermediate and 1 hybrid (possibly with PP). Geographically, accessions originated in Mexico (57), Guatemala (30), Colombia (5), Yugoslavia (3), Rumania, Rwanda, and Turkey (2 each), Costa Rica, Germany, United Kingdom, Honduras, Puerto Rico, Holland and Portugal ( 1 each). The PP accessions included 59 cultivated, 7 wild types, and 3 hybrids. These originated in Mexico (26), Guatemala (29), Colombia ( 1 O), Costa Rica (2), and Venezuela (1 ). An accession of P. vulgaris (cv. 'ICA Pijao') and of P. costarricensis were included for comparison. The technique of AFLP was applied to the PC-PP core collection, using two primer combinations, selected from among 38 combinations of primers tested from a kit obtained from GIBCO BRL ™. Bands were read as present or absent and were analyzed by Multiple Correspondence Analysis (MCA) and by Unweighted Pair Group Method with Arithmetic Averages (UPGMA). Initially the analysis was performed with only accessions produced under controlled pollination (type 1 and type 2 seed), and subsequently accessions with open-pollinated seed were added to the analysis. Results and discussion The four species separated widely from each other in the analyses by UPGMA and by MCA, confirming the independent status of P. po/yanthus as a species apart from P. coccineus (Figure1). 2 P. vu/garis 0.4 P. costarlcensis P. coccineus P. polyanthus 0.5 0.7 Nei-li 0.8 Figure l. Dendrogram of similarity among accessions of P. coccineus y P. Polyanthus with controlled pollination , iocluding P. vulgaris and P. costarricensis, with two primer combinations. 3 1.0 When wild and cultivated PC were analyzed together by MCA, the first dimension separated wild from cultivated (Figure 2). There was no unambiguous overlap between wild and cultivated groups to suggest that a certain wild population might have participated in a unique domestication event. Rather, only two wild accessions from Guatemala occupied an area on the margin of the cultivated. The second dimension showed a geographical gradient, whereby the wild PC from Cultlvat.d • Maxk:o ... Cultlvat.d o A Guat.mala + Mexlc:o ·- Wlld ·Gua t. mala (X Wlld ·M•xlc:o ~~~..,-· o Wlld • Guawmala ! ., 1 ~ Wlld- Mexlc:o '¡ 1 ~ 1 ¡ Wlld·M•xlc:o 1 ~· ¡. i 1 '11 1 1 -:.. ':: D Wlld • Mexlc:o 1' : !J ( ," Wlld • Mexlc:o Ej Wlld- Mexlc:o Figure 2. Three dimensional graph derived from Multiple Correspondence Analysis (MCA) of controlled pollination accessions of P. coccineus, and excluding P. vulgaris and P. costaricensis. The first dimension shows the separation of accessions according to biological state: wilds (to the right) and cultivated (to the left). Guatemala and Mexico separated from each other (Figure 3). However, the wild accessions in general did not result in discrete groups, and the Mexican accessions in particular occupied a very dispersed space. However, in the second dimension, the position of the cultivated was even more ambiguous. lt fell in between the wilds from Guatemala and Mexico, being similar to the Guatemalan wilds on the first dimension and similar to the Mexican wilds on the second dimension. These results suggest that the cultivated PC might not have a direct relationship with any single wild population. lt is possible that the allogarnous nature of the species has resulted in cultivated PC that represents introgression from multiple wild populations, hence its intermediate position between wild populations from Mexico and Guatemala. 4 In the case of P. polyanthus, the first dimension of the MCA (Figure 3) again separated wild from cultivated accessions. Wild PP was surprisingly diverse, considering that all seven accessions of wild PP come from a restricted geographical area of Guatemala. Neither did these present any apparent structure. Among cultivated accessions, it was noted that those that graphed most distantly from the wild beans were those from South America (Figure 4). When these were viewed on the second dimension (Figure S), their differentiation from all other accessions was even clearer. Last year we noted that these occupy an ecological niche that is very different from other PP. Molecular analysis confinns that there is a subtle difference of these accessions compared to others, although the genetic distance compared to other cultivated PP is in fact quite small. When accessions with open-pollinated seed were included in the analysis, no significant changes occurred in the structure of the dendrogram, although the absence of a clear gene pool structure made it more difficult to analyze this effect and to estímate the effect of open-pollination on the genetic structure of the species. However, the allele frequencies of contro\led vs, open-po\linated accessions were compared, resulting in a correlation of r=0.965 for PC and r=0.975 for PP. Therefore, the open-pollination technique has not changed allele frequencies and may not have altered the genetic structure as much as we might have feared. Heterogeneity of different populations was compared (Table 1). In neither species were the wild accessions more diverse than the cultivated accessions. Therefore there does not appear to a serious founder effect in these two species. Neither was one country or region inherently more diverse than any other. The foregoing results probably reflect the allogamous nature of the two species. Outcrossing has probably erased any possible tendency for the fonnation of gene pools by creating a relatively homogenous intennating population. Similarly, the allogamous nature has overcome any possible founder effect. Given the ambiguity of results with nuclear DNA in our attempts to relate wild populations of PC to cultivated accessions, an additional study was undertaken with chloroplast DNA. Chloroplast- specific primers were used to amplify products from 30 wild and 10 cultivated accessions of PC. However, results continued to be ambiguous. Wilds from both Guatemala and Mexico grouped with cultivated PC. Surprisingly, PC presented very diverse chloroplast DNA, with sorne wild PC being identical with P. costaricensis, others grouping close to P. vulgaris, and others fonning groups that were quite divergent among themselves. Neither did parsimony analysis reveal clear tendencies. In contrast to PC, a similar attempt with PP failed to find any polymorphism at all in chloroplast DNA. 5 :) Guatemala - WUdo Guatemala - Wlldo Ñ ·' 1' ·-;~ Guatemala + Mexlco Cultlvated r ~. Q Moxleo - WIIdo ,, .V : ; o Moxleo -WIIdo ¡ ¡1 Moxleo - WIIdo Mexleo - Cultlvated :L EJ Molleo- Wlldo Molleo - Wlldo Moxleo - Wlldo Figure 3. Multiple Correspondence Analysis (MCA) of controlled pollination accessions of P. coccineus (excluding P. vulgaris and P. costarricensis), showing in the second dimension, separation of accessions by geograpbic origin. • 1 J ¡ 1 ¡ ¡ Wlldo - Guatemala Q Wildo - Guatemala Wlldo - Guatemala @ Wlldo - Guatemala Q CultiYated - Guatemala E'l CultiYated - South Amortca • .• .- Cultlvated - South Amortca Cultlvated - Guatemala O Cultlvated - Guatemala e·, Cultlvoted • South Amerlca -~' CultiYated - Soulh A menea ~ CultiYated -Guatemala Figure 4. Multiple Correspondence Analysis (M CA) of controlled pollination accessions of P. polyanthus (excluding P. vulgaris and P. costaricensis}, showing in the first dimension the separation ofwilds (to the left) and cultivated (to the right). 6 ~ 11esoamertca • Culllvat.d 1 Q llesoamertca • Wlldo :> J) MuoiiMifca ' 1 :> Wllda+Culllvated : O llesoamertca • Culllvat.d ' M-· WIIda i ¡ ' Muo......tca • Culllvat.d 1 1 1 Muoam-.tca • Wllda 1 ¡, ·• . South Amerlca • Culllvated South Amerlca • Cultlvated (_, South Amerlca • Cultlv- c:::l South Amerlca • Cultlv- Figure 5. Multiple Correspondence Analysis (MCA) of controlled pollination accessions of P. polyanthus (excluding P. vulgaris and P. costaricensis), sbowing in tbe second dimension tbe separation of accessions by geograpbic origin. Table l. Heterogenei:i values of P. coccineus and P. polyanthus calculated within groups according biological sta e and geographical zone. Heterogeneity Groups H Hs Hst Gst Value t Significance P. cocciMus Wild 0,162 Cultivated 0,139 0,152 0,01 0,07 0,725 0,469 Wi1ds GTM + SMEX 0,124 WildsMEX 0,166 0,147 0,015 0,093 0,99S 0,322 Cultivated GTM + SMEX 0,113 Cu1tivated MEX 0,132 0,131 0,008 0,056 0,322 0,749 Total species 0,164 P. polyanthus - -·----·---Wild 0,031 Cultiva~cd 0,054 0,048 0,006 0,122 0,897 o,3n Mesoamerica O,OS South America 0,051 0,051 0,004 0,077 0,041 0,967 Total species O,OSS H: total heterogeneity, Hs: heterogeneity with groups, Hst: heterogeneity between groups, Gst: proportion of heterogeneity due to differentiation between groups. GTM ,. Guatemala, SMEX • south of Mexico, MEX • rest ofMexico. 7 to Cooclusions Neither P. coccineus nor P. polyanthus showed any clear gene pool structure, and on the contrary, the genetic variability within any groups formed was greater that that between groups. Therefore, we conclude that both species have only one gene pool. Neither did either species show any evidence of a founder effect. On the contrary, present-day cultivated P. coccineus appears to have resulted from ample introgression from different wild populations in Guatemala and Mexico. The allogamous nature of the species has probably contributed to all of these results. However, P. coccineus presents remarkable variability in chloroplast DNA. In the case of P. polyanthus, an ecotype exists in South America which, although it can be distinguished with AFLP markers, is in fact genetically very similar to other cultivated polyanthus. 1.1.2 Genetic diversity of the CIAT tepary bean (Phaseo/us acutifolius) collection measured with AFLP and microsatellite markers M.W. Blair, C. Muñoz, W. Pantoja, D. Debouck SB-2 Project Introduction Tepary bean (Phaseolus acutifolius A.Gray) is in the tertiary gene pool of common beans (P. vulgaris L.) and as such represents a potential but difficult to use genetic resource for the improvement of common beans. The two species have been crossed, despite high embryo abortion, using congruity or recurrent backcrossing and these interspecific hybridizations have been used to incorporate common bacteria) blight resistance into common bean. Notwithstanding their utility for the improvement of other species, tepary beans are a useful and interesting crop in their own right, especially for dryland agricultura) systems. They are known to have high drought and salinity tolerance, good nutritional quality and a tradition of cultivation in Mexico, South westem United States and Central A.merica, that goes back 5000 years. What is needed is to create improved varieties as none exist. For this it is critica) to have a baseline data on the diversity that exists within tepary beans. Previous authors (Scinkel and Gepts, 1988, 1989; Garvin and Weeden, 1994) have suggested that tepary beans seem to be less diverse than common bean or lima beans. They are thought to have had a single center of origin and to have been distributed from a few original sites across the present distribution. The objective of this research was to study the pattems of diversity within the species and its placement relative to other Phaseolus species that were used as an outgroup, using two types of molecular markers : 1) AFLPs which tend to be evolutionarily conserved markers and serve to reference different species relative to each other and 2) microsatellites which are by nature hypervariable loci that can distinguish between varieties. Additional objectives of the study were to determine if P. acutifolius and P. parvifolius merit being separate species and if molecular markers can distinguish between the botanical varieties var. acutifolius and var. teniufo/ius within the species P. acutifo/ius. AFLP markers have been applied before to study wild species of Phaseolus (Tohme et al., 1996) and lima bean, P. lunatus, accessions and their close relatives (Caicedo et al., 1999). We have developed new microsatellites that are beginning to be used in characterization studies. This is the first application of both marker systems to study tepary bean diversity. 8 Materials and Methods Genotypes and DNA extrae/ion: A total of 127 genotypes from the Genetic Resources Unit of CIA T were analyzed in the experiments. The outgroup consisted in 1 O genotype from the Phaseolus genus including 4 P. vulgaris (common bean); 4 P. lunatus (lima bean); 1 P. coccineus (scarlet runner bean); and 1 P. glabellus genotype. For both common and lima beans a wild anda cultivated representative from both the Andean and Mesoamerican genepools was included in the analysis. For the other species only a single representative was analyzed. A total of 117 tepary beans and their close relatives were analyzed, consisting in 49 cultivated P. acutifolius var. acutifolius; 44 wild P. acutifolius var. acutifolius; 12 P. acutifolius var. tenuifolius; and 12 P. parvifo/ius accessions. The genotypes were grown in the greenhouse and total genomic DNA was extracted from 2 g of fresh Jeaf tissue with a large preparation method (CIA T protocols ). AFLP analysis: Amplicon-template preparation, pre-amplification, and selective amplification were as described for the protocol of the Gibe o BRL AFLP analysis system 1 kit for small genomes. In a previous study, we determined which were the best primer combinations based on the EcoRI (E) - Msei (M) adapters and primers with 3 selective nucleotides each. One combination, based on E- AAG and M-CTT primers was analyzed for this study PCR products were run on 4% silver- stained polyacrylamide gels for 1, 1.5 and 2 hours to resol ve as many fragments as possible. Bands were sized by comparison to a SObp ladder molecular weight size standard. All the polymorphic AFLP bands between 1 00 and 400 bp were scored for presence or absence among the lines and used to calculate the similarity matrix. Larger or smaller bands were not considered. Microsatellite analysis: A total of 1 O microsatellites were amplified for the study; of which six were cONA based (BMdl, BMd7, BMdlO, BMdlS, BMyl and BMy6) and four were genomic (BMdll, BMdl2, BMd36 and BM114) (Table 1). PCR product were run on 4% silver-stained polyacrylamide gels and the alleles sized by comparison with 1 O and 25 bp molecular weight Jadders. Alleles were considered separate taxonomic units for the purposes of calculating shared bands and similarity. Data analysis: Genetic similarities between genotypes was determined with the Dice coefficient using the software packages SAS (SAS Institute, 1989) and NTSYS 2.02 (Rohlf, 1993). Results AFLP analysis: The AFLP combination used in this study had a good polymorphism rate, clear amplification profile and well-distributed range in PCR product sizes. The AFLP combination produced a total of 167 bands. Of these 99.5% of the bands were polymorphic across all species although there was substantial monomorphism within the cultivated P. acutifolius. Both monomorphic and polymorphic band were used to determine the genetic similarity between genotypes. Figure 1 a shows the dendrogram created for the AFLP bands. The structure of the dendrogram agrees with known taxonomic relationships for the six species represented in the study. P. /unatus was the most distant group, followed by P. glabellus and P. coccineus. P. vulgaris was the closest to the P. acutifolius- parvifolius clade. The level of similarity was around 35% between the five groups. Within both P. vulgaris and P. lunatus the distinction between Andean and Mesomerican genepools was clear. The level of similarity between genepools was higher in P. vulgaris (68%) than in P. /unatus (62%). Within the P. acutifolius- parvifolius clade, all the accessions shared up to 54% similarity. Five groups could be distinguished within the clade: 1) cultivated P. acutifolius from Central and North 9 America 2) cultivated P. acutifo/ius from North America (mainly Sonora and Sinaloa), 3) wild P. acutifo/ius var. acutifolius 4) wild P. acutifolius var acutifolius and tenuifolius; and 45) P. parvifo/ius. These five groups could be organized hierarchically into two supergroups, consisting of groups 1, 2 and 3 together and groups 4 and 5 together. The frrst supergrouping contained all the cultivated P. acutifolius, while the second supergrouping contained all the P. acutifolius var. tenuifo/ius and P. parvifolius accessions. The wild P. acutifolius accessions were distributed among the two supergroupings, with sorne more allied to the cultivated accesssions of the same species and others allied to the P. parvifolius group. Within the first supergroup, the two cultivated groups ( 1 and 2) were related at 80% simi1arity and these were related to the wild accessions (group 3) within that supergroup at 68% similarity. Within the second supergroup, the P. parvifolius and P. acutifolius (both var. acutifolius and tenuifolius) were related at 64% similarity. The supergroups were distinguishable at 54% similarity. A multiple correspondance analysis confirrned the groupings observed in the dendrogram, where five clusters could be found in the P. acutifolius- prvifolius clade, corresponding to the groupings described above . Microsatellite Analysis: The ten microsatellites detected a total of 75 alleles which were scored as bands (present and absent) to determine genetic similarity between genotypes. The average number per locus of alleles produced across the range of genotypes was as high for the cDNA (7.8 alleles) as for the genomic (7 alleles) microsatellites. Figure 1 b shows the dendrogram created for the microsatellites. Genetic similarities were much lower on average than with AFLP data. The structure of the dendrogram agrees with that of the AFLP results, however their were severa! important differences. In agreement, the microsatellite inforrnation predicted that P. lunatus, followed by P. vulgaris were the most distant groups from P. acutifolius, with low similarities of 9 and 15% similarity, respectively. Within both P. vulgaris and P. lunatus the distinction between Andean and Mesomerican genepools was clear as described above, and these group shared only 55% similarity. In contrast to the AFLP data, the P. parvifolius and P. acutifolius cluster were separate and shared only 32% similarity. Two P. acutifolius var. tenuifolius accessions were found within the P. parvifolius cluster, while the others were found mixed with the wi1d accessions of P. acutifolius var. acutifolius. Surprisingly, P. coccineus was found between the P. parvifolius and the P. acutifolius clusters. All the cultivated P. acutifolius formed a group with the highest genetic similarity of around 85%. One group of cultivars ha ve similarity of 100% and are all accessions from Sinaloa. Despite the differences in scale and detail, correlation between the matrices generated for AFLP and microsatellite datasets was high (r=0.703) and significant as indicated by the approximate Mantel test (t= 12.15, P=.O 198) · Discussion and Future Plans Microsatellites detected much greater differences than AFLPs, probably because they were less conserved than AFLPs. The use of two marker systems to samp1e different part of the geno me that evolve at different rates hopefully gave us a more accurate picture of the relationships within and between species . The high similarity among all the cultivated tepary beans, even with the microsatellites, seems to indicate that the crop may have arisen from a single domestication event that led to a genetic bottleneck which limits diversity within the cultivars. From this study, there is very little evidence for introgression from wild relatives into the cultivated genepool after the initial domestication event. Tepary beans are known to have a very low crossing rate that limits the creation of new diversity within the crop. The lack of diversity within the cultivated tepary bean is a serious limitation for improvement of the crop. The lack of diversity within the cultivated tepary bean belies sorne ofthe variability found for disease and insect resistance within the species. These are also fast evolving characteristics so could be expected to have been generated by mutation even without a lot of initial diversity or inter-crossing. However, that lack of diversity in other 10 characteristics such as plant morphology, adaptation range has serious implications for improving the species agronomically and using the species in inter-specific hybridization. The relationships within the P. acutifolius - parvifolius clade has been controversia!. The AFLP data presented here suggest that the P. acutifolius and P. parvifolius probably do not deserve to be different species, but could qualify as possible subspecies or varieties within the species. The microsatellite data meanwhile show that wild P. acutifolius accessions and the P. parvifolius accessions are the extremes of a continuum, with all of the P. a. var. tenuifolius accessions as intermediates between these two clusters. The high amounts of diversity found in the wild P. acutifolius and P. parvifolius accessions are an interesting resource for breeding tepary bean cultivars. Table l. Outgroup used in the study of Tepary bean diversity Identification Species status region country Gl9889 P. vulgaris Wild Jujuy ARG G23508 P. vulgaris Wild Jalisco MEX ICA PIJAO P. vulgaris Cultivar breeding line COL CALIMA P. vulgaris Cultivar breeding line COL G24390 P. vulgaris Wild na COL Gl9892 P. vulgaris Wild na ARG G24404 P. vulgaris Wild na COL G25704 P. lunatus Wild Jalisco MEX G25914 P: lunatus Wild Cajamarca PER G25832 P. lunatus Cultivar Cajamarca PER G25577 P. lunatus Cultivar Sn Martín PER G40594 P. glabellus Wild SI potosi MEX G35583 P. coccineus Cultivar na GTA Tab1e 2. Microsatellites used in the study ofTepary bean diversity. Type SSRs No. alleles CONA BMd1 11 BMd7 6 BMd10 7 BMd15 5 BMyl 8 BMy6 10 Genornic BMd11 7 BMd12 5 BMd36 9 BM114 7 Total no. ofalleles 75 11 r O. Wild P.acuifolius wilhin cultivated P.acutifolius cluster or cultivated P.acutifolius within wild P.acutifollus cluster. P.a. var. Jenutjo/ius wilhin wild P.acutifolius cluster or P.paJ'\Iifolius cluster. P. glabel/us ri 1 1 1 f J 0.0 0.3 + Wild P.acutifoliu.s within cultivated P.acutifolius cluster r 1 1 1 O.S DICE 1 ~ ~~ ~ ~ i::E ~· ~ r- .-- ~ . ,...-- . . *= ••• rfy~ "' 1 0.8 • - P.a. var. tenuifoliu.s within wild P.acutifoliu.s cluster. .. " - 1.0 .,__ ·-- ~-- I'.«Wifo/iou cvhivn 1'.«011/olius wdd acc. Figure l . Dendrograms showing the associations among 114 accessions of cultivated and wild tepary beans (Phaseolus acutifolius and P. parvifolius) and 10 other Phaseolus species, based on UPGMA clustering using the Dice genetic similarity coefficient for: AFLP banding pattern for the combination E-AAG, M-CTT Microsatellite alleles identified for 10 loci (6 cDNA- derived and 4 genomic) References Caicedo AL, Gaitán E, Duque MC, Toro Ch. O, Debouck DG, Tohrne J (1999) Crop Science 39:1497-1507 Garvin DF, Weeden NF ( 1994) Crop Science 34: 1390-1395 Scinke1 C, Gepts P (1988) Plant Breeding 101 :292-301 Scinkel C, Gepts P ( 1989) Plant Breeding 102: 182-195 Tohme J, Gonzalez DO, Beebe S, Duque MC (1996) Crop Science 36: 1375-84 12 1.1.3 Genetic diversity of microsatellites in common bean M. Blair1, M.C. Giraldo\ H.F. Buendía2, F. Pedraza1, E. Gaitán1, J. Tohme1 1SB-2 Project; 2IP-l Project Introduction Microsatellites markers are based on short segments of ONA in which a specific simple sequence motif of 1-6 bases is repeated in tandem, multiple times. Oue to the innate variability at microsatellite loci, these markers have been ideal for characterizing genetic diversity in crop species at the inter-specific, inter-subspecific, inter-varietal and even intra-varietal levels. Microsatellites have been found to vary in the polymorphism they detect depending on the length and sequence of the repeat motif they contain and their location along the chromosomes, specifically whether they reside in gene-coding or non-coding segments of the genome. The objective of this study was to evaluate all new Phaseolus microsatellite markers developed at CIA T and elsewhere for their allelic variability on two panels of 18 common bean genotypes representing diverse germplasm, both cultivated and wild, Mesoamerican and Andean, which have been used as parents in the bean breeding program. Materials and Methods The genotypes consisted in 30 common bean genotypes arranged in two panels; The previous panel of 18 was described in last year's annual report (SB-02 report 2000). This year a new panel was instituted with 14 individuals, including two Andeans, nine mesoamericans and one tepary bean (Table 1) which are the parents of six mapping populations being studied at CIAT for the inheritance of disease resistance (common bacteria! blight (CBB), bean golden mosaic virus (BGMV), angular leaf spot and anthracnose), insect resistance (Apion) and abiotic stress tolerance (low phosphorous adaptation, drought tolerance and adventitious rooting). Bulked segregant analysis (BSA) was carried out simultaneously with the parental survey for the disease and insect resistance mapping populations for CBB, BGMV and Apion. The parents of an additional mapping population (BAT93 x Jalo EEP558) were included because this population has been the basis for creating an integrated genetic map for the bean genome (Freyre et al., 1999). The populations included three intra-genepool Mesoamerican x Mesoamerican crosses and two inter- genepool Mesoamerican x Andean populations (Table 2). The genotypes were evaluated with a total of 131 microsatellite markers ( of which 65 were derived from genomic libraries and 66 were derived from cONA or gene sequences). The markers were amplified at different annealing temperatures according to the estimated melting temperatures of the primers. The PCR products were resolved on silver-stained polyacrylamide gels and microsatellite alleles were sized by comparison to the 10 and 25 bp molecular weight standards (Promega). Results and Discussion Genomic microsatellites detected more polymorphism than cONA microsatellites in the intra- genepool crosses but were about equally effective in uncovering polymorphism in the inter- genepoo1 and interspecific crosses. The rate of polymorphism was much higher (77.9%) in the interspecific crosses than in the intra-specific crosses (37.5%). The average po1ymorphism rate between the parents of the inter-genepool crosses (44.3%) was higher than that of the intra- genepool crosses (33.1%). Among the Mesoamerican x Mesoamerican crosses, the intra-racial cross Jll7 (race Jalisco) x Jamapa (race Mesoamerica) was more polymorphic (43.5%) than the other crosses OOR476 x SELl309 and VAX6 x MARI which were within race Mesoamerica. 13 Among the inter-genepool crosses BAT93 x Jalo EEP558 was more polymorphic than 02333 x G 19839. The genomic microsatellites were more polymorphic markers than the cONA derived microsatellites. Overall the average polymorphism rate for the genomic microsatellites was 51 .5 % versus 37.2% for the cONA microsatellites. Significantly fewer average alleles per locus were found for microsatellites from genes (3.3 alleles) than for microsatellites from non-coding sequences ( 4.5 alleles). The gene-derived microsatellites frequently were bi- or tri-allelic and distinguished the difference between Andean and Mesoamerican genepools and the difference between Phaseolus vulgaris and P. acutifolius. Meanwhile the genomic microsatellites detected more alleles and were thus able to resolve sorne within-genepool variation. The polymorphism information content (PIC) of the gene-derived microsatellites was lower (0.402) than for the genomic microsatellites (0.553). The PIC values were positively correlated with the number of alleles produced at the locus. Null alleles were uncommon in both types of microsatellites. The allele size range was generally a good predictor of the number of alleles present for a locus. The allele range was 67% wider for the genomic microsatellites (28.8 bp) compared to the gene-derived microsatellites (17.3 bp). However there were severa! microsatellites with large size ranges but few alleles. The differences in allelic variability observed at specific bean microsatellite loci are probably due to the differences in the mutation rate inherent for each locus. Microsatellites mutate when they add or subtract a small number of perfect repeats or undergo changes in the flanking regions of the SSR. These changes can occur due to polymerase slippage, unequal crossing-over and/or insertion- deletion events. Although microsatellites are believed to have sorne of the highest mutation rates observed at any type of molecúlar loci, sorne microsatellites will evidently be more stable than others. In this study as in others before, microsatellite variability seems to be influenced by the structure, motif, SSR length and genomic context ofthe locus. The more polymorphic genomic microsatellites may well become the mainstay of mapping studies since they will be useful even in narrow intra-genepool crosses. They will also be very useful for to analyzing recent changes in population structure and selection history in closely-related germplasm from a given area or from a specific commercial class. Meanwhile the more conserved and stable cDNA-derived microsatellites may find their greatest utility in mapping in wide inter- genepool or inter-specific crosses and in the phylogenetic analysis ofthe genus Phaseolus. Conclusions and future plans We planto continue testing all new Phaseolus microsatellites on the existing panel and ifthe need arises, will create another panel of varieties to survey for polymorphisms in the parents of additional populations. In the future it will be very useful to genotype many ofthe common parents and genetic sources used at CIA T, as this will allow us to implement whole-genome marker assisted selection that is specific to the genetic crosses made in our bean breeding program. 14 Table l. Mapping parent genotypes used for assessment of genetic diversity of common bean microsatellites Varie!X Geneeool Purpose Origin 1 DOR476 Mesoamerican Disease resistance (BGMV) CIAT line 2 SELI309 Mesoamerican Disease rcsistance (CBB) CIAT line 3 BAT93 Mcsoamerican Integrated Map CIAT line 4 Jalo EEP558 Andcan Integrated Map Brazil 5 ICA Pijao Mesoamerican Cultivar Colombia 6 G40001 Tepary bcan Abiotic stess CIA T accession 7 VAX6 Mcsoamerican Diseasc resistance (CBB) CIAT line 8 MARI Mesoamerican Disease resistance (ALS) CIAT line 9 Jll 7 Mesoamerican Insect resistance Mexico 10 Jamapa Mcsoamerican Cultivar Mexico 11 G2333 Mcsoamerican Abiotic stcss CIA T accession 12 Gl9839 Andcan Abiotic stcss CIA T acccssion Table 2. Polymorpbism rate among 11 parent combinations for 70 microsatellite loci (36 cONA and 34 genomic). by marker name Population BM % BMy % BMd % PV % CLONES % All All 9 All 17 gen cONA genomi cONA genomic o mi e 14cONA e 35 cONA DOR 476 X SEL 20 51.3 3 33 .3 6 13.6 12.5 9 29.0 1309 BAT 93 x JALO 31 79.5 8 88.0 17 38.6 3 37.5 10 32.3 ICA PIJAO X G 33 84.6 9 89.0 38 86.4 5 62.5 17 54.8 40001 VAX6xMAR 1 14 35.9 4 100 9 20.4 2 25.0 6 19.3 J 117 x JAMAPA 27 69.2 5 44.4 12 27.3 2 25.0 12 38.7 G 2333 X G19839 31 79.5 6 55.6 6 13.6 2 25.0 9 29.0 Total evaluatcd 39 9 44 8 31 by marker class Popu1ation Cross cONA % Geno mi e % Total % derived DOR 476 X SEL Mcsoamerican X 19.7 1309 Mcsoamerican 13 25 38.5 38 29.0 BAT93 xJALO Mcsoamerican x Andean 32 48.5 37 56.9 69 52.7 ICA PIJAO X G Interespecific 40001 52 78.8 50 76.9 102 77.9 VAX6xMAR 1 Mcsoamerican X Mcsoamerican 16 24.2 19 29.2 35 26.7 Jll7xJAMAPA Mcsoamerican X Mesoamerican 23 34.8 34 52.3 57 43.5 G 2333 X G l9839 Mcsoamerican x Andcan 11 16.7 36 55.4 47 35.9 TOTAL Total (no.) 1 Average(%) EVALUADOS 66 37.2 65 51.5 131 44.2 15 1.1.4 ldentification of common bean genotypes using "Fingerprints" of metabolism enzymes and seed proteins: the case of the En ola variety César H. Ocamgo•, Orlando Toro1, D.G. Debouck •-2 1SB-l Project; SB-2 Project Introduction Crop varieties must successfully fulfill the criteria of newness, distinctness, uniformity, and stability, in order to be registered at nationaJ or international leve) under any PBR regime. Traditionally, morphological data have been used to define the parameters of certification. However, morphological characters whose expression is affected by environment and wich exhibit continuous distribution are notoriously poor taxonomic descriptors. Therefore, there is growing interest in using biochemical and DNA-based tests to provide sharply defined and reproducible genotypic descriptions. On the other hand, the F AO-CGIAR agreement states that no designated germplasm can be protected under any PBR or patent. We are comparing a patented bean variety with 21 bean genotypes in order to check the condition of newness, using seed proteins (phaseolin) and 16 isoenzymes (with 21 monomorphic and 9 polymorphic loci). Our selection work was facilitated by existing prior art about yellow bean varieties selected in Mexico from Peruvian germplasm (Lépiz & Navarro 1983; Voysest 1983). Results The biochemical data (enzyme bands) were interpreted as dominant markers and were compiled in a data matrix on the basis of presence (1) or absence (O) of selected bands. A pair-wise similarity matrix was calculated using the simple matching coefficient. This similarity matrix was employed to construct a dendrogram by the Unweighted pair group group method with arithmetical averages (UPGMA), using the SAHN-clustering and TREE program from the NTSYS-pc, version 2.02i package. The dendrogram obtained from the isoenzyme profile analysis (Figure 3) shows at the 0.87 similarity level six groups. The first and largest group was formed by 12 varieties, which in eludes En ola, while the other 1 O varieties are separated from the main gro u p. These results will help to selecta group of designated germplasm for microsatellite analysis. 16 0.4 0.61 0.14 Si..UtntyCocft'i citnl r l OJ1 ~ ~ ~ b o o o o o, o o o o o o o V o VARIUY_X_T IIIPI_MEX_T 13094_MEX_T lDI.l_M!X_T l2ll7_W!X_T lDJO_W!X_T IUJO_MEX_T .l6SP_MEX_T 4414_COL_T 6.l77_PlR_T 7ClP_PlR_T 7ClO_PlR_T 14114_MEX_S 14ll4_PER_T .l707_PlR_T 7319 _PER_T I:JlJl_PER_T 14HJ_MEX_T .l703_PER_T lClO_MEX_S 1411.l_MEX_S W6_MEX_S 1.00 Fig. l. Dendrogram derived from a UPGMA cluster analysis, using the Dice similarity index based on isozyme banding pattems. Six clusters were resolved at the 0.87 similarity level. References Lépiz R. & Navarro F.J. 1983. Frijol en el Noroeste de Mexico. Instituto Nacional de Investigaciones Agrícolas. Secretaria de Agricultura y Recursos Hidraúlicos, México, Mexico. 69p. Voysest O. 1983. Variedades de frijol en América Latina y su origen. Centro Internacional de Agricultura Tropical, Cali, Colombia. 87p. 17 1.1.5 Simple Sequence Repeat (SSR) marker diversity in cassava landraces: genetic diversity and differentiation in a predominantly asexually propagated crop Martín Fregene1, Heneriko Kulembeka2 ,Steve Kresovich3 1SB-2 Project, 2ARI-Tanzanía, GDI, 3Comell University Introduction The study of cassava land races from two Southem Tanzanian districts reported last year was extended to assessing genetic diversity and differentiation of cassava land races from 5 countries in South America, 2 in Central America, and 2 in Africa. A number of elite lines developed at CIA T and liTA were íncluded in the analysis to evaluate the effect of breeding on genetic diversity. SSR marker variation at 67 loci was assessed in 314 accessions of cassava land races from Brazil, Colombia, Peru, Venezuela, Argentina, Guatemala, Mexico, Tanzania, and Nigeria. Accessions from the Neo Tropics were from the CIA T germplasm collections and those from Tanzania were the same field collection made in 1999 in a key introduction point of cassava into Africa (South Westem Tanzania) and described in the annual report last year. The Nigerian Land races were from a collectíon held at liT A, Ibadan. The main reason for the assessment of genetic diversity and differentiation found in cassava land races is to delineate heterotic pools for a more rational approach to choosing parents for cassava improvement and the exploitation of combining ability via reccurent reciprocal 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, were later found between these populations (Melchinger et. al. 1990). Methodology Plant materials, DNA isolation, and SSR marker analysis have been described elsewhere (CIA T 2000; Fregene et. al. 200 1 ). Genetic diversity within and among accessions was estimated by the software package GEN-SURVEY (Vekemans and Lefebvre 1997) using the following statistics: percentage of polymorphic loci, mean number of alleles per polymorphic loci, average observed heterozygosity, Ho, and the average gene diversity, He (Nei 1978). For all loci and for all accessions the total heterozygosity, (HT) and the proportion of among accession differentiation (GsT) were estimated according to Nei (1978). Standard deviations for the above parameters were estimated over loci and samples by Jackknife (Quenoille 1956; Efron 1982). Given the small evolutionary divergence times for the accessions, the infmite alleles model (IAM) (Goldstein et. al. 1995) was assumed for all calculations. Genetic differentiation was quantified by the F statistics estimator FsT (theta) (Wright 1951) as described by Weir and Cockerham (1984) using FSTAT 2.9 (Goudet 1998). GsT gives the same estímate of genetic differentiation as FsT but takes into account variation in sample sizes, as is the case in this study. Confidence intervals were calculated per locus over samples, and over loci by Jacknife, and by bootstrapping over loci. Pairwise values ofFsT between samples (land race group) was also estimated and the pairwise matrix analyzed by cluster analysis, using Ward's hierachical clustering of JMP (SAS Institute 1995). 18 To assess if random genetic recombination created by fanner selection from spontaneous seedlings have played a part in the evolution of genetic diversity, parent-offspring relations were sought in the SSR data from the Southern Tanzanian collection using the computer program CERVUS (Marshall et. al. 1998). CERVUS simulates a maternal and a paternal genotype from allele frequencies observed in the study population, and derives an offspring genotype by Mendelian sampling of the parental alleles. The simulation also alters the genotypic data to reflect the existen ce of un-sampled males, missing loci and incorrectly typed loci, according to the values ofthe simulation parameters. Each candidate parent is considered in turnas the alleged father, and LOD scores are calculated for all males for whom genetic data exists. Once all males have been considered, the most likely and second most likely males are identified and the Delta score (difference in LOD scores) calculated. The final stage of the simulation is to find critica) values of Delta so that the significance of Delta values found in paternity inference in the study population can be tested. Results The large number of unlinked SSR loci employed in this study enabled a rigorous estimation of genetic differentiation and diversity structure of cassava land races from the primary and secondary center of diversity not previously carried out for cassava. The reliability of estimates for genetic variation, such as He, Ho, FsT and genetic distances, depend more on number ofloci than the number of individuals sampled (Baverstock and Moritz 1996). Estimates of genetic differentiation ranged widely from loci to loci, underscoring the danger of assessing SSR diversity using a small set of SSR markers. The genetic diversity of maize as a sub-set of diversity found in its teosinte progenitors vary from 25% to 75% based upon what location of the genome the diversity analysis was based on (Eyre-Walker et. al. 1998). Principal findings of the study is genetic diversity, as assessed by the average gene diversity, He, was high in all countries with an average heterozygosity of 0.5358±0.1184. (Table 1 ). Highest genetic diversity was found in Brazil and Colombian, although genetic diversity between Latín American and African land races is comparable. No unique alleles with a frequency of more than 25% was found within country samples with an exception of Guatemala and Nigeria. The genetic differentiation estimator FsT (theta), revealed a low leve) of differentiation (Fsr=0.091±005) between country samples compared to the average for crop species - FsT =0.34 (Hammrick and Godt 1997). Nonetheless pair-wise FsT data between countries reveals high genetic differentiation (FsT =0.26) between accessions from Nigeria and Guatemala, anda moderate to high differentiation between country accessions of the primary and a secondary center of diversity (Table 2 and Figure!). A total of 51 parent-offspring relationships were found in the 96 accessions collected from Southern Tanzania using a delta threshold leve) of 1.0 (Fregene et. al. 2001, Appendix2). Analysis ofparent- offspring are confounded by closely related offspring, the statistic delta calculated by CERVUS compares LOD scores of the two best putative parents to reduce the confounding effects of full- or half-sibs. Results of the parent-offspring relationship successfully identified a known parent of TMS 30572, an improved line from liTA, which was included as an interna) control. The genotype 58308 from the Moor plantation, Ibadan, Nigeria, breeding program ofthe 1950s, served as a parent source of cassava mosaic disease resistance (CMD) for TMS30572. The overall low leve) of genetic differentiation in cassava is comparable with that found in perennial forest trees, 0.084 on an average (Harnrick and Godt 1996; Le Corre et. al. 1997). Forest trees have experienced many foundation events after the expansion from a few Southern refuges 15,000 years ago after the last glacial period (Huntley 1990). Austerlitz et. al. (2000) demonstrated that the unexpected low differentiation and high genetic of trees events can be explained by high 19 gene flow, both seed and pollen flow, and the length of their juvenile phase. Cassava was likely domesticated from populations of Mesculenta sub spp jlabellifolia along the Southem rim of the Amazonían basín wíthín the last 10,000 years ago (Oisen et. al. 1999). Its expansion ínto other regions of Latín America, Africa and Asia would have led to founders effect of reduced diversíty andan increase in genetic dífferentiation . The unexpected low level of genetic differentiatíon and the high genetic diversity of cassava land races in all countries may therefore be due to high genetic diversity of original populations, extensive movement of germplasm and spontaneous genetic recombination. The common practice of using volunteer plants and the circulation of woody planting material, often to replace varieties destroyed by herbivores, biotic and abiotic stresses would have lead to highly heterogenous cassava fields after domestication. Diversity found in a single fanner's field has also been shown to be equal to the core-of-the core collection of 38 accessions representative of the world cassava collection at CIA T (Elias et. al. 2000). Future Plans • A larger sample set of land races from Nigeria and Guatemala regions will be analyzed to confirm results obtained here. • Genetic crosses between and within Guatamalan land races and Nigerian land races to test correlation between differentiation and heterosis. References Austerlitz F., Mariette S., Machon N., Gouyon P., Godelle B. 2000. Effects of colonization processes on genetic diversity: differencés between annual plants and tree species. Baverstock PR and Moritz C 1996. Chapter 2 Project Design, pp 17-27 in Molecular Systematics, edited by DM Hillis, C Moritz and BK Mable. Sinauer Associates Inc. Sunderland, MA. Bowcock A.M. Ruiz-Linares A., Tomfohrde J. , Mich E., Kidd J.R., Cavalli-Sforza L.L. 1994. High resolution ofhuman evolutionary trees with polymorphic microsatellites. Nature 368:455-457. CIAT 2000. Annual Report 2000. CIA T, Cali, Colombia. Efron B. 1982. The Jackknife, the Bootstrap and Other Resampling Plans. CBMS-NSF Regional Conference Series in Applied Mathematics, Monograph 38. SIAM, Philadelphia. Elias M., Panaud 0., McKey D.B., and Robert T. 2000. Traditional cultivation of cassava among Amerindians: consequences on genetic diversity assessed with AFLP markers. In: Carvalho L.J.C.B, Thro A.M .. and Vilarinhos A.(eds.), Cassava Biotechnology. IV Intemational. scientific Meeting of the Cassava Biotechnology Network. EMBRAPA, Brasilia. pp111-117. Eyre-WalkerA., Gaut R. L., Hilton H., Feldman D., and Gaut B. 1998. lnvestigation of the bottlenecks leading to the domesticatíon ofmaize. Proc Natl Acad ofSci USA 95,4441-4446 Goldstein D.B., Ruiz-Linares. A.R., Cavalli-Sforza L.L., Feldman M.W. (1995). An evaluation of genetic distances for use with microsatellite loci. Genetics 139:463-471. 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 Predominantly Asexually Propagated Crop (Submitted to Genetics). 20 Goudet J, 1995. FSTAT (vers. 1.2): a computer program to calculate F-statistics. J. Hered. 86:485-486. Hamrick J .L., Godt M.J.W.1996. Effects oflife history traits on genetic diversity in plants. Philos. Trans. R. Soc. Lomd. Ser. B Biol. Sci. 351:1291-1298. Hamrick J.L., Godt M.J.W. (1997) Allozyme diversity in cultivated crops. Crop science 37:26-30. Huntley B. European vegetation history:paleovegetation maps from pollen data -13000yr BP to present. J. Quat. Sci. 5: 103-122. Keeratinijakal V., Lamkey K.R. 1993. Responses toa reciproca! recurrent selection in BSSS and BSCBI Maize populations. Crop Science 33:73-77 Le Corre V., Machon N., Petit R.J., Kremer A. 1997. Genetic variation at allozyme and RAPD locus in sessile oak Quercus Petarea (Matt.) Liebl. : the role ofhistory and geography. Mol. Ecol. 6:1-11. MarshaJI, TC, S late, J, Kruuk, LEB, Pemberton, JM ( 1998) Statistical confidence for likelihood-based patemity inference in natural populations. Molecular Ecology 7, 639-655. Melchinger, A.E, Lee, M., Lamkey, K.R., and Woodman, W.L., (1990). Genetic diversity for restriction fragment length polymorphisms: relation to estimated genetic effects in maize inbreeds. Crop Sci. 30 p 1033-1040. Nei M. ( 1978).Estimation of average heterozygosity and genetic distances from a small number of individuals. Genetics 89: 5-83-590 Olsen K, Schaal B. 1999 Evidence on the origin of cassava: Phylogeography of Manihot esculenta Proc NatJ Acad Sci 96:5586-5591 . Olsen K. and SchaaJ Barbara 2001. Microsatellite variation in cassava (Manihot esculenta, Euphorbiaceae) and its wild relatives: evidence for a southem Amazonian origin of domestication. Amer Journa1 of Botany 88: 131-142 Quenoille M 1956 Notes on bias in estimation. Biometrika 43 :253-260. R.aymond M and Rousset F, 1995. An exact test for population differentiation. Evolution. 49:1280-1283 SAS Institute, Inc., 1995. JMP (version 3.1). SAS Inst. Inc., Cary, NC. ShuJI G.F. 1952. Beginnings ofthe heterosis concept. P 14-48. In: JW. Gowen (ed) Heterosis, Iowa Sate College Press, Ames. Tomes D. 1998 Heterosis: performance stability, adaptability to changing technology and the foundation of agriculture as a business. In: Lamkey K. and Staub J.E. (eds) Concepts and Breeding ofHeterosis in Crop Plants. CSSA Special Publication Number 25. Crop Science Society of America, Madison, Wisconsin Vekemans X. Lefebvre C., 1997 On the evolution of heavy-metal tolerant popuJations in Armería marítima: evidence from allozyme variation and reproductive barriers. J . Evo l. Biol., 10: 175-191 Weir BS and Cockerham CC (1984) Estimating F-Statistics for the analysis ofpopulation structure. Evolution 38:1358-1370. Wright S. 1951 The geneticaJ structure ofpopulations. Annals ofEugenics 15:323-354. 21 Table l. Genetic diversity within groups of cassava land races classified accordiog to country of origio. Standard deviatioos were estimated by jack- koife over loci. Ht, Hs, Dst aod Gst are given over loci and over populations (couotry collections). Population Sample size No. ofLoci No. ofPol. Percent of Poi. Mean No. of Mean No. of Ho He He_p Fis_p Loci Loe. alleles/loc. alleles!Eol.loc. Argentina 3 67 57 85.1 2.6 2.9 0.5174 0.4635 0.5672 0.0596 Brazil 20 67 67 100 5.2 5.2 0.5311 0.6129 0.6285 0.146 Colombia 32 67 66 98.5 6 6 0.5012 0.6177 0.6277 0.1875 GCA 15 67 65 97 4.5 4.6 0.5244 0.5754 0.5952 0.1072 Guatemala 4 67 57 85.1 2.4 2.7 0.4925 0.396 0.4554 -0.1269 Mexico 5 67 64 95.5 3.6 3.7 0.4915 0.56 0.6251 0.1987 Peru 7 66 62 93.9 3.7 3.9 0.4892 0.5596 0.6067 0.1771 Venezuela 5 66 64 97 3.5 3.6 0.4297 0.5692 0.634 0.2975 Tanz-Mtwara 84 67 65 97 5.3 5.4 0.543 0.558 0.5616 0.0295 Tanz- 23 67 64 95.5 4.5 4.7 0.5448 0.5545 0.5667 0.034 Naliendele Tanz-Kibaha 56 67 64 95.5 5.1 5.3 0.5274 0.5334 0.5382 0.0144 Nigeria 19 66 62 93.9 3.9 4 0.5965 0.5296 0.544 -0.1245 liTA 6 67 61 91 3.2 3.4 0.4915 0.4866 0.534 0.061 Moor 4 67 63 94 2.7 2.8 0.51 0.4852 0.5596 0.0572 Mean 94.23 4.03 4.17 0.5136 0.5358 0.5745 0.0799 std deviation 4.45 1.11 1.06 0.0378 0.0602 0.0495 0.1184 Ht Hs Dst Gst Mean 0.6499 0.5812 0.0687 0.1075 std deviation 0.1595 0.147 0.0318 0.0565 95%CI 0.61 0.5463 0.0621 0.0953 99%CI 0.6871 0.6136 0.0758 0.1195 Ho Average observed heterozygosity within country He Average expected heterozygosity within country He Average expected heterozygosity within country corrected for small sample sizes (Nei 1978) Ht TotaJ Heterozygosily in Jhe enlire data sel Hs Gene diversity within country averaged over the entire data set Dst Average gene diversity between populations Gst Coefficient of gene differentiation. 22 Table 2. Pairwise estimator of Fst (tbeta) between pairs of country groupings of cassava 1and races Population Arg. Brazil Colombia GCA Gua. Mexico Peru Ven. Tanz- Tanz- Tanz- Nigeria liTA Moor Mtwara Naliendele Kibaha Argentina o 0.0632 0.0708 0.068 0.2364 0.0429 0.1279 0.0283 0.1453 0.127 0.1654 0.2067 0.1447 0.1631 Brazil 0.0632 o 0.0572 0.0442 0.1454 0.0397 0.1119 0.0297 0.0791 0.0787 0.115 0.1076 0.1296 0.1054 Colombia 0.0708 0.0572 o 0.0114 0.1122 0.0379 0.0782 0.0123 0.0922 0.0827 0.129 0.129 0.1237 0.1137 GCA 0.068 0.0442 0.0114 o 0.1227 0.0283 0.0967 0.0119 0.0809 0.0682 0.1077 0.1388 0.1275 0.1117 Guatemala 0.2364 0.1454 0.1122 0.1227 o 0.1468 0.1914 0.1205 0.1682 0.1638 0.2103 0.2696 0.2689 0.2506 Mexico 0.0429 0.0397 0.0379 0.0283 0.1468 o 0.0853 -0.0059 0.0984 0.082 0.1185 0.1405 0.117 0.1118 Peru 0.1279 0.1119 0.0782 0.0967 0.1914 0.0853 o 0.0538 0.1417 0.1212 0.1529 0.1877 0.1687 0.1612 Venezuela 0.0283 0.0297 0.0123 0.0119 0.1205 -0.0059 0.0538 o 0.0567 0.0476 0.0839 0.1147 0.0795 0.0593 Tanz- 0.1453 0.0791 0.0922 0.0809 0.1682 0.0984 0.1417 0.0567 o 0.0076 0.0452 0.1402 0.118 0.1081 Mtwara Tanz- 0.127 0.0787 0.0827 0.0682 0.1638 0.082 0.1212 0.0476 0.0076 o 0.0097 0.1358 0.0888 0.0919 Naliendele Tanz-Kibaha 0.1654 0.115 0.129 0.1077 0.2103 0.1185 0.1529 0.0839 0.0452 0.0097 o 0.1625 0.1121 0.1215 Nigeria 0.2067 0.1076 0.129 0.1388 0.2696 0.1405 0.1877 0.1147 0.1402 0.1358 0.1625 o 0.1605 0.1142 liTA 0.1447 0.1296 0.1237 0.1275 0.2689 0.117 0.1687 0.0795 0.118 0.0888 0.1121 0.1605 o 0.0182 Moor 0.1631 0.1054 0.1137 0.1117 0.2506 0.1118 0.1612 0.0593 0.1081 0.0919 0.1215 0.1142 0.0182 o 23 Argentina Peru Brazil Colombia GCA Parents (CIA n Mexico Venezuela Guatemala Tanzania·Mtwara Tanzania Naliendele Tanzania Kibaha Nigeria liTA Moor Plantation Figure l. Hierarchical clustering (Ward) of Fixation lndex (FST) pair-wise distances between cassava land races, grouped by country, and elite parents by source. 24 1.1.6 Assessment of genetic diversity among African cassava accessions resistant to the Cassava Mosaic Disease using SSR markers Yvonne Lokko2, Martín Fregene\ Alfred Dixon2 1SB-2 Project; 2IITA Introduction The cassava mosaic virus disease (CMD) is considered the most devastating disease of cassava in Africa causing severe yield losses ranging from 20-95% (Thresh et al., 1994). It is caused by the cassava mosaic begmoviruses, which are transmitted by the whitefly (Bemisia tabaci Genn) and spread through propagation of infected vegetative propagules. lt is estimated that total crop yield losses due to CMD cost the African continent about $2 billion per annum (liT A, 1997). The most effective means of controlling CMD is by host plant resistance and resistance was first identified in third back cross derivatives between cassava and its wild relative Manihot glaziovii Muller von Argau (Nicholas, 1947). Despite the progress made in resistance breeding, there still is the need to increase the levels of resistance, particularly against aggressive recombinant strains that can spontaneously occur (Zhou et. al., 1997). Recently a novel source of resistance controlled by a single dominant gene was found in sorne Nigerian land races (Mignouna and Dixon 1996) and this has lead to a more systematic evaluation of African land races. To facilitate choice of parents for breeding more durable CMD resistance while maintaining a good leve! of genetic diversity, 18 SSR markers were used to evaluate genetic diversity within a collection of 78 African cassava accessions resistant and susceptible to cassava mosaic virus disease (CMD). The accessions include 5 improved accessions, 68 resistant and 10 susceptible land races A total of 18 SSR markers were employed to determine genetic relationships. The second objectives of this study was to predict possible novel sources of resistance to CMD based upon SSR marker clustering which can then serve as a basis for further genetic studies. Methodology The cassava accessions and their source used in this study are shown in Table l. The land races had previously been evaluated in severa! location and years for their reaction to CMD based on their phenotypic expression of symptom severity using the standard five point scoring scale system for CMD (liTA, 1990). DNA isolation was from 1-3 g ofyoung leaves per accession after Dellaporta et al., (1983). Thirty-six SSR markers, two each from 18 linkage groups of the cassava genetic map (Fregene et. al 1997; Mba et. al. 2000) (Table 2), were employed in the initial SSR analysis. SSR analysis was as described in Mba et al. (2000). Individual accessions were scored as diplotypic data "O 1 02" and as haplotypic data "1" presence of a band, and "O" absence of a band for the SSR data. individually and the different alleles were recorded for each sample screened. The haplotypic data was used to calculate genetic distances between pairs of cassava accessions, using the Dice algorithm, and to draw a dendogram using the Unweighted pair-group mean average (UPGMA) cluster method ofNei's genetic distances (Sneath and Sokal 1973). The genetic distances and dendogram were computed with the NTSYS-PC computer programrne, ver. 2.02 (Rohlf 1997). The diplotypic data was employed to calculate estimates of genetic diversity estimates: percentage of polymorphic loci, mean number of alleles per polymorphic loci, average observed heterozygosity, Ha, and the average gene diversity, lL, (Nei 1978), using the computer program Gen-Survey (V ekemans and Lefebvre, 1997). 25 Confidence intervals, at the 95% leve!, were obtained through 200 bootstraps over loci for the means of the above parameters. Results The overall leve! of polymorphism, 92%, is better than that found a previous AFLP study of CMD resistance and susceptible land races, 69%, (Fregene et al., 2000) confirming the superiority of SSR markers for genetic diversity studies. A dendogram of genetic distances grouped the 78 accessions into 5 groups at coefficient of similarity of 0.4. The first group has nine members including the line 58308, the principal parental line for the Mglaziovii source of CMD resistance, and its top progeny TMS 30572, the improved accession 91 /02324, four resistant and one susceptible land race (see Table 1 for accession groupings). The next group, which was the largest, was made up of two improved accessions M94/0583 and 29 land races, including one susceptible accession. All the resistant land races from the Republic of Benin and the majority of resistant land races from Nigeria and the Togo were in this group. The group also included a resistant land race from Angola and one from Ghana. The third group consists of the improved accession TMS3000 1, 17 resistant and one susceptible land races. Group four was made up of seven susceptible and five resistant land races, and group five, made up of the improved accession M94/012land eight other resistant land races. Duplicates were detected between sorne of the Nigerian land races such as TME581 and TME12, TME5 and TME3, TME62, TME6 and tME4, between TME242 and TME240, TME435 and TME288, TME479 and TME470 and between TME480 and TME225. The clustering pattem of the land races and the leve! of duplication is in agreement with the AFLP study of Fregene et al., (2000). Overall genetic diversity ofthe land races was high, 0.512, comparable to that described for a larger set of land races from 7 African, South and Central American countries, although the large difference in number of markers makes this comparison inadequate (M. Fregene et. al. 2001, CIA T 2001, this report). Gene diversity was highest among the land races and accessions in cluster group 3 followed by those in group 4 then group 2 and the lowest was detected in cluster group 1 (Table2). Of the total genetic diversity, 0.47 was dueto within cluster diversity and genetic differentiation between cluster was low (Gst = 0.096). The amount of genetic differentiation which has been reported for cassava, (Gsr=0.43, Fregene et al., 2000) is higher than that found in this study. Discrepancies in gene diversity estimates have been attributed to nature of markers systems (Djé, et al, 2000), but it may also be due to the smaJl set of African land races and an inclusion of Latín American Iand races in the Fregene et. al. (2000) study. Results of this study reveal a substantial amount of genetic diversity in CMD resistance gerrnplasms appropriate for genetic improvement of CMD resistance as well as other traits, particularly yield. It also suggest that there maybe other sources of resistance to CMD other than the known ones based on the clustering pattem of the resistant accessions. The Nigerian land races that have the novel source of resistant cluster together away from land races from other African nations and from the older source of resistance, 58308. This result suggests that resistance to CMD may have arisen independently several times in the past. This result will be conftrmed by genetic analysis of crosses between resistant and susceptible land races from clusters other than those with land races bearíng currently known sources o resistance. Future Plan Marker-assisted genetic analysis of crosses between resistant and susceptible land races from clusters other than those with land races bearing curren ti y known sources of resistance. 26 References Dellaporta SL Wood J, Hicks JR (1983) A plant DNA minipreparation: version II. Plant Mol Biol Rep 1:19-21 Dje, Y. M. Heuertz · C. Letebvre · X. Vekemans. 2000. Assessment of genetic diversity within and among gennplasm accessions in cultivated sorghum using microsatellite markers. Theor Appl Genet 100:918-925 Fregene M, A. Bema1, M Duque, A. Dixon and J. Tohme, 2000. AFLP analysis of African cassava (Manihot esculenta Crantz) gennplasm resistant to the cassava mosaic disease (CMD). Theoretical and Applied Genetics 100: 678-685. Fregene M., Suarez M., Mkumbira J., Kulembeka H., Ndedya E., Kulaya A., Mitchel S., Gullberg U., Rosling H., . 1_ Dixon A., Kresovich S. (200 1) Simple Sequence Repeat (SSR) Diversity of Cassava (Manihot escu/enta 0"! Crantz) Landraces: Genetic Structure in a Predominatly Asexually Propagated Crop (Submitted to Genetics). Hamrick JL, Godt MJW (1997) Allozyme diversity in cultivated crops. Crop Sci 37 : 26-30 liTA, 1997. Annual Report ofthe lntemational Institute ofTropical Agriculture. IITA lbadan Nigeria. Jennings, D.L. and C. H., Hershey, 1985. Cassava breeding: a decade of progress from intemational prograrnmes. In: Russe/1 G. E. (Ed) Progress in Plant Breeding J. Butterworths, Cambridge pp 89-1 16. Mba, R.E.C, P. Stephenson, K. Edwards, S. Melzer, J. Nkumbira, U. Gullberg K. Apel, M. Gale, J.Tohme and M. Fregene. 200 1, Simple sequente repeat (SSR) markers survey of the cassava (Manihot escu/enta Crantz) genome: towards an SSR-based molecular genetic map of cassava Theoretica/ and Applied Genetics 102:21- 31 . Mignouna H.D., and Dixon, A.G.O., 1997. Genetic relationships among cassava clones with varying Jevels of resistan ce to the African mosaic disease using RAPD markers. African Journal of Root and Tuber Crops. 2 (1 &2):28-32. Nei M (1978) Estimation of average heterozygosity and genetic distance from a small nurnber of individuals. Genetics 89:583-590. Nichols, R. F. W., 1947. Breeding cassava for virus resistance. East African Agricultura/ Journa/ 15: /54-160. RohlfF.J., 1997. NTSYS-PC, Numerical Taxonomy and Multivariate Analysis System ver 2.02. Sneath, P.H.A, R.O Sokal (1973) Nurnerical taxonomy. Freeman, San Francisco. Thtesh, J.M., Fargette, D and Otim-Nape, G.W. 1994. Effect of African cassava mosaic geminivirus on yield of cassava. Tropical Science 34: 26-42. Thresh, J.M., G.W. Otim-Nape, J.P. Legg and D. Fargette, 1997. African cassava mosaic virus disease: The magnitude ofthe problem. African Journal ofRoot and Tuber Crops 2: 13-18. Vekemans X. and C. Lefebvre, 1997 On the evolution of heavy-metal tolerant populations in Armeria marítima: evidence from allozyme variation and reproductive barriers. J . Evo!. Biol., 10: 175-191 Zhou, X., Y. Liu, L. Calvert, C. Munoz, G.W. Otim-Nape, D.J. Robinson and B.D. Harrison, 1997. Evidence that DNA-A of a geminivirus associated with severe cassava mosaic disease in Uganda has arisen by inter- specific recombination. Journal o/ General Virology 78, 2101-211 J. 27 Table l. List of Cassava accessions tbeir pedigree/local na me ( or assigned code by Country collectors}, country of origin CMD status (R= resistant, S= susceptible) and assigned cluster group of genetic similarity. Accession Pedigree/Local name Country Status Group 58308 M.esculenta x M. glaziovii liTA R 1 M94/0583 liTA R 2 130001 liTA R 3 91102324 TMEI OP liTA R 1 M94/0121 liTA R 5 TME638 EJ 79 Ghana R 4 TME635 MNN55 Ghana R 4 TME631 SE 210 Ghana R 3 TME630 Amin Ghana R 4 TME581 Oke Local Nigeria R 2 TME572 Udoh Local Nigeria R 5 TME568 Mundele Paco (ANG-3) Angola R 4 TME565 Prescose de Angola (ANG-4) Angola R 3 TME546 SS4 (T8) Uganda R 4 TME526 Ka13 (Kenya Ostrom) Cote d'Ivoire R 1 TME498 R.A 16 Nigeria R 2 TME480 RB92/0119 Benin R 3 TME479 Agric Benin R 2 TME478 RB92/0123 Benin R 3 TME477 RB92/0104 Benin R 3 TME474 CAP94/064 Benin R 5 TME470 CAP94066 Benin R 2 TME461 RB92/0188 Benin R 3 TME456 CAP94062 Benin R 3 TME455 RB92/0116 Benin R 2 TME451 CAP94067 Benin R 2 TME449 RB92/0182 Benin R 2 TME446 RB92/0204 Benin R 2 TME443 CAP94090 Benin R 2 TME437 RB92/0103 Benin R 5 TME435 RB92/0175 Benin R 2 TME434 RB92/0155 Benin R 5 TME431 MAlN 11 Togo R 1 TME429 MAIN4 Togo R 1 TME419 Gbazekoute Togo R 3 TME379 Ofegbe Nigeria R 2 TME288 Akano Nigeria R 2 TME287 Power Nigeria R 2 TME282 Alice Local Nigeria R 2 TME279 O basan jo Nigeria R 2 TME278 Oko W arangbala Nigeria R 2 TME258 25 Ghana R 1 TME243 Toma26 Togo R 3 TME242 Toma 76 Togo R 2 TME241 Toma136 Togo R 3 TME240 Toma 75 Togo R 2 TME236 Toma37 Togo R 5 28 Accession Pedigree/Local name Country TME232 Toma63 Togo TME230 Toma36 Togo TME229 RB92/0130 Benin TME228 Toma 97 Togo TME225 92/0099 Togo TME209 1254(880887) Cameroon TME204 RB98/0113 Benin TME199 RB89/59 Benin TIMES Bagi Wawa Nigeria TME62 Yau Rogor Nigeria TME13 MS-20 Nigeria TME12 Tokunbo Nigeria TMEll Igueeba Nigeria TME9 Olekanga Nigeria TME8 Ama la Nigeria TME7 Oko-lyawo Nigeria TME6 Lapai-1 Nigeria TME4 A tu Nigeria TME3 2ND Agric Nigeria TMEI Antiota Nigeria 130572 58308 X Branca de Santa Caterina OP UTA TME401 Toma 141 Togo TME59 Dandualla-2 Nigeria TME60 Darazo Rogor · Nigeria TME104 Rogor-5 Nigeria TME107 Danwara Nigeria TME117 Isunikankiyan Nigeria TME123 Pan ya Nigeria TME218 881260(882160) Cameroon TME382 Suleja-5(92/0 163) Nigeria TME557 Lossakpleh C6te d' Ivoire Table 2. Gene diversity analysis within and among cassava accessions by cluster group Cluster Group PLJ>A A6 A e e Ha SI Cluster Group 1 88.9 2.7 2.9 0.4979 Cluster Group 2 94.4 2.6 2.6 0.6431 Cluster Group 3 100 2.8 2.8 0.6446 Cluster Group 4 94.4 2.7 2.7 0.6034 Cluster Group S 100 2.7 2.7 0.5663 Mean 95.56±4.65 2.68±0.06 2.73±0.09 0.591±0.061 Ht Hs Dst Gst Mean 0.514 0.466 0.048 0.096 Std 0.138 0.135 0.041 0.083 95%CI 0.446 0.404 0.032 0.062 95%CI 0.583 0.533 0.065 0.135 • Percentage of polymorphic loci at the 5% leve! within accessions b Mean number of alleles per locus within accessions e Mean number of polymorphic alleles per locus within accessions d. Average observed heterozygosity within accessions. • Average gene diversity within accessions corrected for small populations 29 H • ¡;; 0.4165 0.4369 0.4828 0.4378 0.4624 0.4473±0.026 Dm 0.0634 Status Group R 5 R 5 R 5 R 2 R 3 R 1 R 3 R 3 R 2 R 2 R 2 R 2 R 3 R 3 R 3 R 2 R 2 R 2 R 2 R 2 R 1 S 4 S 4 S 2 S 3 S 4 S 4 S 4 S 4 S 4 S 1 • H~ 0.444 0.4459 0.5006 0.4605 0.5193 0.4741±0.034 Rst 0.1417 11 Ili N V [1 L 1~ 1 t~~ l l J [ l=cJ tt-IJJ 1 R 1 1 1 , 11 1 1 j ) 1 1 1 ) 1 1 1 ) ) 1 1 1 1 1 1 1 1 OJI) t.ll 1.21 14) l.ll QBida1 Figure 1. Dendrogram ot geneuc OIStance snowmg tne assocJauon oetween tne 111 cassava accessaons oasea on ~~R using UPGMA cluster analysis. 30 1.1. 7 Analysis of genetic diversity in Cassava landraces from the coastal, andean and forest regio o of Peru Jorge Alcantara1 Luigi Guarino2 Martín Fregene3 1INIA-Peru; 2IPGRI-CIAT; 3SB-2 Project Introduction Peru is consídered one of severa( countries with enonnous amounts of biodiversity and it is a center of diversity for numerous cultivated crop, including cassava. The National Genetic Resources and Biotechnology Program (PRONIRGEB, its Spanish Acronym) of the National lnstitute for Research (INIA, its Spanish acronym) has as its principal objectives the conservation of native gennplasm for use by the scientific community and local fanners. PRONIRGEB has two cassava gennplasm banks located in Donoso, Lima, with 240 accessions, and El Porvenir, Tarapoto, forest agroecology, having 260 accessions. These accessions are kept as field collections, with a significant portian as tissue culture collections, and they have been characterized morphologically. Under the project "Models of diversity and genetic erosion of traditional cultivars in Peru: rapid assessment and early detection of risks using GIS tools" funded by the BMZ and executed by IPGRI and INIA, genetic diversity is being assessed using molecular tools combined with GIS methods to provide indicators of genetic erosion. In the first phase of this project a national laboratory for characterizing genetic resources has been set up and training of personnel, in the area of molecular markers for genetic diversity assessment, is being implemented to run the laboratories. The objective of this study was to train a national scientist from INIA in simple sequence repeat (SSR) analysis in cassava, one of crops addressed under the BMZ project. Methodology One hundred accessions from the PRONIRGEB-INIA cassava gennplasm collection was used in this study. The accessions were selected based upon the place collected: coastal, Andean or forest, to provide a representative sample of cassava grown in Peru (Table 1 ). About 1 OOmg of young leaf tissue obtained from field grown plants was used to isolate DNA using a mini CT AB preparation (CIP 1997). Eighteen SSR markers, one from each linkage group, and selected for to their high heterozygosity in a previous SSR study of cassava land races (Fregene et. al. 2001, CIAT 2001, this report) were used for SSR analysis as described by Mba et al. (2000). Gel analysis of PCR arnplification product ís also as described by Mba et. al. (2000). Raw SSR data was scored as "1" and "O" for presence and absence of DNA bands respectively or haplotype data. The bands were then numbered and the data transfonned by Excel to "O 1 02" or diplotype data.. The haplotype data was used to calculate genetic distances between pairs of cassava accessions, using the Dice algorithm, and to derive principal components (PC) (Sneath and Sokal 1973). The f1rst and second components were presented in a graphical fonn using Excel. The genetic distances were computed with the NTSYS-PC computer programme, ver. 2.02 (Rohlf 1997), while the PC analysis was done using SAS (SAS lnstitute). The diplotype data was employed to calculate estimates of genetic diversity estimates: percentage of polymorphic loci, mean number of alleles per polymorphic Ioci, average observed heterozygosity, Ho, and the average gene diversity, He (Nei 1978), using the computer program Gen-Survey (Vekemans and Lefebvre, 1997). Confidence intervals, at the 95% level, was obtained through 200 bootstraps o ver loci for the means of the abo ve pararneters. 31 Results Average genetic diversity was very high 0.68, and there was no significant difference between diversity found in the three regions (Tablel). Genetic differentiation as estimated by Gst was very low (0.0074) and confirms the same pattem found in the study of cassava land races from 7 countries (CIA T200 1, this report) although the large difference in markers makes this comparison ineffective. PCA of land races from the coastal, Andean and forest region also did not reveal any distinct clustering pattem among the land races with the exception that a single accession from Brazil included in the analysis was separated from the Peruvian accessions (Figure!). The present study reveals an unexpected low level of genetic differentiation and high genetic diversity of cassava land races in regions as diverse as the Andean, coastal and forest region of Peru. lt strongly supports the hypothesis of extensive movement of germplasm between regions, the high genetic diversity of original populations, as well as the wide adaptation of cassava. Future Plans A set of 36 primers, a gift from CIA T, will be used at PRONIRGEB-INIA to continue SSR analysis of all germplasm accessions held in the National collection. Conduct a comparative analysis between Peruvian germplasm held at PRONIRGEB-INIA and at CIA T for exchange of land races currently not present in either collection. References CIP 1997. Centro Internacional de Papa Annual Report, La Molina, Peru. Fregene M., Suarez M., Mkurnbira J., Kulembeka H., Ndedya E., Kulaya A., Mitchel S., Gullberg U., Rosling H., Dixon A., Kresovich S. (200 1) Simple Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta f(J~ Crantz) Landraces: Genetic Structure in a Predominatly Asexually Propagated Crop (Submitted to Genetics). Mba, R.E.C, P. Stephenson, K. Edwards, S. Melzer, J. Nkumbira, U. Gullberg K. Apel, M. Gale, J. Tohme and M. Fregene. 2001, Simple sequence repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: towards an SSR-based molecular genetic map of cassava Theoretical and Applied Genetics 102:21- 31 . Nei M ( 1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583-590. RohlfF.J., 1997. NTSYS-PC, Numerical Taxonomy and Multivariate Analysis System ver 2.02. Sneath, P.H.A, R.O Soka1 (1973) Numerical taxonomy. Freeman, San Francisco. Vekemans X. and C. Lefebvre, 1997 On the evolution of heavy-metal tolerant populations in Armería marítima: evidence from allozyme variation and reproductive barriers. J. Evo!. Biol., 10: 175-191 32 Table l. List ofCassava Accessions from the PRONIRGEB-INIA Germplasm Collection and Source [t. Códi. Región Departamento It. Códi. Región Departamento l. 181 Selva Ucayalí 51. 020 Sierra Cusco 2. 145 Selva San Martín 52. 001 Sierra Amazonas 3. 174 Selva Ucayali 53. 037 Sierra Huánuco 4. 049 Sierra Junin 54. 002 Selva Amazonas 5. 085 Costa Lima 55. 024 Sierra Huánuco 6. 163 Selva Sap Martín 56. 035 Sierra Huánuco 7. 101 Costa Lima 57. 033 Sierra Huánuco 8. 157 Selva San Martín 58. 038 Sierra Huánuco 9. 180 Selva Ucayalí 59. 034 Sierra Huánuco 10. 153 Selva San Martín 60. 029 Sierra Huánuco 11. 143 Costa Piura 61. 030 Sierra Huánuco 12. 108 Costa Lima 62. 026 Sierra Huánuco 13. 104 Costa Lima 63. 027. Sierra Huánuco 14. 082 Costa Lima 64. 043 Sierra Junín 15. 048 Sierra Junín 65. 022 Sierra Cusco 16. 139 Costa Piura 66. 042 Sierra Junín 17. 112 Costa Lima 67. 031 Sierra Huánuco 18. 117 Costa Lima 68. 044 Sierra Junín 19. 097 Costa Lima 69. 023 Sierra Cusco 20 138 Costa Piura 70. 036 Sierra Huánuco 21. 165 Selva San Martín 71. 004 Selva Amazonas 22. 175 Selva Ucayalí 72. 003 Selva Amazonas 23 089 Costa Lima 73. 027 Sierra Huánuco 24. 142 Costa Piura 74. 045 Sierra Junin 25. 148 Selva San Martín 75. 046 Sierra Junfn 26. 179 Selva Ucayalí 76. 113 Costa Lima 27. 155 Selva San Martín 77. 116 Costa Lima 28. 160 Selva San Martín 78. 121 Costa Lima 29. 144 Costa Piura 79. 125 Costa Lima 30. 047 Sierra Junin 80. 129 Costa Lima 31 094 Costa Lima 8!. 131 Selva Lo reto 32. 183 Selva Ucayali 82. 135 Selva Lo reto 33. 178 Selva Ucayalí 83. 133 Selva Lo reto 34. 173 Selva Ucayall 84. 134 Selva Lo reto 35. 182 Selva Ucayall 85. 136 Costa Piura 36. 158 Selva San Martín 86. 137 Costa Piura 37. 164 Selva San Martín 87. 140 Costa Piura 38. 176 Selva Ucayali 88. 141 Costa Piura 39. 086 Costa Lima 89. 146 Selva San Martín 40. 172 Selva Ucayalí 90. 149 Selva San Martín 41. 040 Sierra Huánuco 91. ISO Selva San Martín 42. 032 Sierra Huánuco 92. !51 Selva San Martín 43. 019 Sierra Cusco 93. 152 Selva San Martín 44. 005 Selva Amazonas 94. 154 Selva San Martín 45. 025 Sierra Huánuco 95. 156 Selva San Martln 46. 018 Sierra Cusco 96. 159 Selva San Martín 47. 081 Costa Lima 97. 161 Selva San Martín 48. 028 Sierra Huánuco 98. 162 Selva San Martín 49. 039 Sierra Huánuco 99. 199 Bolivia 50. 021 Sierra Cusco 100. 228 Brasil 33 Table 2. Gene diversity analysis of 100 accessions from the Peruvian Coastal, Andean and Forest regio os Regio o Number PLP" A6 Coastal 27 100 5.2 Selva 39 100 5.2 Sierra 30 100 4.9 Mean 100 5.13±0.18 Ht Hs Mean 0.7032 0.6980 Std 0.0865 0.0866 95%CI 0.6573 0.6506 A e e Ha 2 5.2 0.7259 5.2 0.6178 4.9 0.6879 5.13±0.18 0.6718 ±0.0540 Dst Gst 0.0052 0.0074 0.0086 0.0124 0.0009 0.0012 Pe ru SSR by so un: 1 PC 1 HE e 0.7031 0.6683 0.6879 0.6864 ±0.0174 HE¡;; e .7165 0.6770 0.6994 0.6976 ±0.0198 + Cuto Pt ru .s. ... Sil rr• Figure l. Principal component analysis (PCA) showing the genetic relatiooship between 100 cassava accessioos from the coastal, Andean and Forest regio os of Peru. 34 1.1.8 Root quality and pest resistance genes from wild relatives of Cassava for broadening the crop genetic base Nelson Morante3, Teresa Sanchez3, Daniel Debouk1, Antony Bellotti2, Martín Fregene2 1 SB-1 Project; 2SB-2 Project ; 3IP-3 Project Introduction As a major staple food crop across the tropics, cassava can serve as a cheap means of deploying adequate protein requirement amongst the poor and for feeding animals. But cassava's starchy roots are very low in protein compared to other crops; less than 2% dry matter of protein in cassava compared to 9.1% in potato, there is therefore a need to in crease the pro te in content in roots of cassava. Cassava is also an important source of starch, 70-90% of cassava dry root matter is starch, the rest being fibers . Raw or unmodified cassava starches are increasingly important raw materials in textile, alcohol, animal and human food industries world-wide and this is expected to grow (Henry 1995). An in crease therefore in starch dry matter content ( equivalent to starch content) translates into higher income per unit land, per unit labor (investment) for farmers growing cassava. Severa) wild relatives of cassava are known to possess up to 15% protein and more than 50% dry matter in their roots. These gennplasm resources are a useful source of genes for the improvement of protein in cassava. Reports of crosses between cassava and M tristis revealed root protein content of more ·than 8% in F1 hybrids (Bolhuis 1953; Asiedu et. al. 1992). Unfortunately the high protein content was lost during back crossíng to recover the desired characteristics and high root yield of cassava (Asiedu 1992). For severa) years now, it has been shown that the "tremendous genetic potential locked up in gennplasm banks can be released by shifting the paradigm from searching for phenotypes to searching for superior genes using molecular genetic maps and an advanced back cross mapping scheme (Tanksley and McCouch, 1997). An evaluation of protein and dry matter content, amylose/amylopectin ratio and white fly resistance was therefore conducted on gennplasm resources of 7 wild Manihot species held at CIA T genetic resources unit (GRU). Methodology More than 800 sexual seeds representing accessions of M esculenta sub spp jlabelifolia, M esculenta sub spp peruviana, M tristis., Mcarthaginensis, M walkerae, Mbrachyloba and Mjomentosa were planted in seedling trays at CIA T. Of these number, 695 accessions were transplanted to the field at the Centro Experimental de la Universidad Nacional, Palmira (CEUNP). From six month after planting, sequential evaluation of white fly resistance was conducted on all genotypes. At 8 months after planting 3 roots were milked from 678 accessions and evaluated for protein content, dry matter percentage, amylose/amylopectin ratio, and storage root size according to standard procedures established at CIA T. Another set of 400 sexual seeds of inter-specific hybrids between cassava and M esculenta sub spp jlabelifolia, M esculenta sub spp peruviana, M tristis., Mcarthaginensis, M chlorosticta and M pseudo glaziovii were genninated. A total of 322 were successfully transplanted to the field and 3 roots evaluated at 8 months after planting for the above traits. To confinn results obtained in the frrst year, 6 woody stakes were obtained from wild species accessions and inter- specific hybrids high in protein, dry matter, white fly resistance or low in amylose/amylopectin ratio, and planted in single row clonal observation plots at CIA T. At the same time 6-1 O stakes of 35 a selected sub set of these genotypes were planted in a hybridization block to initiate the advanced back cross QTL marker scheme to introgress favorable genes for the above traits into cassava. Due to the poor gennination of sorne accessions, stakes were planted in the green house befare transfer to the field. Result The first year evaluation of more than 1000 genotypes of 7 wild Manihot species and inter- specific hybrids revealed a moderate to very high levels for protein and dry matter content, waxy starches (low level of amylose), and white fly resistance. Table 1 shows the data for genotypes with the highest protein content. The best genotypes for white fly resistance were found in inter- specific hybrids with M ch/orosticta. (Table 2). A second year evaluation of six plants from the top genotypes will be conducted, but at the same time selections have been made from the top genotypes for genetic crosses. A selection index program developed by the cassava breeding unit (ClA T annual report 2000) was used to select the best 12 genotypes for protein content, dry matter content, and white fly resistance and the best 4 genotypes Jow arnylose content (Table 3). Table 1. Best genotypes of wild Monillo/ accessions for protein content from evaluation of 3 roots. Accession Number Mother Plant Bulking. %dry % Crude %e %ash matter pro te in Crude fibre ow 27- 1 CTH XXX- 1 3 10.17 7.00 4.27 3.10 ow 62- 2 FLA 433- 2 o 27.04 13.08 2.74 ow 130- 4 TST XXX- 42 2 17.92 11.77 3.56 ow 132- 2 TST XXX- 3 4 22.40 11.71 4.69 3.14 ow 132- 4 TST XXX- 3 1 43.82 13.12 12.52 1.85 ow 139- 1 TST XXX- 38 4 29.59 11.69 8.06 1.07 ow 141- 1 TST XXX- 41 o 30.43 12.15 14.72 3.08 ow 143- 1 TST XXX- 54 o 34.30 12.17 4.79 ow 145- 2 TST XXX- 77 o 24.30 13.58 5.41 ow 145- 3 TST XXX- 77 o 34.88 14.71 22.91 2.99 ow 153- 4 CTH 409- 1 o 16.94 13.17 3.45 ow 170- 3 CTH XXX- 106 29.03 12.12 2.65 ow 172- CTH XXX- 121 6.47 18.99 6.59 %Amylose 10.33 13.98 16.31 18.61 ow 180- 4 FLA 423- 5 4 27.47 13.50 5.38 14.56 ow 183- 4 FLA 423- 8 2 21.44 14.59 3.61 3.32 9.71 ow 185- 2 FLA 423- 10 28 .97 11.77 6.97 3.56 17.19 ow 231- 3 FLA 444- 7 4 30.31 11.84 3.15 1.98 13.98 ow 235- FLA 508- 1 35.46 12.07 5.84 1.91 13.94 ow 276- TST XXX- 26 30.48 17.34 25 .53 2.64 ow 278- TST XXX- 40 1 41.69 16.33 36.39 1.57 ow 284- 2 TST XXX- 77 2 17.15 13.45 13.78 4.62 Bulking 1 =Fibrous roots 2=Poor storage root formation 3 = good but small sized storage root formation 4 = Commercial sized storage roots 5 = Very big commercial sized roots 36 Table 2. Best 21 genotypes for wbite fly resistance evaluated in inter-specific bybrids. All bybrids are with M.chlorosticta. Extent of white fly on lea ves Extent ofDamage Accession Number ADULTS EGGS NYMPH PUPAE UPPER MED. LOWER CW 14- 2 1.5 1.5 1.5 1.5 l. O l. O 1.0 CW 14- 3 l. O 1.0 l. O 1.5 l. O l. O l. O cw 14- 4 2.0 2.0 2.0 l. O l. O l. O l. O CW 14- 6 2.0 1.5 1.5 l. O l. O l. O l. O CW 14- 7 l. O l. O 1.0 l. O l. O l. O l. O cw 14- 8 2.0 2.0 l. O l. O l. O l. O 1.0 cw 14- 9 l. O l. O l. O l. O l. O l. O l. O cw 14- 10 2.0 2.0 2.0 2.0 l. O l. O l. O CW 14- 11 l. O l. O l. O l. O 1.0 1.0 l. O CW 14- 12 2.0 l. O l. O 1.0 1.0 l. O l. O cw 14- 13 l. O 1.0 l. O l. O l. O l. O l. O cw 14- 15 l. O l. O l. O 1.0 l. O l. O l. O cw 14- 16 2.0 1.0 l. O 1.0 1.0 l. O l. O cw 14- 17 2.0 1.5 2.0 2.0 l. O l. O l. O CW 20- 1 2.0 l. O l. O l. O l. O 1.0 l. O cw 20- 2 2.0 1.5 1.5 1.0 l. O l. O l. O cw 21- 1 2.0 1.5 1.5 l. O l. O l. O l. O cw 21- 2 l. O l. O l. O 1.5 1.0 l. O l. O cw 21 - 3 l. O l. O l. O l. O 1.0 1.0 l. O cw 21- 4 2.0 2.0 1.5 1.5 1.0 l. O l. O CW 21- 5 l. O 1.0 l. O l. O 1.0 l. O l. O Extent of damage and white fly on 1eavesis in a scale of 1-5, 1 being no damage or presence and 5 being maximum damage. Table 3. Genotypes of wild Manihot accessions and inter-specific bybrids with bigb protein, dry matter content, low amylose (Waxy) and good wbite fly resistance selected for crosses to elite parents of cassava genepools. Trait Waxy Accession Number OW 30- 3 OW 179- 7 OW 183- 4 ow 262- 4 Protein and Dr Matter ( combined) ow 186- 2 OW 186- 5 ow 189- l OW 230- 3 OW 230- 4 OW 231- 2 OW 240- 7 ow 257- 1 ow 261- 1 ow 262- 7 ow 263- 4 ow 263- 9 ow 284- Mother genotype CTH XXX- 62 FLA 423- 4 FLA 423- 8 PER 416- 1 FLA 426- 3 FLA 426- 3 FLA 427- 3 FLA 441- 5 FLA 441 - 5 FLA 444- 7 PER 406- 2 PER 413- 5 PER 415- 4 PER 416- 1 PER 416- 2 PER 416- 2 TST XXX- 77 37 Number of plants 10 10 8 10 10 8 lO 10 10 10 10 10 4 10 8 10 10 Trait Accession Mother genotype Number of plants Number Protein ow 66- 5 FLA 430- 5 10 ow 132- 2 TST XXX- 3 10 ow 139- 1 TST XXX- 38 10 OW 179- 1 FLA 423- 4 10 ow 180- 1 FLA 423- 5 10 ow 180- 4 FLA 423- 5 10 ow 181- 2 FLA 423- 6 10 ow 181- 3 FLA 423- 6 JO ow 182- 8 FLA 423- 7 10 ow 230- 6 FLA 44J- 5 lO ow 231- 3 FLA 444- 7 10 OW 231- 4 FLA 444- 7 5 ow 235- 3 FLA 508- l 5 ow 236- 2 FMT XXX- 4 6 OW 248- 3 PER 411- 5 10 ow 280- 1 TST XXX- 51 10 ow 284- 2 TST XXX- 77 10 Dry Mattter content ow 95- 1 PER 4J2- 4 JO ow 146- 1 TST XXX- 12 10 ow 213- 4 FLA 437- 1 10 ow 213- 5 FLA 437- 1 10 ow 234- 2 FLA 496- 1 10 ow 240- 6 PER 406- 2 10 ow 240- 8 PER 406- 2 10 ow 248- 7 PER 411- 5 lO ow 252- 2 PER 412- 4 10 OW 262- 3 PER 416- 1 10 ow 262- 5 PER 416- 1 9 ow 269- 4 PER 417- 6 10 ow 280- 2 TST XXX- 51 10 White fly resistan ce ow 36- 2 10 ow 61- 4 10 ow 95- 2 PER 412- 4 10 ow 96- 2 PER 412- 8 10 ow 100- 2 4 ow 103- 8 10 ow 105- 6 5 ow 105- 7 10 ow 238- 1 10 Aithough only 4 genotypes from each group will be used for crosses a larger number was selected to accommodate variation that may occur due to the environment. Evaluation for the above traits will therefore be conducted befo re crosses are made to cassava. ). At least 100 seeds are expected for each of the 48 families (cross combination). The cassava parents for crosses are the elite 38 parents of the 4 agro-eco1ogica1 genepools (Table 4). Sorne of these lines are high carotene lines eg. SM 1433-4, and willlead toa combinatíon of high carotene with high protein content. The advanced back cross QTL identification and introgression scheme to be followed is briefly described. The F1 families obtained from above will be evaluated twice at the seedling and clona! observation tria! stage. The best 1 O inter-specific hybrids, 3 for protein, dry matter content, white fly resistance, and one for low amylose, will be selected for each agro-ecology and backcrossed to their respective recurrent parent. At least 200 BC1 seeds will be generated per family, ora total of 6000 seeds from 30 families. The BC1 will be evaluated twice as above and marker genotypíng will be for the best BC1 families from each agro-ecology for each trait for QTL analyisis. BC1 lines for the different agro-ecologies bearing favorable QTLs will be inter-crossed with one another,. Genotypes from the BC1F1 will be planted and evaluated in six-plant rows as described above. Lines found to have high protein and dry matter content will be selected and introduced into the normal breeding program at CIA T. More crosses will be made from the best families identified and sexual seeds will be shared with the lntemational lnstitute for Tropical Agriculture (liT A), lbadan and NARs collaborators in Latín America and Asia. Table 4. Elite Parents of Cassava Genepools for Advanced backcross QTL mapping scheme Elite Parents by their prospective agro-ecologies CLONES FOR THE LOWLAND HUM/D AGRO-ECOLOGY: MTAI -8 SM 14\\-5 CM 3306-4 SM 1433-4 CLONES FOR THEACIDSAVANNAH AGRO-ECOLOGY CM 523-7 CM 6740-7 CM 4574-7 SM 1821-7 CLONES FOR THE MID ALTITUTDE AGRO-ECOLOGY MBRA383 CM 7951-5 SM 909-25 SM 1219-9 Future Plans • Genetic crosses of wild Manihot species accessions and inter-specific hybrids to elite parents of genepools by agro-ecology • Second year evaluation of protein, dry matter, white fly resistance or low in amylose/amylopectin ratio in wild Manihot species accessions and inter-specific hybrids. 39 References Asiedu R, Bai KV, Terauchi R., Dixon AGO, Hahn SK 1992. Status ofwide crosses in Cassava and Yam. In: G.Thotttapily (Ed.). Biotechnology; enhancing research on tropical crops in Africa: proceedings of an intemational conference held at the lntemational Institute of Tropical Agriculture, 26-30 Novl990, liTA, Ibadan, Nigeria. Bolhuis, G.G. 1953. A survey of sorne attempts to breed Cassava with a high content of proteins in the roots . Euphytica 2: 107-112 CIAT 2000. Annua1 Report 2000. CIAT, Cali, Colombia Henry, G. 1995 Global trends in cassava production. CIAT cassava program working document. CIA T, Cali, Colombia Tanksley S D, and McCouch S 1997. Seed Banks and Molecular Maps: Unlocking Genetic Potential from the Wild. Science, vol277: 1063-1066. 1.1.9 Assessing the genetic variability of Xanthomonas axonopodis pv. manihotis in Togo by using RFLP G. Mosquera1, S. García1, S. Restrepo2, and V. Verdier2• SB-2 Project Introduction Xanthomonas axonopodis pv. manihotis (Xam) is the causal agent of cassava bacteria! blight (CBB). The disease was first reported in Brazil in 1912 and has been reported in Colombia and Venezuela and in most African countries. RFLP has been extensively used to evaluate Xam populations using different probes (Verdier et al., 1993; Restrepo and Verdier, 1997; and Restrepo et al., 1999). Based on previous work, it has been suggested that African strains originated from South Ameríca (V erdier et al., 1993 ). 23 5 strains collected in 14 JocaJities representing the different ecological zones in Togo (West Africa) were analyzed using RFLPs with different probes: pthB (a probe containing a pathogenicity gene), pBSS and pBS6 (two genomic and repetitive probes). This study aimed to assess the origin and the genetic diversity of the Xam population within and between ecological zones in Togo. Methodology Bacteria! isolation from infected le aves. Strain isolation was performed as described by Verdier et al. (1998). Briefly angular spots were cut out from lea ves and resuspended into eppendorf tu bes containing sterile water, then incubated at 4°C for 30min. lOOJ.d were spread in agar plates and incubated at 30°C for 48h. Single colonies were purified and stored at -80°C into a 20% glycerol solution. 40 A B e D E F G H 1 J K L M N Analysis with different probes. DNA purification, restriction and hybridization conditíons were done as descríbed by Restrepo and Verdier ( 1997). Statistical analysis. Banding patterns of hybridization obtained with the RFLP/pthB were used to compare the re1atedness of each strain. Each band showing different electrophoretic mobility was assigned a position number after its size was detennined in base pairs. The presence (coded 1) and absence (coded O) of each fragment was recorded for each DNA sample. Similarity among strains was estimated by using NTSYSpc 2.01 program (Rohlf 1994). The díversity of Xam strains from each locality and from the entire collection was calculated by 1 the equation H = [ n/(n-1 )]( 1-SX¡ ), where X¡ is the proportion of the ith distinct pthB haplotype within a group and n is the number of strains in each group (Nei and Tajima, 1981). The percentage of total variance due to differences within and among localities was calcu1ated using Arlequín 2.0 program (Schneider and Excoffier 2000). Results RFLP ana1ysis For the 218 Xam strains analyzed, 17 bands were observed using pthB fragment as a probe. The molecular weight ranged from 4.6 to 14 kb. Nine different haplotypes (Ht) were defined Ofthese groups, one included only one strain while two others are represented by 49 and 31 strains respectively (Table 1). Genetic diversities (H) varied from O (localities C and F) to 1 (locality 1). The total diversity for Toga was 0.66. Table l. RFLP aoalysis showing strains, number of strains, number of RFLP groups and genetic diversity found in eacb locality. Locality No of Haplotypes Haplotype b No. of straios per locality 1 T7 24 o 3 T3, T5, T7 21 0.63 1 T7 5 o 3 T3 , T7, T8 8 0.46 2 T3, T7 13 - 0.28 1 T3 6 o 4 TI,T2,T3,T8 13 0.68 3 T3, T6, T7 49 0.41 2 T6, T9 2 1 5 TI, T2, T3, T6, TI 23 0.75 3 TI, T3, T6 16 0.64 1 T7 1 ND• 5 T2, T3 , T4, T6, T8 6 0.93 4 T3, T6, T7, T8 31 0.55 H " Localíties: A, Tove; B, eraf-Kpa1irne; e, Beme; D, Danyi-Apeyeme; E, Danyi-Apeyeme b; F, Zogbeguan; G, Adeta; H, SotovBova a; 1, Blitta; J, SotovBova; K, Piya; L, Kpaka-Doutelbou; M, Landa- Pagouda; N, Davie. bDesignation ofthe haplotype, T = Togo. • ND: not detennined. 41 The cluster analysis revealed the existence of 7 groups at 70% similarity (data not shown). Clusters grouped strains collected from different localities, except cluster 2 that grouped strains from Craf-Kpalime and clusters with only one strain. In general, no correlation between the geographical origin ofthe strains and DNA polymorphism was observed. No polymorphism among strains was obtained with pBS6 and pBS8 with all strains showing a unique haplotype with each probe. Analysis oj molecular variance Percentage of total variation was determined among and within populations (localities). We observed that there was more variability within population, 63.6%, that among them, 36.4% (Table 2). Diversity is distributed in a microgeographical scale at the locality leve!. Table 2. Pairwise differences (AMOV A) calculated from RFLP results within and among populations (loca li ti es). Source of d.f. Varían ce Percentage Variation components of variation Among Populations 13 0.41268 Va 36.41 Within Populations 207 0.72064 Vb 63.59 Total 220 1.13332 Fixation lndex FST: 0.36414 Future plaos • To test the pathogenicíty of al! the strains and characterize the pathotypes of representative strains. • To recommend a set of strains representative of the genetic diversity to be used for breeding assays. This work was done with the support of the European project on Cassava bacteria! blight in Africa that will be finished at the end of 200 l . Refereoces Nei, M., and F. Tajima. 1981. DNA polymorphism detectable by restriction endonuc1eases. Genetics 97:145-163. Restrepo, S., Valle, T.L., Duque, M.C., and Verdier, V. 1999. Assessing genetic variability among Brazilian strains of Xanthomonas axonopodis pv. manihotis through restriction fragment Jength polymorphism and amplified fragment length polymorphism analyses. Can. J. Microbio!. 45 :754- 763. Restrepo, S., and Verdier, V. 1997. Geographical differentiation of the population of Xanthomonas axonopodis pv. manihotis in Colombia. Appl. Environ. Microbio!. 63 :4427-4434. Rohlf, F.J. 1994. NTSYS-pc (Numerical Taxonomy and Mu1tivariate Ana1ysis System), version 2.02. Exeter Software, New York. 42 Schneider, S., Roessli, D., Excoffier, L. 2000. Arlequin ver. 2.000: A Software for population genetics data analyses. Genetic and Biometry Laboratory, University of Geneve, Switzerland. Verdier, V., Boher, B., Maraite, H., and Geiger, J.P. 1994. Pathological and molecular characterization of Xanthomonas campestris strains causing diseases of cassava (Manihot esculenta). Appl.Environ. Microbiol. 60:4478-4486. Verdier, V., Dongo, P., and Boher, B. 1993. Assessment of genetic diversity among strains of Xanthomonas axonopodis pv. manihotis. Joumal ofGeneral Microbiology 139:2591 -2601. Williams, R.J., Agboola, S.D., and Schneider, R.W. 1973. Bacteria! wilt of cassava in Nigeria. Plant Disease Reporter 57:824-827. 1.1.10 Microsatellites to study genetic diversity in Indica and Japonica rice G. Gallego1, P. Rocha2, A. Albarez3, M. Duque1-4 and J. Tohme1 1SB-2 Project~ 2John Innes Center ~ 3CEADEN, Cuba~ 41P-4 Project Introductioo Microsatellites markers have become the markers of choice for a wide spectrum of genetic, population, and evolutionary studies (Jame and Lagoda 1996, Powell et al. 1996). Microsatellite markers are highly infonnative and they can be rapidly and reliably visualized using silver staining. Although they are expensive to develop, once primers are available, they are technically easy and inexpensive to use . .In rice, microsatellites markers are distributed relatively unifonnly throughout the genome and detect a high level of allelic diversity in cultivated varieties and distantly related species. The objective of this study is to initiate the establishment of a sub set of microsatellite primers that can be used to differentiate rice genotypes for selection of parentals in breeding programs. Metbods We selected 120 accessions as follows: 93 Indica, 24 Japonica and 3 of the Surinam type (Table 1 ). Most of the accessions are used Genetic Base of lrrigated Rice in Latín America and the Caribbean 1971 to 1989 for breeding programs . They were characterized on the basis of parentage coefficient (Cuevas et al 1992) and using RAPOs technique (Escobar 1994). We used 12 microstellite primer pairs (Tab1e 2) to characterize accessions. DNA extractions were carried out according to Dellaporta's modified method . PCR reactions were run according to BRU's methods and gels stained using silver staining. Data were collected in two ways: 1) reading presence ( 1) or absence (O) of alle1es to obtain a binary matrix, or 2) using the concept of "pattem" for each primer pair, which finally constitutes a genotype. The statistica1 analysis was done using SAS, according to the criteria ofTessier et al. 1999. The tree (not shown) was developed using NTSYS. Results Polymorphism of microsatellites may be analyzed with two parameters: 43 2 3 4 5 6 7 8 9 10 ]] 12 13 14 15 The Polymorphism Infonnation Content (PIC), where each band corresponds to one allele. PIC is equivalent to H (heterozygosity) and equals = 1-L[¡2 • The average PIC may be calculated as the average of all bands analyzed for all microsatellites. The parameter O (Oiscrimination capacity), which is actually derived from PIC. O estimates the discrimination power of primer pairs, by estimating the probability that two randomly chosen individuals have different band patterns. O may also be used for comparisons between different markers (Tessier et. al. 1999). A total of 89 different bands, or alleles, were found using the 12 primer pairs. The number of different bands par primer pair ranged from 4 in RM7 to 10 in RMII, RM202 and RM225. The estimated average H, using PICas estimator, was 16.7% for all 120 accessions. Within Indica, the average H was 14.3%, while for Japonica and Surínam it was 15.6% and 15.7% respectively. The results support a low variability within rice gennoplasm used for breeding program in Latín Ameríca. The most discriminatory band was number 51, corresponding to primers RM19, which showed 8 alleles. With the analysis of O we can conclude that most primers have a good discrimination capacity (Figure 1), except primers RM7 y RM167 which had low O (30 and 32% respectively). These primers have already already been used for genetic diversity studies among Cuban traditional varieties that are useful for the breeding program in Cuba (Ceaden- CIA T ,2001 ; data not shown). Ongoing Activities • lncreasing the number of microsatellites in the sub set • Establishment of a database for eventual selection of parental fines based on molecular diversity Table l. Rice accessions. Japonicas Altamira-7 24 El Paso L-1444 47 Porvenir 70 lcta Motagua BR- 25 CR-5272 48 Panama 1048 71 Juma 61 IRGA410 BR- 26 CR-1821 49 Palizada A-86 72 Juma 62 IRGA412 Chanca y 27 CR-1707 50 IR665-23-3 73 Metica l Altomayo88 28 CR-1113 51 IR1529 74 Morelos A-88 Amazonas 29 Cimarron 52 In ti 75 Navolato A-71 Amistad 82 30 Cica 9 53 INIAP 7 76 Tocumen 5430 Anayansi 31 Cica 8 54 INIAP 6 77 Vi flor Araure 1 32 Cica 7 55 INIAP 415 78 CT6458-9-3-6-M Patselrec Araure 2 33 Cica6 56 INIAP lO 79 CT5756-3-5-1-M Patselrec Araure 3 34 Cica4 57 ICT A Virginia 80 C46-l5 Araure 4 35 Chetomal A-86 58 ICT A Tempisque 81 DGWG Bamoa A-36 Juma 58 59 ICT A Quirigua 82 Ecia 122-58-1-2-l Patselrec 75 BR-IRGA 37 Juma 57 60 ICT A Polochi 83 P4076F3-2-2-4 Patinlen 409 BR-IRGA 38 Juma 51 61 PA-2 84 P285IF4-145-9-58-IB-l0 WC97 415 44 16 17 18 19 20 21 22 23 Camponi 39 J-104 Cea 1 40 IR8 Centa A-l 41 IR84-63-5-18 Cea 3 42 Tanioka Empasc 103 43 San Pedro E m pase 1 02 44 San Martín 86 Empasc 1 O 1 45 Saavedra El Paso L- 46 Rustí e 227 Indicas 1 El Paso L-94 2 El Paso L-48 3 Diamante 4 Guay Quiraro 5 Villguay 6 CT6196-33-li-I-3-M WC 5177 7 Tox 340-1-7-3 (ITA 133)- Ace 4 8 CT6240-12-2-2-l -1 P WC5178 9 Monolaya JO Oryzica - Sabana 6 ll Bluerose 1268 12 Bluerose 1269 Surinam 1 Diwani 2 Ciwini 3 P5589-1-I-3 P-4-MP Patselrec 62 Oryzica Llanos 5 85 P3055F4-3-4P-I P-1 B WC-1 06 63 Oryzica Llanos 4 86 IR.3541 0-16-3-2-2-2-2- Patir BN 64 Oryzica 2 87 IR.35353-94-2-1-3 Patir BN87 65 Oryzica 1 88 P4725F2-9-6-1X WC153 66 Empasc 104 89 Khao Dawk Mali 105 67 Empasc 105 90 L3 68 Huarangopampa 91 C039 69 IAC 1278 92 MCVA 93 Carreon 13 CT6261-5-7-2P-5-1P WC 5212 14 CT7242-16-9-2-M Patselrec 15 lrat 146 Acc 406 16 Tox 1859-102-GM-3 WC 5036 17 CT6393 M-9-2-5-M Patinlen 18 Marog Paroc Rexoro 19 Pachian Fortuna 20 Takao IKU 18 21 Irat 13 22 Colombia 1 23 Fanny 24 Lac 23 Table 2. Microsatellite primers used for genetic diversity analysis. Primers were selected from the kit Primer for Genetic Mapping of the Rice Genome, Research Genetics. Chromosome Chr. 1 Chr.2 Chr.3 Chr.5 Chr.6 Chr.7 Chr.JI Primername RM5 RM6 RM7 RMI3 RM225 RM4 RMI67 RM168 RMll RM202 RM18 45 Chr.I2 RM19 Discriminatlon capacity (O) from each primer according to the number of patterns :;. 08 ········································· ·-- ····+ ·~:f:r ············- RM-20z ·-········· € . -RM-19 RM-22~-11 a_ 8 RM-18 - RM-4 ~ 0.6 ······· ·· · ·-········ . RM:6. RM-5- ---· ········ ·········· --- ·-· ... .. ·· · -·-- ·-···· · ·· e o ;¡ • .E 0.4 E -e u .. e o.2 ;( RM-168 + RM-7 e RM-167 0+------r------r-----~-----,------,-----~----~ o 2 4 6 8 10 12 14 Number of pattems Figure l. Discrimination capacity (D) for each primer pair according to the number of band patterns observed References Jame P, Lagoda PJL (1996) Microsatellites, From molecules to populations and back. Trends Ecol Evol 11: 424-429. McCouch SR, Chen X, Panaud O, Temnykh S, Xu Y, Cho YG, Hunag N, lshii T, Blair M (1997) Microsatellites marker development, mapping and applications in rice genetics and breeding. Plant Mol Biol35:89-99. Chen X, Temnykh S, Xu Y, McCouch SR (1997) Development of microsatellites framework map prividing genome-widw coverage in rice (Oryza sativa L.). Theor Appl Genet 95: 53-567. Tessier C, J. David, Boursiquot J M, Charrier A. (1999) Optimization of the choice of moleGular markers for varietal identification in Vitis vinífera L. Theor Appl Genet 98: 171-177. Escobar fabio.1994.Caracterización la base genética del germoplasma élite de los programas de mejoramiento de arroz de America Latina y el Caribe, durante el periodo de 1971 a l989.TesisPregerado, Universidad del Valle-CIAT. Federico E. Cuevas-Perez, Elcio P. Guirnarares, Luis E. Berrio, and Daniel l. Gonzalez. l992.Genetic Base oflrrigated Rice in Latín America and the Caribbean 1971 to 1989 Crop Sci. 32: l 054-l 059 46 1.1.11 Gene flow analysis for assessing the safety of transgenic rice in the Tropics E. Gonzalez1, L. F ory1, A. Mora2, E. Corredor3, P. Ruiz1, J.J. V asquez1, Z.Lentini 1 1SB-2 Project; 2IP-4 Project; 3Fedearroz -Saldaña Introduction Hybridization between crops and their wild relatives sometimes brings genes into wild populatíons, occasionally resulting in the evolution of aggressive weeds and/ or endangerrnent of rare species. Transgenic crops may result in similar outcomes. The likelihood of crop-to-wild hybridization depends on the out-cross rate, and on distance and direction between wild and crop populations. Cultivated rice, O. sativa L., is an autogamous plant, with a low outcrossing rate of 0-1%. In wild relatives of rice, rates as high as 56% ha ve been reported (Roberts et al. 1961 ). Hybridization can be expected within the genomic group that includes O. sativa, viz., the AA group. The wild re1atives of AA genome which are found in Central and South America and may hybridize with the rice crop include O. rufipogon (AA, hybrid seed set 19% without and 73% with embryo rescue), and O. glumaepatula (AA, hybrid seed set 39% without embryo rescue) (Oka and Chang, 196l;Vaughan and Tomooka, 1999). Gene transfer from O. sativa to O. rufipogon under field conditions has been documented in Asia and is not restricted by reproductive barriers. Spontaneous interrnediates between cultivated rice species and their wild relatives occur frequently in and near rice fields when wild taxa are present. Natural rates of hybridization can be sometimes substantial, and the hybrids usually demonstrate heterosis (hybrid vigor) (EIIstrand et al., 1999). Red rice (Oryza sativa f. spontanea) is a weedy rice with a red pericarp and dark-colored grains. The seeds shatter readily and possess dorrnancy characteristics~ the plants typically are tall, late maturing, and have pubescent leaves and hulls. In contrast to Asia where manual transplanting is still predominant, in tropical America direct seeding of red rice- contaminated seed source is common for a high proportion of rice farmers in Latín America, ensuring field re-infestations and making it one of the most serious weed problems in this region. There are indications that genes placed in cultivated varieties of rice have transferred quickly into red rice. The natural rates of hybridization can range from 1% (with early season variety, flowering at 72-76 days) to 52% (with late season variety, flowering at 82-96 days)( Langevin et al. 1990; Clegg et al. 1993 ). Thus, cultivated varieties that flower and mature late, like those mainly grown in Latín America, may enable hybridization with red rice to occur throughout severa! generations. This work is part of a project directed to analyze the gene flow from non- transgenic or transgenic beans and rice into wild/weedy relatives in the Neotropics, and its effect(s) on the population genetic structure of the recipient species. The research will provide guidelines for evaluating the potential risks of using transgenic plants in the tropics, and describe potential areas of gene(s) flow. The inforrnation will contribute to improve the risk assessment procedures in the region, in particular for the partner countries Colombia and Costa Rica, which both rank among the countries with the highest biodiversity in the world. The current report summarizes the progress attained during the initiation of the project, towards setting up the tools to assess gene flow from transgenic and non-transgenic rice into wild Oryza species and red rice under experimental plots or narura1 field conditions. 47 Materials and Methods A preliminary set of 50 microsatellite markers (at least 4 per each chromosome) were selected. Their selection was based on their location on the chromosome (McCouch et al., 1997). At least two markers located distal from the centromere per each chromosome arm were chosen to increase the likelihood of finding recombination between the experimental genotypes. The genotypes includes 9 rice commercial varieties (Cica 8, Cimarrón, Fedearroz 50, Fedearroz 2000, Fedearroz Victoria 1, Iniap 12, Oryzica 1, Oryzica Llanos 5, and Palmar). Sixteen homozygous transgenic Cica 8 rice lines carrying the RHBV -N transgene for rice hoja blanca virus resistance. Four hand made crosses each between one transgenic Cica 8 line and non-transgenic Cica 8, Iniap 12, Fedearroz 50 or Oryzica 1, respectively. One hand made cross each between non-transgenic Cica 8 and Iniap 12, Fedearroz 50 or Oryzica 1, respectively (controls). One hundred and sixty accessions of red rice collected from commercial rice fields in Saldaña, Tolima (Colombia). One accession each of O. rufipogon, O. barthii O. glaberrima. All these genotypes were included in order to select the microsatellites detecting the highest leve! of polymorfisms among genotypes, and the most polymorphic pairs from each class to conduct the gene flow study from transgenic Cica 8 under experimental conditions, and the corresponding gene flow analysis from commercial varieties under commercial rice field conditions. The markers were amplified at different annealing temperatures according to the estimated melting temperatures ofthe primers. The PCR products were resolved on silver-stained polyacrylamide gels and microsatellite alleles were sized by comparison to the 10 and 25 bp molecular weight standards (Promega). Results and Discussion Microsatellite analysis is still in progress. Here is presented the preliminary results with a sub-set of the population using 23 microsatellite loci . The ftrst data analysis was design to identify macrosatellites allowing to detect polymorphism generated by potential gene flow from transgenic Cica 8 lines into rice varieties, wild species and red rice. The average rate of polymorphism between Cica 8 and the different varieties ranged from 30% with the variety Fedearroz 50 to 87% with the variety Palmar (Table 1). Between 74% and 83% of the microsatellite markers analyzed were polymorphic between Cica 8 and O. rufipogon, O. galberrima, and O. barthii, respectively (Table 1 ). O. barthii and O. glaberrima were al so inc1uded in this study to generate information that may be useful for Asia and Africa. The red rice accession analyzed showed a polymorhism of 56% respect to Cica 8. Only one entry of red rice has been analyzed so far because it was the only one fully characterized morpholog_ically at the initiation of the work. The analysis of the other 159 accessions collected from Tolima (Colombia) is in progress. As expected, results indicate that no polimorphism is detected between non-transgenic Cica 8 variety and transgenic Cica 8 lines (Table 1). Results indicate that the transgenic lines are true-type Cica 8 with the exception ofthe transgenes introgress in the rice genome. Results also suggest that the change incorporated by transgenesis is so small that it is not resolved by microsatellite analysis. In order to detect gene flow from transgenic Cica 8 into non-transgenic Cica 8 will be by tracing the transgenes. In contrast, polymorfism from 30% to 39% is detected in hand made crosses between Cica 8 and the selected varieties (Table 1 ). One interpretation ofthese results is that more genetic changes are introduce by conventional crossing within O. sativa, and even more with wild species, than by transgenesis itself. 48 Table 1.-Polymorphism rate between rice variety Cica 8 and various r ice genotypes for 24 microsatellite loci TI T2 Cl F50 C/ I C/ 0 V CM p F2000 o OL5 Ob No o o 7 9 8 12 10 20 13 11 11 11 19 (%) o o 30 39 35 52 43 87 57 48 48 78 83 Og 18 78 TI= transgenic Cica 8 line A3-49-60-12-3 ; T2= transgenic Cica 8 line A3-49-60-4-13 ; Cl F50 = Cica 8 x Fedearroz 50 ; C 1 I = Cica 8 x lniap 12 ; C 1 O = Cica 8 x Oryzica 1; V = Victoria 1; CM = Cimarron ; P =Palmar ; F2000 = Fedearroz 2000; O= Oryzica 1, 1 = Iniap 12 ; OL5 = Oryzica Llanos 5 ; Ob = Oryza barthii ; Og = Oryza glaberrima ; Or = Oryza ruffipogon ; RR = red rice Future Activities Studies to define the red rice/ rice wild relatives complex in the crop contact zone is important to design biosafety guidelines for the Neotropical region. The spatial distribution of alleles will be used to study local gene flow, including polleo dispersa! distances. Microsatellite will be used to trace crop-to wild/red rice gene flow and red rice/wild-to-crop hybridization rate under confined experimental settings as well as under natural conditions. Similar analyses will be conducted to assess transgenic-to non-transgenic variety gene flow. In arder to select the red rice genotypes to use for the gene flow studies, a detailed phenology analysis of the 160 red rice accessions is underway to determine those genotypes more prompt to hybridize with cultivated rice. An accession of O. glumaepatu/a was just received from Costa Rica. Future analysis will also include individuals from this species. References Ellstrand, N.C., Prentice, H.C., and Hancock, J. F. 1999. Gene flow and introgression from domesticated plants into their wild relatives. Annu. Rev. Ecol. Syst. 30: 539-563. Langevin, S. A., Clay, K. and Grace, J. B. 1990. The incidence and effects of hybridization between cultivated rice and its related weed red rice (Oryza sativa L.). Evolution 44 (4): 1000-1008. McCouch S.R., X. Chen, O. Panaud, S. Temnykh, Y. Xu, Y.G. Cho, N. Huang, T. Ishii, and M. Blair. 1997. Microsatellite marker development, mapping and applications in rice genetics and breeding. Plant Molecular Biologyu 35: 89-99. Oka, H.I., and W.T. Chang. 1961. Hybrid swarms between wild and cultivated rice species, Oryza perennis and O. sativa. Evolution 15: 418-430. Roberts, E.H., R.Q. Craufurd, and F. Le Cochet. 1961. Estimation of percentage of natural cross- pollination: experiment on rice. Nature 190: 1084-1085. Vaughan, D.A., and Tomooka. 1999. Varietal Differentiation and Evolution. Wild rice in Venezuela. Rice Genetics Newsletter 16: 15-17. 49 Or RR 17 13 74 56 1.1.12 Genetic diversity and core collection approaches in the multipurpose shrub legumes Flemingia macrophylla and Cratylia argentea M. Anderson1, M. Peters1, J. Tohme2, G. Gallego2; R. Schultze-Kraft1, L. H. Franco3 1University ofHohenheim; 2SB-2 Project; 3IP-4 Project 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 strategy of the CIA T Forages team. C. argentea is increasingly adopted and utilized, particularly in the seasonally dry hillsides of Central America. However, research and development are based on only few accessions and hence activities to acquire and test novel germplasm of C. argentea is of high priority. F. macrophyl/a is also a highly promising shrub legume with excellent adaptation to infertile soils. In contrast to C. argentea, whose adaptation is limited toan altitude below 1200 m as!, F. macrophylla can successfully be grown up to altitudes of 2000 m asl. However, the potential utilization ofF. macrophylla is so far limited by the poor quality and acceptability of few evaluated accessions. The project aims to investigate the genetic diversity ofF. macrophylla and C. argentea with three maln objectives. Firstly, to identífy new, superior forage genotypes based on conventional germplasm characterization/evaluation procedures (morphological and agronomic traits, forage quality parameters, including IVDMD and tannin contents). Secondly, to optimize the use and management, including conservation, of the collections. For this, different approaches to identify core collections for each species will be tested and compared based on, respectively: (a) genetic diversity assessment by agronomic characterizatíon/ evaluation; (b) germplasm origin information; and (e) molecular markers (AFLPs). Thirdly, to create a planning basis for future germplasm collections with respect to methodology, geographical focus and genetic erosion hazards. Methods Agronomic characterization and eva/uation: 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 for (Flemingia macrophyl/a, 73 accessions). Additionally two replications were sown for seed production and morphological observations. The following parameters will be measured in the trials: vigor, height and diameter, regrowth, incidence of diseases, pests and mineral deficiencies, dry matter yield during wet and dry seasons. For the morphological evaluation qualitative and quantitative parameters are measured, such as days to frrst flower, days to first seed, flower color, flowers per inflorescence, flowering intensity, pod pubescence, seeds per pod, seed color, branching capacity, leaf length and width, ratio, peduncle 1ength, etc. For the analysis of nutritive value, crude protein content and in vitro dry-matter digestibilíty (IVDMD) of the entire collections will be analyzed. For F. macrophylla, a more detailed analysis will be conducted of a representative subset which will include high nutritive value accessions (high crude protein content, high IVDMD) as well as intermediate and low nutrítive value accessions. The subsequent analysis wiH comprise fiber 50 (NDF, ADF), condensed tannin and hydrolysable tannin contents, tannin purification, calcium, phosphate, ash and organic matter contents. 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 o[ available origin information: Based on ecogeographical inforrnation of accession origins, a core collection will be created, hypothesizing that geographic distances and environmental differences are related to genetic diversity. The analysis will be conducted with FloraMap™, a GIS too! developed by CIA T, which allows the production of climate probability models using Principal Component Analysis (PCA) and Cluster Analysís. Genetic analvsis by molecular markers (AFLPs): The genetic analysis is conducted using AFLP molecular marker technique (Vos et al, YEAR). Based on the results a core collection will be created, using multivariate statistic tools (PCA and Cluster Analysis). Data analysis and zynthesis: Individual and combined data analyses of all generated inforrnation will be perforrned, including the use of GIS tools and multivariate statistics. In the analysis of each of the different approaches (agronomic characterization, origin inforrnation, molecular marker analysis), PCA and Cluster Analysis is utilized to create core collections. Eventual correlation between the different approaches and clusters obtained is evaluated. The resulting concept is expected to help deciding which of the three methods or which combination is most appropriate (time and cost efficient) to crea te a core collection, depending on availability of time and fmancial resources. E.g., if an agronomic evaluation is not feasible because of time constraints, a core collection may be created using origin inforrnation and/or molecular marker analysis. Based on molecular marker similarities and the GIS analysis, suggestions will be provided for focussing future collections on areas with particularly high diversity, and for collection (= sampling) strategy improvements (e.g., regarding sampling frequency; roadside collections). Accession duplicates in the world collections will be identified. The project is expected to last two growing seasons and to be terrninated at the end of2003. Results and Discussion Agronomic characterization and evaluation: Preliminary data of one evaluation cut in the dry season and one in the rainy season show considerable phenotypic and agronomic variation for Craty/ia argentea (Table l) and Flemingia macrophylla (Table 2). For Cratylia argentea IVDMD varied between 61 and 67% and crude protein content between 18 and 21%. Mean dry matter production of Cratylia argentea was 45 glplant in the wet and 60 glplant in the dry season. According to these initial results the accessions 18674, 22375, 22406, 22408 and 22409 had the highest dry matter yields with between 68 and 100 glplant. Productivity of these accessions were substantially higher than yields of the material advanced for cultivar release in Costa Rica - an accession mix of 18516/18668. The tria! is on-going and quality analyses are pending but preliminary results indicate the potential to identify materials of superior performance to accessions 18516/ 18668. Principal component analysis perforrned with the agronomic data of 39 accessions of Cratylia argentea revealed high correlations between total dry matter production, diameter, rebrotes and 51 vigour (>70%). Cluster analysis (UPGMA) resulted in 9 clusters. 5 ofthe clusters contained only one accession, among them three of the most productive accessions ( 18674, 22406 and 22408) (Table 3). For Flemingia macrophylla IVDMD varied between 31 and 51% and crude protein content between 16 and 24%. Mean dry matter production of Flemingia macrophylla was 60 glplant in the wet and 42 glplant in the dry season. The most productive accessions were C 104890, 21090, 21241,21529 and 21580 with a total dry matter production > 100 glplant. Principal component analysis performed with the agronomic data of 73 accessions of Flemingia macrophy/la revealed high correlations between total dry matter production, hight, diameter and vigour (>70%). Cluster analysis (UPGMA) resulted in 7 clusters. Two of the clusters contained only one accession, among them one of the most productive accessions (21 090) (Table 4). Based on these preliminary feed quality results the following subset has been chosen for subsequent analysis of NDF, ADF, condensed tannin and hydrolysable tannin contents, tannin purification, calcium, phosphate, ash and organic matter contents: 17403, 17407, 18437, 18438, 19457,20065, 20616,20621,20622, 20744,20975,20976,21083,21087,21090,21092,21249, 21529, 21580, 21982, 21990, 21992, 22082, JOO 1 (total of 24 accessions; 9 erect, 11 semierect, 4 prostrate ). Analysis o[ available origin in{ormation: Cluster analysis (UPGMA) was performed with FloraMap™ on the data of 37 accessions of Cratylia argentea and 62 accessions of Flemingia macrophy/la (Figure 1). A first comparison with the clustering according to agronomic data at the leve) of9 (Cratylia argentea), resp. 7 clusters (Flemingia macrophylla) showed no correlation. Genetic analysis by molecular markers (AFLPs): Samples of 5 g ofyoung leaves have been taken of all Craty/ia argentea and Flemingia macrophylla accessions and the DNA has been extracted and . quantified (Table 5). To identify efficient primers for the AFLP analysis, 2 supposedly genetically contrasting accessions of each F. macrophy/la and C. argentea (21990, 21529 and 18672 and 18516 respectively) have been tested with different primer combinations and the resulting polymorphic bands have been counted (Table 6). 52 Table l. Agronomic evaluation of a collection of Cratylia argentea in Quilichao. Preliminary data of four cuts (two in the dry season and two in the wet season). Treatment Height Diameter Regrowing Mean dry matter i:ields IVDM Crude points Wet Dry Total D protein No. CIAT ~cm~ (cm~ (No.~ (~El) (%) (%) 18516 112 105 19 55 78 66 65.03 20.68 18667 112 101 18 45 68 56 64.62 20.36 18668 106 110 17 48 68 58 65.20 19.88 18671 111 106 20 54 55 54 64.30 18.26 18672 96 83 13 34 39 37 62.12 20.10 18674 liS 122 23 91 109 lOO 63.88 19.98 18675 ll2 97 15 47 63 55 63.34 19.03 18676 105 93 14 46 50 48 61.16 19.66 18957 lll 102 16 50 76 63 62.47 20.08 22373 109 93 15 38 57 48 64.37 20.18 22374 116 102 17 55 71 63 66.39 19.57 22375 125 98 16 59 76 68 67.03 2l.l8 22376 95 70 ll 23 36 29 64.14 19.61 22378 103 81 12 34 39 36 61.75 19.76 22379 lll 89 16 47 65 56 63.51 19.60 22380 107 90 ll 31 43 37 61 .31 20.42 22381 105 85 ll 34 46 40 63.98 19.15 22382 llO 92 12 41 62 52 64.16 20.44 22383 99 90 13 34 43 39 62.57 18.60 22384 ll3 91 9 43 47 45 64.52 18.89 22386 lll 86 12 39 47 43 64.67 18.62 22387 lll 90 12 41 57 49 62.55 19.10 22390 99 92 13 45 47 46 64.81 18.49 22391 108 96 15 44 62 53 63 .36 18.88 22392 ll4 83 13 33 53 43 63.19 21.04 22393 IIO 92 17 41 58 49 63.52 20.65 22394 ll2 88 13 33 46 40 63.96 20.52 22396 lO l 79 10 30 43 36 63.78 21.32 22399 102 86 13 35 42 39 66.13 19.85 22400 119 104 16 52 74 63 61.66 20.72 22404 110 97 13 42 68 55 67.00 20.91 22405 111 96 16 41 61 51 62.86 19.91 22406 113 112 20 63 86 74 62.59 20.96 22407 111 99 16 46 59 53 65.28 20.81 22408 120 109 18 69 88 79 67.16 20.06 22409 113 115 17 57 81 69 66.46 21 .25 22410 ll6 96 14 42 60 51 64.28 19.80 22411 103 88 14 37 58 47 64.46 20.23 22412 ll6 90 ll 42 63 52 64.91 18.66 Mean 110 95 15 45 60 52 64.06 19.93 Ranse 95-125 70-122 9-23 23-91 36-109 29-100 61-67 18-21 53 Table 2. Agronomic evaluation of a collection of Flemingia macropllylla in Quilichao. Treatment Height Diameter Regrowing Mean dry matter ~ields IVDMD Crude points Wet Dry Total protein No. CIAT ~cm) ~cm~ (No.~ c~eQ (%) ~%~ J 001 (e) 125 85 30 102 58 80 40.07 22.33 801 (e) 125 90 29 103 62 82 36.27 22.89 7184 (e) 124 95 34 101 82 92 33.97 21.35 e 10489 108 99 34 121 79 lOO 33.65 22.74 (e) 1 15146 (e) 98 70 24 103 58 80 39.88 22.85 17400(s) 63 98 33 55 52 53 33.23 21.5 17403 (s) 67 96 32 68 57 62 35.84 22.17 17404 (s) 58 79 32 46 45 45 32.91 22.46 17405 (s) 65 94 36 71 67 69 36.12 21.93 17407 (s) 78 106 39 87 62 74 32.85 21.87 17409 (s) 56 109 35 87 66 77 33.00 20.21 17411 (s) 55 86 33 56 54 55 35.55 22.36 17412 (s) 73 96 39 61 63 62 38.63 20.21 17413(s) 58 93 35 51 39 45 35.19 20.1 l8048(s) 32 43 19 12 8 10 42.84 20.38 18437 (s) 54 101 37 57 55 56 47.85 22.55 18438 (s) 58 71 31 36 22 29 51.46 23.46 18440 (s) 59 87 38 65 44 55 33.39 21.35 19453 (e) 105 78 20 65 33 49 36.03 21.64 19454 (e) 115 82 24 73 52 63 39.14 19.66 l9457(e) 116 85 25 52 64 58 33.09 21.3 19797 (s) 57 90 22 58 46 52 38.50 20.98 19798 (s) 55 95 27 61 55 58 38.27 20.92 19799 (s) 50 69 19 28 39 33 37.34 21.68 19800 (s) 65 85 29 34 48 41 32.00 20.71 19801 (s) 82 91 40 68 57 63 35.74 21.75 19824 (e) 62 93 35 54 61 58 35.57 21.33 20065 (p) 15 21 4 o 1 l 32.14 18.86 20616 (s) 67 108 34 86 54 70 32.10 21 .99 20617 (s) 72 92 27 51 44 48 30.64 20.14 20618(s) 74 95 31 57 60 58 33.23 21 .57 20621 (e) 84 88 32 58 54 56 31.59 21.57 20622 (e) 146 88 30 105 76 91 42.81 22.86 20624 (s) 74 122 39 IOJ 91 96 34.51 19.83 20625 (e) 128 86 26 105 69 87 42.49 22.81 20626 (e) 115 92 28 88 70 79 39.55 22.3 20631 (e) 121 90 25 97 75 86 41.46 20.92 20744 (e) 125 87 27 102 65 84 42.95 23.09 20972 (p) 24 56 31 12 14 13 39.65 23.42 20973 (p) 24 45 17 4 lO 7 34.17 19.65 20975 (s) 52 83 45 44 24 34 45.28 20.31 20976 (s) 45 57 27 17 ll 14 40.79 20.01 20977 (s) 33 35 9 5 4 4 46.14 18.46 20978 (s) 52 56 24 21 ll 16 46.61 22.08 20979 (s) 48 76 38 27 25 26 38.90 2l.l7 20980 (s) 43 55 26 27 18 22 41.80 20.95 20982 (s) 49 61 28 26 23 25 40.96 19.94 54 Treatment Height Diameter Regrowing Mean dry matter ;tields IVDMD Crude points Wet Dry Total pro te in No. ClAT ~cm2 (cm) ~o.) ~~EQ ~%2 ~%2 21079 (s) 47 78 44 51 25 38 37.87 20.17 21080 (s) 41 58 13 32 11 21 39.21 15.51 21083 (e) 93 79 36 71 43 57 45.78 21.43 21086(s) 27 29 4 o 3 3 21087 (s) 64 66 46 47 32 39 42.32 20.37 21090 (s) 88 106 48 135 66 100 50.02 21.33 21092 (s) 72 81 23 57 39 48 49.15 18.01 21241 (e) 133 93 27 134 66 100 36.17 20.19 21248 (e) 127 92 30 106 77 91 33 .52 23.58 21249 (e) 129 104 34 167 85 126 40.86 21.95 21519(e) 127 101 28 109 67 88 39.52 22.32 21529 (e) 132 102 31 145 71 108 42.01 23.11 21580 (e) 131 1 o 1 32 184 86 135 39.08 19.83 21982(p) 19 62 38 26 11 19 42.13 20.86 21990 (p) 35 66 43 27 19 23 31.94 19.13 21991 (p) 29 52 24 13 10 11 37.47 22.59 21992 (p) 29 50 24 12 9 11 48.52 20.24 21993 (s) 42 77 45 34 24 29 43.90 19.88 21994 (p) 24 42 9 8 7 8 34.53 16.43 21995 (p) 29 50 26 11 8 9 40.80 19.6 21996 (p) 23 44 14 7 6 6 42.49 21.8 22058 (e) 84 58 13 41 29 35 37.20 18.48 22082 (s) 79 82 58 69 37 53 48.35 20.01 22087 (p) 27 51 17 15 4 10 40.34 17.78 22090 (s) 44 47 10 10 5 7 41.1 o 17.52 22285 (s) 43 75 42 32 21 27 38.72 20.43 22327 (s) 41 62 33 20 21 21 48.41 19.28 Mean 70 78 29 60 42 51 39.01 20.91 Range 15-146 21- 122 4-58 0-184 1-91 1-135 31-51 16-24 Preliminary data of two cuts (one in aech season). Growth habit: e= erect, s = semierect, p = prostrate. 55 Table 3. Cluster ana lysis for Cratylia argentea according to agronomic and climatic (Fioramap) data. ID CIAT-No. Elevation Longitude Latitude Cluster Cluster (agron. data) (Floramap) 10 22373 780 -46.3833 -14.0833 1 3 12 22375 620 -46.6167 -13.0000 1 3 36 22411 550 -46.8833 -13.1167 4 24 22392 240 -56.0000 -15.6333 6 26 22394 270 -57.8167 -15.8667 6 27 22396 510 -5 1.0500 -16.3833 6 25 22393 210 -56.1333 -15.7000 8 31 22405 700 -46.8833 -13.1167 8 33 22407 500 -46.8833 -13 .1167 1 8 35 22409 540 -45.0000 -20.0167 1 8 1 185 16 800 -46.4167 -13.3667 2 1 4 18671 230 -56.9333 -14.6833 2 3 11 22374 660 -46.4167 -13.2833 2 3 2 18667 460 -55.6667 -15.6333 2 5 3 18668 180 -56.2167 -15.4167 2 5 15 22379 390 -49.2000 -13.4167 2 5 22 22390 300 -54.8833 - 16.0167 2 5 5 18672 140 -55.2333 -3.7500 3 3 9 18957 350 -48.6 167 -6.5000 3 3 14 22378 390 -49.0333 -13.6333 3 3 7 18675 380 -52.3167 -14.9167 3 5 16 22380 360 -49.5167 -13.2333 3 5 17 22381 365 -50.0667 -13.2500 3 5 18 22382 330 -50.2000 - 13.2833 3 5 20 22386 320 -52.3500 - 14.5667 3 5 19 22384 360 -52.1667 - 14.2333 3 7 23 22391 660 -55.5000 - 15.8167 3 7 21 22387 370 -52.4167 -15.8333 3 8 29 22400 560 -46.6667 -13.1667 3 8 37 22412 400 -51.7333 -15 .8667 3 9 28 22399 660 -46.4000 -13.5000 4 8 30 22404 700 -46.8833 -13 .1167 4 8 13 22376 580 -49.1500 -14.3833 5 5 6 18674 320 -52.3333 - 14.5667 6 5 32 22406 780 -46.8833 -13 .1167 7 8 34 22408 810 -46.8833 -13 .1167 8 8 8 18676 450 -51.6333 -16.5667 9 2 56 Table 4. Cluster analysis for Flemingia macropllylla according to agronomic and climatic (Fioramap) data. ID CIAT-No. Elevatíon Longitude Latítude Cluster Cluster ~a~on. data~ ~Floramae2 lO 18048 150 109.1000 19.3833 1 3 19 19799 150 102.0333 -2.1333 1 3 32 20972 50 110.4333 18.9167 1 3 58 22087 160 103.6833 18.3333 1 3 35 20976 330 109.5000 18.7500 1 4 36 20977 220 109.4667 18.9167 1 4 51 21991 40 108.8333 15.4000 1 4 54 21994 700 108.2167 14.0000 1 4 55 21995 800 107.7667 12.4167 1 4 59 22090 170 104.2500 17.8833 1 4 33 20973 70 110.3333 18.7833 1 5 39 20980 200 109.1000 19.3833 1 6 40 20982 140 109.5667 19.5000 1 6 42 21080 370 99.1333 17.6500 1 6 56 21996 450 107.4000 11.9333 1 6 1 17400 160 102.8000 16.6833 2 1 5 17407 50 99.9167 8.3000 2 1 17 19797 180 100.0167 0.5667 2 1 2 17403 40 99.0667 10.0667 2 2 3 17404 50 99.1833 9.4167 2 2 4 17405 70 99.3833 8.9833 2 2 6 17409 70 100.2833 6.6500 2 3 7 17411 110 102.4333 3.3333 2 3 8 17412 50 102.4167 5.7667 2 3 9 17413 30 102.1833 6.0500 2 3 13 18440 30 102.4833 12.2833 2 3 18 19798 190 100.8333 -0.7333 2 3 20 19800 140 103.6000 -3.7833 2 3 21 19801 520 103.4167 -3.9500 2 3 22 19824 130 100.6000 -0.6500 2 3 23 20616 690 96.7500 4.7333 2 3 24 20617 560 96.7000 5.0167 2 3 25 20618 250 95.4833 5.4000 2 3 26 20621 1100 97.6167 3.7000 2 3 28 20624 30 99.7333 l.l833 2 3 38 20979 370 109.3833 19.2333 2 4 49 21982 70 107.5333 16.4000 2 4 50 21990 80 108.3667 15.4667 2 4 60 22285 80 108.9833 12.5500 2 6 41 21079 250 99.4500 14.7667 2 7 14 19453 1100 146.5667 -6.9500 3 1 15 19454 650 146.5833 -7.2167 3 l 16 19457 1630 145.3667 -6.0333 3 3 27 20622 1350 98.5333 2.5667 3 3 29 20625 270 100.0833 0.2500 3 3 31 20631 550 98.7000 3.1000 3 4 62 22058 370 101.2000 17.0667 3 4 30 20626 250 99.2500 2.7667 3 5 48 21529 750 110.3333 -7.7500 3 6 l1 18437 190 100.8667 -0.4833 4 3 57 ID CIAT-No. Elevation Longitude Latitude Cluster Cluster (agron. data) (Floramap) 12 18438 40 101.7333 12.7833 4 3 57 22082 190 103.1667 18.2333 4 3 45 21087 700 98.8667 18.8500 4 4 53 21993 150 109.0000 14.4000 4 4 61 22327 80 106.6000 11.0833 4 4 34 20975 230 109.2833 18.8167 4 6 37 20978 250 109.4667 19.1667 5 4 43 21083 620 97.9333 18.6667 5 6 47 21092 500 99.9000 20.4333 5 6 52 21992 40 108.8667 14.8500 5 6 46 21090 550 99.5167 19.2000 6 4 44 21086 490 98.8667 18.8333 7 Table 5. Total DNA c:onc:entrations (ng!ml) of Cratylia argentea and Flemingia macrophylla. Cratylia argentea Flemingia macrophylla No. CIAT Total DNA No. CIAT Total DNA No. ClAT Total DNA 18516 82 801 430 20977 328 18667 888 7184 165 20978 275 18668 318 17400 630 20979 413 18671 275 17403 620 20980 280 18672 233 17404 495 20982 282 18674 148 17405 235 21079 584 18675 138 17407 281 21080 222 18676 608 17409 271 21083 334 18956 456 17411 443 21086 22373 245 17412 370 21087 230 22374 339 17413 411 21090 548 22375 200 18048 603 21092 361 22376 529 18437 596 2124 1 246 22378 197 18438 601 21248 403 22379 100 18440 347 21249 191 22380 357 19453 167 21519 ·726 22381 247 19454 411 21529 321 22382 607 19457 443 21580 299 22383 147 19797 294 21982 267 22384 427 19798 357 21990 413 22386 282 19799 417 21991 131 22387 222 19800 572 21992 397 22390 127 19801 379 21993 286 22391 177 19824 389 21994 185 22392 181 20065 159 21995 129 22393 361 20616 339 21996 442 22394 180 20617 354 22058 352 22396 83 20618 255 22082 491 22399 215 20621 467 22087 237 22400 476 20622 329 22090 375 22404 208 20624 265 22285 330 22405 242 20625 442 22327 472 58 Cratylia argentea Flemingia macrophy/la No. CIAT Total DNA No. CIAT Total DNA No. CIAT Total DNA 22406 234 20626 193 C-104890 518 22407 306 20631 316 1-15146 625 22408 438 20744 393 JOOI 463 22409 115 20972 469 22410 242 20973 381 22411 476 20975 447 22412 282 20976 304 Table 6. Polymorphic bands of different primer combinations for Flemingia macropllylla (accessions 21990 and 21529) and Cratylia argentea (accessions 18672 and 18516). Primer combination Polymorphic bands F. macrophy//a C. argentea Total E-AAC 1 M-CAA n.a. n.a. n.a. E-AAG 1 M-CAA n.a. n.a. n.a. E-AAG 1M-CA T 28 / 24 2 / 4 58 E-A CA 1M-CA T 18 120 7 / 9 54 E-A CA 1 M-CTG 15 1 8 4 / 5 32 E-ACT 1 M-CTG 13 1 8 4 15 30 E-ACC 1 M-CAG . 20 1 15 9 / 8 52 E-ACG 1 M-CAG 16 124 2121 62 E-ACG 1 M-CAC 26 / 24 19 1 12 81 E-AGC 1 M-CTA 11 1 18 3 /3 35 E-AGG 1 M-CTC 24121 9 1 12 66 E-AAC 1 M-CTI 45 / 20 18 1 3 86 1.1.13 Use of molecular techniques for the studies of the genetic diversity and conservation studies of endangered palms in Colombia E. Gaitán1, R. Bemal2 and J. Tohme1 1SB-2 Project; 2Universidad Nacional Introduction There are approximately 200 genera and 1500 species ofpalms in the wor1d. The palm genera are endemic to major continental areas, none of them being pantropical. In the Americas 67 genera and 550 species occur naturally (Henderson et al., 1995). A broad spectrum of uses have been described, ranging from foods and nutritional beverages, sugar and starch to construction materials, oil, fuels, fibers, rattan and omamentals. It is precisely because of their usefulness that sorne ofthese species have become endangered. Examples ofthis are three endangered species in Colombia. Two of these species belong to the Ceroxylon genus: C. alpinum and C. sasaimae. The genus Ceroxylon contains sorne of the most spectacular American palms, including sorne of the tallest palms in the world and those growing at the highest elevations. Ceroxylon has 11 species, distributed throughout the Andes-from Venezuela and Colombia to Ecuador, Peru and Bolivia. The young leaves of C. alpinum and C. sasaimae, known as "palma de cera, palma de 59 ramo or palma real," are used for religious celebrations on Palm Sunday. The habitat of these species has been extensively deforested and transformed into agricultura! land, mainly coffee plantations. Given the foregoing, the survival of these species is severely threatened. The third endangered species is Atta/ea amygdalina or "almendrón del Rio Cauca," whose habitat is found in the Cauca River Valley of Colombia. Its seeds, which are edible, were recommended by Ruiz ( 1984) as promising economically because of their high oil content. Most of its habitat has been converted into coffee plantations, and the species is endangered (Berna!, 1989). In collaboration with the Universidad Nacional de Colombia-Bogotá and the Instituto Alexander von Humboldt we started a project to generate microsatellites makers for palms and to study the diversity of these three species. MateriaJs and Metbods Construction of an enriched microsatellite library of C. alpinum, C. sasaimae andA. amygdalina. Enriched microsatellite libraries for all three species were constructed as described by Edwards et al. ( 1996). This in volved the digestion of 200 ng of genomic DNA with Rsa l. An Mlul adaptar was ligated to the digested fragments. Filter-immobilized oligonucleotides representing the CT20 and GT2o SSR marker classes were used to select for the genomic fragments containing SSRs. Enriched fragments were amplified by PCR, using the 21-mer adaptor primer. The enriched DNA was then digested with Mlul and ligated into a modified pUC 19 vector, pN 1 containing a BssHI site (K.J. Edwards, unpublished). Plasmids were transformed into DHSa. Genomic libraries were screened with a mixture of radio-labeled oligonucleotides (CT20 and GT20). Putative positive colonies were ·cultured, and plasmid DNA was isolated from the culture wíth a QIAGEN plasmid purification kit. Sequencing of the purified plasmid DNA fragments was done on the Applied Biosystems' ABI 377 sequencer model. This was from the M13 primer sites, using a BigDye Terminator Cycle Sequencing Kit (Applied Biosystems), according to the manufacturer's instructions. Each sequence was aligned against all the others, using the SEQUENCHER program in arder to eliminate redundant clones. Primers were then designed for the unique clones using the PRIMER3 .0 software program (available at http:/ /waldo. wi .mit.edu/cgi-binlprimer/primer3 ). Microsatellite primer characterization. The SSRs were first amplified from plasmid DNA and the source genotype to standardíze the PCR conditions. The PCR reaction was carried out in a 20-~l final volume containing 20 ng of genomic DNA, 0.1 J.lM of each of the forward .and reverse primers, 10 mM Tris-HCI (pH 7.2), 50 mM KCI, 1.5-2.5 mM MgCh (depending on the primer combination), 250 mM of total dNTP, and 1 unit of Taq DNA polymerase. The temperature cycling profile involved an initial 2-min denaturation step at 94°C. This was then followed by 35 cycles, each of which consisted of denaturation at 94°C for 15 s, an annealing phase of 48-65°C (depending on the annealing temperature for the given primer pair) for 15 s, andan extension at 72°C for 15 s. The PCR products run on 6% denaturing polyacrylamide ge1s (19:1 acrylamide: bis-acrylamide) contained 5 M urea and 0.5 XTBE. Electrophoresis was at 100-W constant power for 2-2.5 h. PCR amplifications were visualized by silver staining according to the manufacturer's guide. Results of Isolating SSR clones and PCR evaluation C. a/pinum library. Of the 1152 clones screened with oligonucleotide probes CT20 and GT20, a total of 198 putative positive colonies (17.2%) were isolated, 30 of which were sequenced. Of these, only 17 were suitable for designing primers. We found 50% for both CA and GA. On 60 average, dinucleotide motifs had a maximum number of repeats (30). A total of 27 primer pairs including sorne primer pairs from the C. sasaimae library are suitable for amplification with 123 individuals. Currently, 10 ofthem have been amplified in a whole population, but only 3 ofthem show polymorphism. C. sasaimae library. Ofthe 1152 clones screened with oligonucleotide probes CT20 and GT20, a total of 99 putative positive colonies (8.5%) were isolated, and 27 were sequenced. Of these, only 11 were suitable for designing primers. We found 50% for both CA and GA. On average, dinucleotide motifs hada maximum number of repeats (38). Ten primer pairs were suitable for PCR amplification. A total of 99 genotypes of a C. sasaimae population from Sasaima (Cundinamarca) were amplified with these 1 O primer pairs and another 1 O from the C. alpinum library. Four primer pairs were monomorphic for this population, and 17 were polymorphic (from 2-7 alleles), with a total of 56 alleles in a whole population. A. amygda/ina library. Ofthe 1152 clones screened with oligonucleotide probes CT20 and GT20, 290 putative positive colonies (25%) were isolated. Of the 29 sequenced, 23 were suitable for designing primers. We found 50% for both CA and GA. On average, dinucleotide motífs hada maximum number of perfect repeats (30). We have started the evaluation of a set of 123 individuals. Ongoing Activities • Continue evaluating the. rest of the primer set in 123 individuals of C. a/pinum • Evaluate a set of23 primer pairs obtained from A. amygdalina on 123 individuals • Establish a database for the molecular data to be linked with ecological information References Bemal, R.G. 1989. Endangerment ofColombian palms. Príncipes 33(3): 113-128. Edwards, K.J. et al. 1996. Microsatellite libraries enriched for severa[ microsatellite sequences in plants. Biotechniques 20(5):758-760. Henderson, A.; Galeano, G.; Bemal, R. 1995. Field guide to the palms ofthe Americas. Princeton Univ. Press, Princeton, NJ. Henderson, A. 2000. Bactris (Palmae). Flora Neotropica Monograph 9:1-181. Ruiz, M. 1984. Contribución al conocimiento de la palma de almendrón. Cespedesia 13: 139-151 . 61 1.1.14 Molecular and agro-morphological characterization of the genetic variability of Soursop (Annona muricata L.) accessions and related Annonaceus species Nelson Royero2, Alvaro Mejía Jiménez1, Inés Sánchez3; Raul Saavedra3, Jorge Cabra2, Joe Tohme1• 1SB-2 Project; 2Corporación BIOTEC; 3CORPOICA; Project funded by Colciencias Introduction Soursop is the most tropical of the 60 species of the genus Annona (Morton, 1987). It has a horticultura] value comparable to those of "chirimoya" (Annona cherimola), "anón" (A. squamosa), "papauca" (A. diversifolia) and "atemoya" (A. squamosa x A. cherimola) (Escobar y Sánchez, 1992). It's origin is northem South America, possibly Colombia or Brazil (León, 1968). This is why Colombia may have the world's greatest genetic diversity of soursop, which has not being used in breeding programs to improve agronomic traits. As a matter of fact in Colombia there is only one commercially available clone, fairly characterized (Ríos Castaño et al., 1996). C.l. Corpoica - Palmira has the only one Colombian annonas germplasm bank with 36 soursop accessions and 7 annonaceus species accessions, which have not been characterized at all. This project is the first attempt to study the genetic diversity of annonaceus species from Colombia. The objective is to know the genetic variability of the germplasm bank by applying DNA molecular marker technology (AFLP). We would like to know if this variability is representative of Colombian diversity. The genetic variability analysis will identify possible duplicates an40%), but very low content in May (<25%). In contrast, clone "B" showed a mediocre performance in March (<35%), but it was outstanding (about 37.5%) after the rains arrived. Because the crop needs to maintain high dry matter content after the rains arrive to supply the industrial sector of the North Coast, we proceeded to selecta group of materia1s that stood out for this trait. Many were already among the 215 genotypes selected by their good general performance, but others were characterized on1y for their high dry matter content, even after the rains arrived (Table 2). The average dry matter content across all clones evaluated was 32.41% and 26.74% for March and May, respective1y, thus indicating the significant progress expected for this trait. 76 On going work The group of selected clones will be evaluated again for their dry matter content in march and may 2002. ln the mean time the plants brought back to headquarters will be used to generate progenies for further genetic analysis along with their agronomic performance. Table l. Results of the selection carried out in the Clona! Evaluation Tria! at Santo Tomás, Department of Atlántico, from 1350 families evaluated during May 2000 to May 2001. Parameter o Yield (t/ha) Harvest lndex Plant type Dry matter content (%) Genotye Fresh roots Dry matter (Oto 1) 1 (l to 5) § March Results from the 1350 clones evaluated Mínimum 0.00 0.00 0.00 1.00 0.00 Maximum 57.02 18.15 0.84 5.00 45.96 Mean 20.23 5.77 0.45 3.02 32.41 Results from tbe 215 clones selected Mínimum 19.76 6.23 0.39 1.00 25.96 Maximum 57.02 18.15 0.84 5.00 45.96 Mean 32.20 9.76 0.54 2.59 34.41 Results from the 8 cbecks iocluded in the trial Mínimum 6.90 1.84 0.40 3.00 0.00 Maximum 34.17 11.02 0.69 5.00 37.45 Mean 23.77 7.29 0.55 3.88 31.62 Best 10 clones selected across tbe three strata for bigh medium and 1ow areas in the field SM 2546-44 53.09 18.15 0.64 2.00 34.31 SM 2546-32 54.88 16.24 0.61 2.00 32.95 SM 2771-5 51.55 16.21 0.64 2.00 35.49 SM 2615-13 47.62 14.82 0.48 4.00 38.57 SM 2615-28 51.31 14.77 0.64 2.00 35.12 SM 2629-36 38.09 12.17 0.60 2.00 37.07 SM 262 1-1 34.29 11.34 0.45 2.00 38.38 SM 2769-15 39.64 10.95 0.62 3.00 31.81 SM 2775-2 41.43 9.71 0.69 3.00 27.08 SM 2769-11 30.24 8.64 0.63 1.00 33.04 1 The harvest index is obtained by dividing the production of commercial roots by total biomass (roots + aerial parts). Preferred harvest indexes are > 0.5. 1 P1ant type integrates under one value, plant architecture, leaves health, and capacity to produce stakes on a scale where 1 = excellent and 5 = very poor is used. 77 M ay 0.00 36.76 26.74 16.64 36.76 29.05 21.83 36.76 29.25 35.00 28.53 30.12 28.22 26.73 30.21 31.33 26. 19 22.20 27.09 Table 2. Dry matter contentas measured in March and May 2001, of a group of clones selected for their good performance in this trait from the Clonal Evaluation Trial. D.IY_ matter content_io/ol Percentage of Pedigree In March In M~ Retention 1 SM 2545-20 39.73 35.77 0.90 SM 2546-52 37.16 35.96 0.97 SM 2546-54 37.36 34.74 0.93 SM 2618-16 40.13 36.70 0.91 SM 2619-1 37.97 34.00 0.92 SM 2619-5 39.78 36.52 0.92 SM 2619-6 38.46 34.58 0.90 SM 2619- 12 37.68 34.02 0.90 SM 2621-4 43 .16 34.60 0.80 SM 262 1-14 39.54 35.16 0.89 SM 262 1-25 37.45 33.85 0.90 SM 262 1-28 38.74 36.71 0.95 SM 2622-1 40.92 36.46 0.89 SM 2623- 1 37.01 33.92 0.92 SM 2772-2 40.41 33.74 0.83 SM 2772-7 38.54 35.89 0.93 SM 2772-8 39.30 34.20 0.87 SM 2773-46 34.3 1 37.98 1.11 SM 2775- 17 38.07 34.54 0.91 SM 2603-9 40.86 34.67 0.85 Promedio 38.83 35.20 0.91 Mínimo 34.3 1 33.74 0.80 Máximo 43 .16 37.98 1.11 ' Ratio between measurement in may over that ofmarch. 78 Below sorne interesting results and observations which are relevant to both SB2 and IP3 projects is sumrnarized Time' Stage and time (old system) Stage and time (new system) o Crossing of selected parental genotypes Crossing of selected parental genotypes t t Fl (SOOO) 6 (6 months] Fl (SOOO) 1 plant 1 1 site 1 1 rep [10 months] 1 plant 1 1 site 1 1 rep t t FlCl (4000) Clonal evaluation (2500) 18 [1 year] [1 year] 1 plant 1 1 site 1 1 rep 8 plants 1 1 site /1 rep t t Clona! evaluation (700) Prellminary yleld trlal (2SO) 30 (1 year] [1 year] 6 plants 1 1 site 1 1 rep t 1 O plants 1 lsite /3 rep t Prdlmlnary yleld trlal (160) Advanced yield tria! (SO) 42 [1 year] [2 years] 20 plants /1-2 sites /1 rep 2S plants / 2-3 sites /3 reps Advanced yield trial (SO) f 66 (2 years] 2S plants 12-3 sites /3 reps • ELITE GERMPLASM Germplasm collection ~ 1 Crossi~ ~ Participatory trials blocks research ,Time in months after germination ofbotanical seed. Figure l. Basic assava brccding schcmcs applied for cach of the priority ecosystcms. On thc rigbt is tbe new schemc currcndy undcr implcmcntation (shaded arca). La ter stagcs or sclcction are made followiog tbe old system (sbaded area oo left). 79 Time 1 o 10 22 34 58 45.00 40.00 >. ~ 35.00 .!: Q¡ 30.00 :::: ro E ~ 25.00 o 20.00 15.00 10.00 10 Relationship between dry matter content (%) measured in March and May 15 ·. . •. . . . . . ·• _.~a.,, 15 years ago has identified severa! sources of resistance to A. socialis (CIA T, 1999). The clone MEcu 72 has consistently expressed the highest levels of resistance. A. socialis feeding on resistant clones had less oviposition, longer development periods, reduced size and higher mortality than those feeding on susceptible ones. Whitefly- resistant clones, in field trials, showed no significant differences in yield between insecticide- treated and non-treated plots (Bellotti et. al. 1999). Whitefly resistance in agricultura! crops is rare; therefore, given the importance of these pests, there is a need to understand the genomics of the resistance that we are observing in MEcu 72 and other resistant clones. lt would be especially advantageous to map whitefly resistance genes and understand their segregation in F 1 progeny. Crosses were, therefore, made between MEcu 72 and a susceptible genotype to map resistance genes by using molecular markers. This will aid in a more rapid selection of resistant germplasm and also isolate those genes in volved in resistance. Materials and Methods A cross was made between the resistant genotype, MEcu 72 and the susceptible genotype MCol 2246. The latter cultivar was selected because of its high leve! of susceptibility toA. socialis, but also having tolerance to mites and thrips, two additional important pests of cassava. In addition MCol 2246 has good floration, an advantage in obtaining the high numbers of progeny necessary for genetic studies. This cross produced 282 F 1 individuals. The sexual seeds produced in the cross were grown in sterile soil, in 67 plastic trays, and held in the screen house for 6 to 8 weeks (Temp. ± 30°C). Seedlings were subsequently planted in the field for multiplication. Greenhouse evaluations were done by in vitro multiplication consisted of cutting plant species, removing to the laboratory, and disinfecting by washing in deionized sterile water, then 70% alcohol, then 0.25% hypochlorite and finally three additional washings in deionized sterile water (Escobar, 1991). The apices were planted in 4E media (Roca, 1984), in 16mm test_ tubes. The growth period was 60-80 days and a second propagation in 4E media resulted in 5 tubes per clone. Later, apices of each clone were cut and planted in 17N media (Roca, 1984) to obtain root growth; a period of 30-40 days. The plants were then ready for removal to the greenhouse for evaluation. The afore-mentioned methodology permits maintaining plants in optimal sanitary conditions, in addition to having sufficient material available on need in a reduced area or space. Greenhouse eva1uations were done with the parents MEcu 72 and MCol 2246, and the progeny using the leaf snap-cages and infected with A. socia/is adults from the CIA T colony. Field trials were carried out at two sties, CIA T, Palmira, and in Nataima, El Espinal, Tolima. The parents and progeny were planted 1 x 1 meter in the field and exposed to natural whitefly infestations. Cassava microsatellite (Simple Sequences Repeat, SSR) were used. DNA visualization was done by the tincture of sil ver ni trate technique to observe allelic segregation of the markers. 110 Results and Discussion In vitro propagation: Through in vitro propagation, 224 genotypes from the MEcu 72 x MCol 2246 cross, were obtained and grown in test tubes on a 4E media. From each ofthese genotypes, 5 clones were multiplied and propagated in a 17N media in the greenhouse. The 58 remaining genotypes are being collected for multiplication. The resistant (MEcu 72) and susceptible (MCol 2246) cultivars were evaluated with 343 cassava microsatellites, including 116 new SSRs of e DNA (Mba et. al., submitted). Approximately 60% of the microsatellites were polymorphic (Table 1); thereby obtaining 180 polymorphic microsatellites from the two parents. With the extraction of DNA from the 282 individuals progeny, polymorphic microsatellites were run (Figure 1). .. .. '"',.... . ..... -~ ~ ~,.......11'::;-M • ; ~ ( • •• • ••••• • ~ •••..••• -; ••• • ' •. • ·f"· •• o' ••• ,,J. .. . ....... . Figure l. Microsatellite 7 (F) of 282 Fl individuals from an MEcu 72 x MCol 2246 crosses. 111 Table l. Cassava microsatellites fro the parental cultivars Mecu 72 and M Col 2246. SSR# Size (bp) T. Anneal oc Polymorphic 5SR# Size (bp) T . Anneal oc Polymorphic SSRYI 197 45 X SSRY51 298 50 X SSRY2 225 55 X SSRY52 266 55 X SSRY3 247 45 X SSRY53 138 55 Monomorphic SSRY4 287 45 X SSRY54 151 55 X SSRY5 173 55 X SSRY55 145 50 X SSRY6 298 45 X SSRY56 137 50 Monomorphic SSRY7 250 45 X SSRY57 293 55 X SSRY8 288 45 X SSRY58 217 55 X SSRY9 278 55 Monomorphic SSRY59 158 55 X SSRY10 153 55 X SSRY60 137 55 X SSRY11 265 55 X SSRY61 233 55 Monomorphic SSRY12 266 55 Monomorphic SSRY62 250 55 Monomorphic SSRY13 234 50 X SSRY63 290 55 Monomorphic SSRY14 300 55 Monomorphic SSRY64 194 55 X SSRY15 2 15 50 Monomorphic SSRY65 299 55 X SSRY16 218 55 X SSRY66 261 55 Monomorphic SSRY17 277 50 X SSRY67 278 55 Monomorphic SSRY18 198 44 Monomorphic SSRY68 287 55 X SSRY19 2 14 50 X SSRY69 239 55 X SSRY20 143 55 X SSRY70 249 55 X SSRY2 1 192 55 X SSRY71 21 7 55 X SSRY22 299 43 Monomorphic SSRY72 141 55 X SSRY23 247 45 X SSRY73 265 50 Monomorphic SSRY24 100 45 Monomorphic SSRY74 114 55 X SSRY25 296 45 Monomorphic SSRY75 284 55 X SSRY26 121 55 X SSRY76 273 55 X SSRY27 277 50 X SSRY77 275 55 X SSRY28 180 55 Monomorphic SSRY78 248 55 X SSRY29 281 55 Monomorphic SSRY79 210 55 X SSRY30 220 50 X SSRY80 299 55 X SSRY31 188 50 X SSRY81 204 55 Monomorphic SSRY32 298 50 Monomorphic SSRY82 211 55 X SSRY33 273 50 Monomorphic SSRY83 239 55 -Monomorphic SSRY34 279 55 X SSRY84 203 55 X SSRY35 282 55 Monomorphic SSRY85 292 50 X SSRY36 134 55 X SSRY86 296 50 X SSRY37 187 50 Monomorphic SSRY87 102 55 X SSRY38 122 55 X SSRY88 243 55 X SSRY39 293 50 X SSRY89 120 55 X SSRY40 231 50 X SSRY90 193 55 Monomorphic SSRY41 271 X SSRY91 300 55 Monomorphic SSRY42 221 50 X SSRY92 171 55 Monomorphic SSRY43 255 43 Monomorphic SSRY93 289 55 X SSRY44 194 50 Monomorphic SSRY94 268 55 X SSRY45 228 50 X SSRY95 282 55 X SSRY46 268 50 Monomorphic SSRY96 149 55 X SSRY47 244 55 X SSRY97 194 55 X SSRY48 178 50 Monomorphic SSRY98 209 55 Monomorphic SSRY49 300 50 Monomorphic SSRY99 192 55 X 112 SSR# Size (bp) T. Anneal oc Polymorphic ;sR# Size (bp) T. Anneal oc Polymorphic SSRY50 271 50 X SSRY100 210 55 X SSRYI01 213 55 X SSRY153 117 45 X SSRY102 179 55 Monomorphic SSRY154 318 55 X SSRY103 272 55 X SSRY155 158 55 X SSRY104 258 52 Monomorphic SSRY156 160 44 Monomorphic SSRY105 225 55 Monomorphic SSRY157 500 45 Monomorphic SSRY106 270 55 X SSRY158 224 45 Monomorphic SSRY107 120 45 X SSRY159 159 45 Monomorphic SSRY108 203 55 X SSRY160 151 50 X SSRYI09 125 55 X SSRYI61 220 55 X SSRYllO 247 55 Monomorphic SSRY162 126 43 X SSRY111 235 55 Monomorphic SSRY163 231 44 Monomorphic SSRY112 117 55 X SSRY164 187 55 X SSRY113 187 45 X SSRY165 243 55 X SSRY114 167 55 X SSRY166 244 55 X SSRY115 296 Non-amplified SSRY167 183 45 X SSRYII6 167 Non-amplified SSRY168 277 55 Monomorphic SSRY117 142 55 X SSRYI69 lOO 55 X SSRY118 169 55 Monomorphic SSRY170 299 55 X SSRY119 155 55 X SSRY171 291 55 X SSRY120 139 55 X SSRY172 201 55 X SSRY121 168 43 X SSRY\73 281 NO SSRY122 273 45 X SSRY174 136 43 X SSRY123 136 55 X SSRY175 136 55 X SSRY124 146 55 Monomorphic SSRY176 112 45 Mooomorphic SSRY125 247 55 Monomorphic SSRY177 268 55 X SSRY126 245 55 Monomorphic SSRY178 104 55 Monomorphic SSRY127 130 44 Monomorphic SSRY1 79 226 55 X SSRY128 243 45 X SSRY180 163 55 X SSRY129 205 55 Monomorphic SSRY181 199 55 X SSRY130 223 55 X SSRY182 253 50 Monomorphic SSRY131 111 45 Monomorphic SSRY183 22 1 50 X SSRY132 196 45 Monomorphic SSRY184 163 so X SSRY133 29S 55 Monomorphic SSRY185 243 50 X SSRY134 213 55 Monomorphic SSRY186 297 55 SSRY135 253 55 X SSRY187 160 55 SSRYI36 296 55 Monomorphic SSRY188 198 55 Monomorphic SSRY137 157 55 Monomorphic SSRY189 185 55 X SSRY138 129 50 Monomorphic SSRY190 164 55 SSRYI39 129 44 Monomorphic SSRY191 186 55 Monomorphic SSRY140 212 43 Monomorphic SSRY192 183 55 X SSRY141 262 55 X SSRY193 218 55 X SSRY142 206 55 X SSRY194 196 55 SSRY143 153 SS Monomorphic SSRY195 186 SS X SSRY144 117 55 X SSRY196 188 55 SSRY145 143 45 X SSRY197 209 55 X SSRY146 139 45 X SSRY198 219 55 SSRY147 113 45 Monomorphic SSRY199 205 55 SSRY148 114 55 Monomorphic SSRY200 205 55 X SSRY149 500 45 X SSRY201 197 55 X SSRY150 175 45 Monomorphic SSRY202 191 55 11 3 SSR# Size (bp) T. Anneal oc Polymorphic 5SR# Size (bp) T. Anneal oc Polymorphic SSRY151 182 55 X SSRY203 246 55 X SSRY152 233 45 X SSRY204 182 55 X SSRY205 201 55 X SSRY257 280 55 Monomorphic SSRY206 219 55 SSRY258 400 55 Monomorphic SSRY207 199 55 SSRY259 220 55 Monomorphic SSRY208 198 55 SSRY260 100 55 SSRY209 195 55 SSRY261 210 55 X SSRY210 219 55 Monomorphic SSRY262 140 55 Monomorphic SSRY211 202 55 Monomorphic SSRY263 n.a. SSRY212 238 55 SSRY264 n.a. SSRY213 199 55 SSRY265 230 55 X SSRY214 234 55 SSRY266 220 55 Monomorphic SSRY215 204 55 X SSRY267 265 55 Monomorphic SSRY216 210 55 SSRY268 2 15 55 solo SSR55 SSRY217 181 55 X SSRY269 200 55 SSRY218 203 55 X SSRY270 220 55 SSRY219 190 55 X SSRY271 280 55 Monomorphic SSRY220 190 55 X SSRY272 220 55 SSRY221 n.a. SSRY273 n.a. SSRY222 150 n.a. SSRY274 280 55 SSRY223 170 55 X SSRY275 50 X SSRY224 n.a. SSRY276 260 55 X SSRY225 n.a. SSRY277 2 10 50 Monomorphic SSRY226 n.a. SSRY278 210 55 Monomorphic SSRY227 200 55 Monomorphic SSRY279 170 55 Monomorphic SSRY228 210 n.a. SSRY280 180 55 Monomorphic SSRY229 200 55 X SSRY281 195 55 Monomorphic SSRY230 185 55 X SSRY282 200 55 X SSRY231 260 55 Monomorphic SSRY283 215 55 X SSRY232 n.a. SSRY284 210 55 Monomorphic SSRY233 205 55 Monomorphic SSRY285 290 55 X SSRY234 n.a. SSRY286 220 55 Monomorphic SSRY235 250 55 X SSRY287 220 55 Monomorphic SSRY236 220 55 X SSRY288 180 55 Monomorphic SSRY237 200 55 X SSRY289 195 55 -Monomorphic SSRY238 225 55 X SSRY290 300 55 Monomorphic SSRY239 220 55 X SSRY291 210 55 X SSRY240 200 55 X SSRY292 n.a. SSRY241 220 55 X SSRY293 50 Monomorphic SSRY242 280 55 X SSRY294 175 55 Monomorphic SSRY243 400 n.a. SSRY295 185 55 X SSRY244 220 55 Monomorphic SSRY296 175 55 X SSRY245 300 55 Monomorphic SSRY297 180 55 X SSRY246 210 55 X SSRY298 170 55 Monomorphic SSRY247 300 55 Monomorphic SSRY299 190 55 X SSRY248 250 55 X SSRY300 260 55 Monomorphic SSRY249 400 55 Monomorphic SSRY301 265 55 Monomorphic SSRY250 200 55 X SSRY302 200 55 X SSRY251 220 55 SSRY303 190 55 Monomorphic SSRY252 220 55 X SSRY304 240 55 Monomorphic SSRY253 190 55 X SSRY305 300 55 X 114 SSR# Size (bp) T. Anneal oc Polymorphic )SR # Size (bp) T. Anneal oc Polymorphic SSRY254 220 55 Monomorphic SSRY306 265 55 X SSRY255 190 55 Monomorphic SSRY307 n.a. SSRY256 210 55 Monomorphic SSRY308 280 55 Monomorphic SSRY309 220 55 Monomorphic SSRY327 n.a. SSRY310 50 Monomorphic SSRY328 240 55 X SSRY3 ll 200 50 Monomorphic SSRY329 210 55 X SSRY312 200 55 X SSRY330 52 X SSRY313 205 55 X SSRY331 52 X SSRY314 190 55 Monomorphic SSRY332 52 X SSRY315 230 50 X SSRY333 52 Monomorphic SSRY316 50 Monomorphic SSRY334 52 Monomorphic SSRY31 7 50 Monomorphic SSRY335 52 Monomorphic SSRY318 50 Monomorphic SSRY336 52 Monomorphic SSRY319 50 X SSRY337 52 Monomorphic SSRY320 50 Monomorphic SSRY338 52 Monomorphic SSRY321 50 Monomorphic SSRY339 220 55 X SSRY322 50 X SSRY340 55 Monomorphic SSRY323 50 Monomorphic SSRY341 200 55 X SSRY324 200 55 X SSRY342 210 55 Monomorphic SSRY325 240 55 Monomorphic SSRY343 300 55 Monomorphic SSRY326 n.a. Conclusions A high percentage of polymorphism (more than 60%) was obtained from the two parents, MEcu 72 and M Col 2246; this g uarantees a high number of markers for the construction of a legitimate rnap. However still lacking are the in vitro propagation of sorne individuals (58) progeny, and field and greenhouse screening with A. socialis. Upon completing the running of the polymorphic microsatellites with the 282 individual progeny, a cornparison will be made with greenhouse and field data. References Arias, B. 1995. Estudio sobre el comportamiento de la "Mosca Blanca" Aleurotrachellus socialis Bondar (homoptera: Aleyrodidae) en diferentes genotipos de Yuca, Manihot esculenta Crant'z. Thesis. Bellotti, A.C., L. Smith and S.L. Lapo in te. 1999. Recent advances in cassava pest management. Annual Review ofEntomo1ogy. 44 : 343-370. Centro Internacional de Agricultura Tropical. 1999. Annual Report: lntegrated Pest and Disease Management in Major Agroecosystems. CIA T, Cali, Colombia 136pp. Escobar, R.H. 1991. Estudio comparativo de dos métodos de propagación de la yuca Manihot esculenta (Crantz) in vitro. Tesis de grado, Universidad Santiago de Cali, Cali, Colombia. Fregene, M., Angel, F., Gomez, R., Rodríguez, F., Chavarriaga, P., Roca, W., Tohme, J. and M. Bonierbale. 1997. A molecular genetic map of cassava (Manihot esculenta Crantz). Theor. Appl. Genet. 95: 43 1-441. Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S ., Mkumbira, J., Gullberg, U., Apel, K., {'J GaJe, M., Tohme, J. and Fregene, M. 2000. Simple Sequence Repeat (SSR) Markers o 115 ~ Survey of the Cassava (Manihot escu/enta Crantz) Genome: Towards an SSR-Based Molecular t() Genetic Map of Cassava. Theoretical and Applied Genetics Journal. Accepted. Roca, W.M. 1984 Cassava. In: Sharp WR, Evans DA, Amirato RV, Yamada, Y (eds). Handbook ofplant cell culture: crop species. Vol2. MacMilliam Pub!, New York. 269-301. 1.2.18 Mapping and pursuit of QTLs for yield and yield components in rice populations derived from backcrosses between wild species and cultivated rice Almeida, A.\ J. López1, O. Giraldo1, A. Bohórquez., C. Castaño., J. Borrero2, J. Carabali2, G.Gallego\ M.C.Duque •-2, C. Martínez1 and J. Tohme1 1SB-2 Project; 2IP-4 Project Introduction Oryza wild species represent a potential source of new a11eles for improving yield, quality and stress resistance of cultivated rice. Still, effective use of wild species genes remains largely unexplored. Advanced backcross-breeding schemes (Tanskley and Nelson, 1996) using molecular mapping techniques represent an alternative to reduce the genetic background from wild species parentals and to rapidly discover and transfer valuable alleles from the wild species into elite rice varieties. Even though wild an~ unadapted germplasm have phenotypically less desirable than modern varieties in their overall appearance and performance, breeders have long recognized the intrinsic value of wild species for the improvement of simple inherited traits (Xiao et al. 1998). The possibility of selectively introgressing useful genes from O. rufipogon to elite rice cultivars suggests a way for improving the performance of O. sativa while simultaneously broadening the genetic base of cultivated rice (Moneada, P. et al. 2001). For this reason, advanced backcross QTLs analysis is proposed as a method to discover and transfer valuable QTL alleles from wild species into established elite inbred lines. BC2 or BC3 populations are used along with negative selection of undesirable characters to reduce the frequency of deleterious alleles present in the donar parental. QTL-NILs can be derived from advanced backcross populations in one or two additional generations and used to verify QTL activity. These same QTL-NILs also represent potential commercial inbreeds improved for one or more quantitative traits. If successfully employed, advanced backcross QTL analysis can open the door to exploiting unadapted and exotic germplasm for the quantitative trait improvement of a number of crop plants (Tanksley, S. D. and Nelson, J. C. 1996). This report focuses on progress made in identifying quantitative traits loci (QTLs) associated wíth yield íncrease in populations derived from crosses with Oryza rufipogon, O. barthii and O. glaberrima, and the introgression oftrait-enhancing QTLs in near-isogenic lines (NILs). Materials and methods Experiments to identify segregating alleles in advanced backcrossed populations were set up in the field, greenhouse and the CIAT Biotechnology laboratory. Deve/opment ofNILsfrom a BC2F2 popu/ation derivedfrom the cross BG90-2 x Oryza rufipogon . Eighty-five BC2F2 families were selected for NIL development based on the molecular and quantitative analysis (Nelson 2000.Qgene 3.0 software) of the BC2F2 population derived from the cross BG90-2 x O. rufipogon. As reported earlier (Ann.Report 2000) 66 SSRs were found associated with putative QTLs 116 detennining yield and yield components. These putative QTLs were derived from O.rufipogon. Selected families were evaluated with known SSR markers and homozygote or heterozygous individuals for alleles derived from O.rufipogon were crossed with the recurrent parent (BG90-2 maJe parent) to generate the BC3Fl population. Based on field perfonnance and agronomic data only 75 families out ofthe 85 initially selected were eh osen for NIL development. A total of 70 BC3Fl plants in each family was planted in the green house for molecular characterization. Leaf discs (5 mm in diameter) were collected from each plant (25-30 day-old seedlings) for DNA extraction using the Alkali method (Klimyuk et al. 1993). So far 11 SSRs have been analyzed in a total of 100 PCR assays done with 63 out of 75 families to verify which of the O. rufipogon alleles has been introgressed in this population (Figure 1 ). Afterwards plants were transplanted in the field to gather agronomic data . ... s •- - - • ... 1' 1' 1' 1' 1' Figure l. CT17211-12 pedigree lioe evaluated with RM234. Pareotals (BG90-2 aod O. rujipogon) are placed oo tbe first two laoes oo the left, followed by the individuals of the family. BC3F2 population derived from the cross Lemont x Oryza barthii . lt was reported earlier that the BC3F2 population derived from the cross Lemont x O. barthii had been evaluated with 85 SSRs. This year we completed 126 SSRs markers in this population. Currently, we are selecting the appropriate statistical analysis method to generate the QTL map. BC3F 1 Double haploid population derived from the cross Caiapo x Oryza glaberrima. Anthers from BC3Fl plants were harvested and run through anther culture (Ann. Report 2000). The response was very good and 695 doubled-haploid plants were obtained. Based on agronomic data 312 OH lines were selected for molecular characterization along with the parentallines. Five plants from each family were planted in the greenhouse for tissue collection and DNA extraction to be used in the development of molecular pro bes and the posterior QTL map for yield and yield components. Based on the screening of 280 SSRs in the parents ( Caiapo and O. glaberrima ) following the method described by Temnykh et al. 2000, the PCR assay protocols for 120 SSRs of them were standardized for use in the screening of the 312BC3Fl DH lines (Figure 2). DNA of young leaves from the parental genotypes and segregating population was extracted using the Dellaporta Method (McCouch et al. 1988) and modified for the PCR assay by the CIA T Biotechnology Research Unit. 117 --RH:m RM'3ol .... Figure 2. Screening of SSRs in the parents Caiapo and O. glaberrima Results and discussion We are following the molecular behavior of each QTL in BC2F2-NILs and BC3F2-NILs generations, developed from the cross BG90-2 x O. rufipogon. Currently, 93 assays were completed corresponding to evaluation of 11 SSRs (Table 1). In 62 cases there was previous molecular data available with regard to the genotype of the BC2F2 plant used as a female parent in crosses with BG90-2 (male-recurrent parent). However, molecular data was not available at crossing time in 31 cases and plants were selected based on phenotype. A test for goodness of fit (Chi-square test) was used to detennine whether the observed data confonns to a specified probability distribution. A 50% homozygous and 50% heterozygous segregation was expected when the BC2F2 plant crossed to BG90-2 was heterozygous for the "wild allele" ata given locus. On the contrary, 100% of the BC3Fl plants were expected to be heterozygous when the BC2F2 plant crossed to BG90-2 was homozygous for the "wild allele". Table 1 shows the observed, adjusted and expected values whilst Table 2 shows data corresponding to segregation obtained with RM242. The segregation pattern for each SSRs was kind of similar to that of RM242 in tenns of the Chi-square results. A summary of the Chi-square test for al! SSRs is shown in Table 3. In 51 cases (55%) the observed values did not differed significantly (5%) from the expected values, whilst in 42 cases (45%) there was a significant difference and the hypothesis was rejected. Table 3 also indicates that in cases (31) where molecular data was not available at crossing time, in 19 cases (61%) the resulting BC3Fl progenies segregated in an expected ratio whilst in 39% ofthe instances did not. Over all it can be seen that results differ a lot from expectations, since in only 55% of the instances results agreed with expectations. A much greater efficiency is expected when molecular markers are used to follow up introgression of specific alleles in a given background. As illustrated with RM242 non e of the SSRs gave 100% accuracy suggesting that the problem has to do with the entire population. This is further supported by the behavior of case where no previous molecular data was available. 118 There are several factors that could explain the 45% inaccuracy observed in this study due to experimental errors of various nature, namely field and laboratory conditions. There are severa) steps where mistakes can occur under greenhouse/field conditions such as errors in plant identification for crossing, selfing, poli en contamination, etc. In tenns of the laboratory protocols for DNA extraction and stability, PCR conditions, silver staining procedures, etc could produce unclear signals difficult to read. Several quality control measures are being implemented to reduce the experimental error. Another ímportant factor has to do with the genetíc distance between the putative QTL and the SSR marker. The greater the distance between them the higher the probability of getting a crossover which could give rise to individuals with non-expected genotypes. Table l. Chi-square test for goodness of fit in 93 BC2/BC3 assays carried out in the development of NILs from the cross BG90-2/0.rufipogon. PEDIGREE SSR obscrved val u es (%) ajusted observed values (%) expected values (%) B H R - B H R B H CT17172-3 RM1 100 o o o 100 o o - - CT17175-10 RM122 100 o o o 100 o o - - CT17184-18 RM122 o 98.6 o 1.43 100 - - CT17204-56 RM122 5.55 90.7 1.85 1.85 5.66 92.45 1.88 - - CT17143-20 RM13 81.03 19 o o 81.03 18.97 o - - CT17143-21 RM13 30 55.7 \4.28 o 30 55.71 14.28 - - CT17144-19 RM13 38.57 60 o 1.43 39.13 60.86 - - CT17145-14 RM13 91.43 7.14 o 1.43 92.75 7.24 - - CTI7145-29 RM13 54.29 41.4 o 4.28 56.71 43.28 - - CT17148-l3 RM13 55.38 43.1 o 1.54 56.25 43.75 - - CT17148·27 RM13 61.43 27.1 o 11.4 69.35 30.64 - - CT17163-20 RM13 87.14 o o ll.9 l OO - - CT17165-10 RM1 3 o 58.5 12.31 29.2 82.6 17.39 100 CT17171 -9 RM13 15.94 81.2 o 2.9 16.41 83.58 - - CT17172-3 RM13 91.67 8.33 o o 9 1.67 8.33 o so so CT17174-7 RM13 17.65 82.4 o o 17.65 82.35 o 100 CT1717844 RM13 o 94.3 o 5.71 100 100 CT17181-3 RM13 80 o o 20 100 - - CT17183-2 RMI3. 100 o o o 100 o o - - CT1718S·ll RM13 100 o o o 100 o o - - CT17186-8 RM13 o 44.3 o 55.7 100 100 CT17189-S5 RM13 82.86 o o 17.1 100 100 CT17142-l3 RM2 12 7.14 90 2.86 o 7.14 90 2.86 100 CT17142·32 RM212 35.21 64.8 o o 3S.21 64.79 o so 50 CT17143-20 RM21 2 100 o o o 100 o o - - CT17143-21 RM2 12 92.86 7.14 o o 92.86 7.14 o - - CT17146-32 RM212 o 100 o o o 100 o 100 CT17146-34 RM21 2 o 100 o o o 100 o so so CT17146-8 RM212 34.78 58 2.9 4.35 36.36 60.6 3.03 so 50 CTI7159-2 RM212 52.17 47.8 o o 52.17 47.83 o so so CT17162-20 RM212 so 48.5 o 1.52 50.76 49.23 - - CT17172-12 RM212 54.29 45.7 o o 54.29 45.71 o so so CT17182-21 RM2 12 98.08 o o 1.92 l OO - - CT17208·12 RM212 o 100 o o o 100 o - - CT17139-12 RM215 o 90 10 o o 90 10 100 CT17139-14 RM215 o 92.9 7.14 o o 92.86 7.14 100 CT17146-32 RM21 S 20 80 o o 20 80 o 50 so 119 PEDIGREE SSR observed values (%) ajusted observed values (%) expected values (%) B H R - B H R B H CT17146-34 RM215 o 98.6 1.43 o o 98.57 1.43 100 CT17146-8 RM215 40.58 21.7 1.45 36.2 63.63 34.09 2.27 50 50 CT17148-23 RM215 o 75.4 o 24.6 100 100 CT17148·27 RM215 46.27 50.8 2.99 o 46.27 50.75 2.99 100 CTI7149-20 RM215 28.57 52.9 7.14 11.4 32.25 59.67 8.06 so so CT I7159-2 RM215 o 71.2 7.58 21.2 90.38 9.61 100 CTI7168-14 RM215 15.1S 80.3 o 4.SS 1S.87 84.1 so so CTI7172-12 RM21S 90 o o 10 100 100 CT17193-16 RM21S o 98.S 1.54 o o 98.46 i.S4 100 CTI7170-12 RM234 42.86 35.7 o 21.4 S4.S4 45.45 - - CTI7192-2 RM2J4 100 o o o 100 o o 100 CT17196-59 RM234 98.44 o o i.S6 100 50 50 CT17206-48 RM234 o 95.4 4.61 o o 95.38 4.61 100 CTI7211-12 RM2J4 48.S7 51.4 o o 48.57 51.42 o 100 CT17138-14 RM242 1.64 96.7 1.64 o 1.64 96.72 1.64 100 CTI7138-9 RM242 o 97.1 2.94 o o 97.06 2.94 100 CTI7139-12 RM242 97.14 1.43 o 1.43 98.55 1.44 - - CT17139-14 RM242 100 o o o 100 o o - - CT17143-20 RM242 53.45 44.8 1.72 o 53.45 44.83 1.72 so 50 CT17143-21 RM242 o 81.4 18.S7 o o 81.43 18.57 - - CTI7147-22 RM242 o 94.3 o 5.71 100 100 CT17147-37 RM242 23.88 26.9 5.97 43.3 42.1 47.36 10.52 100 CTI7148-23 RM242 o 95.4 1 .. 54 3.08 98.41 1.58 100 CT17148-27 RM242 50 44.3 5.71 o 50 44.29 5.71 50 50 CTI7149-20 RM242 32.86 51.4 5.71 10 36.5 S7.14 6.34 so so CT17149-7 RM242 2.86 91.4 2.86 2.86 2.94 94.11 2.94 100 CT171SO-SO RM242 3.08 95.4 1.54 o 3.08 9S.38 1.54 100 CT171S3-14 RM242 18.06 81.9 o o 18.06 81.94 o 50 so CTI7153-8 RM242 47.14 51.4 o 1.43 47.82 52.17 so 50 CT1715S-30 RM242 40 1.54 o S8.5 96.29 3.7 100 CTI71S8-21 RM242 o 95.7 o 14.3 100 100 CT17160-2 RM242 o 58.6 4.29 37.1 93.18 6.81 100 CT1716S-10 RM242 3S.38 3.08 1.54 60 88.46 7.69 3.84 so so CT17165-39 RM242 100 o o o 100 o o 100 CT17172-12 RM242 68.57 o o 31.4 100 100 CT17179-12 RM242 o 22.9 4.28 72.9 84.21 1S.78 1:00 CT17187-9 RM242 51.79 30.4 10.7 1 7.14 55.76 32.69 11.53 - - CT17188-19 RM242 o 90 o 10 100 100 CT17189-5S RM242 o 84.3 1.42 14.3 98.33 1.66 100 CT17146-32 RM2S7 30 67.1 2.86 o 30 67.14 2.86 100 CTI7146-34 RM257 Sl.43 48.6 o o 51.43 48.S7 o so 50 CTI7146-8 RM2S7 47.83 37.7 13.04 1.45 48.S2 38.23 13.23 so 50 CT17177-4 RM257 100 o o o 100 o o - - CTI719S-IO RM2S7 o 92.8 2.9 4.3S 96.96 3.03 100 CT17201-20 RM257 S2.78 47.2 o o 52.78 47.22 o 100 CT171S4-20 RM27 o 93.9 o 6.06 100 100 CTI7139-14 RM44 78.71 14.3 o lO 84.12 IS.87 - - CTI7142-23 RM44 47.14 41.4 o 11.4 53.22 46.77 so 50 CTI7142-32 RM44 40 38.6 1.43 20 so 48.21 1.78 so 50 CT17144-29 RM44 35.29 38.2 2.94 23.5 46.1S so 3.8 50 so CT17148-23 RM44 50.77 o o 49.2 100 100 CT17148-27 RM44 8S.71 7.14 o 7. 14 92.3 7.69 100 120 PEDIGREE SSR observcd valucs (%) ajustcd observcd valucs (%) expcctcd valucs (%) 8 H R . B H R 8 H CTI7159-2 RM44 97.22 2.78 o o 97.22 2.78 o 100 CT17167-31 RM44 92.42 o o 7.58 100 - . CTI7168-14 RM44 96.97 o o 3.03 100 . . CT17203-23 RM44 7.15 57.1 35.71 o 7.15 57.14 35.71 . . .. 8 .Homozygotcs hkc 8g90-2; H:Hctcrozygotcs ; R: Homozygotcs hkc O.rufipogon; • m1ssmg data Table 2. Chi~square test to pro be the hypothesis of 50-SOe;. segregation of heterozygote genotype in BC3~NILs families. RM 242. IPEDIGREE SSR ~bscrved values (%) Ajusted observed values (%) xpected valuesJo/o) X"2 <= 3.84146 ts H ~ B ~ R ts IH 0.05, 1 DF t'Tt7138-14 RM242 1.64 96.72 1.64 o 1.64 ~6.72 1.64 100 k:Tt7138-9 RM242P 97.06 2.94 o o ~7.06 2.94 100 tTt7139-12 RM242 ~7.14 1.43 o 1.43 98.55 1.44 PI7139-14 IRM242 100 o o p 100 o o pJ7143-20 [RM242 [3.45 ~4. 83 1.72 p ~3.45 44.83 1.72 50 ~o 0.772628 ~S ~17143-21 ~420 181.43 18.6 p p 81.43 18.57 PI7147-22 IRM242 p ~4.29 o ~.71 lOO 100 Crl7147-37 ~42 ~3 .88 t26.87 5.97 ~3.3 ~2.1 47.36 10.52 100 Cfl7148-23 IRM242 o ~5.38 1.54 ~.08 98.41 1.58 lOO CT17148-27 RM242 ~o ~4.29 ~.7 1 o 50 ~4.29 !5. 71 150 50 0.652082 NS CT17149-20 RM242 32.86 ~1.43 ~.7 1 JO 36.5 ~7. 14 6.34 50 50 4.664592 ¡. CT17149-7 IRM242 2.86 ~1.43 ~.86 ~. 86 ~.94 ~4.11 ~.94 100 CTI7150-50 RM242 3.08 ~5.38 1.54 o 3.08 ~5 .38 1.54 100 CTI7153-14 RM242 18.06 81.94 p o 18.06 181.94 o ~o 50 40.8065441* CT17153-8 RM242 47. 14 51.43 o 1.43 47.82 ~2.17 ~o ~o 0.189226 INS P17I55-3o RM242 ~o 1.54 o ¡:¡8.5 96.29 ~ . 7 100 P17158-21 RM242P 95.71 p 14.3 100 lOO P17160-2 RM242º 58.57 4.29 b7.1 ~3. 18 6.81 100 P17165-IO RM242 ~5.38 3.08 1.54 ~ 188.46 7.69 3.84 50 !so ~5.3861541- P17165-39 IRM242 100 p o p 100 o p tOO Cr17172-12 lRM242 68.57 p p ~ 1.4 100 100 q17179-12 IRM242 p ~2.86 ~.28 :r2.9 ª-4 .21 15.78 100 CrJ7187-9 IRM242 :S 1.79 ~0.36 10.7 ~.14 55.16 !2.69 11.53 t- CT\7188-19 lRM242 o ~o p 10 100 100 - CT17189-55 lRM242 o 84.29 1.42 14.3 ~8.33 1.66 100 Ns: SSR without statistical significance for the goodness offit test; •••: SSR with statistical significance B: Homozigotes 1ike BG90-2; H: Heteroz.igotes; R: Homoz.igotes like O. rufipogon; "-": rnissing data ACEPTABLE IYES YES YES2 YES YES N02 YES IN03 !YES3 !vES NO IN o IYEs NO YES NO YES NO NO NO !vES INO INO tyEs !vES 1 : Ajusted observed values, without missing data 1 : Instead of unknown genotypes of the parents, the segregation is compared with the three expected genotypes of the F,ssible parents. : Maximum level of tolerance: 5%, respect to the expected genotype according to the parent. 121 Table J. Summary of Chi-square test obtained in BC3Fl-NIL population developed from the cross BG90-2 and O. rufipogon. Acceptable Parents information 50-50 (a) 100-0 {a) 0-100 (a) total With information 62 10 5 17 32 Without information 31 4 13 2 19 TOTAL 93 14 18 19 51 Reiected Parents information 50-50 100-0 0-100 total With information 62 12 1 17 30 Without information 31 - - - 12 TOTAL 93 42 a. Expected percentages of B and H Ongoing activities • Complete the molecular analysis in BC3-NIL population derived of BG90-2 x O. rufipogon. • Follow up introgression of QTLs in BC4-NILs families derived from crosses between BC3-NILs families and Bg90-2. • Complete the statistical analysis of BC3F2 Lemont x O. barthii population. • Map F2 populations derived from BG90-2 x O. rufipogon and Lemont x O. barthii crosses. • Initiate the characterization of agronomic and molecular data, and QTL analysis to determine the number of QTLs associated with yield and yield components for Caiapo x O. glaberrima cross. References Klimyuk VI, Carroll BJ, Thomas CM and Jones JDG (1993) Al.kali treatment for rapid preparation of plant material for reliable PCR analysis. Plant Journal3 ,3: 493-494. Moneada P, Martfnez C P, Borrero J, Chatel M, Gauch Jr H, Guimaraes E, Tohme J, McCouch SR. (2001) Quantitative trait loci for yie1d and yield components in an Oryza saliva x Oryza rufipogon BC2F2 population evaluated in an upland environment. Theoretical Applied Gene ti es 102: 41-52. Tanksley, S. D. and Nelson, J. C. ( 1996) Advanced backcross QTL analysis: a method for the simultaneous discovery and transfer of valuab1e QTLs from unadapted germplasm into elite breeding lines. Theoretical Applied Genetics 92: 191-203. Temnykh S, Park WD, Ayres N, Cartinhour S, Hauck N, Lipovich L, Cho YG, Ishii T and McCouch S. 2000. Mapping and genome organization of microsatellite sequences in rice (Oryza saliva L.). Theoretical Applied Genetics 100: 697-712. Xiao J, Li J, Grandillo S, Nag Ahn S, Yaun L, Tanksley S and McCouch S. (October 1998) Identification of Trait-Improving Quantitative Trait Loci Alleles From a Wild Rice Relative, Oryza rufipogon. Genetics 150: 899-909. 122 1.2.19 Evaluation and selection of inter-specific populations via conventional breeding methods CP Martínez1, J .Borrero2 , James Carabali2 1SB-2 Project; 2IP-4 Project Introduction Genetic variability is an essential requirement to make progress in plant improvement prograrns. Traditionally, plant breeders make use of diverse genetic resources to come up with improved varieties. Most of the time, rice- breeding prograrns are under high pressure to deliver superior varieties to meet demands coming from diverse users. Crosses using good and well-known progenitors have better probabilities of producing the kind of breeding populations from where superior genotypes could be selected. Unfortunately, not very many good donors are available and the continued use of them lead to reshuffling of genes reducing at the sarne time the genetic variability. It has been estimated that around 25% of the total genetic variation available in rice are actually used by rice improvement programs. The use of un-improved gene pools, like wild rice species, represents a difficult task to conventional breeding prograrns; however, this kind of research is more appropriate to programs more strategic in its scope like the ClA T Rice Project. Therefore, activities described in this section deal with the introgression of alleles from wild rice species into the Latin America gene pool using conventional breeding methods. The main objective is to develop potential parents to be used by national rice prograrns. A backcross scheme to diverse improved breeding lines or varieties is used. Starting in the F2 generation single plant selections are made for further evaluation in pedigree rows. Around 250 crosses were made of which 189 were evaluated under upland conditions in Villavicencio. Many populations (119) were discarded because of high sterility and/or poor plant type, low yield potential, and susceptibility to major diseases. Only 476 plant selections for evaluation as F3 lines were made. Table 4 presents a summary of inter-specific populations evaluated in ClA T-Palmira and Villavicencio. Advanced lines with good plant type, early vigor, and excellent grain type were selected in the cross Oryzica 3/0. rufipogon. lt was shown (Rice Program Annual Report 2000) that both Oryzica 3 and O. rufipogon bad a good level of resistance to Rhizoctonia sp, a disease that is becoming more important in severa! areas in Latín America. In collaboration with Rice Pathology and FEDEARROZ, evaluations are underway, under greenhouse and field conditions, for resistance to Rhizoctonia. Advanced lines (F6-F7) with good plant type and yield potential and tolerance to sorne diseases were selected in the cross Bg90-2/0.rufipogon. Sorne of these lines have good grain quality and 25 were selected for yield tests in diverse environments through collaboration with national rice prograrns, including Colombia, Argentina, Venezuela, Costa Rica and Honduras. High sterility has been observed in crosses with O. glaberrima, even after 2-3 backcrosses. However, F4-Fs progenies with high fertility, early plant vigor, good tillering, strong stems and 123 good grain type have been selected. Severa! accessions of O. glaberrima possess high levels of resistance to the rice stripe virus disease (Rice Program Annual Report 2000); consequently, selected lines are being screened for resistance to rice stripe virus in collaboration with Rice Pathology and FEDEARROZ. Crosses between O. barthii and improved cultivars have resulted in progenies with low yield potential, poor plant type and high sterility. Nevertheless, 2-3 backcross to Lemont have produced advanced lines with excellent grain type, cooking and milling quality, early maturity, and, tolerance to major diseases, except hoja blanca virus disease. Sorne lines have better yield potential than Lemont. Severa! lines were included in the VIOAL nurseries for distribution to national rice programs. Two populations (Caiapo/0. glaberrima and Progresso/0. barthii) were evaluated under upland savanna conditions in Villavicencio, in collaboration with the CIRAD-CIAT program. Caiapo/ O. glabe"ima showed better performance in terms of adaptation to acid soils. Besides, doubled haploid lines were obtained, which will be used to identify and characterize alleles derived from O. glaberrima that are associated with traits of agronomic importance. In addition, 2704 F 4 lines from a collaborative project between CIA T and Peru were evaluated for resistance to rice hoja blanca virus in Palmira and tolerance to major diseases in Santa Rosa, Villavicencio. In spite of the high disease; pressure 1097 single plant selections were made for further evaluation as F s pedigree rows. Table 4. Inter-Specific Populatio.ns Evaluated in the Pedigree Method for the Development of Improved Germplasm Population Oryzica 3/0. rufipogon Bg90-210. rufipogon Bg90-210. glaberrima Bg90-210. barthii O. llanos 5/0. barthii Lemont/0. barthii Caiapo!O.glaberrima Progresso/0. barthii Target Ecosystem Irrigated Upland Acid Soils Generation BC3F1 BC3F• BC1F• BC F BC1F• BC1F$ BC3F~ BC R. 124 Lines Evaluated 1.123 JO 952 93 338 203 1.224 1.014 533 304 111 633 614 119 354 488 29 49 422 No.Piants Selected 613+26 55 1.970 63 203 124 1.014 234 53 20 9 613 561 154 17 45 2240H 695 OH 24 Observations Excellent plant type, good grain quality, potential source of R genes for Rhizoctonia 25 Advanced lines selected for yield tests outside CIA T Excellent plant vigour, good tillering and grain type. Sorne sterility. Low yield potential Poor vigour, low yield potential Excellent cooking and milling earliness, long panicle Good variability, OH lines to be used for mapping purposes Late maturing, low yield 1.2.19 Influence ofwild rice species on the grain quality, nutritional val u e, the eating and cooking quality of inter-specific progenies CP. Martínez1, J. Borrero2, MC Duque 1"2 1SB-2 Project; 2IP-4 Project Introduction The impact of modero agriculture in improving the life and well being of billion of people around the world is impressive. The prospect of mass starvation was avoided due to increased food production through the Oreen Revolution's push for food security. However, little thought was given to nutritional value and human health, and almost none to the concentration of iron and other micronutrients in the new cereal varieties bred (Welch and Graham, 2000). Research at IRRI has shown genetic variation for iron and zinc concentration in brown rice; improved cultivars contain about 12 mg of iron and 25 mg of zinc per kilogram, while sorne traditional cultivars ha ve doubled these amounts (Gregorio et al, 2000). Results from W ARDA ( 1998) indicates that inter-specific crosses with O. glaberrima gave rise to progenies with higher protein content, good eating quality, and high nutritive values. As mention earlier, we at CIA T are looking at wild rice species as potential sources of new alleles associated with traits of agronomic importance. Emphasis was given to alleles associated with grain yield and its components. In this section, we are dealing with the nutritional and grain quality aspects. Materials aod Methods Seed of the advanced lines CT14938-30-5-M-3 and CT14938-36-1-M-1 derived from the cross Lemont/0. barthii was harvested, dried and milled. Samples were taken to the quality lab for evaluation. Remnant seed was bulked up and milled and 2-kg samples were given to 64 people for cooking and eating evaluations. People were advised to cook the rice sample the same way they used at borne and to compare its behavior with that of the rice they usually buy and eat. Eleven rice cultivars (Tablel) including O. barthii, O. glaberrima andO. rufipogon were choosen to determine its iron and zinc content. Brown and milled rice samples were obtained from field plots grown at CIA T and 5 gram each were sent to the lab for chemical analysis using the method proposed by Isaac and Kerber, 1971 . The experiment was replicated three times. Results Data from the quality 1ab showed that both lines had long and slender translucent grains (0.2 white center), with amylose content around 26-29 %, and excellent milling recovery (60% head rice). Data from the cooking!eating tests are presented in Figures l. Forty-seven (75%) people reported that the rice sample was dry and fluffy after cooking whilst 96% of people said that the grain appearance was good before cooking; 4% reported that the appearance was fair (Figure 1). Only 4% found the rice sample to be sticky. It is important to keep in mind that the ratio rice/water used by people was different (it ranged from 1/2, 2/3, 3/4 to 1/ 1 ). 125 Figure 2 shows that 34% of people detected sorne kind of aroma after cooking and a different taste compared to the rice they usually consume. Besides 51 % of surveyed people reported that the sample given produced more cooked rice than the one they usually consume whilst 41% was willing to paya little bit more for that kind of rice( data not shown). Results suggest that O. barthii did not affect in a negative way the eating and cooking quality of rice; on the contrary, sorne people detected special features in the quality of the rice derived from an inter-specific cross. Data also confirm differences in people's preference in terms of grain quality opening up opportunities for the development of special types of rice. so cu 40 DGood •Fair -c. o 30 cu D. ..... o 20 . o z 10 o Dry/Fiuffy Intermediate Moist/Sticky After Cooking Figure l. Grain Appearance Before and After Cooking 50.--------------------------------------------- 40 cu -c. o 30 cu c. ..... o 20 . o z 10 DSome • Off-Fiavor 0+-----~--------~------r-----~--------~---- Slight Off-Aroma Figure 2. Assessment of MiJJed Cooked Rice by Consumen lnfluence of Wild Rice on the Nutritional Value 126 Data are presented in Table l. Duncan's multiple test was used to determine statistical differences at the 1% level of significance. There were significant differences among cultivars with regard to iron and zinc content in both brown and milled rice, as well as in the effect dueto milling. As expected, brown rice contained higher amount of both iron and zinc than milled rice. O. glabe"ima had the highest content of iron with regard to brown rice followed by Fedearroz 50 and Oryzica 1, whilst O. barthii had the highest content of zinc, followed by Fedearroz 50 and three accessions of O. g/abe"ima. The effect of milling in reducing the contents of both iron and zinc is seen in Table 8. Milling reduced by 59 and 26%, respectively the content of iron and zinc. However, there were genotype differences. O. glaberrima lost 88% of its iron followed by the breeding line CT 13956-29-29-M (Bg90/0 . g/aberrima), IG 1 O (an acc. O.glaberrima), and the line P 1274-6-8-M-1-M. It is encouraging to see that CG 14 ( different acc. O. g/abe"ima ), O. rufipogon and Oryzica 1 had the highest content of iron after milling, suggesting its potential use as parents in a breeding program to increase the content of iron in commercial rice varieties. In terms of zinc, O. barthii, CG14, IGlO andO, glabe"ima had the highest content after milling. Results suggest that wíld rice species can contribute to improve the nutritional quality of commercial rice varieties. lt is noteworthy to mention that Oryzica 1 and Fedearroz 50 are good examples of improved varieties developed in Colombia out of the CIA T/FEDEARROZ breeding programs and no breeding effort was made to improve its nutritional value. However, given the fact that these varieties posses a good level of iron and zinc suggest that it should not be difficult to develop improve varieties with a better nutritional value. This is supported by the fact that the correlation coefficient to breed simultaneously for iron and zinc was 0.53. These results are in agreement with data from IRRI (Gregario et al 2000) and W ARDA.l998. Table l. Effect of milling on grain Iron and Zinc content of selected rice cultivars lron Zinc Cultivar Brown Rice Milled Rice Milling Brown Rice Milled Rice Milling Effect Effect% % Bg90-2 7.2 5.1 29.1 17.3 13.9 19.5 Barthii 10.4 4.2 60. 1 27.9 22.0 21.1 CG-14 10.8 6.3 41.3 24.8 19.7 20.4 CT13956-29-M-3-M 10.8 3.0 72.1 18.4 11.9 35.3 Fedearroz 50 14.0 4.8 65.9 25.6 16.7 35.0 IG10 12.3 3.7 70.1 24.8 18.1 27.0 O. g/aberrima 30.4 3.6 88.0 25.0 19.2 23.3 Oryzica 1 13.5 6.1 61.8 16.5 11 .0 24.2 Oryzica Llanos 4 13.0 4.9 54.4 20.8 15.7 33.3 P1274-6-8-m-1 -3-M4452 12.3 3.2 74.2 13.7 10.5 23.3 O. rufipogon 10.5 6.2 41.3 20.5 15.7 23.6 TOTAL 13.2 4.6 59.3 21.4 15.9 26.0 127 References Gregorio GB, D.Senadhira, H. Htut, RD. Graham 2000. Breeding for trace mineral density in rice. Food and Nutrition Bulletin, 21:382-386. W ARDA 1998. Highlights of 1998 Activities. W ARDA, Bouaké. Cote d'Ivoire, 24pp Welch RM, RO Graham 2000. A new paradigm for world agriculture: Productive, sustainable, nutritious, healthful food systems. Food and Nutrition Bulletin 21: 361-366. 1.2.20 M.A.S. for BGMV resistance in the common bean Constanza Quintero1, H. Terán2, S. Beebe1; and J. Tohme1 1SB-2 Project; 2 IP-1 Project Introduction In October 2000, F 1 plants derived from multiple crosses (BGMV 222 to BGMV 268) were planted at Santander de Quilichao (Cauca, Colombia). These plants were screened for the presence of the bgm-1 marker (DOR21 ). A number of 1191 plants out of 2971 were found to have the marker and were then selected for generating new crosses and generation advancing. From January to August 2001, three separate plantings were done. In January five nurseries were screened for the presence of the bgm-1 marker. The first was a group of 12 F 1 multiple crosses (BGMV 269 to BGMV 280), in which 536 plants out of 904 had the marker and were used to generate new crosses. The rest of the nurseries were screened in order to evaluate the segregation of this DNA fragment. The main purpose was to introduce the bgm-1 gene in those bean varieties that are widely planted in Central America. Then eight plants per row of each F4 backcrosses with DOR 364, DOR 390, DOR 500, ICT A OS TUA or ASO 1 were sampled individually. Families that had a high proportion of the bgm-1 marker were harvested and advanced to the next generation. Al so 14 F 3 populations including the variety MD-2324 (Bribri) were evaluated. This time four individual plants of each population were screened with the bgm-1 marker. Only the two that were positive (marker present) were harvested. A wide number of F5 families bred for Central America were also selected with the BGMV marker. Small red (168) and small black lines (166) were sampled for DNA amplification. Four plants per line were sampled as a bulk, and DNA was extracted using the alkali method with s light modifications in the amount of solutions used. The volumes of 0.25M NaOH and 0.25M HCI were increased from 40 to 60 ¡ .. d, and 30 }11 of 0.5M Tris-HCl (pH 8) was used instead of 20 }11. The rest of the protocol remained the same. After PCR, 94 small red families and 41 small black families having the bgm-1 marker were se lected. In addition, 84 F4 families of G 22041 (Garbancillo Zarco) were selected among a group of 157. In March another nursery ofF 1 multiple-cross hybrids was planted. High-throughput M.A.S. was improved with the evaluation of the BGMV SCAR in a nursery of 7085 plants. A total of 3253 128 (46%) plants were found to have the bgm-1 marker and selected for drought evaluation in the next generation. A total of28.7 days was spent in screening the marker in this nursery (7085 plants) {Table 1) in order to compare the efficiency of screening twice the usual number of plants at one time. Given that a different number of people are involved in each trial, we defined person-days (p-d) as the amount oftime (days) that one person would spend while carrying out each activity (no. ofpeople times no. of days). The screening of a set of 7000 plants took 55.2 p-d (or 23.7 p-d for 3000 plants). Thus, in comparíson to a previous tria! of 3000 plants that required 30.7 p-d, the p-d required to evaluate the BGMV SCAR was reduced significantly (Table 1). Thís was accomplished through the utilization ofhigh-technology equipment that accelerated alllaboratory processes such as electronic pipettes (Finnpipette Biocontrol) and the Hydra 96 micro-dispenser (Robbins Scientific) often u sed for large-scale genomics and high-throughput screening. Table l. Improving large-scale screening for selecting bgm-1 marker. Previous Tria1 Last Tria1 Task (N==30002 (N=70002 Persons Time Persons Time {no.} {Da;rs} {e- Transposon, containing a selectable marker (kanamycin resistance) and sequencing primer binding sites, into BAC DNA. In vitro reaction conditions have been optimized to maximize transposon insertion efficiency while minimizing multiple insertion events. In brief, about 200 ng of BAC DNA were subjected to transposition with the EZ::Tn Transposon as recommended by the manufacturer, and 0.5 J.ll of the reaction were used to electroporate 20 J.ll of DH10B E.coli competent cells. After recovery, aliquots of the cells were grown on plates containing chloramphenicol (Cm, 12,5 J.lg/ml) and kanamycin (Km, 50 J.lg/ml) to select for transposon insertion BAC clones. These were grown in LB broth with Km and Cm for BAC DNA isolation using the same protocol as befare (BRU, 132 Annual Report, 2000). BAC DNA was used for sequencing employing primers annealing the EZ::Tn transposon. Sequencher (Gene Cedes, Ann Arbor, MI).) is used to edit sequences and assemble contigs. Database searches are performed with the BLASTX and BLASTN algorithms (Aitschul et al., 1997). Results and discussion Molecular geneticists have seen multi-copy probes as disadvantageous molecular markers; however, as they may constitute complex gene families, their importance has been vindicated because they are more abundant than expected (M. Delseny, seminar at CIA T). This is the case for severa! of the common bean RGAs. Assuming that the multiple genes containing RGA 7 are physically linked, we used it as a unique probe to screen the common bean BAC library expecting the identification of overlapping clones. Seventeen clones were identified and subjected to fingerprinting with EcoRI, southem hybridization with RGA 7 and BAC End Sequencing. Two contigs were assembled using data from these procedures. One of the contigs included 11 out of the 17 BAC clones and 7 members ofthe RGA7 family (Figure 1). BAC clone 57-M14 contains four of these members and was selected for complete sequencing by the transposition insertion approach. Complete sequence of the BAC is important to unveil the structural organization of the cluster. As shown by BLASTX homologies of BAC End sequences, the genomic region containing the RGA 7 family is especially rich in retroelement-type sequences. Association between transposable elements and a cluster of R-genes, the Xa21 from rice, has already been reported and evidence for the involvement oftransposition in the evolution ofthe cluster provided (Song et al., 1997). ~~----~~~~~~~~ ----~~~~------~ ~--~ ~~~ ----~~ ~----- 3 Contig 7All 66-12 67-L9 57Ml4 51123 llA20 23A20 2IG2l 70Nl2 5N04 4N05 Figure l. A conrig of 11 BAC clones containing 7 members (stripped boxes) of the RGA 7 cluster was constructed based on fingerprinring, bybridization and BAC End Sequencing data. Key for BLASTX homologies of BAC End Sequences: 1 Maiu transposon En!Spm; 2 Mala te deshidrogenase; 3 Retroelement-type sequences; 4 Resistance Gene Analog. Ends witbout a number do not bave significant bomologies. BAC clone na mes are sbown at rigbt. 133 We have successfully sequenced 179 different transposed BAC clones either by one or both ends with an average reading of 400 bp per sequencing reaction, that account for 47.5 Kb of total sequence achieved. Nineteen contigs ha ve been assembled. The majority of the sequen ces are part of retroelements as indicated by BLAST searches. Another portion corresponds to sequen ces with no significant homologies. Finally, four contigs of 5.3, 3.5, 1.1 and 1.1 kb encode for R-genes. Interestingly, the R-gene contig of 5.3 kb contains a retroelement-type sequence in the middle of the coding sequence. About a half of the BAC remains to be sequenced. This work is currently under way. Future Work • Finishing and annotation of complete sequence of BAC clone 57-Ml4. • Identification and comparison of the complete sequen ce of the four RGA 7 members contained in BAC clone 57-Ml4. Oesign of primers to isolate by Jong-range PCR the remaining members ofthe family contained in other BAC clones. • Isolation of members of the RGA 7 family from parental line 619833 which contains resistance specificities in this family. References Altschul, S.F., Madden, T.L., Schllffer, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. 1997. Gapped BLAST and PSI-BLAST: a new generation of pro te in database search programs. Nucleic Acids Res 25: 3389-3402. Hammond-Kosack, K.E. and Jones, J.D.G. 1997. Plant disease resistance genes. Annu Rev Plant Physiol Plant Mol Biol. 48: 575-607. Song, W. Y., Pi, L. Y., Wang, G. L., Gardner, J., Holsten, T., and Ronald, P. C. 1997. Evolution ofthe rice Xa21 disease resistance gene family. Plant Cell. 9: 1279-87. 1.3.2 Screening of a common bean cDNA library to isolate full-length Resistance Gene Analogs (RGAs) I.F. Acosta and J. Tohme SB-2 Project Introduction Our set of Resistance Gene Analogs (RGAs) from common bean has provento be usefullanding markers of resistance Joci (BRU, Annual Report, 1999; López et al., manuscript in preparation). As RGAs may correspond to a truly functional Resistance Gene (R-gene), we are interested in obtain the fuii-Iength coding sequence of as many RGAs as possible. Screening of a cONA library with RGAs is not only a way to obtain their fuii-Iength sequences but also to establish if an RGA is actually expressed in the plant. We present the results of screening a custom-made cONA library ofbean (GffiCO-BRL) using RNA from young leaves. 134 Results Last year, attempts were made to screen the custom-made cONA library of common bean (BRU, Annual Report, 2000). However, tested approaches were not apparently efficient to isolate full- length clones. This year, 55296 clones were picked to 384 plates and arrayed on high-density filters (M. Muñoz and M. Blair). A set of these filters was kindly provided to our group by M. Blair and we screened them with three multiplexes of our set of 15 RGAs as hybridization probes using the same protocols to screen BAC library high-density filters (BRU, Annual Report, 2000). A total of 14 clones were identified. Their 5' and 3' sequences were obtained and compared to the GenBank database. Seven clones were considered as false positives because they were homologous to sequences other than R-genes. The remaining 7 clones corresponded to ROAs. Figure 1 depicts the expected structure of a cONA clone corresponding to a typical R-gene from the NBS-LRR class. 5' and 3'sequences of our cONA clones have been located according to their sizes and homologies in this scheme. As shown, none of the 7 clones corresponds to a full-length cONA. The same observation was made last year using different screening protocols. Therefore, we can conclude that the presence oftruncated cONA clones is a characteristic ofthis library. An R-gene as the one shown in Figure 1 may be 4 kB long. Thus, obtaining a full-length cONA is a complicated task and high-quality procedures must be done. · Surprisingly, all 7 clones corresponded to only two out of the 15 ROA probes used for the screening: ROA 1 and ROA2. Creusot et al. ( 1999) al so isolated members of this family in their cONA screenings. These ROAs probably constitute a large family of sequences in common bean that expressed at higher levels if compared to other R-genes. Indeed, R-genes are probably expressed at low levels (Scott Hulbert, pers. comm.). This may explain why it was not possible to isolate cONA clones for every ROA u sed as pro be. Screenings in the order of 1 0,000-1 00,000 clones are far of representation for rare transcripts. On-going work We are currently designing a strategy to isolate full-length cONA clones corresponding to R-gene homologues using rapid amplification of cONA ends (RACE, Frohman, 1988). This method does not require the construction and screening of libraries and is based on PCR amplification procedures. References Creusot F, Macadre C, Ferrier Cana E, Riou C, Geffroy V, Sevignac M, Dron M, Langin T. (1999) Cloning and molecular characterization ofthree members ofthe NBS-LRR subfamily Jocated in the vicinity ofthe Co-2 locus for anthracnose resistance in Phaseolus vulgaris. Genome 42(2):254-64 Frohman MA, Dush MK, Martin GR. ( 1988) Rapid production of full-length cON As from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci U S A 85(23):8998-9002 135 Figure l. Schematic representation of an NBS-LRR type R-gene and location of the 5' and 3' sequences (black bars) of7 cDNA clones corresponding toRGAs. Dotted lines represent portions of the clones that have not been sequenced. Numbers in parenthesis besides clone names are sizes in kb. Abbreviations in the R-gene scheme are as follows : 5' UTR, 5' untranslated regions; TIR, toll and interleukine-1 receptor domain; CC, coiled-coil motifs; NBS, nucleotide-binding site; LRR, leucine rich repeats; 3' UTR, 3' untranslated region. 5'UTR TIR orCC NBS 1 1 LRR 1 1 3' UTR -···············- -···············- -·················- -· ................ ~ ..... ·.· ........................... . _. .............................................. ·- _., .................................... -- -· ...................... . R-gene cRGA3 ( 1.7) cRGA4 ( 1.7) cRGA6 ( 1.7) cRGA7 (4.0) cRGA2 (4.0) cRGAI (3.0) cRGA5 (?) 1.3.3 Microsatellite repeats in common bean (Phaseolus vulgaris): Isolation, characterization and cross-species amplification in Phaseolus Gaitán, E1., M.C. Duque1•2, K.J. Edwards3 and J. Tohme1 1 SB-2 Project; 2 IP-4 Project ; 3 IACR, UK Introduction Phaseolus beans are distributed worldwide and are cultivated in the tropics, sub_tropics, and temperate zones. However, while more than 30 Phaseolus species have been described, only four are cultivated for human consumption. Of these, the common bean, P. vulgaris, is the most widely grown, the remaining four being the runner bean (P. coccineus), year-long bean (P.polyanthus) lima bean (P. /unatus) and tepary bean (P. acutifolius). In the present study we evaluated 68 primers pairs on a 21 P. vu/garis genotypes. In addition to these, 9 accessions of P. acutifo/ius, P. coccineus, P. polyanthus P. /unatus were also analyzed to check cross-specific amplification, . Materials and Methods Microsatellite primer characterization A total of 21 P. vu/garis genotypes were chosen for the evaluation of the primer pairs. This was made up of 14 wild accessions from CIAT's wild core collection and 7 cultivated accessions from the Mesoamerican and Andean pools. In addition to these, 3 and 2 accessions each of wild and cultivated P. acutifo/ius and P. coccineus, respectively, were also evaluated. Also, two genotypes of wild and cultivated P. polyanthus and two wild genotypes from P. /unatus were also analyzed 136 to check cross-specific amplification, bringing the total number of bean accessions use in the study to 30. ::- The PCR reaction was carried out in a 20¡.!1 final volume containing 20ng of genomic ONA, 0.1 ).1M of each of the forward and reverse primers, 1 O mM Tris-HCL (pH 7 .2), 50 mM KCL, 1.5mM to 2.5mM MgCL2 depending on the primer combination, 250mM of total dNTP and 1 Unit of Taq ONA Polymerase. Temperature cycling pro file involved an initial denaturation step of 2 min duration at 94°C. This was then followed by 35 cycles with each cycle made up of denaturation at 94°C for 15 sec, an annealing phase ofbetween 48°C and 65°C (depending on the annealing temperature for the given primer pair) for 15 sec and extension at 72°C for 15 sec. A volume of 6).11 fonnamide containing 0.4% w/v bromophenol blue and 0.25% w/v xylene cyanol FF were added to each reaction, denatured and then 4)llloaded on 6% denaturing polyacrylamide gels (19: 1 acrylamide: bis-acrylamide), contained SM urea and O.SXTBE. Electrophoresis was at 100-W constant power for between 2 and 2.5 hours. PCR amplifications were visualized by silver staining according to the Manufacturer's guide (Promega Inc., USA) with sorne modifications. Data analysis Gels containing PCR amplifications were scored for the presence or absence of alleles generated by each pair of primers for all 30 individuals. In this study, we used the Discriminating Power, (O) to compare the efficiency of the microsatellites to differentiate between genotypes . "O" value represents the probability that two randomly chosen individuals show with the microsatellite locus different allelic pattems, and thus are distinguishable from one another. If Pi is the proportion ofthe population carrying th banding pattem, and calculated for each microsatellite locus (Tessier, et al, 1999), then D=1-1:p/. This is an extension of the Polymorphism Infonnation Content or PIC (Anderson ET. Al., 1993) available from the frequencies of the different banding pattems (or genotypes) generated by a primer. Results Characterization of selected microsatellite sequences into P. vulgaris Sixty-eight primers were used to investigate the polymorphism detected among 21 iñdividuals of P. vu/garis. Fourteen loci were monomorphic in the materials tested. A total of 584 alleles were detected at the 68 microsatellite loci. The number of alleles per microsatellite locus ranged from 1 to 14 with an average of 6.3 alleles per primer pair. Also, 1-19 different banding pattems were generated. We used the data from the microsatellite loci and their corresponding alleles and pattems per loci to calculate the Discrimination Power (O) in arder to examine the extent of infonnation on diversity that these markers can provide for P. vulgaris. The O va1ue in the present study ranged from 0.09 to 0.94, with an average of0.73 for loci with more than 1 banding pattem. We observed that 73% ofthe loci had more than 50% probability discriminating between two individuals. As reported by Tessier, et. al., 1999, we found that the analysis of discrimination power (O) revealed that the efficiency of a given primer does not depend only on the number of pattems it generated. For example, primers BM170 and BM140 produced the same number of pattems and alleles, but they also had different discriminating powers. On the contrary, primers BM153 and BM164 with quite a different number ofpattems (7 and 11, respectively) had similar discrimination powers. Primers BM 188, GA TS 91 and BM 143 ha ve D val u es higher than 0.90, which means that they would be very useful for genotyping P. vulgaris gennplasm accessions. 137 The high Discriminating power values ex.hibited by microsatellites make them the markers of choice to saturate ttie bean map. Conservation of microsatellite loci across Phaseolus species The conservation of these 68 loci in Phaseo/us was examined by evaluating them with P. coccineus, P. polyanthys, P. acutifolius and P. lunatus . The first two species are more closely related to P. vulgaris as they, belong to the same lineage than to P. lunatus, which is more distant and belongs toa different lineage (Fofana, et. al., 1999). Primers BM16, BM 138 and BM195 fai led to amp1ify any non-source; GATS11B, BM150, BM167, BM188, BM195, BM199 and BM 200 amplified in almost all species, just 50% or less of the total of individuals . Thirty three microsatellite markers produced amplification products in all samples with sorne of them showing length variability between species. This indicares that a considerable leve! of sequence conservation exists within the primer regions flanking the microsatellite 1oci. Eighteen microsatellite loci produced non-specific amp1ification products in P. lunatus. This occurred mainly with primers BM16, GATS11B, BM16?, BM195, BM138, BM25, BM150 and BM200 that failed to amplify in 50% or less of the tested species. Nevertheless, the uti1ity of the bean primers in producing PCR-amplified products across the genus has been demonstrated. It must be noted however thatjust as has been pointed out in other studies on SSR loci conservation in plant species, a decline of amp1ification success was observed with increase of genetic distance (White and Powell, 1997, Roa, 2000). References Anderson, J.A., G.A. Churchill, J.E. Autroque, S.D. Tanksley, and M.E. Swells. 1993. Optimizing selection for plant linkage map. Genome 36:181-186. Fofana, B., J.P. Baudoin, Vekemans, X., Debouck, D.G. and du Jardin, P. 1999. Molecular evidence for an Andean origin and a secondary gene pool for the Lima bean (Phaseolus lzmatus L.) using chloroplast DNA. Theor. Appl. Genet. 98:202-212. Promega.l995. Technical Manual: Silver sequence DNA sequencing System Roa, A.C., P. Chavarriaga-Aguirre, M.C. Duque, M.M. Maya, M. W. Bonierbale, C. Iglesias, and J. Tohme. 2000. Cross-species amplification of cassava (Manihot escu/enta) (Euphorbiaceae) microsatellites: allelic polymorphism and degree ofrelationship. Am.J. Bot. 87:1647-1655. Schoonhoven, A., andO. Voysest.(Eds). 1991. Common Beans: Research for Crop Improvement. 980 pp. Redwood Press. Scott, K.O., P. Eggler, G. Seaton, E.M. Rossetto, L.S. Lee, and R.J. Henry. 2000. Analysis of SSRs derived from grape ESTs. Theor Appl Genet 100:723-726 Tessier, C., J. David, P. This, J.M. Boursiquot, and A. Charrier. 1999. Optimization of the choice of molecular markers for varietal identification in Vitis vinífera L. Theor. Appl. Genet. 98: 171-1 77. Whie G., and W. Powell. 1997. lsolation and characterization of microsatellite loci in Swietenia humilis (Meliaceae): and endangered tropical hardwood species. Mol Eco! 6:851-860. 138 1.3.4 Microsatellites isolated from common bean_cDNA and small- insert genomic libraries M.W. Blair1, F.Pedraza\ M. Muñoz1, M. C. Giraldo1, E. Gaitan1, J. Tohme1, D. Frisch2, R. Wing2 1SB-02 Project; 2Clemson University Genome Institute. Introduction Microsatellites based on simple sequence repeats have been developed for a wide range of plant species. V arious techniques exist for discovering new microsatellite markers from anonymous genomic sequence. AH of these techniques rely on the availability of DNA libraries. It is best if these libraries consist of small-insert clones (< 1 kb) because they are easier to screen for microsatellites and to sequence if they are shown to be positive. V arious enrichment procedures have been used to increase the prevalence of simple sequence repeats in genomic 1ibraries. One method relies on the selective capture of small fragments with oligonucleotides on hybond membranes (Edwards et al., 1996). An enrichment method has been used at CIA T to develop two GAn and CAn microsatellite-containing libraries for common beans (CIA T Annual Report 1999- 2000). In this study, we were interested in screening non-enriched libraries with simple sequence repeat motifs that have not been screened in the enriched libraries. The screening of non-enriched random genomic libraries will aUow us to investigate the normal frequency with which different microsatellite motifs occur in common bean and to develop a new set of genomic microsatellites for mapping and tagging projects in common bean. Materials aod Methods Libraries: Four libraries were screened for microsatellites. The largest library was a leaf cONA library (pCMV Sport 6.0 vector) consisting of 64,128 clones. In addition, three small-insert genomic libraries were screened. These are named after the restriction enzyme that was used to generate them. They were: (1) Rsai library (19,584 clones) 2) Alui library (18,432 clones); and 3) Haeiii library (19,968 clones). These libraries were made with frequently cutting restriction enzymes as described in more detail in the 2000 Annual Report. All three small insert libraries were made with DNA from the genotype DOR364 that was digested with the appropriate restriction enzyme and size selected to the range of 0.4-1.2 kb. The small-insert libraries for A fui and Haem were made with the standard pBiuescript II KS+ vector while the Rsal library was made with pGEMt-easy vector. The clones for each library were picked by a Q-bót robot and spotted onto gridded Hybond N+ membrane filters with six fields each with a 96 position double- replicate 4 x 4 pattem. Each of the small insert libraries was contained on a single filter while the cONA library was spread over a set of four filters. AH the clones were stored in 384-well plate format glycerol stocks that were copied twice into working and master copies ofthe library. Filter hybridization and library screening: Six simple sequence repeat motif oligonucleotide probes were used to screen the cONA and small-insert library filters . The probes included two that targeted dinucleotides repeat motifs, namely (CAho and (GA) 20 ; and four that targeted trinucleotide repeat motifs, namely (AA n 14; (CAG) 14; (CAA) 14; and (ACG) 14, where the sequence within the parenthesis indicates the repeat and the number outside the parenthesis indicates the number of copies of that repeat. Hybridization consisted in end labeling the simple sequence repeat motif pro bes with T 4 DNA Kinase and hybridizing this pro be to the DNA contained on the filters with standard protocols (Edwards et al., 1996). Briefly the filters were pre-hybridized in 1 00 mL of hybridization buffer for 4 - 6 hours at 60°C. Meanwhile, 1 O pmoles ofprobe was end labeled with 1 ul ofT4 polynucleotide kinase and S ul ofyP32-ATP in a total 139 reactíon volume of 20 ul that was incubated at 3 7°C for 80 minutes and stopped at 65°C for 20 minutes. The labeled oligonucleotide was ádded directly to the pre-hybridizing filters and incubated at 60°C for 12 hours. After the hybridization step, the filters were washed twice at 60°C with 6X SSC 1 0.1 % SOS for 5 minutes each. Longer washes were used when signa! was intense (> 100,000 cpm). The filters were blotted dry, covered with saran-wrap and arranged face-up in cassettes along with three sheets of X-ray film. The films were taped to each other so that they would not shift during the Oto 8 day exposure in a -80°C freezer. Films were removed at three exposure intervals: after 2 hours, after 1 day and after 1 week to identify high, medium and low signal positive clones, respectively. Filters were re-used for sequential screening of different oligonucleotide repeats by stripping them between each use. Stripping consisted in washing the filters at room temperature in 1 OOmM NaOH, 1 OmM EDT A, 0.1% SOS twice, followed by a 5X SSPE rinse for 1 O m in. Clone identification: Positive clones were identified by which filter they were on; which field within the filter they were in, and what address they had within the field. Each filter contained six fields and each field contained the equivalent of eight 3 84-well pi ates worth of clones, for a total of 48 plates per filter. Clones could be identified by their position in the double-replicate 4 x 4 pattem found at each grid axis in the address system. Only double-spotted clones were selected. Any spots for which the replicate did not hybridize were considered false positives and were not selected. Putative simple sequence repeat-containing clones were picked from their appropriate position in the 384-well plate format glycerol stocks Clone sequencing: The positive clones were sequenced initially from one end of the insert. In the case of the cONA library, the 5'side of the clone was sequenced using either Sp6 or Ml3 Reverse primer, while in the case of the small insert libraries the clones were sequenced with either T7 or TI high temperature primers. The sequences were searched for vector segments to check for insert integrity and were screened for simple sequence repeats with the program Sequencher. Any cDNA clone that did not contain an SSR in the 5'end was sequenced again from the opposite end with the T7 or Ml3 Forward primers. Results and Discussion We were interested in calculating the overall frequency of di- and tri-nucleotide repeats in coding versus non-coding DNA and in the frequency of GA versus CA repeats in common beans . Therefore our hybridization strategy was the following: the small insert library-filters were hybridized with each dinucleotide probe separately, the cDNA filters were hybridized with a mix of both dinucleotides probes; and the trinucleotides probes were hybridized simultaneously in both libraries. We limited the comparison between the di-nucleotide probes to the small~insert genomic libraries rather than the cONA library, because we reasoned that di-nucleotide repeats occur with greater frequency in non-coding DNA than in coding regions. This screening technique allowed us to calculate the frequencies of di- and tri-nucleotide motifs in each library. Table 1 shows the number of positive clones found in each hybridization experiment with each of the libraries and with each set of oligonucleotide probes. The total number of clones screened in the three small insert libraries was 57,984, while the total number of clones in the cDNA library was 64,128. Surprisingly the CA probe hybridized to more clones (190) than the GA probe (24) in the two libraries that were evaluated for this purpose, Alui and Haeiii . These two libraries were chosen because they would be contrasting, given that the Haem library represents the GC rich fragment of the genome due to its restriction enzyme recognition site, while the Alui library does not. The results indicate that either GA motif is more common than CA motif in the bean genome or that the hybridization worked better for this probe. Neither hypothesis is satisfactory, 140 since the first possibility contradicts the higher frequency found for GA motifs in many other plant species, while the second possibility is unusual giveñ that both oligonucleotides have the same melting temperature. The frequency of the tri-nucleotide positive clones was as high as that of the di-nucleotide positive clones, except in the Rsai library and in the cONA library, were trinucleotide positive clones were more frequent than di-nucleotíde positive clones. These results may indicate that trinucleotide repeat are more common in coding sequences than across the entire genome. Sequencing results confirmed whether the hybridizing clones from each library were false positives or not and the number of repetitions that each positive clone contained. The rate of false posítives was higher for clones selected by placing the hybridized filters on film for one week of exposure versus those that were on film for two hours .(Figure 1 ). Meanwhile, the sequencing confirmed that as might be expected, the clones with the highest intensity signa! had a larger number of repeats. The sequencing results showed that dinucleotide clones were more common than trinucleotide clones in the small insert libraries and about equal in the cONA library (Figure 2), contradicting the earlier results based solely on hybridization. Among the 264 dinucleotide repeat containing clones identified so far, the CA motifs were 50% more common than the GA repeat in the small insert library, but only 30% as common as the GA rei>eat in the cONA library (Figure 3). The AT motifs were more frequent in the small insert library than in the cONA library. When AT microsatellites did occurr in the cONAs they were more likely to be found in the 3' end than in the 5'end ofthe clones. This was in contrast to the GA and CA microsatellites which were three times more frequent in the 5'end than in the 3'end ofthe cONA. The cONA library hada higher proportion oftetra- and penta-nucleotide simple sequence repeats than the small insert libraries, although these motifs had not been screened for specifically in the hybridization experiments. We believe that the screening of the small insert libraries has given us an accurate picture of the relative frequency of different SSR motifs in the bean geno me based on the fact that these libraries each represent a total of 1 O Mb of ONA. Considering that the genome of common bean is 650 Mb in size, each library is equivalent to 0.015 X genome equivalents and taken together the three libraries contain approximately 30 Mb of bean DNA which should be equivalent to 5 % of the total bean genome. An estímate of the relative abundance of microsatellites on an absolute scale per kb of genomic sequence could therefore be obtained. Similarily these results give us a good idea of which motifs are common in cONA se_quences and where these are located. The large majority of the repeats occurred in the S'untranslated region (UTR) upstream of the start codon for the open reading frame (ORF) where these could be identified. Another proportion of the SSRs occurred in the 3' UTR, while fewer were found within the ORF. The total percentage of ESTs that might be expected to contain SSRs based on this study is approximately 0.7 %. Non-SSR containing cONA sequences are a useful resource ín their own right since they add to the body of ESTs available for the crop. The present work brings to a total of approximately 2000 the nurnber of ESTs for beans generated by this project. The sequen ces represent both 5' and 3' ESTs, sin ce they come from a directionally cloned cONA library. The sequences have been BLAST searched against each other to check for sequence redundancy (redundant clones consist in 90% identity over half the nucleotides in a sequence) and the vast majority of the sequences were unique. The sequences were also compared to the Swissprot database and to all the soybean proteins downloadable from Genbank. This collection of new gene sequences for common bean represents many more individual sequences than are currently in the Genbank public database for all Phaseolus species together. 141 Table l. Number of positive clones detected by hybridization of each Iibrary, at each exposure time and with each oligonucleotide probe. LIBRAR Y PROBE EXPOSURE1 Alul Haelll &a e cDNA2 2nt CA 6 2 63 54 3 22 44 GA o 29 55 2 15 3 11 231 3 3 2 28 87 tota12nt 109 105 68 373 3nt o 6 25 2 21 10 4 95 3 8 32 53 13 tota13nt 30 42 63 133 TOTAL 139 147 131 506 ll exposure times: 1 =2 hours; 2=a day; 3= a week 'J/ both oligo probe were added in the same hyb reaction % positive clones "' : ;;<. • ""- '"'"" """ •. .. 40 :., ...... :lli' .. 1C~ ~ ... 30 11! ~ ~ 20 r 1¡,, ,. "..., • '!' 10 ., ... ~ ~"' - .. o ... .. 111,. "'"'""' ... 21v 1 dly Ume of l x.pMutl on 111m lweek number of repetitions 18.00 l""'i\1 .J'I !' ~íi .. &:'.&'l.;.E"" ':!! ..!1 :itl' "'" 18.00 .•• "'""":.,p.. ~ \'1 = 14.00 ""-·· ...... - " .A1:. 12.00 ""' SS 10.00 .... " -~ .. -- Rs 8.00 o ~ ·'Al" : ~- ...__ ~ e.oo .111 ~ .f!l r w ~ 4 .00 ~ ··¡¡¡; ---,;:- 2.00 Ir. "" 0.00 ,,. 21v 1 dly 1wHI< timt of expoe\R on n1m Figure l. Percentage positive clones and number of repetitions for clones selected from hybridized filters that were left for three exposure times of 2 hours, 1 da y and 1 week. TETRA TRINUCLEOTIOE cONA library TRINUCLEOTIOE 22% Small lnsert Libraries TETRA 2% OINUCLEOTIOE 76% Figure 2. Percentage positive clones from the cDNA and small insert libraries that contained di-, tri-, tetra- and penta-nucleotide simple sequence repeats u pon sequencing. 142 • ! o ü o ci e llg(ct/galtc cONAiibrary ec/gt/~tg atila SSR motlf total a amall inaert •5'endc0NA o3'end cONA Figure 3. Number of GA, CA and AT motif dinucleotide simple sequence repeats found in tbe sequenced small insert fibrary clones and in tbe S'and 3' ends of the cONA clones. Future plans We are sequencing all positive SSR containing clones to design microsatellite primers for the development of new markers and hope to recover a large number of polymorphic markers for the crop. The sequenced cONA clones represent the first substantial number of EST sequences in beans. Ultimately many of these cONA sequences can be genetically mapped as RFLPs or SNP (single nucleotide polymorphism) based assays, especially as simple procedure such as dense chips, become available. Reference Edwards KJ, Barker JH, Daly A, Jones C, Karp A (1996) Biotechniques 20: 759-760. 1.3.5 Legume microsatellites tested in common bean Blair MW1, Pantoja W1, Pedraza F1 ,Cregan P2 ;Fatokun C3 1SB-2 Project ; 2USOA-Beltsville, USA 3IITA-Ibadan, Nigeria Introduction A large number of microsatellite markers have been developed for soybean by the USOA (Cregan et al., 1999) and at companies such as Oupont (Peakall et al., 1998). Relatively fewer microsatellites ha ve been developed for other tropical legume crops. However at liT A an effort is underway to develop cowpea microsatellites (C. Fatokun, pers. Comm.), and an initial set of peanut microsatellites is also available (Hopkins et al., 1999). The large number of common bean 143 microsatellites made at CIA T recently would also be useful for mapping in other species of legumes. Our objective in this study was to try to adapt the microsatellites available for other Phaseoleae legume crops ( especially soybean and cowpea) to common bean and to test common bean microsatellites in a panel ofthese other legumes. Materials and Methods The microsatellites tested included 423 from Gly cine max (408 with ATin motif, 3 with GAn motif and 12 from coding sequences) and 118 from Vigna (mixture of GA, CA, AT and compound motif, genomic microsatellites and two cONA based markers). An additional four cowpea microsatellites were designed from sequences in Genbank (Table 1 ). The markers were tested against a panel of nine legumes, that included the soybean, Mesomerican and Andean common bean genotypes (Williams, OOR364 and G 19833, respectively) that were the sources of microsatellite libraries made for these crops. For the other legume species we used genotypes that were representative of varieties grown in the Andean region. Primer amplification was tested with a range of conditions. The soybean microsatellites were tested initially with lax amplification conditions using 45 to 47°C for annealing temperature and 2.0 to 2.5 mM final concentrations of MgCI2. The cowpea microsatellites were amplified with 52°C annealing temperature and 2.5 MgCI2. Both sets of markers were analyzed on 2.0% agarose gels with ethidium bromide staining. The microsatellites with single amplification products were analyzed on 6% polyacrylamide gels with silver staining. Results and Discussion Soybean and common bean appeared to be especially divergent in regard to their microsatellite loci. The soybean genomic A ITn microsatellites generally did not amplify well in cornmon bean. The cONA based microsatellites were also poorly conserved. Table 2 shows the nine most transferable microsatellites and the molecular weight of the amplified products. Most other microsats amplified multiple bands from common bean, tepary bean, lima bean, cowpea or mung bean DNA that were completely different in size compared to the soybean allele. The likelihood that these represent homoelogous microsatellite loci was deemed low and these microsatellites were not investigated further. The total rate of transferability to the six species was between 2 and 0.5%. Therefore, it seems that soybean primers may be less useful in comrnon bean than we thought because of the evolutionary distance separating these species. This is in marked contract to studied with animals where microsatellites have been successfully transferred among related species such as birds, tortoises, primates (eg. gorillas/apes), ungulates (eg. horses/cattle) and rodents (eg. rats/mice). Ultimately it is the genetic distance between species and genera that determines the ability of SSR primers to amplify in different genomes and the ability to transfer microsatellites between species must be determined empirically. And it seems that in plants, unlike animals, microsatellite loci are not well conserved over large genetic distance between species. Cowpea genomic microsatellites were slightly more useful than the soybean genomic microsatellies for amplification in Phaseolus species. The transferability rates were between 7.6 and 11 .0 %. The gene-derived microsatellites were more conserved than the microsatellites from non-coding sequences, and four out of six primer pairs from Table 1 amplified well across the subtribe. For example one cowpea microsatellite, VM21 , amplified well in a range of legumes and represented a gene from Vigna radiata for ACC oxidase that contained an A Tn repeat in the 3' untranslated region. These results suggest that the close phylogenetic relationship between cowpea and cornmon bean allowed us to exchange sorne of the genomic microsatellites and most of the cONA based microsatellites. Based on these promising results, we tried the reciproca! experiment of amplifying the cONA based microsatellites we have developed for common bean in cowpeas other species. In these experiments, we found that transferability was much high 144 especially within the Phaseolus genus as expected, but also across into Vigna, G/ycine and Cajanus. In summary, we found that the pattern of amplification of legume microsatellites from soybean, cowpea and common bean, agreed well with the tribe, subtribe and genus designations of the legumes being tested. Why sorne related lineages maintain intact microsatellites over evolutionary time or why microsatellites in certain location such as upstream of genes are more conserved is not known. However, we can postulate that mutation rates for microsatellites are constrained around genes as compared to non-coding regions, just as they are for other forms of sequence variation. Future studies A set of common microsatellites that amplify across a range of tropical legumes, would be useful to study the synteny between species. Many legume genera are thought to have a conserved gene order, however this has been less well studied than in the grass family. Soybean, common bean and mung bean ha ve been shown to ha ve regions of macrosynteny across m u eh of their geno mes, however there has been no analysis of microsynteny or sequence conservation at homoelogous loci in the genomes of these related legumes. Refereoces Cregan PB, Jarvik T, Bush AL, Shoemaker RC, Lark KG, Kahler AL, Kaya N, Vantoai, TI, Lohnes DG, Chung J, Specht JE (1999) Crop Sci. 39: 1464-1490. Hopkins MS, Casa AM, Wang T, Mitchell SE, Dean RE, Kochert GD, Kresovich S (1999) Crop Sci. 39: 1243-1247. Peakall R, Gilmore S, Keys W, Morgante M, Rafalski A (1998) Mol. Biol Evol. 15: 1275-1 287 Table l. Microsatellites developed for cDNAs from Vigna. SSR Genbank gene Predicted Species Motif Repetition Predicted nurnber location MW VMdl U08140 Ca-dep. protein kinase 5'UTR V. radiata CAA (CAA)2-(CAA)5 116 VMd2 AB030294 CPRD86 cds parcial ORF V. unguiculata GAA (GAA)5-(GAA)4 120 VMd3 Y08624 Ted2 5'UTR V. unguiculata CTT (CTT)6- 115 VMd4 Z23083 Endo-1 ,4-beta-glucanase ORF V. radiata TCTTC (TCTTC)4 127 VM21 U06047 ACC oxidase 3' UTR V. radiata AT (AT)7-(AT) l0 179 VM22 M99497 proteina kinasa 5'UTR V. aconitifolia GA (GA) 12 2 17 145 Table 2. Amplification products of soybean, cowpea and common bea n m icrosatellites on a panel of economically-important legumes from the Phaseoleae tribe1• Family Fabacea Tri be Phaseoleae Sub-tri be Phaseolinae Genus Phaseolus Vigna Species P.v (M) IP.v (A) IP.a IP.I V.u V.r Variety DOR364 IG 19833 IG40001 iPeru Molina Mungo Source ot Marker Primer Microsatellite Type Soybean cDNA SoyPRP1 240 240 - - 240 240 cDNA SoySc5 14 260 260 - - 260 260 Genomic Satt 51 1 360 360 - 540 - 270 Genomic Satt 40 1 160 160 - - 500 670 Genomic Satt 206 140 140 - 220 - - Genomic Satt 237 320 320 320 320 320 320 Geno mi e Satt 305 - - - 290 - - Geno mi e Satt 411 - - - - - 290 Geno mi e Satt 275 140 140 140 - - - Transferability (%) 1.89 1.89 0.47 1.18 1.18 1.66 Cowpea Genomic VM26 - - 90 160 160 140 Geno míe VM63 300 300 300 300 300 190 cDNA VM2 1 208 202 208 208 240 240 Geno mi e VM9 1 330 330-510 330 330 330 150 Geno mi e VM98 160 160 160 160 160 160-90 Geno mi e VM114 270 270 - - 270 270 Genomic VMI1 8 280 280 280 280 280 280 Transferability (%) 11.02 11.02 10.17 9.32 7.63 100.00 Common Bean cDNA BMdl 204, 164 204, 164 162, 166 174 204 - cDNA BMd2 108 104 - 100 108 - cDNA BMd3 226 226 - - - - cDNA BMd6 122 122 122 122 - - Genomic BMd l1 160 160 158 164 150 - Genomic BMd12 162, 125 164, 125 - ! 55, 125 164 - cDNA BMd 13 192 192 192 192 - - - cDNA BMdl4 188 188 - 188 205 - cDNA 8 Md16 148, 86 136 180 240, 122 205, 185 - cDNA 8 Md1 7 100 106 106 274, 100 106 - cDNA 8 Md1 8 242 156 242 /58 - 242, 148 242, 204 - cDNA BMd19 154 154 154 302, 164 172, 154 - cDNA BMd20 128 124 - 124 126 - cDNA BMd22 118 130 . 118 - 130, 11 8 - - cDNA BMd23 138. 126 138, 126 - 138, 126 138, 126 - cDNA 8Md26 140 148136 - 148, 136 160 - cDNA BMd30 145 145 150 11 5 - - cDNA BMd32 112 112 - 106 270, 224 - cDNA BMd33 100 110 - 95 - - Transferability (%) 100.00 100.00 22.58 58.06 41 .93 na Glycinin. Cajanin. Glycine Cajanus G.m C.c Williams 1S-10 120-240 240 120 - 250 470 170 - 220 - 250-320 . 320 200 - 100 - 160 - 100.00 0.71 140 140 190 190 260-202 - 330 330 160 160 270 - 280 300 11.02 5.93 204, 166 204 - - 226 - 122 - - - 170, 160 - 192 192 200 - 180 276 222, 170 - 248, 104 250,235 308, 148 154 138 188, 174 138, 126 138, 126 - - 150 - 312 222 - - 45. 16 22.58 1/ P.v. (M) : Phaseo/us vulgaris- Mesoamerican; P.v. (A): Phaseolus vulgaris- Andean; P.a: Phaseolus acutifo/ius; P. 1: Phaseolus lunatus; V.u: Vigna unguiculata; V.r: Vigna radiata; G.m: Glycine max and C.c: Cajanus cajan. 146 1.3.6 Enhanced microsatellite map developed for common bean M.W. Blair, F. Pedraza, E. Gaitán, J. Tohme SB-2 Project Introduction Microsatellites are polymerase chain reaction (PCR) based markers that detect length polymorphisms at loci with simple sequence repeats. They are also single-locus markers that are specific to a given place in the genome. Microsatellites are advantageous because they are readily amenable to relatively high throughput marker assisted selection strategies. The specificity of microsatellite marker for use in MAS selection depends on tight genetic linkage of the marker with a gene that produces a reliable phenotype. Genetic maps are needed to determine where microsatellites are located in the genome and what genes they may be linked to. Microsatellites have been found to be distributed densely throughout the genomes of higher plants, making them very appropriate for genetic mapping. Nearly-saturated microsatellite maps are now available in severa) crop plants including soybeans, rice, wheat, barley, etc. It would be tremendously useful to have a genetic map for common bean consisting entirely of microsatellites. These second- generation markers would be easy to assay and would enable a large number of segregating individuals to be analyzed in gene and QTL tagging studies. For now, we have implemented a set of over one hundred and fifty microsatellite markers in genetic mapping studies for common bean at CIA T. These microsatellites come from genomic sequences, gene or cDNA sequences and database searches mostly of common beans but also of other legumes as described in other sections ofthis annual report. Materials and Methods The parents ofthe DOR364 x Gl9833 RIL population were surveyed for polymorphism with 153 microsatellites. The markers belonged to several different classes as shown in Table l. BM microsatellites are genomic, while BMy and BMc microsatellites are based on cDNA. The BMd and Clone microsatellites are based on both genomic and cDNA sequences. The microsatellites are in varying stages of development. Clone microsatellites will be given a new designation under the BM naming system once they ha ve been further tested. The polymorphic microsatellites were used to amplify DNA from the 87 recombinant inbred line progeny and the parents of the population. PCR product were run on silver-stained polyacrylamide gels and scored for the parental allele that they represented. Segregation data was used to place the microsatellites on a genetic map constructed with RFLP, RAPD, AFLP and SCAR markers (CIA T annual Report, 1998) using the software application Mapmaker. Results and Discussion A total of 81 microsatellites (52.9% of those tested) were polymorphic for the parents of the DOR364 x G 19833 population and could be located on the genetic map of common bean. An additional 5 microsatellites from Yu et al (1999) could be placed by comparative mapping between the DOR364 x G19833 and BAT93 x JaloEEP58 (Freyre et al., 1998) maps. Each chromosome was tagged with at least three microsatellites (Table 1). Two chromosomes, B02D and B04B had a relatively greater number of microsatellites, 14 and 12 markers respectively, placed on them. The average number of markers per chromosome was 7.4 microsatellites each. Considering that the total genetic distance of the entire map was 2453 cM, the average distance 147 between microsatellites was 30.3 cM. However many larger gaps remain and these will need to be filled with additional microsatellites. Other AFLP, RAPO and RFLP markers are available between the microsatellites, Since the entire map consists of 434 genetic markers, the average distance between markers was 5.7 cM. The markers BMy-11, BMd-28 and BM205 consistently amplified two po1ymorphic bands per reaction, which are suspected to represent duplicate loci. In another mapping population, the two BMy-11 loci co-segregate and are presumed to be very tightly linked (under 1 cM). Severa! microsatellites designed for related phytohemagglutinin and lectin gene sequences amplified single bands but were found to co-segregate, indicating that this gene family clusters at a single location in the genome. The microsatellites mapped during the course of this research will be invaluable for marker assisted selection beca use they are simple to analyze, specific for single genes of interest and diagnostic in most crosses due to their high level of polymorphism. Future Plans We will be studying the potential of specific microsatellites to be used in MAS selection for specific genes with which they are linked. The mapped microsatellites will provide a good set from which to chose markers for diversity studies and future QTL analysis. We planto develop a set of fluorescent microsatellites for accurate allele calling and high-throughput mapping. A full set of anchor markers for the bean genome will probably require 300 or more working microsatellites, therefore the work of developing and mapping new microsatellites will continue. Of the 153 microsatellites evaluated, a total of 43 were monomorphic (28.1 %) and 29 did not amplify well. We will try to recover these microsatellites by amplifying with variable PCR conditions and by mapping the monomorphic markers in well-integrated populations such as BAT93 x Jalo EEP558 or in populations where polymorphism is high, such as wild x cultivated crosses across genepools. Table l. Number of microsatellites from each marker designation mapping to each chromosome in the CIAT genetic map based on the cross (DOR364 x Gl9833). BOl 802 803 804 805 806 807 808 809 810 B11 BM 8 7 2 6 2 5 3 1 BMd 4 2 4 5 2 3 Bmy 5 BMc 2 Clone 2 2 2 TOTAL 3 14 10 12 8 4 6 5 7 6 6 1.3.7 Isolation and characterization of Tyl-Copia group retrotransposon LTR sequences in Phaseolus vulgaris L.M. Galindo, E. Gaitán and J. Tohme SB-2 Project Introduction UNLINKED 2 2 Tyl-copia retrotransposons are mobile genetic elements that seem to be ubiquitous in higher plants (Flavell et al., 1992; Voytas et al., 1992). They a1so show high insertional polymorphism 148 (Fiavell et al., 1992; Hirochika et al., 1992; Pearce et al., 1996) and preference for euchromatic regions (Hirochika et al., 1996; Flavell et al., 1998; Garber et al., 1999). In the last few years :- there has been a growing interest in exploding retroelements for molecular marker-based studies (Waugh et al., 1997; Ellis et al., 1998; Flavell et al. , 1998; Kalendar et al., 1999; Pearce et al., 2000), however, most research in the field of transposable elements is aimed at diversity, expression or structural analysis because isolation and characterization of retrotransposons for marker studies imply a long process. High levels of polymorphism, necessary for developing different techniques, are usually encountered in long tenninal repeats (LTRs), which cannot be amplified directly as they are specific for each retrotransposon. The technique developed by Pearce et al. (1999) makes it possible to isolate RNAse-PPT-LTR (ribonuclease-polypurine tract-long tenninal repeat) sequen ces from Ty 1-copia retroelements for molecular marker-based studies such as Sequence-Specific Amplification Polymorphisms (SSAPs), Retrotransposon-Based Insertional Polymorphisms (RBIPs), Inter-Retrotransposon Amplified Polymorphisms (IRAPs) or Retrotransposon-Microsatellite Amplified Polymorphisms (REMAPs). We used this technique to isolate the aforementioned fragments using Phaseolus vulgaris DNA from CIAT accession 019833 . Materials and Methods Previously extracted genomic DNA from CIA T accession G 19833 was used to isolate RNAse- L TR sequen ces by the method of Pearce et al. ( 1999) with several modifications. Sequen ce analysis was carried out using the programs Sequencher 3.0, Blast x 2.0, Clustal W 1.8 and Clustal X 1.62. · Results and Discussion From a library of 1152 clones, 45 were selected for sequencing. Ofthese, 24 were homologous to previously isolated sections of retrotranscriptase, ribonuclease or polyproteins when entered into Blast X using an E value of 1 O. Alignment of isolated sections of ribonuclease using Clustal W showed conserved aminoacids (Figure 1 ), which ha ve been previously reported by Pearce et al. ( 1999). High conservation might be important for enzymatic function, confmning ribonuclease's central role for reverse transcription and reintegration. In addition to the RNAse sequence, representative se(!tions for the polypurine tract and a fraction of the long tenninal repeat were detennined for each sequence (sorne sequences are shown in Figure 2). Variability seems to be low at ribonuclease, but the PPTs and LTR sections show high plasticity due to mutations created by retroe1ement enzymes on nonessential transposon regions (Biusch et al., 1997). A dendrogram constructed by Clustal X using neighbor-joining and bootstrap analyses helped detennine the relationships among the RNAse sequen ces (Figure 3 ). Five different groupings, supported by bootstrap and high percentages of sequence identity between its members were defmed, providing support for the existence of severa! Tyl-copia retrotransposon families. Sequences were also compared to homologous retrotransposon sections from other legume species, indicating a highly heterogeneous population of retroelements in legume species, but smaller differences between Phaseolus vulgaris RNAses (result not shown). 149 P6 ADIL TKALGKERFL TLRHKLGVLDLHLPT Btl004 ADILTKALGKERFLTLRHKLGVHDIHPPT Bt764 ADML TKALGKERFL TLRHKLGVLDLHHP- Pll ADML TKA VGKERFL TLRHKLGFMIFTYQL Bt737 ADML TKPLPK-RFFFLRNELGILDLNNLS Btll36 ADML TKPLPKERFFFLRNELGILDLNNLS Bt602 ADML TKPLPKERFFFLRNELG ILDLNNLS Bt490 ADML TKPLPKERFFFLRNELGILNLNNLS Bt731 ADIL TKPLPKERFFFLRNELGILDLNNLS Btll26 ADIL TKPLPKERIFFLRNELGILDLNNLS Bt966 ADIL TKPLPKERFFFLRNELGILDLNNLS Bt59 ADIFTKPLPKYRFFLSRNELGIL YSHNIS Bt454 ADILTKHLPKDRFFLLRNELGIINSHTLS Bt261 ADIL TKPLPKNRFLLLRNELGIVDSKNLS Pl7 ADML TKGLPTKQFEDL TCKLGMIDIHSPT Btl094 ADML TKGLPTKQFEDL TCKLGM/0/HSPT Bt438 ADML TKPLPSAKFDHCLNLAGIIHT --- Bt443 ADML TKPLPSAKFDHCLNLAGIIHT --- Btlll2 ADMLTKPLPSAKFDHCLNLAGIIHT --- Bt781 ADML TKPHPSAKFDHCLNLAGIIHT - - Bt293 ADIL TKPLPSAKFDHCLNLAGIIHT -- Bt877 ADML TKPLPP AKFDHCLNLASIIHT -- Bt814 ADMLTKVVRAKFEHCLDL VNILHI-- Bt829 ADILTKVVTRTKFEHCLDL VNILHl-- Figure l. Alignment of isolated fragments of RNAse sequences (* = identical or conserved aminoacids,: = conserved substitutions,. = semiconserved substitutions). Element a a nb PPT LTR Nd Pll 29 o GAGGAGAG TATIAAATAAAT 56 P1 7 29 -2 GAGGGGGAG TGTIGCATAATC 83 Bt59 29 40 GAAGAAAAA TAGGGGGAGAT 128 A Bt261 29 55 AGGGAGAAAAT TGTIGGTTI A TI 200 A Bt438 25 15 GAGAGAG TGTICCAGTCAA 182 Bt602 29 11 GAAAG TGTITTCTGTIG 42 4 Bt814 25 -2 GAAG TTGGCGCTCGAA 282 G Figure 2.Regions in isolated sequences: aa = no. of aminoacids from RNAse, nb = nucleotides between RNAse and polypurine tract, PPT = polypurine tract, L TR = long terminal repeat, nd = remaining no. of downstream isolated nucleotides. 150 0.05 960 } 865 Rt11U Rt4:"'4 Rt:"'9 997 Rrll29 } ' Rtll14 } Rtll77 1 Tpv2-6 } <1 P17 .,, } 'i Rtlntl4 p,; Figure 3. Possible families of retroelements as determined by ribonuclease-isolated sectioos; Tpv elemeots were previously isolated (Garber et al., 1999). References Blusch, J.; Halttmeier, M.; Frech, K.; Sander, 1.; Mosch, C.; Wemer, R.; Wemer, T. 1997. Identification of endogenous retroviral sequences based on modular organization: Proviral structure at the SSA VI locus. Genomics 43:52-61. Ellis, T.; Poyser, S.; Knox, M.; Vershinin, A.; Ambrose, M. 1998. Polyrnorphism ofinsertion sites ofTyl- copia class retrotransposons and its use for lin.kage and diversity analysis in pea. Molecular and General Genetics 260:9-19. Flavell, A.; Dunbar, E.; Anderson, R. ; Pearce, S.; Hartley, R.; Kumar, A. 1992. Tyl-copia group retrotransposons are ubiquitous and heterogeneous in higher plants. Nucl Acid Res 20(14):3639- 3644. Flavell, A.; Knox, M.; Pearce, S.; Ellis, N. 1998. Retrotransposon based insertion polyrnorphism (RBIP) for high throughput marker analysis. Plant J 16(5): 643-650. Garber, K. ; Bilic, 1.; Tohme, J.; Bachmair, A.; Schweizer, D.; Jantsch, V. 1999. The Tpv2 family of retrotransposons of Phaseolus vulgaris: Structure, integration characteristics and use for phenotypic classification. Plant Molecular Biology 39(4):797-807. Hirochika, H.; Fukuchi, A.; Kikuchi, F. 1992. Retrotransposons fami!Tes in rice. Molecular and General Genetics 233: 209-216. Hirochika, H.; Sugimoto, K.; Otsuki, Y.; Tsugawa, H.; Kanda, M. 1996. Retrotransposons of rice in volved in mutations induced by tissue culture. PNAS 93:7783-7788. 151 Kalendar, R.; Grob, T.; Regina, M.; Suonemi, A.; Schulman, A. 1999. IRAP and REMAP: Two new retrotransposon-based DNA fmgerprinting techniques. Theor Appl Genet 98:704-711. Pearce, S.; Harrison, G.; Li, 0.; Heslop Harrison, J .; Kumar, A.; Flavell, A. 1996. The Ty1-copia group retrotransposons in Vicia species: Copy number, sequence heterogeneity and chromosomal organization. Molecular and General Genetics 250:305-315. Pearce, S.; Rogers, S.; Knox, M.; Kumar, A.; Ellis, T.; F1avell, A. 1999. Rapid isolation ofplant Tyl-copia group retrpotransposons L TR sequences for molecular marker studies. Plant J 19(6):711-717. Pearce, S.; Knox, M.; Ellis, T.; Flavell, A.; Kumar, A. 2000. Pea Tyl-copia group retrotransposons: Transposítional activity and use as markers to study genetic diversity in Pisum. Molecular and General Genetics 263(6):898-907. Voytas, D.; Cummings, M.; Konieczny, A.; Ausbel, F.; Rodermel, S. 1992. Copia-like retrotransposons are ubiquitous among plants. PNAS 89:7124-7128. Waugh, R.; McLean, K.; Flavell, A.; Pearce, S.; Kumar, A.; Thomas, B.; Powell, W. 1997. Genetic distribution of BARE-1 like retrotransposable elements in the barley genome evealed by sequence specific amplification polymorphism (S-SAP). Molecular and General Genetics 253:687-694. 1.3.8 Analysis ofTyl-copia retrotransposon LTR Sequences and Their Use for genome organization studies L.M. Galindo, E. Gaitán and J. Tohme SB-2 Project Introduction Long terminal repeats (L TRs) are a distinguishing feature of Ty 1-copia retrotransposons. Short sequences in the LTR act as promoters and enhancers of transcription, and the majority of such sequences are in the U3 region. Activity of the different motifs has been well documented through transient reporter gene assays (Casacuberta and Grandbastien, 1993; Suonemi et al., 1996; Takeda et al., 1999). However, as functionality lies only through small stretches, the rest of the L TR is subjected to the high mutation rates of retroelements (Gojobori et al., 1990, cited in Flavell et al., 1992), generating polymorphisms at high levels. Differences in LTR length and sequence can be used to study the evolution of specific retrotransposon families in individuals or in different species; furthermore insertional polymorphism of retroelements can be easily assessed when information about LTRs is available. In other research ( see "lsolation and Characterization of Ty 1-Copia Group Retrotransposon L TR Sequences in Phaseo/us vulgaris," pp. xx of this report) we used the technique developed by Pearce et al. (1999) to isolate 24 different sequences corresponding to RNAse-PPT-LTR (ribonuclease- polypurine tract-long terminal repeat) sections ofTyl-copia retrotransposons. Here we performed an analysis of sequences derived from amplifications using polypurine tract primers (desígned from the afo¡ementioned isolated sequences) and Msel primers. We also initiated a study using primers derived from L TR sections. Primers were also useful in tests for Sequence-Specific Amplification Polymorphisms (SSAPs) and lnter-Retrotransposon Amplified Polymorphisms (IRAPs) on accessions G 19833 and DOR364. Both varieties possess many agronomíc traits of interest. 152 Materials and Methods A modification of the technique developed by Pearce et al. ( 1999) was developed to isolate polypurine-tract downstream sections using seven primers designed from polypuríne tracts and their flanking regions. RNAse-PPT -L TR sections necessary for the primer design were previously isolated and characterized (Galindo, 2001 ). Sequence analysis was carried out using the programs Sequencher 3.0, Clustal W 1.8, Clustal X 1.62, Blast n 2.0, Blast x 2.0 and Matinspector 2.2. PCR products from amplifications using PPT and Msel primers were used for SSAP reactions with PPT primers and 9 Msel primers containing selective nucleotides. Amplification was canied out by a conventional AFLP touchdown protocol, modifying the annealing temperature according to primers. IRAP assays were modified from the methodology developed by Kalendar et al. (1 999). Results and Discussion Of the seven PPT primers, only five provided enough colonies for library construction, with a total of 1920 clones. Of the 72 fragments eh osen for sequencing reactions, only 3 7 showed the expected pair of primers (PPT-Msel) and were used for further analysis with Blast, Clustal and Matinspector. SSAPs were carried out with all the seven primers, and IRAPs with two of them. BLAST search. Severa! sequerices downstream from the polypurine tracts (presumed LTRs) showed high similarity with previously isolated sequences in Phaseolus vulgaris. Analysis of the sequences led to the conclusion that these sections did not correspond to insertion sites of retroelements, but rather to upstream and downstream conserved regions of basal genome reading frames. Residual retrotransposon sections are a common feature of the flanking regions of normal plant genes (White et al., 1994) and play an important role in genome evolution, apparently conceming promoter regions. Clustal alignment. Comparison of sequen ces resulted in a dendrogram of the L TR relations (resu1t not shown). Related sequences (derived from a specific primer) exhibit a tendency to group, but sorne sequences are part of a mixed population. Unspecific grouping is possibly due to high levels of recombination between L TRs or inter-retroelement transposition (P~arce et al., 2000), mutations generated by retrotranscriptase enzyme, causing the loss of overall sequence similarity (Biusch et al., 1997) and unspecific amplification. Matinspector analysis. Conserved regions required for transcription are usually found in L TRs. Isolated LTR sequence L814-90 presented high levels of similarity (3 out of 4 had 100% similarity) in all the basic motifs (CAA T and TATA boxes, polyadenilation signa!, polyadenilation downstream signal) when compared to the consensus sequences ofthe program's database. The rest of the sequences had only a few motifs with conservation. High conservation levels in transcription factor-binding sites have proved useful, but not definitive for retrotransposon transcriptional competen ce (Poteau et al., 1991 , cited in Manninen & Schulman, 1993). Retrotransposon molecular markers. Preliminary SSAP tests with two PPT primers (PPT438 and PPT814) resulted in polymorphism of 22.56 and 12.56%, respectively, between accessions G 19833 and DOR364, confirming the utility of this molecular marker for mapping purposes. 153 Polymorphic bands corresponded to both SSAP and AFLP markers as the retrotransposon- specific primers were not labeled. The IRAPs also showed differential patterns between accessions, g1vmg evidence to recent transpositional activity since the divergence of varieties (Kalendar et al., 1999). Ongoing Activities Standardization of a technique to use polymorphic bands as probes for microarray-based mapping. A modification of AFLP (Vos et al., 1995) is being used to evaluate the leve! of polymorphism between accessions G 19833 and DOR364. PCR reactions include a pre- amplification step with EcoRI and Msel primers, a selective amplification step with specific LTR primers and Msel primer, and a + 3 amplification step with L TR-specific primers and Msel primers with selective nucleotides. Polymorphic bands are eluted from gels, re-suspended in water and re-amplified. After band identity confirrnation in agarose gels, the PCR product is ligated toa PGEMT-easy vector and transforrned into Escherichia coli DH5a cells. Severa! clones corresponding to each eluted band are re-amplified and electrophoresed on acrylamide gels to select the exact clone corresponding to the original eluted band. Selected clones are sequenced to confirrn their retrotransposon origin and are used as probes for microarray tests. IRAPs are being tested with the protocol ofKalendar et al. (1999) with sorne modifications. References Blusch, J.; Haltmeier, M.; Frech, K.; Sander, 1.; Mosch, C.; Wemer, R.; Wemer, T. 1997. Identification of endogenous retroviral sequences based on modular organization: Proviral structure at the SSAV1 locus. Genomics 43:52-61. Casacuberta, J.; Grandbastien, M. 1993. Characterization of LTR sequences involved in the protoplast specific expression ofthe Tobacco Tnt! retrotransposon. Nucl Acid Res 21(9):2087-2093 . Flavell, A.; Dunbar, E.; Anderson, R.; Pearce, S.; Hartley, R. ; Kumar, A. 1992. Ty1-copia groupretrotransposons are ubiquitous and heterogeneous in higher plants. Nucl Acid Res 20(14):3639-3644. Galindo, L. (200 1) Aislamiento y caracterización de las secuencias L TR de retrotransposones del grupo Yl-copia en Phaseolus vulgaris. Tesis de grado Kalendar, R.; Grob, T.; Regina, M.; Suonemi, A.; Schulman, A. 1999. lRAP and REMAP: Two new retrotransposon-based DNA fmgerprinting techniques. Theor Appl Genet 98:704-711. Manninen, I. ; Schulman, H. 1993. BARE-1, a copia-like retroelement in barley (Hordeum vulgare L.).Plant Molecular Biology 22:829-846. Pearce, S.; Knox, M. ; Ellis, T.; Flavell, A. ; Kumar, A. 2000. Pea Ty1-copia group retrotransposons: Transpositiona1 activity and use as markers to study genetic diversity in Pisum. Molecular and General Genetics 263(6):898-907. Suonemi, A.; Narvanto, A.; Schulman, A. 1996. The BARE-1 retrotransposon is transcribed in barley from aL TR promoter active in transient assays. Plant Molecular Biology 31 :295-306. 154 Takeda, S.; Sugimoto, K.; Otsuki, H.; Hirochika, H. 1999. A 13-bp cis-regu1atory element in the LTR promoter of the tobacco retrotransposon Tto 1 is invo1ved in responsiveness to tissue culture, wounding, methyljasmonate and fungal elicitors. Plant J 18(4):383-393. Vos, P.; Hogers, R.; Bleeker, M.; Reijans, M.; Lee, T.; Homes, M. ; Frijters, A.; Pot, J.; Poloman, J.;Kuiper, M.; Zabeau, M. 1995. AFLP: A new technique for DNA ftngerprinting. Nucl Acid Res 23(21 ):4407 -4414. White, S.; Habera, L.; Wessler, S. 1994. Retrotransposons in the flanking regions ofnormal plant genes: A role for copia-Iike elements in the evolution of gene structure and expression. PNAS 91: 11 792- 11296. 1.3.9 Application of the Diversity Arra y Technology (DarT) in beans for mapping and germplasm characterization E. Gaitán and J. Tohme SB-2 Project Introduction The DNA microarray technology has been applied to severa! model organisms including Escherichi coli, yeast and D. melanogaster (Chu et al., 1998; Richmond et al., 1999; White et al., 1999). Currently, two complementary techniques are available: fragment-based DNA microarrays and oligonucleotide-based chips, also referred to as Affymetrix chips. Briefly, DNA microarrays allow the simultaneous hybridization oftwo fluorescent-labeled probes toan array of immobilized DNA fragments (usually PCR-amplified DNA sequences), each corresponding toa specific gene or genotype. After scanning the microarray with a laser scanner, the signal for each DNA fragment reflects the abundance of the corresponding messenger RNA in the sample. Recently a new methodology has been adapted using DNA microarrays: Diversity Array Technology (DarT) at Cambia-Australia (Jaccoud et al, 2001). DarT, which is not reliant on DNA sequence information, has severa! potential applications including germplasm characterization, genetic marker-assisted breeding and tracking genome methylation changes (Jaccoud et al., 2001). A project was initiated to obtain molecular markers using DarT methodology and "demonstrate their potential as markers in germplasm characterization, mapping and tagging of resistance genes. Methodology Two different libraries were constructed to generate different representations: Diversity panels anda mapping panel for beans. The methodology used here followed Jaccoud et al. (2001). Generaling diversity panels. DNA from 8 accessions that included mapping parentals, wild and cultivated beans from Mesoamerican and Andean pools was used to generate diversity panels. One restriction endonuclease (EcoRI) was used to generate representations. About 6 ng of DNA from each genotype were bulked and digested with 2 units of EcoRI. After digestion, the EcoRI adapter was ligated to the fragments using T4 DNA ligase, diluted and amplified using a combination of 4 primers with 3 selective nucleotides in a final volume of 50 J.Ll. Amplicons from representation were ligated to the pGEM T-easy vector, and the competentE. co/i strain DHSa 155 was transformed by electrophoresis. White colonies were transferred onto a freezing medium and grown overnight at 37°C. Dilutions were made using 5 J.ll of bacteria! and 45 f .. Ll of water. A 5- fll dilution was used to amplify by PCR reaction, using T7 and SP6 universal primers. After amplification, the PCR products were precipitated with 1 vol of isopropanol in ice. The DNA was re-suspended in lO fll of TE+glycerol at about 20 ng/¡ll. The purified fragments were arrayed with two replicates per fragment onto poly-lysine microscope slides (homemade) using the SPBio microarrayer from Hitachi After PCR amplifications, representations were column purified, and fluorescent dye (Cy3 or Cy5) was incorporated using a Megaprimer Labeling Kit from Amersham. Probes were purified before hybridization. Then 5 fll of Cy3- and Cy5-labeled representations were mixed with 2 fll of 20 J.lg/fll salmon sperm DNA dissolved in ExpressHyb hybridization solution (Clontech, USA) and denatured at 96°C for 3 min. The denatured probes were then mixed with 50 fll of ExpressHyb hybridization solution, pipetted directly onto the microarray surface, and covered with a glass cover slip. Slides were placed in a 60°C water bath in hybridization chambers (Clontech) and left overnight. After hybridization the cover slips were removed, and the slides were rinsed at room temperature. Slides were dried quickly by centrifugation at 1000 rpm in a slide rack for 1 min. Slides were scanned using a Virtek Chipreader. Spot signa! intensities were analyzed by ArrayPro. Generating a mapping panel. The mapping panel was generated using two mapping parentals: DOR 364 and G 19833. These two genotypes were digested using Msel to generate the representation. The remaining procedures were carried out as for the diversity panel methodology. Results Diversity panel. A total of250 clones were spotted onto poly-lysine slides. Eight DNA fragments were identified between DOR364 and G 19833 as being potentially polymorphic (with the red/green ratio >3.0, Figure 1). Severa! DNA fragments were identified among the other 6 accessions as being potentially polymorphic for diversity studies (red/green ratio >2.5). A total of 47 potentially polymorphic DNA fragment were identified. Mapping panel: A total of 2000 clones were spotted onto poly-lysine slides and hybridized with DOR364 and G 19833, labeled with Cy3 and CyS. Three hybridizations were made to reconfirm the polymorphic clones. A total of 180 polymorphic clones (ratio >3.0) were obtained. Ofthese 180 clones, only 30 of them are contrastive between both parentals. Progeny ha ve been processed for digestion/ligation, and they will be amplified to hybridize with contrastive clones. 156 Fig l. Hibridization pattem on a set of clones WIÍllg OOR364 and G 19833 parentals Labelled with Cy3 and Cy5. Ongoing Activities • Hybridize polymorphic clones with a diversity panel • Hybridize contratingclones with the pro gen y of DOR364 •o 19833 • Sequence the 30 contrasting clones to determine the leve! of duplicates • Construct new libraries using different enzymes and two selective bases to increase the level of complexity References Chu, S.; DeRisi, J.; Eisen, M.; Mulholland, J. ; Botstein, D.; Brown, PO.; Herskowitz, I. 1998. Tbe transcriptional program of sporulation in budding yeast. Science 282:699-702. Jaccoud ,D.; Peng, K.; Feinstein, D.; Kilian, A. 200 l. Diversity arrays: A so lid state technology for sequence information independent genotyping. Nucl Acids Res 29:4. Richmond, C.S.; Glasner, J.D.; Mau, R.; Jin, H.; Blattner, F.R. 1999. Genome-wide expression profiling in Escherichia coli K-12. Nucl Acids Res 27:3821 -3835. Schena, M.; Shalon, D.; Davis, R.W.; Brown, P.O. 1995. Quantitative monitoring of gene expression pattems with a complementary DNA microarray. Science 270:467-470. White, K.P.; Rifkin, S.A.; Hurban, P.; Hogness, D.S. 1999. Microarray analysis of Drosophíla development during metamorphosis. Science 286:2 179-2 184. 157 1.3.10 Application of tbe Diversity Array Technology (DarT) in genepool characterization and marker assisted selection for Cassava improvement Chikelu Mba, Eliana Gaitan, Diego Cortés and Joe Tohme SB-2 Project Introduction Recognizing that the length of time required to develop new cassava vanet1es through conventíonal plant breeding methods (up to 10 years) is a major bottleneck in the development and delivery of new varieties of this food security crop to farmers, CIA T has o ver the years been developing and disseminating molecular genetic tools to add markers assisted selection (MAS). The RFLP-based molecular genetic framework map of cassava (Fregene et al., 1997) marked a major step in making cassava varietal development more efficient. However since most of the cassava breeding in the tropics is carried out by resource poor NARS in Asia, Africa and South America, the application of this map was limited on account of the prohibitive cost of the facilities required to gaínfully and safely apply such a potentially hazardous radioactivity based protocol. In recognitíon of this, CIA T set out to provide Cassava Scientists with a fast and robust method to assess genetic diversity and conduct large scale mapping experiments with a significan! reduction in costs and time thrpugh the development of efficient, cheap and easy to use PCR- based markers. A measure of success has been recorded in this with the development and deployment of about 500 microsatellite or simple sequence repeat (SSR) markers Chavarriaga et al. , 1998; Mba et al., 2001; Fregene et al. , unpublished data; Mba et al. , unpublished data). Germplasm screening using these SSR markers is essentially by way of silver staining the amplification products on polyacrylamide gels or visualizing the PCR products on ethidium bromide stained agarose gels. While this will undoubtedly for a long time remain the marker of choice for these NARS on account of the ease of adoption it fails to address the need for high throughput. In furtherance of CIA T's commitment to the development of cheap and reliable molecular tools for use in high throughput genetic fingerprinting of cassava germplasm collections, genetic fine mapping, gene tagging and molecular marker assisted (MAS) breeding, our Unit has begun applying the Diversity Array Technology (DArT) to eliminate this identified inherent shortcoming of lack of throughput protocols in the use of SSR markers. DArT is one of several applications of the novel DNA microarray technology platform that employs the so lid state, open platform method for detecting DNA polymorphisms such as SNPs. In addition, this method is sequence-independent and requires only a minimal amount of DNA. The simplicity and low-cost of the technology strongly recommends it for use in tropical orphan crops like cassava where comprehensive genome sequences are definitely lacking and may not be available any time soon. The low-cost high-throughput technology of DArT, in addition to its obvious use in germplasm characterization, genetic mapping and gene tagging and MAS, can also be used in tracking genome methylation changes while the composite variant of the diversity panel permits the resolution of complex genomic samples into individual components. In our laboratory, we have employed the DArT technique to identify whole genomic DNA fragments that discriminare between cassava genotypes, including parents of each of our 2 mapping populations (that used for the framework map and for genetic mapping of cassava white fly resistan ce). This report describes a proof of concept pilot application of DArT to develop robes in the form of 158 polymorphic DNA fragments for use in the development of dense maps, cassava germplasm characterization, MAS and in assessing genetic diversity germplasm accessions. Methodology Generation of panels Whole genomic DNA was extracted from the 4 parents of 2 mapping populations, NGA-2 or TMS 30572 and CM2177, for the F1 mapping population; and ECU72 and MCOL2246, for the whitefly resistance gene mapping population, respectively. From this, 9ng of DNA from each of these 4 genotypes was bulked and digested with EcoRI. The digested bulk DNA was ligated, PCR amplified and purified using QIAGEN PCR cleaning kit. The cleaned PCR products were then used to transform competent bacteria! cells. The transformed bacteria! cells were incubated and plated out on LB+agar. Positively transformed cells (white colored colonies) were picked and cultured overnight in 384-well plates containing freezing media. These were then PCR amplified and the extension products alcohol precipitated, completely dried down and re-suspended in water. Equal volumes of each of these and spotting buffer (TE 20% glycerol) were introduced into fresh 384-well plates. These were then arrayed in duplicates onto 3 glass slides for each array using MiraiBio's spotter, SPBIO and the slides processed following Jacoud et al., 200 l. Generation of representations In each case, purified cloned amplicons of 2 representations i.e. the 2 parents of the respective mapping populations that had been bulked to generate the panel were labeled with Cy3 and Cy5 dyes, respective\y, fol\owing usual procedures. The two \abeling reactions were subsequently bulked and cleaned using a QIAGEN PCR purification kit. Hybridization The Cy3 and Cy5 labeled probes were hybridized onto the Diversity Panel overnight in Corning Hybridization chambers using Expresshyb hybridization solution from CLONTECH. These slides were post processed by serially washing in SSC and SOS followed by washes in decreasing concentrations of SSC. The slides were gotten ready for scanning by spin-drying in a tabletop SOIRV ALL centrifuge at room temperature Scanning and image analyses Processed slides were scanned using the Cy3 and Cy5 channels of ChipReader (VIRTEK, Canada) and the images analyzed using ArrayPro4 from Cybernetics, USA. Positive clones, i.e. those with significant Cy3 or Cy5 fluorescence based on the ratio of their median intensities were identified. These values were obtained after normalization ofthe signals. Figure 1 shows a sample sean image with the Cy3 and Cy5 dyes having the green and red pseudo-colors, respectively. The yellow color represents a situation where both dyes hybridize positively to a clone. Clones were considered positive for Cy3 and Cy5 if they had Cy3/Cy5 ratios of more than 2.0 and less than 0.5, respectively. Results and Discussion A total of 350 putative polymorphic fragments ha ve been identified as discriminating between the 2 parents ofthe Cassava F1 mapping population. As a confirmation, a few ofthese would be used as probes to screen the 2 parents. An expected positive result would then lead to usíng these as probes on the 150 progenies that make up the mapping population. 159 Future activities • Confinnation of the polymorphism of the isolated fragments • Producing panels made up of the mapping population and screening these with the identified fragments as probes. • Repeating same for the whitefly resistance mapping population • Using the data for placing the fragments on the map • Use ofthe probes for germplasm characterization • Sharing of the information Reference 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: 43 1 -441. Jacoud, D., Peng, K., Feinstein, D., Kilian, A. 2001. Diversity arrays: a solid state technology for sequence information independent genotyping. Nucleic Acids Research. 29 No.4 e25. Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg, U., Apel, K., Gale, M., Tohrne, J. and Fregene, M. 2001. Simple Sequence Repeat (SSR) marker survey of the cassava (Manihot escu/enta Crantz) genome: towards an SSR-based molecular genetic map. Theor Appl Genet 102:21-31. 1.3.11 Geno me location of SSR markers from a cassava cDNA library Chikelu Mba, Tanya Garcia, Martin Fregene and Joe Tohme SB-2 Project Introduction There has been a concerted effort at the Biotechnology Research Unit (BRU) of CIA T to develop and disseminate molecular genetic tools that would make the development of new cassava varieties and germplasm characterization easier, cheaper and adoptable for cassava scientists especially in the National Agricultura! Research Systems (NARS) where most of the cassava improvement work is concentrated. The first step in this direction was the RFLP based molecular genetic framework map of cassava (Fregene et al., 1997). Recognizing the inherent bottlenecks in the use of RFLP as molecular markers, especially by the NARS, a major thrust of the efforts has been directed towards saturating this map with PCR-based molecular genetic markers. RFLP based techniques are expensive, require the use of hazardous radioactive probes that are not available to many resource-poor developing country research programs, and these probes must be physically transferred from one site to another under strict safety protocols. In contrast, PCR- based markers are robust, inexpensive to assay, easily shared among researchers and readily accessible in public and prívate domains, making this a much more appropriate approach in these countries. With access to a simple text file containing the sequences of the oligonucleotide primers for the PCR-based markers of interest, a breeder can rapidly and efficiently evaluate the germplasm under study. 160 In this regard, our Unit has developed and made publicly available over 500 simple sequence repeat (SSR) or microsatellite markers (Chavarriaga et al., 19998; Mba et al., 2001; Mba et al., unpublished data; Fregene et al., unpublished data). Of these, a total of 157 were sourced from a cassava root and leaf cONA library while the rest were from whole genomic libraries. From the latter, 85 have been located on the cassava genome. The present report describes the characterization and initial efforts at the mapping of the 157 SSR-containing cONA fragments. Methodology Development of SSR 's from a cassava cDNA library A cassava cONA library was constructed commercially by Life Technologies, MD, USA from RNA extracted from leaves and roots tissue (see SB-2 Annual Report, 2000). The sequencing and initial primer designs were also as described for the whole genomic SSR clones using Perkin Elmer's ABI Prism 377 automated sequencer (Mba et al., 2001 ). Parental Survey using the SSR markers Each of the SSR primer pairs were used to amplify the relevant Joci of the cassava genome in the two parents of the F1 mapping population, TMS 30572(female) and CM2177 (maJe), respectively. Those that had at least one unique allele in at least one of the parents were used to screen the !50 members ofthe F1 mapping progeny. Mapping of SSR markers SSR markers that had a unique allele in either or both parents of the mapping population were used to screen the 150 progenies making up the F 1 mapping population. The segregation data of the markers that fitted the expected ratio of 1: 1, presence: absence of the unique parental allele were used to place the markers on the framework map using the linkage analysis computer package MAPMAKER 2.0. (Lander et al., 1987). The data analyses followed the same procedures as described for the mapping ofthe initial36 SSR markers (Mba et al., 2001). Results and Discussion Characterization ofSSR loci types From the approximately 400 putative SSR-containing cONA clones sequenced, a total of 157 clones contained SSR Joci in good enough positions for primer design. Less than 1 0% redundancy has been observed. Many of these clones contained more than one repeat motif_ at times in different loci. However, a great majority of the repeat loci were the CT/GA repeat which accounted for over 81% of the SSR-containing clones for PCR primer design. A breakdown of the loci type is given in the table below. Breakdown of loci types for SSR primers from the cDNA library Type Percentage CT/GA 79 CA/GT 12 TAIAT 8 Others 5 SSR Parental Survey All primer pairs successfully amplified the corresponding SSR loci in the parents of the cassava mapping progeny; even though with different MgCI2 concentrations, and 2 annealing 161 \ " \ temperatures, 55·c and 45·c respectively. In all, about 45 % of all SSR markers tested in the parents, revealed a unique allele in at least one of the parents while less than 20% showed a unique allele for both parents. A pattem of SSR polymorphism between the two parental loci is shown in Figure below. Genome location of SSR markers Figure 2 shows the map positions of 92 SSR loci from the SSR markers analyzed to date on the mate- and female-derived molecular genetic map. The segregation data on the 150 members of the F1 mapping population of another 47 SSR markers sourced from the cONA library are currently being placed on the cassava map. Linkage group nomenclature is as described for the molecular genetic map of cassava by Fregene et a/.(1997) except for groups L, O, and P that have now been merged with other groups. The SSR markers reveal a complete spread over the cassava genome - at least one marker being placed on all but one of the eighteen linkage groups. The existence of markers with unique alleles in both of the parents or "allelic bridges" (Ritter et al. 1991) will assist in the construction of a consensus map of analogous mal e- and female- derived linkage groups for the cassava genome. On-going activities Completion ofthe mapping ofthe rest ofthe SSR markers from the cONA library. Further dissemination of the SSR marker technology to intemational and national programs and other collaborators. The use of the microarray Diversity Array Technology (DArT) to develop polymorphic DNA fragments to further saturate the map and for use in high throughput cassava genome characterization. References Chavarriaga-Aguirre P, Maya MM, Bonierbale MW, Kresovich S, Fregene MA, Tohme J, Kochert G ( 1998) Microsatellites in cassava (Manihot esculenta Crantz): discovery, inheritance and variability. Theor Appl Genet 97:493-50 l. Fregene M, Angel F, Gomez R, Rodriguez F, Chavarriaga P, Roca W, Tohme J, Bonierbale M {1997) A molecular genetic map of cassava. Theor Appl Genet 95:431-441. Lander ES, Green P, Abrahamson J, Barlow A, Daly MJ, Lincoln SE, Newburg L (1987) MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1: 171-178. 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 ofthe Cassava (Manihot esculenta Crantz) Genome: Towards an SSR-Based Molecular Genetic Map of Cassava. Theoretical and Applied Genetics. 102:21-31. Ritter E, Debener T, Barone A, Salarnini F, Gebhardt C (1991) RFLP mapping of potato chromosomes of two genes controlling extreme resistance to potato virus X (PVX). Mol Gen Genet 227: 81-85. Roder MS, Korzun V, Wendehake K, Plaschke J, Tixier M-H, Leroy P, Gana! MW (1998) A microsatellite map ofwheat. Genetics 149: 2007-2023. 162 1.3.12 Annotation of SAGE tags (Transcripts) differentially expressed in CMD resistant and susceptible genotypes Ryohei Terauchi2, Hideo Matsumura2 Maria Christina Suarez3 Edgar Barrera1, Janneth Gutierrez\ Martín Fregene1 1SB-2 Project; 2Iwate Biotech Research Center, Japan; 3Fresno University Introductioo The objective of the serial analysis of gene expression (SAGE) of CMD resistan ce in cassava is to identify candidate genes that are expressed in the CMD2 - mediated response to the cassava mosaic virus (CMV). These candidate genes may be the novel dominant gene or genes expressed down stream that maybe the molecular basis of resistance. SAGE of CMD resistant and susceptible progeny from an F1 mapping progeny yielded 12,700 15bp tags (representing expressed genes), divided into 5733 and 7053 tags for the resistant and susceptible genotypes respectively (CIA T 2000; Fregene et. al. 2001 in preparation). One hundred and seventy ti ve transcripts were expressed 3 to 12 times more in the resistant bulk compared 94 transcripts found 3-5 times in the susceptible bulk. The next step of the SAGE experiment is to identify the transcripts represented by the 15bp tags, a task complicated by the scanty EST data available for cassava. A PCR approach was therefore employed to amplify longer fragments using the 15bp tags as primers and a cONA library from resistant genotypes as template. At the same time an EST project was initiated in collaboration with the Iwate Biotech Research Center, Kitakami, Japan. Methodology A cONA library had earlier been constructed in pYES (Invitrogen Inc.) using mRNA from the CMD resistant bulk (Fregene and Terauchi 2000 unpublished data). The 13 or 15 bp SAGE tags served as forward primers and p YES vector sequen ces as reverse primer in PCR of a dilution of the cONA library. PCR primers were synthesized for 28 tags expressed four times or more in the resistant bulk compared to the susceptible. PCR reaction conditions was 1 O mM Tris-HCI, 50 mM KCI, 2.5mM Mgch. 200~M each dNTPs,10pmol ofthe forward and reverse primer, and 2.5U of"hot star" Taq polymerase (Qiagen Gmbh) in a 50~1 volume. DNA template was 1~1 ofa lOX dilutions ofthe cONA library from the CMO resistant bulk. Therrnal cycling conditions were 35 cycles of 95°C 4min, 94°C 30sec, 45°C 1m in, 70°C 1m in, and a final extension cycle of 70°C for Smin. PCR product was cloned into the PGEMT vector (Promega Inc.) transforrned into E.Coli by electroporation and sequencing was off the purified plasmid template using the T7 forward pnmer. To identify associations between differentially expressed genes and the dominant CMD resistance gene, cloned fragments from the tag PCR amplification were screened as RFLP probes in Southem blots ofONA ofthe parents and bulks ofthe CMD mapping progeny using 4 restriction enzymes namely: EcoRI, EcoRV, Hindlll, and Haelll. ONA isolation, filter preparation and Southem hybridization were as described by Fregene et. al. ( 1997). Transcripts found to be polymorphic in the parental survey were analyzed in the 80 individuals of the resistant and susceptible bulks and if found polymorphic, will be analyzed in a Jarge mapping progeny of about 2500 individuals to identify the precise position relative to the CMD resistance gene. 163 For EST generation, 2ul ofthe cONA library was electroporated into 40ul of E.Coli HBIOI cells (Gibco BRL) and plated on LB agar plates + ampicillin ( 1 OOug/ml). A total of 5,000 colonies were picked into 70ul ofLB media+ ampicillin (IOOug/ml) in 384 well plates. Plasmid isolation was using the MONTAGE 96-well plate system (Millipore Inc), 4 96-well plates or 384 clones were processed at a time. PCR sequence reaction was with the 3' end primer designed from pYES (lnvitrogen Inc.) and the big dye terminator kit (Applied Biosystems) on a 9600 Perkin Elmer Machine or an MJ Research DNA engine (Tetrad). The sequence reaction was cleaned using the multi screen 96-well plate format (Millipore lnc.) and analyzed on a Shimadzu RISA 384-capillary sequencing machine. A target of3000 ESTs have been set for tag annotation. Results Of the 28 tag PCR, 24 gave good PCR products that could be cloned and sequenced, and 18 of these gave good and long enough sequence for BLAST (Atschul 1990) sequence similarity searches. Most of the tags were about 150-300bp long and the tag primer was found for 16 of the 18 sequenced tags, about 40bp from the 5' end. The putative identities of these transcripts and their tag primers are shown in Table l. Parental and bulk filters were screened with all 24 tag PCR products. Results revealed polymorphism in 2 of the tags, 25 and 11. Tag 25 is a transcript showing homology to WPKY transcription factor, while tag 11 shows similarity to a bHLH transcription factor GBOF-1. These tags are being analyzed further in a larger population. So far tag annotation has identi.fied genes known to be involved in systemic acquired resistance (SAR) response to disease in plants. They include a WRKY transcription factor, catalases, a pectin-esterase and reductases. Other genes were also found implicated in plant response to disease but are part of the cell mechanism known to aid virus replication including elongation factor alpha-1. Elongation factor 1-alpha (EF 1 alpha), is an essential component of the translation machinery that delivers aminoacyl-tRNA to ribosomes. Virus proteins such as lllV -1 Gag polyprotein that play key functions at almost all stages of the viral and the conserved 3'-terminal stem-loop (3' SL) of the West Nile virus can bind to EFlalpha and incorporate it into the virus replication machinery (Biackwell and Brinton 1997; Cimarelli and Luban J 1999). The 3' end sequencing of cONA clones is ongoing and is expected to provide 3000 ESTs for tag annotation. Homology with known genes and proteins deposited in public data bases are being sought for as sequences are produced. Preliminary results reveal that that transcripts known to be abundant in cells such as ribosomal and chloroplast sequences constitute only about 10% of all sequences suggesting the cONA library provides a good representation of genes expressed. Sequence data generated will be also be submitted to the Gene Bank. Future Plan • Annotate many more tags using 3' end ESTs • Determine the function of sorne differentially expressed genes by over-expressing them in susceptible cassava plants and challenging them with infectious ACMV clones. References Blackwell JL, Brinton MA 1997 Translation elongation factor-1 alpha interacts with the 3' stem-1oop region of West Ni le virus genomic RNA. Virol 1997 Sep;7 1 (9):6433-44 164 Cimarelli A, Luban J 1999 Translation elongation factor 1-alpha interacts specifically with the human 7" immunodeficiency virus type 1 Gag polyprotein. J Viro! 1999 Ju1;73(7):5388-40 1 Fregene M, Angel F, Gomez R, Rodríguez F, Chavarriaga P, Roca W, Tohme J, Bonierbale M. A molecular genetic map of cassava (Manihot esculenta Crantz). Theor Appl Genet 95: 431-441 Table l. BLAST sequence homology of28 differentially expressed transcripts in bulks ofCMD resistant and susceptible cassava genotypes. TagNo. Sequence No. of transripts No. of transcripts BLAST homology of amplified cONA in resistant bulk in resistant bulk fragment 1 CTAGAATGACCTTGT 12 1 Cytoskeleton related protein 2 CTAGCGCCAGACAGT 11 3 Elongation factor 1-alpha 3 CTAGCTCTGTGTATC 8 2 NIA 4 CTAGCAAAGCAGCGC 7 o Pectin-esterase 5 CTAGGAAACAATCCT 7 Photosystem 1 chain 11 precursor 6 CTAGTACACAATGTA 7 1 NIA 7 CTAGCTCGCCGTAAG 6 o Histone 8 CTAGTTAATATGGTA 6 NIA 9 CT AGTTCAAAGGAAG 6 1 Ribosomal protein 10 CTAGTTAAAATGTGA 5 o Catalase ll CTAGAGCTTTTCACT 5 bHLH transcription factor GBOF-1 12 CTAGCCGGATCTCCT 5 NIA 13 CTAGCGATTAAAAAA 5 Rubredoxin 14 CTAGTGGAGCAATAC 5 ONA binding regulatory protein 15 CTAGTTGCTTTGCAC 5 Initiation factor 3k (A. thaliana) 16 CTAGAAGTGGTGCTT 4 o Nuclear import protein 17 CTAGACTGAAGTCAG 4 o Hypothetical protein (A. thaliana) 18 CTAGAGCACGAGT 4 o Ubiquinol--cytochrome e reductase 19 CTAGATAATAAAAGG 4 o NIA 20 CTAGATCCTTGCCTT 4 o No significant sirnilarity 21 CTAGGCAGGATCAAG 4 o NIA 22 CTAGCTGAATTATAG 4 o NIA 23 CTAGGCAGCCGCCGC 4 o Catalase 24 CTAGGGA TT A TTCAT 4 o NIA 25 CTAGGTGGACGAGAC 4 o Transcription factor WRKY 26 CTAGTAATCGCTCAG 4 o 40S ribosomal protein 27 CTAGTTGGATCTT 4 o NIA 28 CT AGTTGGA TTCTTT 4 o NIA NIA cONA sequence data not yet available 165 1.3.13 A comparison of marker assisted selection (MAS) and conventional selection for the rapid deployment of the novel CMD resistance gene (CMD2) in cassava gene pools. Oluwole Ariyo\ Alfred Dixon2 Edgar Barrera., Martín Fregene1 1 SB-2 Project; 211T A Introduction The principal objective of the project "Molecular Mapping of Genes Conferring Resistance to the Cassava Mosaic Disease (CMD) in African Cassava Germplasm" funded by the Rockefeller foundation is to identify markers tightly linked to different sources of CMD resistance for efficient and cost-effective deployment of resistan ce genes. Molecular markers linked to a novel and high leve! of resistance, designated CMD2 (Akano et. al. 2001 ), have been identified. The effectiveness of CMD2 against a wide spectrum of strains of the virus in sub Saharan Africa, including the aggressive Ugandan variant (UgV) (see section 1.4, this report), makes its deployment very appealing in protecting cassava production against the ravages of CMD both in Africa and Latín America. Of urgent importance is containing the rapidly advancing front of UgV that has now reached the Democratic Republic ofCongo, Kenya, and Tanzania. Conventional breeding for CMD resistance involves at least 4 cycles of selection for resistance at the seedling, clona! and preliminary and advanced yield tria! stages. Disease pressure, which may vary from year to year, may lead to escapes that are carried along, in certain cases, up to the third cycle. Genetic markers for CMD resistance enables the elimination of susceptible genotypes at the seedling stage and reduces significantly, 50% in the case of CMD2, the materials to be evaluated in the field at the crucial single row (clona!) tria! stage, where more than 95% of genotypes are eliminated. Markers are even more important when two or more genes/traits are involved, the reduction in progenies to be evaluated becomes even higher. In many cassava production scenarios, CMD resistance has to go hand-in-hand with cassava bacteria! blight (CBB) resistan ce. A MAS project for CMD resistance was therefore initiated with liT A to enable us to test the fidelity of CMD markers developed at CIA T, a very convenient approach considering that phenotypic and molecular data can be obtained at the same time and compared. The MAS project is also necessary to work out the details for routinely using these markers in cassava breeding. Methodology The original group of CMD resistant Iand races for the study consists of TME3, TME4, TME28, and TME9. TME28 was dropped out dueto very poor flowering, while seeds obtained from crosses with TME4 are not considered here dueto its very close genetic similarity with TME3. Only crosses to TME3 and TME9, crossed to elite liTA parentallines, TMS30572, TMS91934, and TME 117, a land race favored for his good cooking roots, are reported here. TMS30572 is moderately resistant to CMD, while the other two are susceptible. TMS91934 has a very high leve! ofresistance to the all known strains ofCBB (Verdier 1999, pers communication) and can serve as a so urce of CBB resistan ce in progenies of the highly CBB susceptible land races TME3 and TME9. The seedling nursery and field establishment were at the liTA sub-station in Mokwa Iocated in the Guinea Savannah agro-ecology of Nigeria. This si te is characterized by low CMD disease pressure and ideal for the multiplication of cuttings from CMD susceptible lines. 166 Before transplanting, 2 young leaves were harvested from each genotypes for DNA isolation and SSR analysis. The leaves were bagged in small plastic sample bags and carried on ice to liTA head quarters, Ibadan for DNA isolation. DNA isolation was by a miniprep isolation procedure of the Dellaporta et. al. (1983) protocol using 1 00-200mg of fresh leaf tissue and a twenty fold reduction in volumes of the isolation buffers and reagents. DNA isolated was shipped to CIA T head quarters for SSR marker analysis. DNA was not quantified for marker analysis, Sul of a 1 OX dilution was used in PCR reactions. All the genotypes were analyzed with SSR markers tightly linked to CMD2, SSR118 and NS158. SSR marker analysis were as described by Akano et. al. (2001). Results Sexual seeds obtained from crosses to the CMD resistant land races and plantlets, transplanted to the field are summarized in Tablel, a total of 2490 genotypes are currently growing in the field. Harvesting of two young lea ves from plantlets just befo re transplanting was initially thought to be stressful to the young plants, however more than 99% of transplanted genotypes survived and only 2 were lost. This is an important observation as it suggests that molecular assisted selection can be done even while the plants are in the seedling nursery. DNA was successfully isolated from all 2488 genotypes and the parentallines. The SSR markers, SSRY28 and NS 158 have been analyzed in close to half of the genotypes and marker analysis is still ongoing. The large number of plants obtained from crosses from both TME3 and TME9 makes them ideal not only for making comparisons between MAS and conventional selection, and for marker-fidelity studies; but also for fine mapping markers linked to CMD resistance. Great care was therefore exercised in relating plants in the field to raw SSR data, a special template was set up in microsoft excel for this purpose. Once molecular marker data becomes available for all genotypes, those with marker alleles linked to CMD2 will be selected and five 20cm long woody stakes harvested for field establishment in a high CMD pressure location, in this case, liT A Ibadan. CMD resistance will be eva)uated at 3 and 6 months after planting to confirrn the results of marker analysis and for comparison with an unselected population of all genotypes also to be established at liT A, Ibadan. Future Plan • Evaluate all 2490 genotypes for CMD resistance under heavy disease pressure. • Compare selection efficiency of marker-aided selection against phenotypic selection References Akano A., Barrera E., Dixon A.G.O., Fregene M. (2000). Molecular Genetic Mapping of Resistance to the African Cassava Mosaic Disease. (Theor and Appl Genet). Dellaporta SL Wood J, Hicks JR (1983) A plant DNA minipreparation: version U. Plant Mol Biol Rep 1:19-21 167 1'able l. Summary ofseeds and plants obtained from reciprocal crosses ofTME3 and TME 9 Family name Female M ale Seeds harvested Plants in field Total plants in field M1 TME3 TME 117 36 18 M2 TME 117 TME3 220 95 113 M5 TME3 91934 103 49 M6 91934 TME3 60 12 61 M7 TME3 30572 70 49 M8 30572 TME3 846 791 840 M17 TME9 TME 117 368 309 M18 TME 117 TME9 174 107 416 M21 TME9 91934 370 282 M22 91934 TME9 27 12 294 M23 TME9 30572 264 214 M24 30572 TME9 700 552 766 Grand Total 2490 1.3.14 cDNA-AFLP analysis of differential gene expression in the Cassava- Xanthomonasaxonopodis pv. manihotis interaction M. Santaella1, E. C. Suárez1, C. González2, C. López2, A. Badillo3, J. Tohme1, V. Verdie~. 1SB-2 Project, 2IRD, 3Los Andes University. Introduction The cassava bacteria! blight (CBB) is a major disease, endemic in Latín America- and Africa causing serious damage to cassava and resulting in severe yield losses. The causal agent is Xanthomonas axonopodis pv. manihotis (Xam) . The most appropriate and realistic approach for controlling CBB is through host resistance. Resistance to CBB operates from the vascular system, and seems to be polygenic and additively inherited (Hahn, 1978). However, no resistance genes have been identified. Plants in general, have a wide spectrum of cellular and molecular defenses including cell wall fortification, phytoalexins production and development of a hypersensitive response (HR), characterized by a necrotic reaction surrounding the invasion area to restrict pathogen expansion (Baker et al., 1997, Hammond-Kosack & Jones, 1996, Culver & Dawson, 1991). These defense reactions are directly or indirectly activated by resistance genes. Understanding how these genes are involved in the recognition and response against pathogen attack will allow us get into the manipulation ofthe resistance against a wide range ofpathogens. 168 The objectives of the study are to implement the cDNA-AFLP technique to identify differentially expressea bands between two different cassava cultivars, one resistant and one susceptible to CBB (MBRA 685 and MCOL 1522, respectively). We analyzed the pattem of cDNA-AFLP at different times post inoculation with a Xam strain and identified putative molecular disease resistance markers. Material and Methods Sample prepara/ion and cDNA synthesis We compared a resistant (MBRA 685) anda susceptible (MCOL 1522) variety for differential gene expressions o ver time after inoculation with Xam isolate CIO 151. Y oung plants were inoculated by stem puncture. Stem tissues were collected at 24 and 72 hours post inoculation (pi), 7, 15 and 30 days pi. The controls were healthy non-inoculated plants and plants inoculated with sterile water. The tissue was grounded in liquid nitrogen and total RNA was isolated using the Proteinase K method (Hall et al., in Rocha, 1995). Poly (A) RNA was isolated using aligo (dT) coupled to DynaDeads (DYNAL). cONA was synthesized using aligo (dT) primer and SuperScript 11 reverse transcriptase (GIDCO BRL) from 400-500 ng of mRNA, as starting material. cDNA-AFLP profiling analysis The template for cDNA-AFLP was prepared according to Bachem el al. (1996) using EcoRI and Msel restriction enzymes and adapters. Preamplification was carried out with one EcoRI and one Msel single chain adaptar with no (O) or one ( 1) selective bases. This product was checked on agarose gels and a 1130 dilution· was used for subsequent amplifications. These were done with a pair of primers with 2 or 3 selective bases (GIDCO primers and Plant AFLP Kit, GIDCO, respectively). Selective amplification products were separated on a 6% polyacrylamide gel run at 1 OOW, 50°C for 2 and a half hours. lt was then processed with the silver staining technique (Promega). . Isolation, cloning and sequencing of target cDNA bands Bands of interest were marked, cut and eluted in ddHO. AFLP fragments were reamplified by PCR, ligated to pGEM®-Teasy (Promega) and sequenced using the automated sequencer (ABI Prism 377). The sequences were edited using Sequencher 3.0 (Gene Codes Corporation) and compared with the GenBank databases using BLASTx and BLASTn. Results and Discussion The cDNA-AFLP technique was successfully implemented using RNA isolated from stem tissue from cassava plants. We evaluated 32 and 40 combinations of AFLP primers with two (2) and three (3) selective bases, respectively. cDNA-AFLP profiles showed 40 to 70 bands per primer combination, ranging between 100 and 1500 base pair (bp), with -3600 fragments screened (Figure 1). Differential expression was observed for 353 fragments putatively induced by the pathogen in the resistant variety, with an average of -5 bands per combination. These fragments ranged between 130 to 650 bp. The more informative primer combinations were E-AAIM-AG, E- ANM-CT, E-ACG/M-CTT and E-ACT/M-CTT with 11 , 10, 14 and 16 differential bands, respectively. We sequenced 201 bands and compared their homology with GenBank databases. Significant homologies with known genes or putative genes ha ve been found for 149 sequences. 3 7 of these showed homology with plant resistance or defense related proteins or other type of plant proteins (Table 1). Sequences similar to the resistance genes Cf-2 and 12 from tomato were found, and 169 others with homology to putative resistance proteins. These fragments are expressed at 24 hpi in the resistant variety, indicating that they are induced by the pathogen or that their expression increased in the presence of the bacteria. This is in contrast with the hypothesis that resistance genes are constitutively expressed in plant cells (Hammond-Kosack & Jones, 1996, Staskawicz el al., 1995, Lamb, 1994, Keen, 1990, Gabriel & Rolfe, 1990). Sorne of these sequences showed expression in the susceptible variety al so, but Jater after the infection ( 15 dpi, bands E6, E 18 and E45, Table 1), suggesting that these putative genes are present in both genotypes but expressed early in the resistant variety. Figure l. AFLP Polyacrylamide gel electrophoresis stained with silver nitrate. Primer combination E-ACA/M-CAG. ll l=C: l <;;T ANT Table l. Nucleotide homology and probabilities from BLASTx search in GenBank of the differentially expressed cDNA-AFLP bands. Band Bp. Btst Match E Valut Exprtssion Defense RelaJed Prott!ins E6 213 Disease resistance Cf-2 like protein (L. esculentum)2e-05 24 hpi E45 260 Resistance protein 12 (L esculentum) 2e-08 24 hpi M62 830 Putative resistance protein (A . Thaliana) 3e-24 24 hpi E 18 171 Resistance gene analog protein (L saliva) le-05 24 hpi M 56 350 Putative receptor protein (A. thaliana) 2e-1 4 24 hpi M63 360 Putative t:ranslation initiation factor JF-2 7e-24 24 hpi ¡:' 1 11 il l( ;nOC',. nt'nt .. i n c;;:~J fr, ,..., ... ¡ • .,,.,;;,._,,.,, Fragments E6 and E45 showed significant homology with Jeucine rich repeats (LRR) from Cf2 and 12 genes (type LRR and NBS-LRR, respectively). The LRR motif in resistance genes is involved in protein-protein interactions and acts in the specific recognition of avirulence genes from pathogens (Staskawicz et al., 1995). The Cf gene family recognizes the fungus Cladosporium fulvum and activates the defense response in severa! tomato species (Dixon el al., 1996). The 12 gene family confers resistance to severa! species of Fusarium sp._ and shares structural characteristics with NBS-LRR resistance genes (Ori el al. , 1997). The fragment El8 showed homology with a putative resistance protein in lettuce. This putative gene has structural elements characteristic from NBS (kinase V motive and P-loop) and LRR (Shen et al., 1998). Other fragments showed similarity with Serineffhreonine or receptor protein kinases that belong to a different type of resistance genes that modulate the phosphorylation of other proteins. They are involved in the signa! transduction cascarles that actívate defense responses in plants. These results indicate that through the cDNA-AFLP technique we have isolated resistance and defense related fragments induced by Xam, corresponding to three different types of resistance genes in plants. We also found significant homology with transcription factors that act in the last part ofthe signa! transduction pathway that leads to the activation of defense related genes (Dixon el al., 1994). Severa! fragments showed homology to senescence, apoptosis and dormancy associated proteins suggesting that they might be involved in the programmed cell death mechanism included on the hypersensitive response. This defense reaction, very common in plants, creates a toxic media for 170 the establishment and expansion of the invading pathogen (Hammond-Kosack & Jones, 1996, Staskawicz et al., 1995). ~ Another 52 fragments did not show significant homology to known sequences in the databases. These were differentially expressed since 24 hpi indicating that were also induced by the pathogen, and might represent novel sequences in cassava putatively associated with resistance to Xam. Future Plans We are focusing our attentíon on those fragments that showed significant homology with known genes and those that are strongly induced to confmn their differential expression by Northem blot analysis. These fragments will be used as probes to hybridize a cONA library to isolate full-length clones. This information can be used to develop markers associated to resistance. References Bachem et al. 1996. The Plant Journal 9 (5), 745-753. Baker et al. 1997. Science (276), 726-733. Culver, J.N. & W.O. Dawson. 199l. Molecular Plan/ Microbe Jnteractions 4 (5),458-463. Dixon el al. 1996. Cel/ (84), 451-459. Gabriel, D. W. & B.G. Rol fe. 1990. Annual Review of Phytopathology (28), 365-391. Hahn, S. K. 1978. In Diseases of tropical food crops. H. Maraite and Y. A. Meyer (Ed.), Proc. lntemational Symposium at Université Catholique de Louvain, Louvain La Neuve, Belgium. Hall el al. 1978. (n: Rocha, 1995. Evaluacion comparativa de ARN mensajero de Phaseolus vulgaris L. Entre una variedad resistente y una susceptible a Acanthoscelides obtectus (Say). Tesis de Grado, Universidad Nacional de Colombia. Harnmond-Kosack, K.E. & J.D.G. Jones. 1996. Plan/ Molecular Biology (48), 575-607. Keen, N.T. 1990. Annual Review ofGenetics (24), 447-463 . Lamb, C.J. 1994. Cel/ (76), 419-422. Ori el al. 1991. The Plan/ Cel/ (9), 521-532. Shen et al. 1998. Molecular Plan/ Microbe Jnteraclions (11), 815-823. Staskawicz el al. 1995. Science (USA) 270 (5243), 661-667. 171 1.3.15 Study óf gene expression duríng pathogenesis of Xanthomonas axonopodis pv. manihotis using an AFLP-based microarray Lopez, C.E., Mosquera, G., Restrepo, S., Tohme, J., and Verdier, V. SB-2 Project Introduction Systematic profiling of genes that are specifically expressed by a pathogenic bacterium in its host plant would assist in understanding basic virulence mechanisms (Okinaka et al., 2001 ). Xanthomonas axonopodis pv. manihotis is a plant pathogenic bacterium, causal agent of cassava bacteria! blight. A plasmidic sequence containing a pathogenicity gene, pthB, has been described previously (Verdier et al., 1996). However, others sequences should be involved into the pathogenesis process. By now, no new sequences have been reported. The elucidation of those bacteria! expressed genes in infection process is a hard task when the entire genome sequence is not available. Recent advances in functional genomic technologies such as DNA microarrays have provided a unique way to monitor gene expression on a genomic scale and under different conditions. The majority of microarrays have been constructed from organisms for which the whole genome sequence is known or from organism that have an important collection ofESTs. We are constructing a Xam microarray based on the AFLP amplification of the Xam genome in order to identify genes implicated in pathogenesis and to study the global changes in gene expression associated with the process of infection of Xam during its interaction with cassava plants. MetbodoJogy AFLP libraries AFLP was carried out as described previously (Restrepo et al., 1999), using 250ng of Xam genomic DNA from strain Cio-46. The PCR was performed using the EcoRI + C and Msel + A primer combination and the PCR product was cloned into pGEMT-Easy. Another library using AFLP adaptors as primers for the PCR amplification was constructed but only the fragments above 300 bp were eluted from the acrylamide gel and reamplified. The reamplified products were cloned. Plasmids obtained from the two libraries were introduced to E. coli by electroporation. Bacteria were grown overnight in freezing medium and 5J.1l of a 1: 1 O dilution was used for the insert amplification using T7 and SP6 primers. PCR products were precipitated with isopropanol and resuspended in TE ( 1 O mM Tris; 1 mM EDTA) and 50% DMSO. An aliquot of 2J.1l was used to confirm the amplification in a 1% agarose gel. Twenty ¡.ti ofthe PCR resuspended in the TE-DMSO mix were transferred into 384- well plates for slide printing. cDNA-AFLP library In another experiment, Xam strain Cio-46 was used to inoculate the susceptible cassava variety MCOL 1522. Inoculations were performed as previously described (Restrepo et al., 1997) by stem puncture with one bacterial colony. Stems (2 cm around the inoculation site) were collected 24, 172 48 hours and 5, 7 and 15 days after inoculation. Tissue was cut into small slices and resulting fragments were placed in DEPC-treated water. Sample was vigorously vortexed, supernatant was recovered and centrifuged for 5 m in at 8000g. Bacteria} pellet was resuspended in RL T buffer and RNA was extracted using the RNeasy plant mini kit (Qiagen, Valencia, CA). cDNA synthesis was performed using random primers and AFLP was conducted using the AFLP analysis system fro microorganisms (GIBCO). cDNA was also synthesized from ARN extracted from bacteria grown in-vitro in liquid medium. Nine primer combinatíons were assayed for the second AFLP amplification. After the second AFLP amplification, the product was separated in a polyacrylamide gel (6%). Differential bands present in the inoculated bacteria and absent in the bacteria grown in-vitro were eluted from the gel, re-amplified and cloned as described above. A total of 118 clones were sequenced using the automated sequencer (ABI Prism 377). The sequences were edited using Sequencher 3.0 (Gene Codes Corporation) and compared with the GenBank databases using BLASTx and BLASTn. Slide printing and blocking Each DNA fragment was arrayed (arrayer SPBIO ver1.54, MiraiBIO, Inc) with four replicates onto glass slides coated with aminopropyltriethoxisilane (Sigma). The spacing between spots was 0.3mm. S lides were placed on a dessicant chamber wrapped into aluminum foil until use. A set of control genes was printed in the slide. Controls were cassava housekeeping genes, Xam ribosomal genes, pthB, a Xam pathogenicity gene, N 52, an interna! sequen ce of the pthB gene and a repetitive sequence present in the Xam chromosome. Two days after arraying, slides were processed. S lides were baked at 96C on hot plate for 1 min. They were then crosslinked using a UV stratalinker at 650Jll. Slides were treated with amine blocking solution (1.5g of succinic anhydre in 250ml 1-methyl-2-pyrrolidinone and 250m! of 0.2M boric acid pH8) for 15 min, gently shaking at room temperature. Then, slides were placed in boiling distilled water for 2 m in to denature DNA, soaked in 100% ethanol and dried by centrifugation (1 min at lOOOrpm). Results From the AFLP-derived librarles, 768 clones were obtained (384 clones from each library). These clones were all printed in the array. From the cDNA-AFLP-derived library, fragments varied in size from 50 to 800bp. For each primer combination, 20 to 80 bands were obtained. From 6 to 1 O clones were pic~ed for each differential band eluted from the gel. A total of 118 clones were sequenced, most of them showed homology to ribosomal genes. The 17 remaining (not showing a homology to ribosomal genes) did not presented an homoloy when compared with sequences in the Genbank We selected all clones from these 17 sequences to be arrayed in the Xam microarray. Future plans To hybridize Xam microarrays with ARN or cDNA obtained from Xam collected at different time points after inoculation and ARN from bacteria grown in-vitro. The clones showing differential expression will be sequenced to identify genes involved in pathogenesis. The clones, identified and sequenced, will be used as probes against northern blots to confirm their differential expression. 173 References Okinaka, Y., Yang, C.-H., Gavilanes Ruiz, M., and Keen, N. 200 l. Microarray profiling of gene expression in Erwinia chrysanthemi 3937 during plant infection. Abstract 263 In Proc. Int. Congr. Mol. Plant Microbe lnteract.; Madison, Wisconsin. Restrepo, S., and Verdier, V. 1997. Geographical differentiation of the population of Xanthomonas axonopodis pv. manihotis in Colombia. Appl. Enviran. Microbio!. 63:4427-4434. Restrepo, S., Duque, M. C., Tohme, J., and Verdier, V. 1999. AFLP fingerprinting: an efficient technique for detecting genetic variation of X anthomonas axonopodis pv. manihotis. Microbiology 145: 107- 114. Verdier, V., Cuny, G., Assigbetse, K., Geiger, J.-P., and Boucher, C. 1996. Characterization of pathogenicity gene in Xanthomonas campestris pv. manihotis. Abstract G-71 In Proc. lnt. Congr. Mol. Plant Microbe Interact. ; G. Stacey, B. Mullin and P. Gresshoff eds., Knoxville TN. 1.3.16 Gene expression in cassava stems in response to infection by Xanthomonas axonopodis pv. manihotis Santaella, M.\ Lopez, C.E. 2, Restrepo, S. 2, Jaccoud, D. 3, Gaitan, E.\ 1 Verdier. V2• and Tohme, JI. 1SB-2 Project, 21RD, 3Cambia Iristitute Introduction The factors controlling the outcome of host-pathogen interactions where there is no obvious hypersensitive response are yet not well understood. This is the case for cassava bacteria! blight, caused by Xanthomonas axonopodis pv. manihotis (Xam). Xam is a foliar and a vascular pathogen. No mechanism has been observed to limit the multiplication and development of the bacteria in the mesophyll of resistant cultivars during the foliar phase and intercellular multiplication in the mesophyll (Boher and Verdier, 1995). However defense mechanisms against the pathogen have been shown in the vascular system of infected cassava plants (Kpémoua el al., 1996). In order to establish which genes are involved in the expression of resistance of cassava (Manihot esculenta Crantz) to Xam infection, we used DNA microarrays containing 3872 Arabidopsis thaliana stress-related ESTs. Metbodology Plant inoculations, RNA isolation and cDNA synthesis Four weeks old plants (variety MBRA 685) were inoculated by stem puncture with Xam isolate CIO 151. Stem tissues were collected at 24 and 72 hours post inoculation (pi), 7, 15 and 30 days pi. The controls were healthy non-inoculated plants. Tissue was grounded in liquid nitrogen and total RNA was isolated using the Proteinase K method (Hall et al., in Rocha, 1995). Poly (A) RNA was isolated using oligo (dT) coupled to DynaDeads (DYNAL). cDNA was synthesized using oligo (dT) primer and SuperScript 11 reverse transcriptase (GIBCO BRL) from 400-500 ng of mRNA, as starting material. 174 Preparation of labeled pro bes Fluorescence-labeled probes were prepared from cDNA. Each feaction (6~1) consisted of 2-3~g of cDNA, 3~g Random primers, 2mM each of dATP, dCTP, and dGTP, 0.65 mM dTTP, 2nmol of either Cy3-dUTP or Cy5-dUTP, and IOU of Klenow fragment in IX reaction buffer. cDNA from healthy plants was labeled with Cy3 and a pool of cONA from all time points after inoculation was labeled with Cy5. The labeling reaction proceeded for 1 hour at 37°C. After incubation, reactions of the two samples were combined and purified using the Qiaquick kit (Qiagen, Valencia, CA). The sample was then dried until 5~1 were left and resuspended in 26~1 of CLONTECH ExpressHyb buffer (with 40ng of salmon sperm DNA). The probe was denatured at 95°C for 5 min and applied to the microarray. Microarray hibridization and data analysis Hybridizations were performed overnight at 60°C in humidified chambers. The slides were sequentially washed in the following solutions: 1 X SSC and 0.1% SDS for 1 O m in, 1 X SSC twice for lmin, 0.2X SSC twice for 2 min and 0.02 X SSC for 5 to lO sec. Slides were dried by centrifugation. After hybridization and washing, microarrays were scanned with Virtek chipreader®. Spots representing the arrayed genes were identified, and distinguished and analyzed using the Spotfinder and Arrayviewer softwares from TIGR (The Institute for Genomic Research). The average (integral) fluorescence intensity for each fluor and each gene was determined and background fluorescence was calculated as the median fluorescence signa! of non-target pixels around each gene spot. Missing spots, spots with low signal intensity, and spots in high background areas were flagged ·and excluded from the analysis. Normalization between the Cy3 and Cy5 fluorescent dye emission channels was performed using the total intensity for each channel, based on the assumption that under the conditions being tested, most genes will not change in expression. In this experiment, we defined induction or repression of a gene as a mínimum 2.5-fold change in its transcript level. Results Microarrays were used to study gene expression quantitatively after infection of cassava stems withXam. Figure 1 shows an image ofthe microarray after hybridization. Figure 1: Results of gene expression experiments using Arabidopsis microarrays and cDNA obtained from cassava stems, healthy (Cy3) and inoculated (Cy5) as probes. Analysis of data revealed that 20 spots were not flagged as bad spots by the Spotfinder software and of these, 15 ESTs on the microarray showed significant differential expression in response to the infection of cassava stems with Xam. The position of these spots in the array, the Cy3 and Cy5 intensities after normalization and the Cy5/Cy3 ratio is shown in Table l . 175 TabJe l. List of A. thaliana ESTs showing a differentiaJ expression after hybrídízatíon wíth labeJed cDNA obtained from bealthy (Cy 3) and inocuJated (CyS) cassava stems. Their position in the slide and total intensity obtained after scanning both fluors (Cy3 and CyS) as well as tbe Cy5/Cy3 ratio are shown. Row Column Metarow Metacolu Subrow Subcolu Cy3 Cy5 Cy5/Cy3 mn mn 1 83 4 J J7 13468 45156 3.35 6 76 1 4 6 10 13J83 43505 3.30 8 60 1 3 8 J6 13565 63114 4.65 10 77 1 4 JO JI 8538 1426286 167.05 JO 84 1 4 10 J8 14071 76568 5.44 28 65 2 3 6 21 112179 43910 0.39 34 6J 2 3 12 17 14115 532074 37.70 34 63 2 3 12 19 81949 405565 4.95 40 21 2 1 18 21 24604 1833882 74.53 51 84 3 4 7 18 12098 1867275 154.35 64 80 3 4 20 14 26607 9477 0.36 70 24 4 2 4 2 37780 655197 17.34 78 4 4 1 12 4 21540 69820 3.24 79 54 4 3 13 10 192629 1311269 6.81 80 10 4 1 14 10 27473 9325 0.34 Future plans Identification of the ESTs showing a significant induction or repression pattern during the incompatible reaction between cassava and Xam. Confirrn the differential expression of the genes identified with new Arabidopsis microarray hybridizations and through northern blots analysis. Develop microarrays containing cassava genes obtained from compatible and incompatible cassava-Xam interactions. References Boher B, Verdier V, 1995. Cassava bacteria! blight in Africa: The state ofknowledge and implications for designing control strategies. African Crop Sci. J. 2:1-5. Kpémoua K, Boher 8, Nicole M, Calatayud P, Geiger JP, 1996. Cytochemistry of defense responses in cassava infected by Xanthomonas campestris pv. manihotis. Can. J. Microbio!. 42:1 131-1143. Hall et al. 1978. In: Rocha, 1995. Evaluacion comparativa de ARN mensajero de Phaseolus vulgaris L. Entre una variedad resistente y una susceptible a Acanthoscelides obtectus (Say). Tesis de Grado, Universidad Nacional de Colombia. 176 1.3.17 Toward Fine-Mapping of Major Ge~es for Blast Resistan ce in Rice · Gerardo Gallego and Joe Tohme SB-2 Project Introduction We are rnaking progress towards cloning a blast resistance gene using positional cloning. Tightly linked rnarkers are required for rnarker-assisted breeding to ensure higher selection accuracy and efficiency. Our strategy for using positional cloning is based on: The genorne size of the rice is the srnallest among the rnajor crops ( 400 Mb ), The physicallength of the genetic distan ce per unit is srnall, ca 200 Kb cM; Rice has the rnost reliable transfonnation systern among the rnonocots; There are abundant restriction fragment length polymorphisrn (RFLP) markers and genetic infonnation about rice blast resistance (Kawasaki et al 1996). A large set of rnicrosatellites distributed along the en tire chrornosorne set, sorne of thern linked to RFLPs next to Pi genes. YACs and BACs contigs that contain Pi genes (Ciernson University-USA and Rice Genorne Program-Japan). Updated ESTs database ( Wang et al2001). We have started fine mapping a region of chrornosome 6 that contains genes Pi2, Pil3, and Pi9 for rice blast resistan ce, using as a so urce of resistan ce the variety lrat 13. Methods For mapping we used parental Fanny and Irat13, susceptible and resistant respectively for Pyricularia grisea isolates SRL-1 to SRL-6. The linkage rnap was constructed based on phenotypic segregation data frorn 104 doubled-haploid individuals. Plants were inoculated with blast SRL-1 isolates cica9-31-4 and cica9-6. MapMaker, with LOO 4,0, was used to analyze linkage of RFLP, RAPO, SCAR's, RGA, AFLP's and Microsatellites. For fine mapping, we selected a 6-7 cM interval between markers SCAR's B10 and microsatellite RM3 (easy to evaluate by PCR) in chromosorne 6, which rnay involve the centromere. To- search for recornbinants within this interval, a 1500-2000 individuals, F2 population is being grown in the greenhouse. Evaluations with cica9-31-4 and cica9-6 isolates will be carried out with the F3. The centromere is being rnapped using probe N36 (donation of Dr. Q. Zhang from National Key Laboratory of Crop Genetic Improvernent in China). Similarly, 29 RFLP-cDNA probes, rnapped around the centrornere and the interval for fine mapping (Harushima et al 1998), were donated by MAFF DNA BANK through RGP. Sorne ESTs close to RG648 (Z. wang et al 2001 ) and sorne rnapped retrotransposons (S. Wang et al 1999) rnay be used for fine mapping theinterval. We also developed a cONA expression library frorn Irat13, with 13824 clones that will be evaluated with rnicroarrays. 177 Results We obtained flanking markers for major resistance genes on chromosome 6 (figure 1) involved in resistance or susceptibi1ity to SRL-1 isolates cica9-31-4 and cica9-6 in a 104, doub1ed-hap1oid population derived from the cross Fanny x Irat13. Fifteen markers ( 1 SCAR's, 6 RAPO, 3 RFLP's, 2 AFLP's, 2 Microsatellites and 1 RGA), closely linked to resistance are within a 6-7 cM interval. This region may involve sorne Pi-2, Pi-13 or Pi- 9 genes, possibly located in the centromere or within the pericentromeric region. Two PCR- based, flanking markers have been selected for fine mapping (B10 2+/-3cM y RM3 3.5+/-4.0 cM; (Figure 1). Of the 15 restriction enzymes tested to search for polymorphisms for centromere mapping, Clal and Kpnl were selected in the parentals. Mapping the centromere and probes closely linked to it, will allow us to precisely define this region. For this purpose, we can use probes from other rice maps tightly linked to the centro mere that map next to or within the interval B 1 O-RM3. Twenty nine RFLP-cONA probes from the Japanese map, located within the centromeric region , were tested in parenta1s with Xbal, Oral, EcoRI, EcoRV and Hindill. No polymorphism was detected. The same probes are being tested with 15 more enzymes for future mapping to saturate the region. SCAR's B 1 O developed at CIA t is a good starting point for fine mapping. lt has shown perfect co-segregation with Pi-2 when tested in an Isolinea1 x Iso1inea6 segregating population. (Fernando Correa, pers comm). Resistance Genes-lrat13 Chromosome 6 Pil1(1) PUl Pil Pi9 ::~7 ;g~ ~ ~14 Ci Plant Mini Kit protocol of QUIAGEN. mRNA ¡urification involved severa! RNA mini-extractions, which were then run through Oynabeads Oligo(dt)25 from Oynal. Once mRNA was ready, we proceeded to cONA synthesis with the Kit Super Script Plasmid System® for cONA Synthesis and Plasmid Cloning of GIBCO-BRL. To fraction cONA, we used columns from GIBCO-BRL and collected fractions 7,8 and 9 for cloning. Libraries were screened for insert size by endonuclease digestion and PCR amplification. A E Figure l. Screening of cDNA expression libraries of O. Llanos 5 (A) and lratl3 (B) by digesting recombinant plasmids with Noti/Sal l. Results The two libraries showed an average size insert of 1,8 Kb, ranging from 0,5 to 2,0 Kb, which is optimum for cONA Microarray (Xiang et al, 2000). The cONA library of O.Lianos 5 was stored as ligations, at -80°C, for posterior electroporation into bacteria. Irat13 library was used to spot 13824 clones in 384-well plates, by duplicate, and stored at -80°C. Different clones from different plates were PCR-amplified directly from bacteria which guaranteed the presence of inserts in 90-95% ofanalyzed clones (Figure 1) . Both cONA libraries are expression libraries since inserts are directionally cloned respective to the transcriptional polarity of the mRNAs they derive from. Oirectional cloning thus facilitates the construction of subtraction libraries to search for diffrentially expressed genes. 180 Ongoing Activities • Screening expression libraries with markers mapped and closely linked to blast resistance genes. • Development of ESTs to saturate Fanny x Irat 13 in regions of interest. • Microarrays to monitor gene expression. • Development of subtraction libraries to search for differential gene expression. References Ou SH (1985) Rice Oiseases. Commonwealth Agricultura( Bureaux, Wallingford, UK. Xiang, C. and Chen, Y. cONA Microarray technology and its applications. Biotechnology Advances 18 (2000) 35-46. Rneasy Plant Mini Kit, 1999. QIAGEN. SuperScrip Plasmid System For cONA Sinthesis and Plasmid Cloning. Cat No. 18248-013. GffiCO BRL. 1.3.19 Detection of Differentially Expressed Genes Related to Apomixis using cDNA Sub.traction Coupled to Microarray Hybridization Cortés Diego Fernando, Miles John & Joe Tohme SB-2 Project Introduction The identification of differential gene expression between two organisms or cells types is a frequent goal in modern bio\ogical research. The possibility to determine mRNA expression differences between apomictic and sexual genotypes using the novel tools of functional genomics is a powerful too) to access the gene expression involve in this. Severa) recent and rapid PCR-based method, including Subtractive Suppression H_ybridization (SSH) and Representational Differences Analysis {RDA) have been for the cloning of genes that are differentially expressed between genotypes. Recently, cONA microarrays have been developed and used to quantitative differential gene expression by hybridization a complex mRNA -derived pro ve onto an arra y of PCR products. Microarrays allow thousands of genes to be monitored simultaneously for expression leve) and compared between tissues. Here we show the advancement in the merging of cONA subtraction technique with microarray analysis as a potential method for detection of unique differential expressed genes related to apomixes in Brachiaria. Materials and Methods Based in the most common protocols for cONA microarray different consensus protocols were tested in the implementation of this technique for the gene expression study with Brachiaria at CIAT. 181 The plasmids of 1920 selected cONA clones were collected from the cONA subtraction library, which was obtained from the subtractive suppressive hybridization· (SSH) between the cONA of B. decumbes, which contains specific transcrips and the reference cONA of B. ruzziciensis. The inserts of the cONAs were amplified by PCR in 96-well plate format using T7 and SP6 primers pair specifics for the pGEMT-easy vector. PCR reaction of 50¡.d with 5¡.¡.1 of a dilution 1:20 directly from the grow bacteria! as template. The PCR products were precipitated both by adding 50¡.¡.1 of ethanol and 5¡.¡.1 of 3M sodium acetate or 50¡.¡.1 of isopropanol. The precipitated samples were centrifuged at 3000rpm at 4°C for 30 minutes and washed with ethanol 70%. After dry down the samples it were resuspended in 10¡.¡.1 different spotting buffers (TE+glycerol 50%, SSC 3X, SSC 3X + SOS 0.1 %, and TE+OMSO 50%). The yield and qual ity of the PCR products were analyzed by agarose gel electrophoresis. The PCR samples were arrayed in duplicates from 384-well plates onto home made slides coated with poly-L-Lys as well as onto SigmaScreen n.t coated slides. After spotting both types of slides were processed to avoid non-specific hybridization. Two different were used for preparation of the labeled probe in the hybridization. In the first method we label cONA directly by incorporating fluorescently labeled nucleotides during oligo- dT primed reverse transcription and in the second method the cONA was prepared by a normal cONA synthesis reaction and it was used as template for a random primer labeling reaction with exonuclease free klenow (USB). Following the hybridization the slides were submerged in the washing solutions and the fluorescent image was acquired for both fluorescent dyes used by scanning the slides with the Hita,chi Genetic Systems. Results Experiments using cONA microarrays, can encounter technical problems at any step. Oifferent points, which are critica! in the cONA microarray implementation, has been salve, at least on the gene expression experiments with Brachiaria. The possibility provided by the microarray format, to include numerous interna) controls facilitate the recognition and correction of many kinds of problems. The different combination among the cleaning method of the PCR product, the spotting buffer, the type of slide and the probe labeling system allow us to establish a putative protocol for the gene expression analysis in Brachiaria with cONA microarrays. We establish that the precipitation of the PCR product with isopropanol work better than the Ethanol and sodium acetate combination. Using isopropanol the amounts of PCR product recover is higher than with ethanol, and the steps in the precipitation and cleaning process is less time consuming. About the spotting buffers and the type of slide, the combinations between poly-L-Lysine home made slides and TFJGlycerol or TFJDMSO 50% give stronger signa! than any other spotting buffer. Beside that poly-L-Lysine have more ONA affinity the background signal was also higher than when SigmaScreen s1ides were used. When the different spotting buffers were tested on SigmaScreen slides the only one which gives high affinity for the slide surface and regular spot form was TFJDMSO 50%, follow by 3x SSC, the rest of the spotting buffers spread the cONA all over the slide joining each spot to the neighbors. The hybridization results both incorporating fluorescently labeled nucleotides during o1igo-dT primed reverse transcription of the total RNA or random primer labeling reaction with exonuclease free klenow (USB) using cONA already synthesized give almost the same results without significant differences. 182 Future Activities cDNA microarray spotting. 5,000 clones from each library will select after mass excision, PCR amplified and spotted onto replicate glass slides using the SPBIO spotting robot at CIA T. Microarray hybridizations with Cy3 and Cy5 labeled cONA derived from poly(A)+ of genotypes used in the cDNA subtraction library construction. Microarray hybridizations with labeled PCR product of the SCARs markers includes in the Brachiaria genetic map, which are linked to the apomixes loci. Generation of cDNA chips with clones derived from a full-length cDNA library both sexual and apomictic genotypes. Identification of clones of interest. Computer analysis of the microarray hybridization output will be use to identify clones that change its expression patter between sexual and apomictic genotypes. Microarray hybridization and sequencing of the full-length cDNA chips, using as prove cONA clones identified as interesting in the cDNA sudstraction cDNA chip. Interrogating of microarrays. Cy3 and Cy5 labeled cONA derived from poly(A)+ of time points of interest will be used as pro be ·to hybridize cDNA slides. Identification of clones of interest. Computer analysis of the microarray hybridization output will be use to identify clones that change in abundance during the time course ofPPD. 1.3~20 Isolation of Resistan ce Gene Analogs (RGAs) from Brachiaria I.F. Acosta and J. Tohme SB-2 Project Introduction The cloning of resistance genes in different species is providing a wealth of information about the structure, expression and function such genes. Recent genetic studies reporting that the Mi gene for resistance in tomato to the root knot nematode, Meloidogyne incognita is the same gene for resistance to specific isolates of the potato aphid, Macrosiphum euphorbiae (Rossi et al., 1998). Mi was the frrst example of a plant Resistance Gene (R- gene) active against two such distantly related organisms. M oreo ver, it was the first isolate-specific insect resistance gene to be cloned. The gene belongs to the nucleotide- binding (NBS), leucine-rich repeat (LRR) family which includes the majority of cloned R-genes. Sequence similarity between cloned disease R-genes, especially those from the NBS class, allowed us to use degenerate primers in rice, cassava and common bean to isolate NBS-containing sequences that are potentially part of R-genes and are called Resistance 183 Gene Analogs (RGAs) (BRU Annual Report, 1998, 1999). RGAs have proved to be useful as "candidate genes" to map resistance loci, mainly in common bean (BRU Annual Report, 2000, 2001 ). Spittlebug is the most harmful pest of Brachiaria in America. Different methods must be integrated to achieve effective control of this pest. An efficient method is the use of cultivars that are naturally resistant. However, the molecular mechanisms underlying resistance are not fully understood. The objective of this study is to initiate a candidate gene approach using degenerate primers to isolate RGAs the NBS-LRR class from Brachiaria. Sequence similarity between cloned disease R-genes, especially those from the NBS class, allowed us the use of PCR with degenerate primers in rice, cassava and common bean to isolate NBS-containing sequences that are potentially part of R-genes. and are called Resistance Gene Analogs (RGAs) (BRU Annual Report, 1998, 1999). RGAs have proved to be useful as "candidate genes" to map resistance loci, mainly in common bean (BRU Annual Report, 2000, 2001). Spittlebug is the most harmful pest of Brachiaria in America. Different methods must be integrated to achieve effective control of this pest. A low cost method is the use of cultivars that are naturally resistant. However, the molecular mechanisms underlying resistance are not fully understood which does not really allows to take advantage of it. The objective of this study is to initiate a candidate gene approach degenerate primers to isolate RGAs the NBS-LRR class from Brachiaria. Methods and results DNA templates were used from B. brizantha CIA T 6294 and B. ruziziensis BR4x44-02, which are resistant and susceptible to the spittlebug, respectively. A mapping population of 215 individuals has been derived from the interspecific cross of these species. Degenerate primers based on the Nucleotide Binding Site (NBS) which is conserved in R-genes have been used successfully in rice, cassava and common bean to isolate RGAs of the NBS type (BRU Annual Report, 1998, 1999). For Brachiaria, we assayed degenerate primers designed by Leister et al. ( 1996, 1998), Lopez and A costa (BRU Annual Report, 1999) and Silvia Peñuela (unpublished results). Amplifications were obtained for the primers indicated in Table l . Primers targeting the TIR dornain of R-genes or the TIR-type NBS were also tested but, as expected, no amplification was obtained because monocotyledonous does not have this type ofR-genes (BRU, Annual Report, 1999; Meyers et al., 1999). PCR products from B. brizantha CIA T 6294 were separated and bands of the expected size were purified, cloned and transformed into E.co/i electrocompetent cells. A total of 144 clones (36 of each combination) were obtained and grouped by their restriction pattems with the 4-bp cutter enzyme Alul. Twenty two groups were identified and one clone of each group was sequenced using the Oye Terminator Cycle Sequencing Kit and the Applied Biosystems Prism 377 DNA sequencer (Perkin-Elmer). Sequences of 19 clones corresponded to RGAs and were classified in 7 classes according to their similarity: BRGA 1 to BRGA 7. 184 On-going work The next step is locate these RGAs in the genetic map that is currently being constructed at the BRU (Oiga X. Giraldo and Jaime Vargas, BRU, Annual Report, 2001) using the mapping population ofthe interspecific cross between B. brizantha CIAT 6294 and B. ruziziensis BR4x44- 02. RGAs are usually mapped as RFLPs and we have already obtained good polymorphic band between the parental DNA. Screening of the mapping population will be conducted next year. We have also designed primers that specifically amplify each BRGA class to convert them in PCR-based markers that are easier to evaluate. BRGAs will be amplified from each parental and cut with A/ul. Digestion products will be separated on acrylamide gels and polymorphic bands between parents will be used for mapping. Resistance to one species of spittlebug has been evaluated quantitatively in each individual of the progeny. After location of BRGAs in the map, QTL analysis will be performed hoping to find one BRGA explaining sorne leve! of resistance to spíttlebug. T!lble l. Summary of the use of degenera te primers to isolate RGAs from Brachlf!ria Primer combination S2 + AS3 S2 + AS4 S2 + GLPL3 S2 + PRS3 Expected size (bp) 500 500 500 700 RGA class obtained BRGA1 BRGA2 BRGA2 to BRGA 7 Primer sequences correspond to those in BRU Annual Report ( 1999) References Leister, D., Ballvora, A., Salamini, F. and Gebhardt. C. 1996. A PCR-based approach for isolating pathogen resistance genes from potato with potential for wide application in plants. Nat Genet 14: 421-429. Leister, D., Kurth, J., Laurie, D.A., Yano, M., Sasak.i, T., Devos, K., Graner, A. and Schulze-Lefert. P. 1998. Rapid reorganization ofresistance gene homologues in cereal genomes. PNAS 95: 370-375. Meyers, B.C., Dickerrnan. A.W., Michelmore, R.W., Sivaramakrishnan, S., Sobra!, B.W. and Young, N.D. 1999. Plant disease resistance genes encode members ofan ancient and diverse protein family within the nucleotide-binding superfamily. Plant J 20: 317-332. Rossi, M., Goggin, F. L., Milligan, S. B., Kaloshian, 1., Ullman, D. E., and Williamson, V. M. 1998. The nematode resistance gene Mi oftomato confers resistance against the potato aphid. PNAS 95: 9750-4. 185 1.3.21 Construction of a molecular genetic map of Brachiaria and QTL analysis of spittlebug resistan ce O. X. Giraldo1, J. Vargas1, E. Gaitán1, M.C. Duque1, J. Miles1, C. Cardona2, J. Tohme1• 1 SB-2 Project ; 2IP-1 Project Introduction The genus Brachiaria Griseb. belongs to the tri be Paniceae, comprises aproximately 100 species, mostly of African orígin. Sorne of these have found commercial use as forage in tropical America, with approximately fourty million hectares of Brachiaria pastures in Brazil alone (Valle and Miles 1992). The commercial species of B. brizantha and B. decumbens are tetraploid apomitic (Valle 1986), The Construction of a Brachiaria molecular map was initiated (BRU annual report pp 123-127, 2000), using a population of215 F1 individuals derived from a cross between an autotetraploid spittlebug susceptible individual B. ruziziensis and a tetraploid spittlebug resistant individual B. brizantha. The objective of the study is to increase the saturation of the map using SCARs and SSRs developed at CIA T, AFLPs, RFLPs probes from other grases species and tag the quantitative trait loci (QTLs) controlling spittlebug resistance in Brachiaria. Materials and Methods Plant Material: A sexual tetraploid B. ruziziensis (Swenne et al., 1981 ), susceptible to spittlebug (Clone 44-3), was used as a female parent in a cross with natural and apomitic tetraploid genotype B. brizantha resistant to spittlebug (accession CIA T-6294). DNA Extraction: DNA was extracted using the protocol described by Carlos Colombo (personal communication) with sorne modifications. 1g of tissue was dried at 48 oc for 20 hours and ground to fine power; 15 mi ofextraction buffer (0. 1M Tris-Hcl pH8.0, 0.05M EDTA pH8.0, 0.7 M NaCI, 4% CTAB and 1% BMe) was added and incubated at 65 °C for 10 min; 15 ml of chloroform:isoamyl alcohol (24: 1) was added and centrifuged at 3000 RPM for 30 min. The aqueous phase was transferred toa new tube and 8 mi of chloroform:isoamyl alcohol was added and centrifuged at 3000 RPM for 30 min, repeated twice. A volume of cold isopropanol was added to the supernatant and incubated over night at - 20 °C. The isopropanol mixture was centrifuged at 3000 RPM for 30 min at 4 °C. The DNA pellet was washed with cold 75% ethanol and dried at room temperature, and then resuspended in 300 ul of TE. Pancreatic RNAse was added to a final concentraction of 20 ug/ml. DNA was quantified on a DYNA QUANT 200 fluorometer (Hoffer Scientific Instruments, San Farancisco CA). Microsatellites:The isolation of the microsatellites and the methodology for PCR amplification and evaluation of polymorphism have been described previously (BRU annual report pp 123-127, 2000). An additional set of 26 new SSRs was evaluated this year. AFLP, RFLP RAPO: All 215 individuals were evaluated using the combination (E-ACG/M- CTA), The screening methodology was described in (BRU annual report pp 123-127, 2000). Protocols for RFLP and RAPD, markers in Brachiaria were described previously (BRU Annual report pp 105-11 O 1997). Linkage Analysis: Segregation of markers as single dose restriction fragrnent (SDRF) markers according to the genetic model was determined by departure from the hypothesized 1: 1 ratio by the Chi-square test. The data matrixes obtained for presence or absence of bands were analyzed 186 with MAPMAKER v 3.0b for PC (Lander et al. 1987), using LOD score of 6.0 and recombination fraction 0.3. Recombination was translated to genetic distances using the Kosambi map function. Results and Discussion Phenotypic screening: The Tropical Forage Entomology section screened the population of the average damage of individual hybrid plant (C. Cardona et al., 1999). The results indicate that approximately 74.5% ofthe population can be classified as resistant or susceptible individuals to the spittlebug damage (table 1). The average damage values cover a continuous range from 1 to 5 suggesting a quantitative trait. Three different ranges were derived allowing the classification of the population as resistant, intermediate or susceptible individuals. The Genetic Linkage Map: Sixty-eight SSRs, 5 combinations AFLPs (116 markers), and 35 RFLPs segregating in the male parent (CIAT-6294), were tested for linkage using MAPMAKER V.3 .0b. Polymorphisms were scored for presence (H), and absence (A), and analyzed for dosage among Fl progeny using Chi-square tests (P60 in vitro FEC, not for plants 92 > 20 in vitro S ame S ame Positive for (Oct 20, 2000) FEC and plants• 270400 1 in GH LBA4404- 7 months Negative _{AJ>r 27, 2000) pBIGCry (*) Unes 55 and 92 (Figure l) are high and fast expressors of the gus gene. Usually it takes less than an hour, at room temperature (25°C), for them to develop and homogeneous blue color in stems, shoots and leaves. · Molecular confinnation of the transgenic status of all four lines is currently done through PCR and RT-PCR for cry gene insertion and expression. Plants are still young for DNA extraction for Southem and Northem analysis. PCR has shown bands of the expected size for the cry gene. However, most probably due to the high GC content of primers, non-specific, amplified bands appeared in non-transgenic cassava lines as well. The identity of non-specific bands is being confmned by Southem hybridization ofPCR products. Further standardization ofPCR conditions to amplify cry is required to report accurate data. Transient expression of plasmid pBIGCry has been observed severa! times in MCol2215 after transfonnation with Agrobacterium strain C58Cl. However, it has not been possibl~ to recover FEC expressing gus (stable FEC transformants) on media with antibiotics as it was reported in the previous Annual Report (2000). We have found that besides the sensitivity to Agrobacterium infection, FEC from MCol2215 grows much slower than that of TMS60444 on non-selective media (Figure 2). This may partially explain the lack of success with the former cultivar. Rapid growth helps the cells to recover fast and survive on selective media after transfonnation. We are therefore intensifying our efforts with non-antibiotic, selectable markers like pmi for transfonnation with Agrobacterium and the particle gun. Although the gun may not be the most desirable method for transfonnation dueto multiple and incomplete insertion events (which may speed up gene silencing), cells are less stressed after transfonnation, and do not undergo long treatments with antibiotics to eliminate supervirulent Agrobacterium strains. The results with the gun show that pre-treating cells with high sucrose concentration before shooting them increases transient or stable expression of the gus gene after 2 to 6 weeks on selection with antibiotics or mannose. The expression of the gus gene from plasmid pBIGCry is usually stronger than that from pSGManCry, so more blue spots are observed with the former. 203 After shooting, cells from all three cultivars resume growth easily in liquid medium, maintaining their yellow color, indicating that shooting is notas stressful as infection with Agrobacterium. Figure l. Transgenic cassava plant lines 55 (left) and 92 from cultivar TMS6044, resistant to selection with geneticin and paramomycin, expressing gus and probably carrying the gen crylAb (inserted between the gus-intron and nptll genes in the same construct). Both lines were obtained through Agrobacterium-mediated transformation. Growth Curvu of TM$50444 and MCol2215 a.aoo ---··-·----- ---7 01000 0.7000 L ~ 0&000 / 1: / ...o()-lMB. G02 "' ~ TMS • G02•Ñ•CH ~ 0.5000 ~ .....a--MI* •G02 ~ ~~·002• .... •CH 0 4000 ~.....- 03000 0.2000 ' 7 ,. 21 D•ys Conclusions and Ongoing Work Figure 2. Growth curves of FEC from cassava cultivars TMS60444 (TMS) and Mcol2215 (Mcol) on Gresshoff and Doy (1974) medium (GD2). Note that the doubling time for the former is less than 14 days, while for the latter it takes more than 21 days. (AS = asparagine; CH = ca.~eine hvrlrnlv7.ate) Transformation of cassava cultivar TMS60444 using Agrobacterium has been achieved at BRU. Full transgenic plants, growing in the greenhouse and expressing inserted genes, can now be obtained in about 11 months. TMS60444 becomes then a model to fast-test genes of interest in cassava. We will perform molecular tests to confmn transgenesis ofTMS lines and the expression of the cry gene. These transgenic plants may now be used for bioassays to test if they can control Lepidopteran insect pests of cassava like the stem borer (C. clarkei). Transformation experiments will continue using Agrobacterium and the gun with emphasis on commercial cultivars for which FEC has been established (i.e., for Mcol221 S, CM3306-4 and SM 1219-9). We will al so emphasize on the use of non-antibotic markers ( i.e., p mi) for selection of transgenic lines. 204 References James C. (2000). Global status of commercialized transgenic crops: 2000. ISAAA Briefs No. 21: Preview. ISAAA: lthaca, NY. Gresshoff, P. and Doy, C. (1974). Derivation ofa haploid cellline from Vitis vinifera and the irnportance of the stage of meiotic development of anthers for haploid culture of this and other genera. Z. pflanzenphysiol. 73: 132 - 141 . 2.1.3 Transformation and Regeneration of so me CIA T Elite Cultivars: Towards Testing Candidate CMD Resistance Genes Nigel Taylor3, Claude Fauquef; Tunji Akano2, Martín Fregene1 1 SB-2 Project ; 2IP-3 Project; 3IL T AB Introduction Following the discovery of a single dominant gene(CMD2) that controls the novel form of resistance to CMD and efforts to identify candidate genes by the serial analysis of gene expression (SAGE), the stage is set for gene function analysis of CMD2. However genetic complementation of gene function is severely hampered by a lack of routine and robust transformation and regeneration .methods for Latín American cultivars, the target group of CMD susceptible cultivars. A series of experiment was therefore set up in collaboration with the Danforth center to examine and refine the different steps for cassava regeneration and transformation The first step in the development of tissue for transformation and regeneration experiments is the induction of embryogenesis. Cassava varieties are known to respond differently to embryogenesis. There is _ therefore a need to first screen for responsive lines and to identify optimal conditions for embryogenesis. A set of elite materials from CIA T were selected for the experiments based on their success with farmers and as parents for cassava breeding. Methodology A list of the seven CIA T cassava elite lines used in this study can be seen in Table l. The lines were multiplied in-vitro in MS2 (Murashige and Skoog's basal medium supplemented with 2%sucrose) medium to generate large quantities of plantlets. This was to facilitate production of young leaf lobes for explanting to perform primary embryogenesis experiment. Unopened young leaf lobes were excised from these plantlets at 4weeks old and placed on Murashige and Skoog's media supplemented with 2% sucrose and 50Mm Picloram(MS2,50P). Alongside this work, explants from 4 week-old plantlets from model cultivar for cassava embryogenesis, TME 60444 were cultured on MS2, SOP and placed at different light intensity. This was to identify the optimal condition for embryogenesis, since light is known to be a crucial factor. Three light levels were tested-dark, low light (0.42 Jlmol m"2 s"2) and high light (7.2 Jlmol m"2 s"2) . Globular embryogenic structures from model cultivar 60444 and the 4 responsive lines from CIA T were excised and placed on Gresshorf and Doy (1974) medium, supplemented with 2% sucrose, 50Mm Picloram. These cultures were in culture rooms and at 3-4 weeks, the new globular embryogenic structures produced were transferred to fresh GD2, SOP medium. 205 Results The result of the embryogenesis experiment in the CIA T lines are summarized in Table l. Three CIAT lines, CM523-7, CM6740, CM2177-2 had a conversion rate to embryogenic structures from explant comparable to the control cultivar TMS60444. Pretreating in vitro plants in different light regimes did not have a significant influence on the induction of embryogenic tissues. However, when young unopened Jeaf lobes were explanted onto Murashige and Skoog basal media supplemented with 2% w/v sucrose and 50 J..LM picloram, low light was found to be significantly superior no light and high light treatments. Low light lead to an increase in the fonnation of embryogenic structures by a factor of two. After two cycles FECs were produced in the model cultivar. These were then multiplied to generate more target tissues for transfonnation experiment. After 3 cycles FECs were generated in 3 out of the 4 lines from CIA T. Work is on going to increase the conversion rate among the CIA T lines and al so produce more OES from explants in order to increase the generation of FEC Future Plan Conduct transformation and regeneration efficiency experiments, with FEC from model cassava cultivar TME 60444 as target tissues, to compare particle gun bombardment and Agrobacterium mediated transformation Translate the findings from above to elite lines from CIA T from which FEC have been generated to achieve a transgenic lines with genes of interest. Table l. Conversion rate of explants from CIAT's elite cassava lines to embryogeoic structures Serial Genotype Percent Putative Percent Organized No of explant used number Embryogenic Embryogenic Tissue Tissue l CM523-7 20 70 80 2 CM6740 20 50 90 3 CM2 177-2 20 75 80 4 CM4574-7 1 l 80 5 Tai1 1 1 -80 6 Tai8 o 60 80 7 MBRA383 5 1 80 8 Control TME60444 20 75 206 2.1.4 Control of RHBV (Rice Hoja Blanca Virus) through nucleoprotein mediated cross protection in the greenhouse and in the field. L. Fory1, A. Mora2, E. Tabares1, G. Delgado2, T. Agrono2, C. Ordoñei, C.Dorado\ M.C.Duque1-2 J. Silva2, L. Calverr, Z. Lentini1 1SB-2 Project; 2IP-4 Project Introduction Rice hoja blanca virus (RHBV) is a major virus disease of economic importance affecting rice in northern South America, Central America and the Caribbean. Rice transformed with the RHBV nucleocapsid protein (N) gene had a significant reduction in disease development. Reactions to inoculation with RHBV ranged from susceptible to completely resistant plants (immunity). The most frequent reaction was characterized by local necrotic lesions (hypersensitive reaction) followed by the production of new leaves without symptoms. Other plants developed chlorotic lesions in the inoculated leaves, but recovered producing tillers free of virus (recovery phenotype). These transgenic RHBV resistant rice lines expressed the N gene RNA at low levels that could be detected using RT-PRC but not by Northern blots analysis. The nucleocapsid protein could not be detected in any of the transgenic plants either by Western or ELISA tests. These results suggest that resistance conferred by the N gene is RNA mediated. Earlier reports indicated that besides the resistant phenotype when challenged with RHBV, the resistant transgenic lines showed significant increased performance for important agronomic traits including number of tillers, number of grains per plant, and yield as compared to the susceptible control. Upon inoculation sorne ofthe resistant transgenic plants showed agronomic traits similar to the uninoculated non-transgenic Cica 8 control. Using both agronomic traits and disease severity as criteria, severa! of the most resistant lines were followed through the R4 generation in the greenhouse and demonstrated that the N gene and RHBV resistance was ínherited in a stable manner. Last year, results also suggested that the resistance conferred by the N transgene towards RHBV disease is expressed independently of the genotype background. The transgenic resistance could be used to complement the natural resistance source to the virus, when crossing selected transgenic lines with diverse genotypes carrying the breeding resistance gene(s). Results showed that the non-transgenic F1s control plants were susceptible, whereas the transgenic F1s were resistant even when inoculated at 1 0-day-old. These results suggested that the protection conferred by the RHBV -N transgene is inherited and expresses independently of the genotype background, and that the transgene could be used to complement the natural resistance source. This year report includes the characterization of mode expression of the transgenic resistance conferred by RHBV -N. The evaluation in the field for two semesters of transgenic lines representing various generations, and F2 populations derived from crosses with Fedearroz 50, Oryzyca 1, Iniap 12, and Cica 8. Planting was coducted at CIA T headquarters u pon approval from the Colombian Biosafety Committee on September 2000. lt is also reported the progress generating transgenic rice containing the RHBV non-structural 4 (NS4) gene from the RNA 4. Materials and Methods RHBV Resistance Assays For the greenhouse experiments, in order to characterize the RHBV -N resistance according to the plant age plants of 15 or 28 days old were inoculated with RHBV using four 2nd or 3rd instar T. 207 orizicolus nymphs per plant, from a colony of 80% virulence. lnsects fed on the test plants for 5 days when insecticide is applied. Controls consisted of transgenic plants of Cica 8 carrying only the hygromycin resistance (hpt)gene, which was used as the selectable marker to generate the RHBV-N transgenic plants, or non-transgenic plants of Cica 8. Plants were scored for the development of RHBV disease symptoms every 3 days for 25 days and then evaluated once a week for 5 weeks. Plants were scored for the date of the first appearance of symptoms, and the percentage of leaf affected by RHBV was determined. For the field evaluations, 280 transgenic lines and F2 plants derived from crosses between resistant transgenic plants with Fedearroz 50, Oryzyca l, Iniap 12, and Cica 8 were planted in November 2000, and 486 transgenic lines and F3 plants derived from the same crosses described above were planted in July 200 l. Lines were planted in a randomized plot design with 3 or 4 replications in the field, for years 2000 and 200 l respectively. Plants were inoculated at 15 days of age with two insects per plant from a colony of 80% virulence. Insects were feeding on the plants for 15 days in the first season, and for lO days in the second season, upon when insecticide was applied as a biosafety control measurement. Plants were evaluated for the development of disease symptoms every two weeks until ' 45 days of age. Disease evaluations were conducted using an scale from Oto 9, were O refers tono disease symptoms, and 9 indicates more than 90% leaf area is affected by the RHBV disease. Plants with rating from l to 3 = resistant, score 5 = intermediate, and 7-9 =susceptible. Results and Discussion Comparative Leve/ of RHBV-N Resistance in Transgenic Rice in the Greenhouse Respect to the Field. Last year we showed that line A3-49-60-l2-3-3 showed the highest leve! of resistance throughout the whole life cycle. Between 74% to 81% of the plants did not show any disease syrnptoms when inoculated either at 15 days or 28 days of age, and only a 22% of the plants showed more than 25% of the leaf area affected when inoculated at 15day-old (Table l ). In contrast, Cica 8 control showed l 00% of the plants with severe disease symptoms at 15-day-old (Table l ). Line A3-49- 60-4-5-8 showed intermediate leve! of resistan ce at 14-day-old (Table l) and 71% of the plants without symptoms at 28-day-old .About 70% of the plants of line A3-49-60-l9 had less than 25% of leaf area affected at 15-day-old (Table 1 ). Sister !in es A3-49-60-12-3-1 (susceptible) y A3-49- 60-12-3-3 (resistant) showed different disease reaction indicating that the resistant phenotype was still segregating at the T4 generation or gene silencing was affecting the expression of the RHBV- N gene in sorne of the plants. This year results indicated that there is a high correlation between leve! of resistan ce seen in the greenhouse with that in the field (Table l ). Field evaluation of RHBV resistance in transgenic rice containing the nucleoprotein gene Advanced generations of transgenic lines with stable RHBV resistance were selected first in the greenhouse until permit for the field test was granted by the Colombian Biosafety Comrnittee on September 2000. Field was first planted on November 2000 to conduct evaluations for RHBV resistance and agronomic traits following International as well as the Colombian environmental biosafety regulations at the biosafety rice field located in CIA T experimental station Palmira. 208 Table 1.- Comparative Disease Reaction in tbe Greenbouse and tbe Field in Plants lnoculated at 15 days old Greenhouse Fietd> Leaf Arca A.ffected % Disease reaction Line o >Q-25 >25-100 2000 2001 AJ-49-60-12-3-3 74 4 22 1 2 AJ-49-60-19 53 12 36 2 3 AJ-49-60-4-S-8 54 o 46 3 AJ-49-60-13 25 4 70 3 AJ-49-56-15 9 13 78 6 6 AJ-49-60-12-3-1 7 o 93 6 7 AJ-49-101-18-19-2 15 o 85 8 6 AJ-49-782 o o 100 ND 7 Cica8 o o 100 9 7 Transgenic control with the hpt gene only. ND = not determined >Mean values ofthree (in 2000) or four (in 2001) replicates Replicated trials were conducted as described in materials and methods. Disease evaluations were conducted using an scale from Oto 9, were O refers tono disease symptoms, and 9 indicates more than 90% leaf area is affected by the RHBV disease. Plants with rating from 1 to 3 = resistant, score 5 = interrnediate, and 7-9 = susceptible. Besides the non-transgenic variety Cica 8, the transgenic Cica 8 line A3-49-78) carrying only the hygromycin resistance (hpt)gene, used as the selectable marker to generate the RHBV -N transgenic plants, and which does not contain the RHBV -N transgene , wasused herein as controls. Other 12 varieties were used as reference for differential disease reaction pattem. This variety differential included: Caribe 8, Capirona, Cimarrón, Colombia 1, Fedearroz 50, Fedearroz 2000, Fedearroz Victoria 1, Fundarroz PN1 , Iniap 12, Linea 2, Oryzica 1, Oryzica Llanos 5, and Palmar. Field evaluations corroborated results obtained previously in the greenhouse. Forty five entries derived from line AJ-49-60-12-3-3 were highly resistant showing scores 1 to 3. A subset ofthese lines are shown in Table l. Resistance in advanced transgenic Iines (T6 and T1) was inherited stably (95% of derived progeny plants were reproducibly resistant) indicating that the RHBV-N transgene seems to be fixed in each of these transgenic lines (Table 1) . The most resistant lines were A3-49-60-12-3-3-79, AJ-49-60-12-3-3-24, A3-49-60-12-3-3-28, A3-49-60-1 2-3-3-32, A3- 49-60-12-3-3-68, A3-49-60-12-3-3-67, A3-49-60-12-3-3-72 (Figure 1 ). These lines were more resistant than Fedearroz 2000 in average over the two field evaluations (Figure 1).0ther transgenic lines were interrnediate giving ratings between 3 to 5, but several transgenic lines were as resistant as Fedearroz Victoria (Figure 1 ). Large number of lines were more resistant than commercial varieties, and the resistance genetic gain from Cica 8 to the transgenic lines was a reaction score change from 7-9 (Cica 8) to 1-3 (best transgenic lines) (Figure 1). Transgenic line A3-49-78 which only carries the hygromycin resistant gene and does not contain the RHBV-N gene was as susceptible as Cica 8 (Table 2) indicating that the resistance noted in the RHBV-N transgenic lines is due to the RHBV N viral gene. Fedearroz 50 , the main variety currently commercially grown in Colombia, was susceptible to the virus (score 7) and more susceptible than Fundarroz PNl (score 5.5) which reacted as interrnediate (Figure l ). Similar progress towards resistance was noted in the crosses between the highest resistant transgenic lines and the varieties. Crosses between non-transgenic Cica 8 and Fedearroz 50, Oryzica 1, and Iniap 12 gave an average of disease reaction of 8.0, 7.0, and 8.5 respectively corresponding to susceptible phenotype, in contrast the corresponding crosses generated with the transgenic resistant plants and the same varieties showed an average disease reaction of 4.5, 4.0, and 4.0 respectively . The crosses are at F3 generation, thus the RHBV-N gene is still segregating in those lines. Resistant 209 plants (score 2-3) within those segregating F3 lines had been identified. These field results corroborate earlier findings indicating that the protection conferred by the RHBV -N transgene is inherited and expresses independently of the genotype background, and that the transgene could betransferred to breeding populations by standar crossing. Sister lines of the resistant transgenic lines and transgenic crosses evaluated last season indicated that sorne ofthe plants showed a yield potential similar to varieties as Fedearroz 50, Fedearroz Victoria 1, and Oryzica l. Currently progeny plants derived frorn resistant plants selected in the first field evaluation are currently being selected for yield potential and other agronomic traits. Table 2.- RHBV Resistance Evaluations ofTransgenic RHBV-N Plants Over Two Consecutive Seasons in the Field Seo re Line 2000 2001 A3-49-60-12-3-3-68 1 3 A3-49-60-12-3-3-32 1 3 A3-49-60-12-3-3-79 1 2 Fedearroz 2000 2 3 Victoria 1 2 4 A3-49-60-12-3-3-24 2 2 A3-49-60-12-3-3-28 2 3 A3-49-60-4-5-8-79 2 4 A3-49-60-12-3-3-72 2 3 A3-49-60-13-2 2 2 A3-49-60-12-3-3-19 3 3 A3-49-60-12-3-3-18 3 4 A3-49-60-12-3-3-78 3 3 A3 -4 9-60-13-8 3 4 A3-49-60-13-1 3 3 A3-49-60-12-3-3-77 3 4 A3-49-60-12-3-3-12 3 3 A3-49-60-12-3-3-59 3 4 A3-49-60-12-3-3-3 3 3 A3-49-60-12-3-3-14 3 3 A3-49-60-12-3-3-67 3 2 A3-49-60-19-8 4 4 A3-49-60-12-3-3-23 5 4 A3-49-10 1-18-19-2 5 6 Fundarroz PN 1 5 6 Palmar 6 6 lniap-12 8 8 Colombia 1 9 7 Oryzica 1 8 7 Capirona 8 7 Caribe 8 8 8 Cirnarron 8 8 Fedearroz 50 7 7 Linea 2 6 7 Oryzica 1 7 7 Oryzica Llanos 5 7 7 A3-49-60-1 0-27 ND l A3-49-78 * ND 7 Cica 8 9 7 (*) Transgenic fine carrying only hygromycin resistance gene .It does not contain the RHBV-N transgene. 1Mean values ofthree (in 2000) or four (in 2001) replicates 210 Fedcanat 2000 Genetic Gain Victoria 1 h \mar Fodearrat SO Oryzica Uanoo S Oryzica 1 Capirma Caribe 1 Cimarrón Colombia 1 lniap 12 L----- Ciea 1 o 2 3 4 5 6 7 8 9 RHBV disease reactioo Figure 1.- Disease reaction of RHBV-N transgenic plant.s and commercial varieties in the field. Mean values of two evaluations in November 2000 and July 2001. 2.1.5 Characterization of Transgenic Rice Containing the RHBV Non- Structural 4 (NS4) Gene from the RNA 4 L. Fory1, l. Lozano2, E. Tabares1, G. Delgado2, T. Agrono2, C. Ordoñei, C. Dorado\ M.C. Duque 1•2, J. Silva2, Z. Lentini 1'2 1SB-2 Project ; 2IP-4 Project Introduction The genome of RHBV consists of four species of ssRNA designated RNA 1, 2, 3 and 4. The RNA 4 consists of 1991 nucleotides with two open reading frames (ORFs). The most important ORFs is located in the 5' proximal region ofthe viral RNA 4. The RNA 4 encodes a major non- structural protein (NS4) , which accumulates in the tissues ofthe infected plants with the RHBV. NS• protein is clearly distinguishable from the nucleoprotein (N) by specific antiserum. Another difference between NS4 protein and N protein, is that the fonner only is expressed in the plant whereas the latter expresses both in the plant and the insect vector. It is inferred from the 21 1 differential expression of these proteins that the major NS4 protein may has a function that is needed in the plant but not in the plant-hopper. The differential plant-insect NS4 expression, and the similarity of NS4 sequen ce with well characterized helper proteins described for other insect transmitted viruses, suggest that NS4 might be involved in the RHBV transmission from the plant to the plant-hopper, or in the virus movement from cell to cell. The main goal for the expression of the RNA 4 in transgenic rice is to determine the function of the major NS4 protein and study the potential for a novel and different method of producing viral resistant plants. Materials and Methods Rice Transformation Mature embryos derived calli from indica varieties CICA 8, Palmar, Cimarrón and Fundarroz PN1 , where used as targets. We used Agrobacterium Agl1 strain mediated transformation to introduce the NS4 gene. Constructs p!C002 and p/C004 contain the RHBV NS4 gene in sense and anti-sense orientations respectively, driven by the 35S CaMV promoter using as backbone the plasmid pCAMBIA 1301 which carries the gus-intron and hygromycin resistance genes. Constructs,PIC007and p/C009 contain the NS4 sense gen, and p/COOB the NS-1 anti-sense gene driven by the ubiquitin or 35 S CaMV promoter. These genes were cloned into pWBVec8 plasmid (from Peter Waterhouse's laboratory at CSIRO, Australia), which carries the hygromycin-cat 1 intron gene as selectable marker. Plants were regenerated after stepwise selection on 30 mgll and 50 mgll hygromycin, followed by 50 mgll hygromycin throughout plant differentiation. Plants were grown to maturity in the biosafety glasshouse. Molecular Analysis of the Transgenic Rice Plants Southern Analysis.15 J.Lg DNA genomic were digested with Eco RI, fractionated in 1.0% (WN) agarose gels, and transferred to nylon membranes (N+ Amersham). The hybridization probe was a radioactively labeled 850 pb PCR fragment amplified using primers RHBV 4 forward and reverse. DNA probes were random primer labeled and hybridization was carried out ovemight at 55°C. RT-PCR and Northern Analysis.Total RNA was extracted from 100 mg of fresh material using the RNAeasy ™ plant total RNA kit (Quiagen, Dorking, UK). The cDNA synthesis was done with the SUPERSCRIPT™ One Step RT-PCR system. Following the manufacturer's instructions and using primers RHBV4 forward and reverse. Northem analysis was carried out wjth 15 J.Lg of total RNA per lane, using denature RNA gels with formaldehyde and forrnamide (Sambrook et al., 1989). Characterization of Sequence of NS4 Gene. Nucleotide sequence of the gene NS4 in four To generation transformed plants was carried out using the ABI PRISM. Dye terminator kit (Perkin- Elmer) with primers RHBV4 forward and reverse. These products of PCR were applied to Biosystems Prism 377 DNA sequencer (Perkin-Elmer) and edited with Sequencher (Genecodes, Ann Arbor, MI). The sequences were analyzed using the BLAST algorithm (Altschul et al., 1997). RHBV resistance assays. lnoculations and evaluations were conducted in the greenhouse following the same procedure as described above for the RHBV N transgenic plants. 212 Results and Discussion Last year we reported the generation of 1 O different constructs canying the NS4 gene in sense and anti-sense orientations. A total of 21 transgenic plants canying the NS4 sen se orientation, and 70 plants canying NS4 anti-sense orientation were produced (Table 1). The plant regeneration effciency varied according to the genotype from 2% to 44%. Southem blot analyses using Bam Hl or Eco R1 which excise the complete NS4 gene in sense or antisense orientation, or using Sal! which does not cut the gene cassette within the right and left borders indicated that between 50% to 1 00% of the regenerated plants analyzed contained the NS4 gene (Table 1 ), and most cases the NS4 gene is integrated as a single non-rearranged copy. Table l. Transformation Efficiency ofThree indica Varieties Using the RHBV NS4 in Sense an Anti- Seose Genes Genotype Plasmid NS4 Plants RE % Plants s• % Plants s• TE % PALMAR piC007 Sense 1 2 ND piC009 Sense l 3 ND piC004 @Sense 13 28 13/1 3 lOO 28 piC008 @Sen se 9 20 6/6 lOO 20 CICA-8 piC002 Sense 14 33 1111 4 79 26 piC007 Sense 6 12 2/3 67 8 piC009 . Sense 18 44 18/ 18 100 18 piC008 @Sen se 10 18 5/10 50 9 CIMARRON piC008 @Sen se 19 48 NO ND ND @Sense = anti-sense. RE = p1ant regeneration efficiency. s• = Southem positive. TE = transformation efficiency . ND = not determined The NS4 gene was also amplified by PCR generating the expected gene size. The PCR product was sequenced, and in all cases the sequence corresponded to the entire NS4 gene indicating that the transgene did not have rearrangements. PCR analysis for the gus-intron, hpt, and NS4 transgene; and Southem blot, RT-PCR and Northern analyses for the NS4 of T1 plants derived from T0 identified as transgenic by Southern blot, indicated that not all T1 plants iñherited the three transgenes (i.e gus-intron, hpt and NSo~ )(Table 2). Variation in the level of gus expression was also noted, and the expression of either gus or NS4 was not always detected although plants contained the corresponding gene. These results suggestthe presence of gene silencing (Table 2). Most plants showed low levels of RNA expression from either the NS4 sense or anti-sense genes. In these plants the RNA was detected by RT-PCR. NS4 gene expression was detected by regular Northern in only one plant so far (Table 2). 213 Table 2. Gus-intron, hpt, and NS4 transgenes inheritance, and NS4 expression in T1 transgenic plants Plant Plant PCR NS4 Genotype Plasmid To Tt ous• gus hpt NS4 SJ RT-PCR Northern Cica 8 PIC002 1 ND ND 2 5 + ND ND 2 11 ND ND 7 +++ + + + ND ND 7 2 +++ + + + ND ND 7 3 +++ + + + + ND ND 7 4 + + + + 7 18 ++ + + + + + 9 14 +++ + + + + 9 15 +++ + + ND ND 12 7 +++ + + + + + 12 11 ++ + + ND ND 12 15 +++ + + ND ND Cica8 None NT NT Palmar PIC004 18 + 24 10 16 ++ + + + 4 3 +++ + + + + + 4 5 +++ + + + + 4 17 +++ + + + + + 4 18 4 20 +++ + + + + + 4 25 +++ + + + + + 7 4 + + 7 16 + + + + 7 22 + + + + 7 23 ND ND Palmar None NT NT 1 Test of Gus in leaves. The expression of GUS gene was scored based on the level of expression. (+) low; (++) intermidate; (+++ ) high. 2 PCR analysis to detect the transgenes: gus ;hpt, hygromycin; NS4 . 3 S = + Positive to Southern blot analysis. ND= Not determined. T1 plants derived from the same T0 plant are sister lines. NT = Not transgenic T1 plants derived from self cross of eight T0 plants, originally identified as transgenic based on Southem analysis, were selected to conduct the RHBV resistance evaluations in the greenhouse. Each T 0 !in e was represented by 13 T 1 plants and were inoculated at 18 days after germination with four insect vectors per plant derived from a colony with 80% virulence. Paralelly, 25 T1 sister plants were used to determine the leve! of GUS expression. Results indicate that most T 1 plants from line 4 transformed with plasmid piC004 showed the expected GUS expression indicating inheritance of the gus-intron gene. In this line, the number of plants with no o minor disease symptoms ( < l 0% leaf area affected) was double respect to the non-transgenic control (Table 3). No difference in disease reaction were noted for the other transgenic lines and the 214 corresponding control (Table 3). However, for these lines there is indication that the transgene inheritance is significantly deviating from a Mendelian segregation and in 5 out of the 8 lines evaluated showed less than 25% T1 plants with the transgene respect to the 75% expected. Beca use of this skewed segregation, commonly found in early generation of transgenic plants, it is necessary to advanced to the T2 generation from those T1 transgenic plants carrying and expressing the corresponding transgenes. Following, T2 plants that inherited and expressed the NS4 gene need to be dentified by molecular analyses, and challenged thern with RHBV. Table 3. Disease resistance on T 1 transgenic plants derived from different T0 plants carrying NS4 sense (piC002) or NS4 anti-sense {piC004) and inoculated with RHBV at 18 day old in tbe glasshouse Leaf Area Affected Genotype Plasmid T0 Piant % plants Gus 1 {% Plants) S 10 >10-100 Cica 8 piC002 1 4 o 100 7 4. o lOO 9 40 15 85 12 40 o 100 None Control o o lOO Palmar piC004 1 12 31 69 4 84 62 38 7 12 46 54 JO 24 46 54 Non e control o 31 70 Eigbteen plants were evaluated per eacb T1 line. 1 25 plants per T 1 line were tested for GUS expression. References Altschul, S.F., Madden, T.L., Schaffer, A., Zahang, Z., Miller, W. and Lipman. D. 1997. Gapped BLAST and PSI-BLAST: A new generation ofprotein database search programs. Nucleic Acids Res. 25, 3389- 3402. Sambrook, J. Fritisch, E. and Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2"" ed. Cold Spring Harbor. NY: Cold Spring Harbor Laboratory Press. 2.1.6 Foreign genes as novel sources of resistance for fungal resistance E. Tabares 1, L.F. Fory1, G. Delgado2, M.A. Santana3, T. Agrono2, C. Ordóiiez 21, N. N. Turner4, Z. Lentini1•2• 1SB2 Project, 2IP4 Project, 3IDEA, Caracas, Venezuela, 4Biotechology Center, Rutgers University. Introduction The fungal Rhizoctonia solani (sheath blight), Helmithosporium, Rhincosporium, and Sarocladium is already causing important rice yield losses in the Southem con e of South Arnerica 2 15 and increasing spreads had been reported in Colombia, Mexico and Venezuela. All rice varieties are susceptible and there are not known sources of stable genetic resistance for these diseases in rice. In the case of sheath blight, IRRI had placed a major effort in developing biological control strategies for this disease without success either. At present, the control of this complex mainly depends on use of fungicides (Dr. Fernando Correa, CIA T Rice Pathologist, Cali, Colombia, personal communication). Recently, FLAR suggested CIA T (Dr. Peter Jennings, personal communication), to develop molecular strategies for incorporating resistance to this fungal complex. However, very little is known about the interaction between the rice and these pathogens in order to designspecific resistance strategies for each of these fungi. Of the four, the plant-pathogen interaction with Rhizoctonia solani is the better known. Work conducted by another principal investigator of this project (Dr. Nilgun Tumer, Biotechnology Center at Rutgers University, USA) showed that a pokeweed antiviral protein (PAP), a 29-kDa protein isolated from Phytolacca americana (a weed naturally found from USA to Argentina), has a ribosome-inactivating ability. Mutated versions of PAP gene has potent antifungal activity (Zoubenko et al., 1997). Homozygous progeny of transgenic tobacco plants expressing these P AP genes displayed resistance to the fungal pathogen Rhizoctonia solani. Transgenic PAP potato showed prótection against Phytophtora infestans, and transgenic PAP turfgrass are resistant to various fungal pathogens. These results suggest the possibility of designing molecular strategies for incorporating fungal resistan ce by introgression of mutant P AP gene(s) in transgenic rice plants. Here we report the progress made during the second year ofthis project. Last year we reported the gerieration of eight new PAP mutations directed to change the aminoacids composition in the PAP protein. These new mutated genes were placed into yeast vectors, and transformed into yeast to check for no toxicity. The non-toxic mutated genes are being transformed into tobacco frrst to check the gene expression and toxicity before using them for rice transformation. Two mutated versions of PAP (1 deleted and 11) already tested for no toxicity in turfgrass (another monocot species) were used as the first approach to transform rice. These genes driven by the ubiquitin promoter were placed in the plasmid vectors p WBVec8, pWB10a, and pBGXiHGFP kindly supplied by Dr. Peter Waterhouse (CSIRO, Australia). These plasmids had been used successfulJy by Waterhouse to transform rice via Agrobacterium. They contain a hpt gene with a CA T -1 intron for increased expression of hygromycin resistan ce and selection in rice, a gus-intron-gene, ora gfp (green fish fluorescent) gene, respectively, to aid the recovery of transgenic plants. A total of 35 independent transgenic events carryin_g the PAPI deletion mutant gene, and 50 independent transgenic events carrying the P APll gene were generated Iast year. A first set of plant tissue was sent to Rutgers this summer for analysis and plants with PAP gene expression were identified based on Western analysis. This year a new version of PAP gene (PAPY123) which include a deletion of3 nucleotides was used to generate another set transgenic plants. Materials and Methods Transformation and Plant Regeneration Mature embryos derived callus of indica varieties CICA 8, Palmar, Cimarrón and Fundarroz PN 1, were used as targets. Palmar and Cimarrón showed high and moderate tolerance to sheath blight, whereas Fundarroz PN 1 and Cica 8 are highly susceptible to sheath blight. The transformation experiments were conducted using Agrobacterium tumefaciens strain Agll (Wang et al. , l997) carrying one of the following plasmids, NT305, NT306 and NT446. Maize ubiquitin or 35SCaMV promoter drives these plasmids, which carry various mutant versions ofthe PAP gene (PAPI, PAPII and PAPY123). The hygromycin resistance conferred by the hpt-cat intron gene 216 was used as the selective marker. Plants were regenerated after stepwise selection on 30 mgll and 50 mg/1, followed by other 30 mg/1 selection throughout plant differentiation. Plants were grown to maturity in biosafety glasshouses. Molecular Analysis ofthe Transgenic Rice Plants Southern Blot and PCR analyses were used to detect the presence of the PAP and hygromycin genes. For this purpose, 15 ~g of DNA were digested with different enzymes. The gels were denatured and neutralized by standard procedures. The DNA was transferred to nylon membranes (Hybond-N, Amersham). The filters were hybridized at 60°C. The presence ofthe kanamycin gen was determined throughout PCR. Results and Discussion A total of 59 transgenic plants carrying the PAPY123 gene were generated and confirmed by Southern blot {Table 1 ). More than 92% of the regenerated plants had integrated the PAPY 123 gene in the genome as determined by Southern blot (Table 1). A similar total number of plants were generated for each of the P AP genes. This year to confirm the integration of the P AP gene, 63 plants of the varieties Palmar and Cica 8 transformed with the plasmids NT 446 were evaluated by Southern Blot and PCR methods. The genomic DNA was digested with different enzymes ( Bam HI!EcoRI and Hind III) which excises the P AP gene. The results indicate the integration of at least one copy of the P AP gene in the rice genome. Moreover, a better analysis on the patterns of integration was obtained with the enzymes BamHIIEcoRI that split by half the PAPY123 gene into two fragrnents (Figure 1). This study showed that 67% of the plants analyzed revealed only one copy of the gene without rearrangements (Figure 1 ). The PCR analysis of sorne plants confrrmed the presence of the PAPY 123 gen. At present, the integration of the P APY 123 of varieties Palmar y Cica 8 varieties has been confirmed. The Southern blot analysis also revealed the presence of the hpt and nptll genes. Western analysis indicated that about 50% ofthe plants analyzed from PAPY123 showed gene expression of P AP protein, whereas 18% of the p\ants tested are expressing either P API or PAPII genes. PAP expressing plants will be evaluated for sheath blight resistance under greenhouse conditions, while detailed molecular analyses will be conducted to determine the number of gene copy. Tabla l. Transformation efficiency of tbree indica varieties using tbe AgU (NT446) Genotype Plasmid Callus Plants RE% Plants Plants % Analyzed by s· Plants Southem S+ PALMAR. NT446 182 48 26.4 36 34 94.4 CICA-8· NT446 220 34 15.4 27 25 92.5 CIMARRON .. NT446 16 5 31.2 NO ND ND • Five replicates. •• One replicate. RE= plant regeneration efficiency. s• = Southem positive. TE= transfonnation efficiency 217 % TE 25 14.2 ND Palmar NT 446 Palmar NT 446 19 20 21 22 23 24 25 26 27 28 29 lO 32 J3 34 H 36 J1 38 NT p .5 kb +-- Figure l. Southern Blot Analysi$ of Genomic DNA of T0 plants Transformed with PAP Y123 Gene. DNA was Cut with Bam HI and Eco RJ. T =Control non Transgenic. P = Plasmid NT446 carrying PAPY123 2.1.7 Development of genetic transformation of Brachiaria mediated by Agrobacterium tumesfaciens C.P. Flores, Z. Lentini SB-2 Project Introduction Brachiaria grasses are the most widely grown pastures in subhumid and humid tropics. B. decumbens cv Basilisk (signalgrass) is important because of its high productivity under intensive use and its tolerance of low soil fertility and relative freedom from pests and diseases, apart from spittlebugs. Forage plant breeding has been largely based on phenotypic selection following sexual recombination of natural variation found between and within ecotypes. Advances in plant genetic manipulation over the last 15 years have provided convincing evidence that these powerful technologies can complement and enhance plant breeding programs. Molecular breeding based on transgenesis to overcome limitations in forage quality may be targeted to the individual subcharacters involved: dry matter digestibility, water-soluble carbohydrate content, secondary metabolites, alkaloids, etc. These molecular breeding approaches may include modification of lignin profile to increase dry matter digestibility and genetic manipulation offructan metabolism to increase non-structural carbohydrate content (Spangenberg et al 200 1 ). Most quality or anti-quality parameters are associated with specific metabolic pathways or production of specific proteins. This allows target enzymes or suitable foreign 218 proteins to be identified, corresponding genes isolated and their expression manipulated in transgenic forage plants. A protocol for genetic transfoimation of Brachiaria will be particularly useful to further improve the quality traits associated with the nutritional value of the pasture. Earlier work at CIA T's included the establishment of tissue culture methods for plant regeneration (Lenis, 1992), and genetic transformation mediated biolistic (Galindo, 1997) of Brachiaria species. Last year we reported the optimization of RITA system for an automated mass production of embryogenic calli of Brachiaria and the effects of medium composition on calli induction and plant regeneration in the same species. This year we report the progress made in the establishment of a protocol for Agrobacterium-mediated transformation of Brachiaria species and how the transformation is highly influenced by genotype. Materials and Methods Evaluation of different accessions of Brachiaria spp for in vitro tissue culture Genetic transformation mediated by Agrobacterium is highly influenced by the genotype. With the aim to increase the number transformation events, two accessíons of B. decumbens (CIA T Nos. 606 and 16497), four accessions of B. brizantha (CIA T Nos. 16316, 25665, 6387 and 16467) and three hybrid materials (CIAT Nos. 36060, 36061 and 36062) wei-e evaluated. Induction of ernbryogeneic callus from mature zygotic embryos of each of these genotypes was assayed. Gene tic Transforma/ion of B. decumbens mediated A. tumefaciens Scutellum-derived embryogenic. callus of B. decumbens (CIA T Nos. 606 and 16497) and B. brizantha (CIA T 25665) were used as target explants. Two A. tumefaciens hypervirulent strains, AGL-1 and C58C1, both carrying the binary vector pCAMBIA 1305.2 (1 1,921 Kbp), kindly provided by Dr. Richard Jefferson, Australia were tested. The pCAMBIA1305.2 contains the hygromycin resistance (hpt) and GUS-Plus-intron genes. GUS-Plus is a new reporter gene isolated form Staphylococcus sp with sorne superior properties to E. coli GUS gene. The GUS-Plus gene contains the intron from the castor bean catalase gene to ensure detection of plant-specific-GUS expression and the glycine-rich protein signa! peptide sequence. The protocol used was basically the same as reported by Florez and Lentini, 2000. An explant was considered GUS-positive when at least one blue spot was observed. Results and Discussion Induction of embryogenic callus was highly dependant on the genotype (Table 1 ). Of the four B. brizantha accesions evaluated, accessions No. 16316 and No. 25665 showed the highest leve) of callus induction. B. decumbens accession 16497 showed the highest response of all the genotypes evaluated (Table 1). Table l. Embryogenic callus induction on different Brachiaria species and accessions Specie B. brizantha B. decumbens Hybrid Hybrid CIA T accession 6387 16316 25665 16467 606 16497 36060 36061 No. of isolated embryos 60 60 45 45 75 66 23 33 219 No. embryogenic caJius o 30 21 o 40 50 9 3 % CaJius induction o 50 47 o 53 76 39 9 Preliminary results indicate a higher response to Agrobacterium infection when using B. decumbens CIA T 16497. This accession showed an average of 26 % transient gus expression. Not difference were noted between the strains used (Table 2). Table 2. Gus transient expression of embryogenic callus co-cultivated witlt Agrobacterium tumesfaciens strains AGL-1 or C58Cl Specie CIATNo. Strain Callus GUS + % assa~ed GUS + B. decumbens 16497 C58Cl 59 17 28.8 16497 AGL-1 77 18 23.4 606 C58Cl 172 4 2.3 606 AGL-1 55 o 0.0 B. brizantha 25665 C58Cl 41 3 7.3 25665 AGL-1 37 3 8.1 Future Plans • To use B. decumbens CIAT 16497 as target for Agrobacterium infection • To increase bacteria activity through utilization of new protocols for higher expression leve! • To conduct assays until regeneration of putative transgenic plants is achieved References Florez C. and Lentini Z. 2000. Brachiaria genetic transforrnation mediated by Agrobacterium tumefaciens. SB2 Annual Report 1999 Galindo, L. 1997. Transformación genética de la gramínea forrajera Brachiaria spp, mediante la técnica de Bombardeo de partículas. Tesis Ingeniería Agronómica, Universidad Nacional de Colombia, Sede Palmira. 130 p. Lenis, S. 1992. Regeneración de plantas de la gramínea forrajera Brachiaria spp a partir de tejidos cultivados in vitro. Tesis Bioquímica, Universidad Santiago de Cali, Facultad de Educación, Departamento de Biología. Santiago de Cali. 80 p. Spangenberg, G., Kalla, R., Lidgett, A., Sawbridge, T., Ong, E.K. and John, U. 200 l. Transgenesis and genomics in molecular offorage plants. Proceedings ofthe XIX Intemational Grassland Congress 200 l . 615-625. 2.1.8 Isolation of lignin biosynthetic genes from Brachiaria decumbens C.P. Florez 1"2, Z. Lentini1, G . Spangenberg2 1SB-2 Project; 2La Trobe University, Australia Introd uction Lignin biosynthesis occurs through a series of reactions involving (1) the shikimate pathway which provides phenylalanine as a substrate, (2) the phenylpropanoíd pathway which results in severa! cinnamoyl CoAs that actas precursors for a wide array of phenolic compounds, and (3) 220 the monolignol pathway which converts cinnamoyl-CoA moieties into monolignols and lignin. All enzymes in the phenylpropanoid and monolignol pathways, except one, have been cloned, usually from multiple plant species. Lignin concentration and composition both control herbage digestibility. Therefore, transgenic technology can be used to increase digestibility of forage crops by down regulation on enzymes in the phenylpropanoid or monolignol pathways. Except for a few highly unusual transgenics, the biggest difference between plant transformation and natural variation may be that the novel-lignin phenotypes occur at higher frequency within transgenic lines, making them easier to identify than novel-lignin phenotypes that occur relatively infrequently in natural populations (Casler and Kaeppler, 2001). Relative small changes in quality of forage crops can lead to large changes in animal performance (Casler and Kaeppler, 2001). For these reason, CIAT jointly with the Plant Biotechnology Centre in Victoria, Australia are working to isolate genes from B. decumbens that are involved in lignin biosynthesís using Lolium perenne OMF-1, 4CL-2, CCR-1 and CAD-1 cONA clones as probes. OMF-1, 4CL-2, CCR-1 and CAD-1 are the key gene in the lignin biosynthetic pathway. Materials and Methods Northern Blot Hybridisation Total RNA from roots and shoots of ten days old L. perenne (ryegrass) and B. decumbens seedlings was isolated using the protocol described by Chang et al 1993. Fifteen to twenty ¡..tg of total RNA per lane was run. Ryegrass cONA clones OMF-1, 4CL-2, CCR-1 and CAD-1 were used as probes. cDNA Library Construction in ZA.P Total RNA was cleaned using the Quiagen RNEasy kit. To yield mRNA from total RNA, Oligotex mRNA kit (Qiagen Company) was used. The Stratagene cDNA Synthesis kit was used for the construction of directionally cloned cONA library. For the packing of the library, Stratagene GigaPack Gold III was chosen. For the screening of the library ryegrass cONA clones OMF-1, 4CL-2, CCR-1 and CAD-1 are being used as probes. Results The Northem hybridisation analysis was made to determine the homology between L. perenne and B. decumbens lignin biosynthetic genes. The results showed that OMF-1, 4CL-2 and CCR-1 transcripts are present in young roots and shoots of B. decumbens (Figure 1 ). These transcripts accumulate to higher levels in the roots than in the shoots. The presence of endogenous OMF, 4CL and CCR genes in B. decumbens is confirmed. Future Plans To complete cONA library screening for key lignin biosynthesic genes of B. decumbens To isolate and characterize key lignin biosynthesic genes isolated from Brachiaria library To produce transgenic germplasm with manipulated lignin metabolism To obtain transgenic cultivars with enhanced forage quality Reference Casler, M.D. and Kaeppler, H.F. 200 l . Molecular breeding for herbage quality in forage crops. In Spangenberg G. (Ed). Developments in Plant Breeding- Molecular Breeding of Forage Crops, 176-188. 221 1 2 3 4 1 2 3 4 2.4- 2.4- 1.4- 1.4- OMT CCR 1 2 3 4 2.4- ~· ~~-~!,;! )" -· 1.4- i'·~~ ... ' ......... 4CL Figure l. Nortbern bybridisation analysis. 1 Shoots of B. decumbens. 2 Roots of B. decumbens. 3 Shoots of L. perenne. 4 Roots of L perenne. 2.1.9 Genetic transformation of tomato variety UNAP AL Arreboles for resistance to udworm (Tuta absoluta) H. Ramírez2, L.F. Fory1, L. Mancilla1, Z. Lentini1 1SB-2 Project; 2Universidad Nacional de Palmira Introduction Tomato (Lycopersicon esculentum Mili) is one of the most important crops in the fresh vegetable market as well as in the food processing industry (Rick and Yoder, 1988). Tomato is the major consumed vegetable crop in Colombia, with a planted area of 15,000 hectares yieldmg 450.000 tons per year (UNAL, 1997). In Colombia, this crop is highly affected by several pests and diseases, and abiotic stresses such as drought, high and low temperatures, and salinity. Since 1985, the vegetable breeding program at the Universidad Nacional de Colombia, Palmira Campus, has as main objective the development of varieties with resistance or tolerance to sorne ofthese traits. In 1997, this program released the tomato variety UNAPAL Arreboles, which has several traits attractive to tomato growers such as fruit firmness and good adaptability specially to the Valle del Cauca region. But this variety is susceptible to one of the major limitations to tomato production in this region: the budworm (Tuta absoluta), which affects the tomato buds and young leaves. It had been difficult to breed tomato resistant to this pest by standard breeding. The only sources of resistance genes is from wild tomato species which are incompatible with the cultivated tomato, and so far the attempts for an inter-specific breeding program has not been successful (Lourencao et al., 1985). The main objective of this work is to transform the tomato variety UNAPAL-Arreboles with the Bt gene cryiA(b), which had been used successfully to obtain resistance against Lepidoptera pests in various economical important crops (i.e. maize, cotton). 222 In previous reports it was described the evaluation of three protocols commonly useci for tomato callus induction and plant regeneration (Fillatti et al., 1987, Narvaez, 1993, and Ultzen et al ., 1995). Results indicated that the highest response for callus induction and plant regeneration is noted on M3 medium sequence (Ultzen et al ., 1995). An increase in response of about 2-fold and 4-fold on callus induction and plant regeneration was noted on M3 media respect to the other media tested. The lowest response was obtained on M2 medium (Fillatti et al., 1987). This year it is reported the progress made on developing UNAP AL-Arreboles transgenic tomato containing the cryiA(b) gene. Materials and Metbods Two Agrobacterium mediated transformation protocols commonly used for tomato (McCormick et al., 1986; Fillatti et al., 1987), were tested using the tomato variety UNAPAL-Arreboles. Agrobacterium strains C58C 1, Agl1 and LBA4404 containing the pBIGCry construct (L.I. Mancilla at CIAT) were used. This gene construct contains the crylAb gene driven by the 35S CaMV promoter, the nptll gene for kanamycin resistance as selection markers, and the gus-intron as a reporter gene. Transgenic plants were identified by Southern blot analysis. Inheritance of gus expression and kanamyzin resistance was evaluated frorn TO to Tl generation. C1onallys propagated plants ofthe original To plants were evaluated fro agronomic traits in the greenhouse. Results and Discusion Preliminary results indicated that the highest leve! of gus transient expression was attained with the Agrobacterium strain LBA4404. A total of four hundred tomato explants ( cotyledonary lea ves of7-10day-o1d plantlets) were infected with LBA4404/pBIGCry weekly. After co-cultivation for 48 hour, about 10% of the exp1ants were analyzed for gus transient expression. The rest of the explants from cultures showing transient expression were transferred to selection media containing kanamycin. After three weeks on selection media, regenerated plantlets were recovered. The number of regenerated plants recovered from kanarnycin containing medium varied among the different experiments. From zero to ten plants were recovered per experiment. A total of 51 putative transgenic plants were produced from 8 experiments ( 400 explants by experiment). This shows an efficiency of 1.6 % for recovering kanarnycin resistant plants from the initial agro-infected explant. Of these plants 15 were transferred successfully to the greenhouse and 6 plants had shown stable gus expression throughout the vegetative and reproductive life cycle. This year it was carried out the morphological and molecular characterization of kanarnycin- resistant and gus-expression clones P28, P33 and P47 at the TO and TI generations. Self progeny derived from TO plants (TI clones) was evaluated for agronomic traits and resistance to kanamycin and Gus expression. No differences were noted neither between the TO and TI plants, nor between the transgenic plants and the tomato control plant for the various morphological traits evaluated (plant height, node lentgth, leaf type, presence of pubescence in stem, tlower color and shape, fruit color and shape). These results indicated that no somaclonal variation is apparent in the transgenic plants. Preliminary results on the molecular characterization of clones P28,P33 and P4 7 by PCR and Southern analyses suggested that both genes are inserted in the geno me of these plants. Molecular analysis for the patterns of insertions of the crylA(b) in these plants is in progress. For the three clones evaluated, a 3:1 segregation ratio for gus expression and kanamycin resistance was noted. Indicating the insertion of an active locus for each of these genes. Variation was noted on the co-segregation ofthe two genes. Clones P28 and P33 showed a 223 95% and 82% co-segregation ofboth genes from TO to TI , whereas P47 showed a cosegregation of 63%. Currently, inmunological and bioassays studies are being· carried out to detect the CrylAb gene expression on the transformed tomato clones and to determine if these plants are insect resistant. References Fillatti, J. J. , Kiser, J. , Rose, R. , and Comai, L. 1987. Efficient transfer ofa glyphosate tolerance gene into tomato using a binary Agrobacterium tumefactions vector. Bioffechnology 5:726-730. Lourencao, A. L., H. Nagai, W. J. Siqueira and M. l. S. Fonseca. 1985. Selecao de linhages de tomateiro resistentes a S. absoluta. Hort. Bras. 3: 57-59. McCormick, S., Niedmeyer, J., Fry, J., Bamason, A., Horsch, K. and Fraley, R. 1986. Leaf disc transformation of cultivated tomato (L. esculentum) using Agrobacterium tumesfaciens. Plant Cell Reports 5: 81-84. Narváez-Vásquez, J. 1991. Expression ofproteinase inhibitor genes in transgenic plants: Effect on insect resistance, levels of accumulation in four plant species, and celluJar compartmentalization. Ph.D. thesis. Washington State University, Pullman, USA. Rick, C. M. and Yoder, J. l. 1988. Classical and molecular genetics of tomato: Highlights and perspectives. Ann. Rev. Genet. 22: 2821 300. SAS Institute, 1988. SAS User's guide. SAS lnstitute, Cary, N.C., USA. Ultzen, T., Gielen, F., Venema, A., Westerbrock, P., Haan., Mei-lei Tan, A., Schram, M., van Grinsven, and Goldbach, R. 1995. Resistance to tomato spotted wilt virus in transgenic tomato plants. Euphytica. 85: 159-168. Universidad Nacional de Colombia-Sede Palmira. 1997. Obtención de un nuevo cultivar de tomate chonto Lycopersicon esculentum Mili . Memoria técnica No. 3. 2.1.10 Resistance to sugar cane yellow leafvirus (ScYLV): Genetic transformation an alternative aiding breeding of sugar cane M.P. Rangel 1•2, E. Tabares1, F. Angei 2;J. Victoria 2, H. Guerrero2, E. Mirkov3 , z. Le~tini 1 1SB-2 Project; 2Cenicaña; 3Texas A&M Introduction Wide spread of yellow leaf syndrome disease caused by Se YL V was introduced in Colombia in 1998 through the Brazilian variety SP 71-6163 (Victoria et al., 1999). The main so urce of the disease is the use of vegetative seeds from clona! propagated infected plants, and transmission by the aphid vector Melanaphis sacchari widely present in the sugar cane region ofthe country. The disease is associated with reduction in sucrose content and crop yield, although plants may be symptom-less. In Brazil, the variety SP 71-6163 infected with this virus showed yield losses of 60% to 80%. The molecular characterization and cloning of ScYL V was performed at Texas A&M (Mirkov et al., personal communication). Severa! constructs were generated, and one gene version containing part of the coat protein gene encoded high levels of protection in transformed sugar cane USA varieties. The main objective of this project is to transform sugar cane with the 224 truncated version of Se YL V coat pro te in gene vi a biolistics, since transgenic sugar cane has already being generated in other laboratories using this gene transfer method. Materials and Metbods Genetic transformation of sugar can e was conducted using the PDS-1 000 He particle accelerator device. A construct (pSc YL V) containing the truncated version of the Se YL V coat protein gene driven by the ubiquitin promoter, and the nptli gene for genetycin resistance will be used. The protocol optimized at CIA T for rice transformation was tested and modifications introduced accordingly. In this regard, preliminary experiments were conducted using plasmids pGV1040, pCAMBIA 1201, and pCAMBIA 1301 all carrying the gus gene driven by the 35S CaMV promoter, to optimize bombardment conditions evaluating GUS transient expression. Simultaneously, callus induction and plant regeneration of the varieties CC 87-434, CC 85-63, ce 85-92, ce 85-96 and ce 84-75 was tested to select the most responsive genotype and use as target. Callus was induced in the dark from apical meristems of in vitro plantlets cultured on MS salts supplemented with 3 mg/1 2,4-D, l 00 ~g/1 inositol, 18% coconut water, 3% sucrose, and 0.2% gelrite. Last year results indicated that the routine protocol used for clona) propagation of sugar in CENICAÑA although efficient for the generation of plants, it was not appropriate for the maintenance of embryogenic callus needed for bombardment since plant differentiation occurs as soon as callus induction is obtained. A modification of the standard tissue protocol was introduced by culturing the induced callus under dim light rather than direct light. This change in light culture conditions restrained callus differentiation by 85%, allowing the maintenance of embryogenic callus at a optimal for bombardment. Callus cultured under dim light conditions, showed about 80% of plant regeneration when transferred to direct light. These results suggest that changes of light conditions at early stages of callus induction, does not have deleterious effects on the plant differentiation process. Results and Discussion Last year results indicated that 51.3 % ± 27.8% of bombarded tissue showed gus transient expression 48 hr after bombardment. After a complete stepwise selection at genetycin 30 mg/1 followed by genetycin 50 mg/1, a total of 68 petri plates with 20 to 25 explants each had been evaluated so far. About 50% ±_29% ofthe bombarded callus were resistant to genetycin 50 mg/1. Of these resistant callus, about 4% ± 5% regenerated plants. Seventy regenerated plants were recovered after selection in genetycin 50 mg/1. Eleven plants have been tested by Southern blot so far. Six out of eleven tested plants were positive showing a 0.45 kb signa) after digestion with Hind III indicating the excision of the complete truncated version of the Se YL V coat protein gene. Plants will be tested after digestion with Pst 1 and BamHI in order to determine number of copies and rearrangments present in transformed plants. References Victoria, J. 1.; Guzmán, M. L.; Cuervo, E.; Lockhart, B. Síndrome de la hoja amarilla en Colombia. Serie divulgativa W 07. CENICAÑA. 1999. Villalobos, V. M. Y Thorpe, T. A. Micropropagación: conceptos, metodología y resultados. En: Roca, W. M. Y Mroginski, L. (Eds.) Cultivo de tejidos en la agricultura: Fundamentos y aplicaciones. CIAT. 1991. P. 127- 141. 225 2.1.11 Expression of recombinant CRYI(A)b Protein in E. coli L.I.Mancilla1, C. Ramírez, C.J. Herrera, P. Chavarriaga1, J. Tohme1 and W.M. Roca 1•2 1SB-2 Project; 2CIP-Peru Introduction Bt genes of the cry family ha ve been used to confer resistan ce to insects in different plants species (cotton, maize, canola, soybean). The toxic effect of the protein CRYI(A)b has been previously reported for severa! lepidopterans. To evaluate the toxic effect of this recombinant protein on cassava stem borer (Chilomima clarket) larvae, the cry I(A)b gene was cJoned into expression vectors and the protein purified. Different artificial diets were evaluated on larvae stages of stem borer from artificial colonies established at Nataima (Tolima) in an attempt to optimize conditions for toxicity tests. Methodology The cryi(A)b gen from PLANTECK was cloned into a series of expression vectors (pGEX-5X-1, pGEX-5X-2 and pGEX-5X-3; Figure l) to restore the open reading frame and to obtain the recombinant protein. The CRY recombinant protein was cloned attached to the 3 '-end of a Gluatahione S-transferase gene ·sequence. The protein expression of severa! clones with these constructions were induced with IPTG in different concentrations, by four or six hours, at 30 oc or 37 °C. Positive clones with recombinant protein were identified by PAGE-SDS 10% (Figure 2). Induced bacteria! cultures were concentrated and CRY protein was purified using Pharmacia chromatography colurnn with Glutathione Sepharose 48 resine. After binding the fusion protein to Glutathion, the chromatography column was washed three times with PBS buffer before elution of proteins with reduced Glutathione 10mM pH 8.0. The concentration and molecular weight of purífied pro te in was evaluated by electrophoresis. Artificial diets were tested with stem borer larvae using different parts of the cassav plant. Flour made of lyophi1ized stems, leaves and rootw were ground and complemented with nutritive components like wheat gerrn, bacto-agar and beer yeast extract (Table 1). Table l. Contents of artificial diets (gil) to rear Cbilomima clarkei. (Reference: Lastra and Gómez, 1988) DIET Carrot Total Cassava Cassava Cassava Compone ni Cassava Xi/e m Root Le aves stem Wheat germ 38.0 Beer yeast extract 40.0 Carrot flour 40.0 0.0 0.0 0.0 0.0 Liophylized tissue 0.0 40.0 40.0 40.0 40.0 Hidrolizated casein 8.0 Powder Sucrose 20.0 Sugarcane leaf flour 80.0 Corn cob flour 11.25 Agar 3.75 Metil paraben 3.5 Stretopmicine 1.5 226 Results and Discussion Recombinant Cry Protein Expression Two of out three constructions with CRY in pGEX-5X vectors have been obtained. Expression of the protein has been detected on PAGE-SDS (Figure 2). pGEX 1 CRY and pGEX2CRY Recombinant CRY protein was detected as two fragments between 60 and 40 Kd. Tac-p Sal 1 Notl 1 Gluthatltione S-tranfrerasa Cry IA(b) pGEXCry (7.72 Kbp) Figurel. Diagram of constructs with cryl(A)b gene in series 1 and 2 of pGEX-5X expression vector. k 205 116- 97.4 66 1 2 3 4 5 6 Cry Figure 2. lnduction of exprrssion of CRY recombinant protein in pGEXICRY and pGEXZCRY clones with 0.2 mM of IPTG by 6 hours at 30°C. l.Molecular marker; 2. pGEX-SX-1 (ntgativt control); J . pGEJ Cry; 4. pGEX2Cry; 5.pGEXICry; 6. pGEX-SX-1. 1 2 3 4 5 Kd 205 116 94.7 66 45 Figure 3. PAGE-SDS with eluted fractions of purification of CRY recombinant protein from pGEXCRY clone. l. Supernatant of bacteriallysate, 2. Elution with wash buffer, 3. Elution with reduced glutathione, 4. Racterial nellet. ~- Molecular marker. 227 Artificial Diets to Rear C. clarkei Three different experiments with 5 repetitions were made with larvae of C. clarkei to evaluate different artificial diets. However, the viability of larval stages of C. clarkei was very low in these assays. C. clarkei larva showed the highest viability in diets with cassava stem flour and carrot flour and survived eighth days after inoculation. These results have to be duplicated. The time of larvae survival wasn't enough to evaluate protein toxicity. Conclusions and undergoing work The CRY lA(b) recombinant protein was obtained from expression vectors pGEX-SX-1 and pGEX-SX-2 and purified by Glutathione Sepharose 4B afinity cromatography. C. clarkei larvae have been difficult to rear on severa! artificial diets. An artificial colony has been established in CORPOICA-Nataima (Tolima). We will continue testing modifications on the diets to rear Chilomima. We will produce enough recombinant protein to test its toxicity on larvae artificially reared, or by sprying cassava cuttings with known concentrations of Cry 1 A(b ). References Glutathione Sepharose 48 Manual of Instructions. Phannacia Biotech (1996). Lastra, L. A. and Gómez, L.A. ( 1988) Colombian Journal of Entomology. (Revista Colombiana de Entomología) 14(2): 9-14. pGEX expression serial vectors (GST Gene fusion) Manual of Instructions. Phannacia ( 1 998). Sambrook, 1., Friisch, E.R. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Edn. Cold Spring Harbor N.Y. Cold Spring Harbor Laboratory. 228 Activity 2.2 Development of cellular and molecular techniques for the transfer of genes for broadening crop genetic base Main Achievements • A novel backcross methodology for producing fertile cornmon x tepary beans hybrids from otherwise incompatible genotypes was developed. • Lines of Friable Embryogenic Callus of cassava commercial cultivars CM3306-4 (lea Negrita, for the Northem Coast) and SM1219-9 (for lnterandean Valleys) were established. • The cryo-preservation protocol was tested with 43,5% of the cassava core collection. 82% of the accessions in the core have >30% recovery rates after freezing. • A cryo-preservation protocol was adjusted for cassava wild relatives. Plants were recovered for M escu/enta subsp.jlabellifolia, M esculenta subsp. peruviana and M carthaginensis. • Friable Embryogeneic Callus from cassava cultivars TMS60444 and MCol2215 were recovered after freezing. • Plants ofthe tropical fruit Tree" tomato were recovered from frozen seeds. • Propagation methods using RITA were validated with 16 cornmercial cassava clones reaching multiplication rates between 1:6 to 1:10. • Cassava plants, produced in vitro by small fanners in Santa Ana (Cauca), were moved to the field for testing in Perico Negro (Cauca). • Four cassava clones, selected for fanners from Santa Ana, have now been included in the propagation scheme. • Simplification of RITA system and construction of cheap transfer hoods to reduce costs of implementing in vitro propagation systems for fanner cornmunities and public schools. • The use of bioreactors from increasing the response to rice anther culture was implemented. • A methodology for the reproducible plant regeneration ofnaranjilla fruit (i.e.lulo) was developed. • Selected soursop (Annona muricata L.) clones propagated in vitro through micrografting, planted three years ago in different locations, showed excelent agronomic behavior regarding general growth, tree architecture, flowering initiation and fruit quality, thus validating this technology as useful for producing planting material. 229 2.2.1 Regeneration of cassava plants from friable embryogenic cal_lus (FEC) by combining conventional solid media and temporary immersion using RITA® D Lopez1, JE Montoya1, R. Escobar1, P Chavarriaga1, J Tohme1 WM Roca 1•2 1SB-2 Project; 2CIP-Peru Introduction Somatic embryogenesis has been the main regeneration method to obtain transgenic plants of cassava. However, the efficiency of embryo-to-plant conversion is still low with the few cultivars that have been transformed. We are investigating the efficiency of plant regeneration from FEC in three cultivars, two of which are of commercial importance in the Northem Coast of Colombia. Combinations of solid and liquid media (using temporary immersion in RITA~) are being tested to improving the rate of embryo-to-plant conversion. Improving this rate will have a positive impact on the efficiency of plant genetic transformation. It will also contribute towards developing synthetic seeds. Methods Three cassava cultivars were used for the experiments: TMS60444, which is a model cultivar for regeneration and transformation, and Mcol2215 (Venezolana) and CM3306-4 (lea Negrita) which are preferred cultivars among small~to-medium scale farrners. Approximately O, 17 g of FEC from each cultivar was explanted for embryo maturation on petri plates containing solid MS2 medium (López 2000) plus either hormone NAA or BAP. From this initial step, differentiating embryos in early torpedo and cotyledonary stages of TMS60444 were tested for further development on treatments A to D (jirst test). Based on the best results, a second test was carried out with TMS60444 plus CM3306-4 and Mcol2215 to check treatment E only (Figure 1). The treatments were as follows: • liquid MS2 medium plus NAA, BAP and GA3 in RITA~, with subsequent transfer to solid 4E propagation medium • solid MS2 medium plus activated charcoal, transfer to solid MS2 plus BAP, and transfer to solid 4E • liquid MS2 plus activated charcoal in RITA~. transfer to solid 4E • solid MS2 plus activated charcoal, transfer to liquid MS2 plus BAP, and transfer to solid 4E • like D, except that the liquid medium contained BAP plus GA3 Results and Discussion For cultivar TMS60444 in the first test, embryos in the torpedo stage were observed eight days after the first maturation treatment on solid medium with NAA. Fourty days later we collected 160 embryos at the cotyledon stage (314 embryos per gram of tissue) from this initial treatment. They were equally split (60 each) among treatments Ato D. Overall, two embryo harvests were made (30 and 90 days) after treatments were initiated, which produced 21 (52.5%), 13 (30.2%), 17 (42.5%) y 23 (53.4%) elongated embryos (plantlets) for each treatment respectively. Based on the quality of embryos, we determined that treatment D was the best from the first test to obtain the largest number of vigorous, elongated embryos. 230 The results reported here for the second test are partial since the experiment is still ongoing. So far the maturation on solid medium with BAP has produced larger amounts of mature embryos for CM3306-4 and Mcol2215, while medium with NAA seemed to be better for TMS60444 (Table 1). Embryos from MCol2215 were few, atypical and vitrified. Those from CM3306-4 were more normal-looking and vigorous (Figure 2). Currently embryos are being transferred to RITA with GA3 as explained in Figure l . GA3 was added to liquid medium to improve shoot elongation of germinating embryos which, in general, from test one, did not seem to have well defined apical shoots. Table l. Total number of mature embryos, from FEC, obtained after treatment with NAA or BAP on solid medium for three cassava cultivars. Cultivar Media Total mature Embryos per gram of embryos FEC CM3306-4 MS-NAA 40 105 MS-BAP 162 450 MS-NAA o -MCOL 2215 MS-BAP 46 85 TMS 60444 MS-NAA 247 1300 MS-BAP 35 66 Conclusions and Ongoing Work Globular embryos from FEC differentiate asynchronously into torpedo and cotyledon-stage embryos at different rates and speeds depending upon genotype and hormones used. BAP seems to be more effective for embryo maturation with CM3306-4 and Mco12215, while NAA produces a larger number of mature embryos in TMS60444. Including RITA - ... ~ L 56 .g-¡¡¡ SM 909-25 60 o -> 8 "' a ., ... L 26 90. c.., SM 1219-9 44 a~ 8 6 • ~~a; wu induced 1>u1 dodn t pn>llterate~; •• FEC wu induced. two hnes eslal llhed. • fEC wu onaucea. one one esubhsloed. 234 References Roca, W.M., Rodriguez, J.A., Mafia, G., Roa, J.C. 1984. Procedures for recovering cassava clones distributed in vitro. CIA T, Cali, Colombia. 8p López O (2000) Inducción de callo embriogénico friable "CEF" y regeneracion de plantas de la variedad de yuca (Manihot esculenta Crantz) MCol 2215. Thesis, Universidad Nacional de Colombia, Palmira, Colombia, 85 pp 2.2.3 Development of a novel backcross methodology for producing fertile common x tepary beans hybrids from otherwise incompatible genotypes Alvaro Mejía Jiménez\ Leonardo Galindo\ Arturo Criollo1, Steve Beebe2, César Cardona2, Joe Tohme1 1 SB-2 Project; 2IP-1 Project Introduction The tepary beans (P. acutifolius A. Gray) possess severa! traits that are important for common bean improvement. Such incompatibility barriers as embryo abortion, hybrid lethality, hybrid weakness and hybrid sterility hinder crossing of common beans with tepary beans. Due to this, the tepary beans have been classified in the tertiary gene pool of the common bean (Debouck, 1991 ), the economically most important species of the genus. Conventional crossing methodologies (such as recurrent backcrosses) have been used to transfer bacteria! blight resistance from tepary to common beans (Mejía-Jiménez et al., 1994; Singh and Muñoz, 1999). This crossing strategy however has not been successful for transferring other important agronomic traits. Congruity backcrosses (CBC; the altemate backcross of the hybrids with genotypes of each species for severa! cycles; Haghighi and Ascher, 1988) promises to produce hybrids with higher genetic recombination and gene introgression rates. However CBC could only be applied to few facilitator genotypes ofboth species (Mejia-Jimenez et al., 1994, and unpublished results). As a result of the necessity to develop fertile common x tepary beans hybrids using the tepary bean genotype NI576 (genotype competent to Agrobacterium mediated genetic transformation, see Mejía-Jiménez et al. in this annual report), which could not be crossed using conventional or congruity backcrosses, a novel backcross methodology called Double Congruity Backcross (DCBC; Mejía-Jiménez et al. in preparation) has been developed. The DCBC methodology has allowed the production of fertile common x tepary beans hybrids involving the above-mentioned genotype. This methodology seems also to be generally applicable for the production of fertile hybrids from otherwise "incompatible" tepary and common beans genotypes. In the year 2001, the DCBC-methodology was applied in crossing the common beans and tepary beans genotypes that have been also difficult to cross using recurrent or congruity backcrosses: G40023 and G40068 (drought), G40199 (bruchids, A. obtectus and Z. subfaciatus) and G40019 and G40036 (leaf hopper, Empoasca kraemeri). 235 Methodology Double congruity backcross hybrids with the cytoplasms of common or tepary bean (DCBCvulg and DCBCacut hybrids) were developed starting from advanced congruity backcross hybrids with cytoplasm of common bean developed previously (Mejía-Jiménez et al., 1994), as described in the annual report 2000 (Mejía Jiménez et al. 2000). Embryo rescue was applied when necessary to recover viable hybrids from aborting embryos (Mejía-Jiménez et al., 1994). Results Genotypes of tepary beans that were hitherto considered incompatible when crossed directly to common bean genotypes or even to advanced CBC hybrids, were crossed easily to advanced DCBC hybrids of the same cytoplasm (DCBCacut hybrids, table 1 ). Several fertile backcross hybrids were obtained from each parental combination. The produced plants were clearly recognizable as hybrids through morphological markers (hypocotyl and flower pigmentation, or primary leave size). - The generated hybrids are currently being used to "carry" the genes from the incompatible genotypes to DCBC hybrids with the cytoplasm of P. vulgaris (DCBCvulg hybrids, table 1). This cross is already generating vigorous hybrids, which are expected to be fertile or cross-fertile. Theoretically 25% of the alleles of the hybrids produced this way will be from the tepary bean genotype that hitherto had been incompatible with common beans when crossed using other methodologies. Table l. Double c:ongruity bac:kc:ross (DCBC) strategy followed to produce fertile of c:ross-fertile hybrids between genotypes of P. acutifolius and P. vulgaris. (1) (2) Advanced hybrids DCBCvulg2 X ..J., (3) [DCBCvulg x (G40023 x DCBCacut)] hybrids (25% of 040023) Fertile or cross-fertile Incompatible P. acutifolius X genotype (i.e. G40023) Advanced hybrids [G40023 x - DCBCacut] hybrids (-50% of the genome of the incompatible genotype) Self-fertile 1 Double congruity backcross hybrid with P. acutifolius cytoplasm 2 Double congruity backcross hybrid with P. vulgaris cytoplasm DCBCacut1 In the case of the genotype 040023, the G40023 x DCBCacut hybrids are being crossed to advanced congruity hybrid lines with common bean cytoplasm that were selected for several cycles as drought tolerant (S. Beebe, personal communication). 236 Conclusions DCBC allows the production of viable and fertile or cross-fertile common x tepary hybrids from genotypes that were before incompatible using other crossing techniques. Future plans To produce fertile hybrid populations with common bean cytoplasm involving each of the selected tepary bean genotypes. To measure the introgression of DNA fragments from the tepary bean genotypes in the fertile hybrid populations produced after DCBC, using AFLP techniques. References Debouck, D. (1991) Systematics and morphology." CIAT. Common beans: research for crop improvement. Ed. A. van Schoonhoven & O. Voysest, Cali, Colombia (1991): 55-117. Haghighi, KR. and PD. Ascher. Fertile, intermediate hybrids between Phaseo/us vulgaris and P. acutifo/ius from congruity backcrossing. Sexual Plant Reproduction 1 (1988): 51-58. Mejfa-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-CIA T Mejfa-Jiménez, A., Mui'loz, 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 Common Bean Improvement. Crop Science 39 (1999): 80-89. 2.2.4 Farmer's cassava seed production using in vitro low cost system R. Escobar1, C. Hemander, J. Restrepo3, G. Ospina3, J. Tohme1 and W.M Roca14 1 SB 1 Project, 2Representative of Farmers community, 3:FIDAR, 4:CIP-Peru Introduction The use of cuttings of unknown phyto-sanitary conditions for cassava propagation reduces yield, farmer's income and seed quality and quantity for next crop cycle. An interdisciplinary group constituted by farmers, biotechnology researchers, and one NGO, sponsored by PRGA, modified and simplified in vitro propagation methods. Using a participatory scheme it was possible to establish a simple, low-cost cassava propagation system to produce certified, healthy and enough planting material (CIA T, BRU Annual Report 2000). Materials and methods A simple in vitro propagation system that minimized externa} inputs was implemented in a pilot experimental station located in Santa Ana, Cauca (Colombia). A representative person of the 237 community (Mr. C. Hernandez), selected by the community itself, trained a group of eleven women in tissue culture. A propagation scheme was set up with MCol 1522 (Algodona), a traditional clone in the zone of which farmers say "it's very good but it gets sick easily". Results and discussions One hundred and thirty one different media have been tested during the last two years of the project. We selected three media with propagation rates (1:3-4) (BRU, Annual Report 2000) similar to that of conventional 4E medium (Roca 1984). Other varieties were incorporated into the propagation system (CM 523-7, MBra 383, CM 6740- 7). We met with farmers to move in vitro plants to the field. Four blocks with 20 plants each were planted in Perico Negro, Puerto Tejada (Cauca, Colombia). One ofthe expected outcomes ofthis exercise is to help determine differences in root production between in vitro- and cutting-derived planting material. Furthermore, these plots could serve as the initial phase of a farmer's nursery system. As an extra income source, we provided them with horticulture seeds to plant in the communal garden and in individual farms (seeds were donated by the Universidad Nacional de Colombia- Palmira). Due to a long dry season in Cauca, and to contamination associated with animal feeding areas (swine and poultry) located too close to rural tissue culture labs, it was necessary to move plants back to CIA T's station and make adjustments to the rural plastic cabinets. Mr. Hernandez initiated a propagation se heme in CIA T to recover and in crease the number of in vitro plants before returning them to Santa Ana. We are currently testing options to replace MS medium. Explants (shoots and nodes) grown on media with Agrimins® show good response (root induction and growth). Tissue grown on media with Urea alone (46:0:0) died. A complete fertilizer (Coljap® desarrollo) induced callus proliferation with no root induction. We built a prototype of a low-cost, sterile transfer cabinet, that reduced the purchasing price 12 times, from 3,000 to 250 US dollars/chamber. CIA T maintains a pilot experimental area in its headquarters to optimize the system and in elude other crops to establish a farmer's seed platform. Multi-crop seed production systems could help farmers to establish a rural seed enterprise. 238 Figure la: Farmers meeting to transplant in vitro plants to the field. (lb) Farmer using low-cost chamber la lb Conclusions Transplanting in vitro plants to the field was a success (99% recovery); it was the first time women from the farmers' group used in vitro-generated planting material. Low-cost transfer cabinets reduce costs, maintaining an efficient propagation activity. We replace all inputs for the conventional in vitro medium. This is an advantage if others farmer groups become interested in propagation activities. Future activities A propagation scheme needs to be established to integrate low-cost components with rapid propagation systems. This will help to initiate a cassava farm nursery system. A cost analysis is critica! for the process. Include other crops in the propagation scheme. A second phase to consolidate the propagation plan will be critica! for the process. References Roca 1984 CIA T Annual report 2000 2.2.5 Estimation of the costs involved in a rapid multiplication scheme based on the use of micro-stakes H. Ceballos 1"2 1IP-3 Project; 2SB-2 Project Introduction The low multiplication rate of cassava has been listed as one of the most important constraints for cassava development. Severa! techniques have been developed for overcoming this problem, 239 including fas ter tissue culture or the use of micro-stakes alternatives. Although the protocols ha ve been worked out, seldom these alternative multiplication rates have been applied at large scale. Therefore there is a need to scale up the process both to evaluate its viability when large volume of vegetative material is handled, and to estímate the costs of production. This information can be used by the private sector if the interest eventually develops for the creation· of commercial seed production endeavors. Specific objectives • To set up a facility for a large production ofvegetative material through the use ofmicro- stakes. • To complete the system with an anti-white flies screenhouse. • To produce clean vegetative propagules of elite clones. • Estímate the costs of production. Materials and methods Severa! key elite clones were selected for multiplication using this system. Most of them are adapted to the mid-altitude valleys, where the common presence of white flies has resulted in a higher incidence of frog skin disease. The process starts with the indexation of "mother plants" to make sure they are free of frog skin and various virus that affect cassava.. The long stems obtained from these disease-free plants were cut into micro-stakes, about 5-cm long and with two buds, which were then put to sprout in moist chambers. Every two weeks the resulting shoots are "harvested" and put to produce roots in glass containers with plain water. Once they produce roots the plantlets are transferred to plastic bags with soil and eventually moved to the screenhouse. for hardening and further grow (about 1-2 months) before being transplanted to the field. Results The process has been used to multiply SM 909-25 (1355), SM 1219-9 (1213), SM 1460-1 (450), SM 1741-1 (272), CM 7514-7 (975), CM 7951-5 (490), MBRA 383 (998). The number ofplants produced for each clone and already transplanted to the field is mentioned within parenthesis. These plants are certified to be disease free and constitute an important set of elite clones for this environment. lt should be pointed out that the multiplication process begun while the indexation took place (frog skin disease indexation requires a grafting with the highly susceptible 'secondina' clone, which may take up to three months for results). As the results from the indexation carne, severa! plants were found to be contaminated and, therefore, the plantlets obtained from them had to be discarded. Otherwise, the systern proved to be operational and able to produce consistently large number of vegetative material with few problems to be solved now and then. The system implemented allowed two people to produce about 10,000 plantlets in a period of three months. Costs have been estimated for a period of continuous production for two years (8 batches of 10,000 plants each). The total for infrastructure costs (mainly the rnoist chambers) was therefore divided by eight in the estímate of costs presented in Table 6. The cost of the special screen house was not included in this cost estimates because it is on1y necessary in areas where white flies are an important vector for cassava diseases. 240 Table 6. Distribution of costs (US dollars,) for the production of a batch of 10,000 plantlets using the micro-stakes multiplication system. ITEM COST Agrochemicals, plastic bags, razor blades, detergent, fertilizers, etc. 253.59 lnfrastructure (distributed in 8 batches) 112.28 Labor (two people) 909.09 TOTAL 1274.97 1 1 US$ -- .200 Colomb1an pesos An important connection to of this activity for the SB2 project is that the later stages of the process described above apply equally to plantlets developed from tissue culture techniques. Particularly relevant in this case is the application that rapid temporary immersion systems could have in commercial multiplication of cassava propagules .. 2.2.6 Rapid propagation of planting material by temporary immersion bioreactors. 1R. H. Escobar, 1L. Muñoz, 1J. Tohme and 1"2W. Roca 1SB-2 Project, 2CIP-Peru Introduction Cassava has been considered the most important altemative crop since it is the only crop that yields acceptably under marginal conditions, with mínimum inputs. Demand for cassava planting material is high and conventional methods of propagation don 't satisfy the needs (Buitrago 1999). We are adapting RITA~ to scale-up and produce enough cassava planting material of desired, indexed, commercial clones, using nodes as initial explants (Annual Report 2000). Methodology Different immersion periods and growth regulators were tested. TDZ at low concentrations, combined with short immersion periods, resulted in better propagation rates than those obtained with BAP. The method is being adjusted with 16 commercial clones. In the last month we initiated RITA with clones MPer 183, HMC 1, CM3306-19 and MCol 1522. Results We increased propagation rates up to 1 :6 to 1: 1 O (Table 1 ), depending u pon the genotype, which was higher ifcompared with rates (1:3-4) ofnormal propagation on 4E solid media (Roca 1984). Plants produced with this system were transferred to the screenhouse and compared with plants produced on solid media. No morphological differences were observed. Carbon dioxide was injected into RITA. Partial results indicate bud activation but not elongation. More time may be required for shoot elongation. We are adapting bioreactors at low cost. RITA vessels are being replaced with recyclable soda glasses, and silicone tubing by aquarium flexible pipes. Hydrophobic, 0,22J.i.m filters were 241 substituted by hand-made, cotton filters. We are setting up plans to run a low-cost, pilot bioreactor system, that can be transferred to small-scale farrner systems. Table 1: Propagation rates of commercial cassava clones using RITA system. Recovered tissue Cassava clone Common name Shoot Nodes Propagation rate CM 3306-4 lea Negrita 50 51 10.1 CM 4574-7 40 25 6.5 CM 523-7 Catumare 46 60 10.6 MBra 383 Brasilei'ia 43 25 6.8 MBra 507 Tu cuma 43 35 7.8 MCol2215 Venezolana 58 30 8.8 MColl505 Verdecita o Pl2 28 25 5.3 Mcub 74 Sei'iorita 31 42 7.3 MEcu 72 Injerta 26 42 6.8 Mtai 8 Rayon 60 26 35 6.1 SGB 765-2 Caribei'ia 37 20 5.7 SBG 765-4 Ro jita 27 32 5.9 CM 6740-7 Reina 32 18 5.0 CM 3555-6 46 40 8.6 CG 1141-1 lea Costeña 48 40 8.8 Mven 25 QuereEa amarga 40 40 8.0 A member of the PBA regional committee (Mr. Eduardo Erazo) discussed with us the importance of including in our propagation schemes sorne regional varieties. Farrners from the Atlantic Coast gave us stakes of local clones "Ramirana", "Yema de huevo" and "Por encima". A GRU staff isolated meristems and treated them with thermotherapy. We are now running RITAs with these materials. We are also adapting RITAs to improve somatic embryo induction in Brachiaria and rice, and to improve rice anther culture. We are also using RITA for FEC regeneration to support cassava transformation activities (see transformation reports in this issue). Conclusions Propagation rates were improved form 1:3 in conventional so lid mediato 1:6-11 in RITA system. Propagation rate is clone dependent. RITA could be adapted to reduce implementation costs. RITA could improve somatic embryogenesis and plant regeneration in other crops. Future activities Test RITA with local varieties (Ramirana, Yema de Huevo and Por Encima) Transplant plants (80-1 00 each) of 1 O commercial clones propagated with RITA and compare to conventional stakes. A meeting with farrners, national programs and PBA-regional committee to exchange experiences. Adjust low cost systems. 242 References Annual Report 2000. Project SB-02 pp 185-187 Buitrago J. 1999. Ingenio yuquero. Estudio de prefactibilidad. FDI, CIA T and FENA VI. Cali- Colombia Escobar R.H., Muñoz L, Tohme J y W.M. Roca. Estado actual de la micropropagacion de la yuca. Seminario Internacional Programa Colombiano de Biotecnologfa Agrícola, Cartagena, Colombia, Febrero 21-23,2001 Escobar R.H., Muñoz L y W.M. Roca. Cassava micropropagation for rapid seed production using temporary inmersion bioreactors. Poster at the symposium "From germplasm bank to farmer fie1ds, role of CIA T biotechnology in research and training in Latín America", December 4 of 2000, CIA T, Cali, Colombia. Roca, W. M. (1984) Cassava. In: Handbook ofplant cell culture, Vol. 2. Crop species. (Sharp, W.R.Evans, D.A.; Ammirato, P.V.;and Yamada, Y.) p(269-30l) MacMillan Publishing Co., New York. 2.2.7 Cryopreservation of cassava shoot tips using the encapsulation - dehydration tec~nique. R.H. Escobar1· \ N.C. Manrique2, D.G. Debouck1, J. Tohme2, and W.M. Roca2-3 1SB1Project, 2SB2 Project, 3CIP-Peru Introduction The encapsulation-dehydration technique allows the direct placing of cassava meristematic tissues into liquid nitrogen, avoiding thus the use of expensive equipment and opening the possibility for large-scale, long-term conservation at low cost (Escobar et al2000). We are testing reproducibility between replications and conservation time in liquid nitrogen as steps to determine logistical aspects in the management of an in vitro base gene bank. Methodology Methods have been previously described in Annual Report 2000, and Manrique 2000, and no modifications have been made since then. Results We have currently tested 280 (43,5%) genotypes of the entire core collection (640 total accessions). With data obtain this year (from 127 genotypes) we could establish three groups based on percentage of shoot recovery after freezing: a high-recovery group of 39 genotypes (up to 70%; Table 1 a), an intermediate group of 65 genotypes with 30-70% recovery (Table 1 b) 23 genotypes with low-recovery (below 30%) (Table le). 243 Preliminary testing showed consistency among results (such as shoot recovery) in time (Table 2). We are planing to test the entire core for different conservation times in liquid nitrogen. We propose a Minimum Shoot Recovery Percentage (MSRP) of 30% as the lower limit for the establishment of an in vitro base gene bank. This means that any cultivar to be included in the bank needs to have at least 30% of shoot recovery. A 30% MSRP has been constant among different experimental years ( 1999-2000) in experiments carried out so far. Of the clones tested, 82% comply with this mínimum MSRP requirement. To scale up cryopreservation of the collection, the limiting step is the initial tissue (shoots). It takes about 3-4 propagation cycles to increase the number of shoots to be frozen . As an average, one plant has one shoot and 3-4 nodes with buds. Nodes could also be used for cryopreservation. For this reason we checked axillary buds for cryopreservation. We found that in all clones it was possible to obtain shoots after freezing (Table 3). We plan to adopt buds as explants because it simplifies the work (reduces time, effort and propagation supplies). Future activities Conserve the entire core on cryotanks and establish a partía! monitoring evaluation protocol for different conservation times Optimized the use of buds as initial explants for cryopreservation Establish a logistical aspect involved in management of cryo-banks Table 1: Response of 127 cassava clones from the core collection, cryopreserved in Iiquid nitrogen using encapsulation-dehydration technique: lA. Clones grouping with highest shoot response after freezing (more than 70%) Cassava clone % % Shoot formation Viability CM 1999-5 lOO 83.3 MARG9 lOO 85 MBOL1 100 76.6 MBRA 190 90.9 90.9 MBRA315 lOO 88.88 MBRA356 96.3 92.6 MBRA699 88.6 72 MBRA 702 96.2 92.5 MBRA 77 100 100 MBRA 781 100 80 MBRA 900 87.7 70.1 MCOL 1178 94.4 88.8 MCOL 1186-A 96.6 96.6 MCOL 1780 100 100 MCOL2032 lOO 90.45 MCOL2061 92.9 82.5 MCOL262 100 87.5 MCOL590 93.3 70 MCOL608 86.6 73.3 MCOL802 93.3 80 244 Cassava clone % % Shoot fonnation Viabili!}: MCOL955 90.5 75.9 MCUB 1 lOO 70 MCUB29 lOO 88.8 MCUB 53 lOO 73 .3 MCUB 58 90.3 87.3 MECU 166 lOO lOO MMAL24 85.9 75.5 MMEX 17 95.75 76.93 MMEX49 100 100 MMEX59 90.6 77.4 MPAN131 92.9 79.2 MPAR 135 100 95 .83 MPAR51 90.6 77.57 MPER279 92.7 86.45 MTAI2 86.6 80 MUSAS 92.9 74.3 MVEN309 88.3 70.5 MVEN 67-B 100 87.87 MVEN 82 93 .3 72.9 Table 1 B: Intermediate response between 30-70% Cassava clone % % Shoot Cassava clone % % Shoot Viability fonnation Viability fonnation CG 1372-5 100 35.45 MCUB 11 93 .3 63.3 CG 915-1 100 65.8 MCUB 36 88.76 50.9 MARG7 100 44.3 MCUB 56 86.6 50 MBRA 162 96.6 67.3 MCUB 74 96.3 63 .4 MBRA674 100 58.56 MECU10 74.4 43 .3 MPER243 100 65.6 MECU 141-A 66.6 46.6 MPER333 100 60 MECU 144 100 53 MPER431 96.6 46.6 MECU 166 90 50 MPER 503 86.87 54.37 MECU 33 87.25 63.45 MPER518 67.42 49.7 MECU 68 95.2 43.8 MPER 556 80.46 57.6 MGUA44 91.9 55.2 MPER569 94.4 53.85 MGUA7 84.4 36.6 MVEN 173 86.3 42 MIND26 90.3 52.4 MVEN 174 97.2 55 MIND3 96.6 60 MVEN208 59 52.3 MIND4 96.9 57.6 MVEN217 81.8 48.6 MMAL26 65.5 62.2 MVEN297-A 48 .16 38.4 MMEX 102 96.26 68.8 MVEN 322 100 43.3 MMEX 54 95.8 42.5 MVEN6l 100 42.7 MPAN 127 100 52.7 SG 455-1 87.5 54.2 MPAR 152 82.5 67.5 MBRA697 100 37.02 MPAR 7 82.5 31.6 MBRA 73 93.3 49. 1 MPER 184 100 58.1 245 Cassava clone % % Shoot Cassava clone % % Shoot Viabi1i~ formation Víabi1í~ formation MBRA 730 94 46.1 MPER241 90 51.8 MBRA 792 92.9 44.8 MCOL 1467 72.5 52.5 MBRA897 100 65.1 MCOL 1736 76.38 47.2 MBRA 916 100 46.6 MCOL 2089 57.4 40.5 MCOL 112 88.8 41.6 MCOL 2157 85 53 .85 MCOL 1398 87.7 52 MCOL 2318 74.24 59 MCOL490 100 67 MCOL 2361 63.3 49.9 MCOL 534-A 87.7 58.86 MCOL 2493 86.9 37.2 MCOL590 100 57.7 MCOL474 100 56.25 MCR 18 92.9 64 MCR84 100 48.6 Table IC: Clones witb lowest sboot formation after freezing (less than 30%) Cassava clone % Viability % Shoot fonnation MBRA589 63.3 23.3 MBRA 891 76 14 MBRA915 91.8 21.8 MCOL2245 38.6 23 .25 MCOL 638 93.3 33.33 MCUB 32 78.8 3.3 MCUB42 50 6.6 MECU 141 35.5 2l.l MECU 29 45.4 9 MECU71 70 21.8 MECU 85 100 5 MECU104 44.4 20 MGUA63 66.6 33.3 MPAR 109 91.9 11.43 MPAR 163 77.1 31.9 MPAR25 96.6 31.6 MPAR35 lOO 26 MPAR36 93.3 28.2 MPER584 31.8 22.12 MVEN200 96.26 20.7 MVEN210 90.09 21.21 MVEN219 32.8 14.7 MVEN 329-A 92.2 24.4 246 Table 2: Response after freezing of different explants (Escobar et al 2001 ). Cassava clone Explant Viability (%) Shoot recovery (%) MEcu 31 Shoots 35.0 25.0 Buds 56.3 39.7 MGua 14 Shoots 92.3 83.0 Buds 78.3 36.6 MCol2215 Shoots 50.0 26.7 Buds 58.3 33.3 MCol1505 Shoots 83.3 81.3 Buds 37.7 22.2 MCol2016 Shoots 70.0 16.7 Buds 96.3 79.6 Table 3: Response of sorne clone cryopreserved across different conservation periods (Escobar et al 2001). Cassava clone MCo122 MPer436 MVen90 References Treatment Control Exp. 1 month 1 report Control Exp. 1 month 1 report ControiExp 1 month 1 report Viability (%) 88.1 100 95 lOO 91 .65 94.1 88.65 95 76.7 Shoot recovery (%) 88.1 100 85.45 87.5 78.75 79.7 88.65 95 50 Manrique, N. C. Respuesta varietal de 95 genotipos de la collección núcleo de yuca a la crioconservación usando la técnica de encapsulación-deshidratación. Tesis CIAT-U. Nacional, sede Palmita. 2000 Escobar R.H., Debouck D and W.M. Roca. 2000. et al2000 In: Cryopreservation of tropical plant germplasm:Current research progress and applications. Engelman F. and H. Takagi (edt). JIRCAS-Tsukuba, Japan!IPGRI. Escobar R.H., Manrique N.C., Debouck D. Tohme J and W.M. Roca. Cryopreservation research at CIA T. Presentation in a workshop: Cryopreservation of vegetative propagated tropical crops. INIBAP, 2-6 july-2001 Annu1 Report 1999. SB-2 pp 89-91 Annual Report 2000. SB2- pp 178-181 247 2.2.8 Costing all expenses about the different methods to conserve cassava germplasm P P d 1 1 fl 2 d 1-2 . ar e~ ; B. Koo , G. Ma a an D.G. Debouck 1IFPRI; SB-1 Project; 3SB-2 Project As part of a systemwide intiative in view of the establishment of an endowment fund for the germplasm collections held in-trust in the CGIAR, the GRU together with IFPRI has carried out a costing study of all its operations. The following table shows the costs per accession for the different conservation methods of cassava. lt also shows how much would be needed to conserve the present collection in perpetuity. Another outstanding figure of this study is the amount of money saved through slow-growth in vitro, a piece of research initiated in 1997. If the GRU can extend the period between each subculturing from 12-14 months (an average between clones) up to 24 months, the saving could be of US$ 230.93 per accession, allowing a saving of US$ 1,404,079 for the entire collection of 6,080 accessions in perpetuity. Costing studies of cassava germplasm conservation (cosl per accession in US S; under 4%) in vitro cryo field a JI per year 10.34/ 67 .6 1 0 .86/ 113 .77 7.18 N.A. in perpetuity 268 .73 144.06 186 .69 599.48 after Koo et al. 200 1 2.2.9 Preliminary studies on the cryopreservation of meristems and seeds of wild Manihot species R.H. Escobar1•2, N.C. Manrique2, F. Gi11, J. Tohme2, and D.G. Debouck1 1 SB 1 and 2SB2 Introduction CIA T maintains wild Manihot species as a field gene bank and as a small in vitro collection. Under field conditions, these species are exposed to pests and diseases, and their flowering and seed set are affected by environmental conditions. Twenty-nine species and 300 genotypes are maintained in vitro with subcultures every 6-12 month (personal communication, G. Mafla). Both options are seen as temporary given their costs, space and manpower requirements. Seeds in cold 248 storage have been kept up to seven years without loss of viability; if maintained at room temperature however, they lose vlability after one year. Marin el al (1 990) established a cryopreservation protocol with seeds and zygotic ernbryos. Later Escobar and Mafla rnodified the wanning step after freezing, expeditíng the process (BRU, data not published). Two other approaches for the long-tenn conservation of Manihol diversity are being tested: cryopreservation of meristematic tissues (conservation of individual genotypes), and storage of botanical seeds ( conservation of population genetic diversity) Methodology Modifications of the encapsulation-dehydration protocol, successfully used for shoot tips of cassava (M esculenla) (Escobar el al 2000), were tested on two genotypes of M carlhaginensis, three genotypes of both M esculenla subsp. peruviana and M esculenta subsp. jlabellifolia. These modifications consisted in different periods of dehydration on silica gel, and pre- and re- growth media components. The in vitro wild material selected by this study had completed passport data. Experiments have been undertaken with seeds of three Manihol species by freezing them in liquid nitrogen during different periods. Recovery of seeds will be monitored. Results and discussion According to Velasquez ( 1995), wild species had different behavior in in vitro conditions. lt was necessary to adapt five propagation and conservation media (12A1, 12 A2, 12A3, WPMI and WPM2) according to wild-group (genotype or species) response. When we tested the response after freezing using 12A3 medium in the process of regrowth, no growth was observed. We are currently using 12A3 only in the propagation scheme. At the beginning of this project we observed that wild frozen shoots responded with callus fonnation and tried to fonn shoots, but the in vitro growth conditions were not optimal. It was then necessary to increase Kinetin content, test 4E medium (Roca 1984) and MS with activated charcoal to obtain shoots after freezing. The quality of cassava shoots is important for a good response after freezing (Escobar et al 2000). The morphology ofwild shoot was different prior to freezing. M carthaginensis showed no shoot growth on 4E medium before freezing (only differentiated leaves); M esculenla subs. peruviana and jlabellifolia showed stronger shoots after 3 days on 4E medium prior to freezing. Probably the lower response of M calhaginensis was influenced by the previous conditions of donor tissues (Table 1). Different pre-growth periods (3,4 y 5 days) and media could be helpful to improve shoot quality before freezing. We are introducing other wild species such as M orbicularis, M cecropiaefolia, /ongipeliolala to validate the method. Table 1: Response of wild Manihot species after freezing Genotype % Viability % Shoot formation M escu/enta subsp. F/abeliifo/ia 439-003 51.35 39.55 M escu/enta subsp. F/abeliifolia 437-007 lOO 83.3 M carthaginensis O 17E O O 249 We compared different desiccation periods on silica gel (16h vs. 24 h) before freezing. We found that with M carthaginensis 17f 417-001 it was possible to obtain plants after freezing (7.7%) with 16 h desiccation. With M esculenta subsp. jlabellifolia it was possible to obtain shoot formation in both conditions (Table 2). Table 2: Recovery of frozen wild Manihot species desiccation during different periods on silica gel. Rep Viabi1ity % Shoot Viabi1ity % Shoot % formation % formation Desiccation duration 24 h 16 h Genotype M esculenta subsp. Peruviana 413-003 R3 35.7 21.4 M esculenta subsp. Peruviana 417-003 R1 o o 12.5 12.5 M esculenta subsp. Peruviana 417-003 R3 - 60 30 M esculenta subsp. Peruviana 417-005 Rl 58.J 16.6 o o M esculenta subsp. Peruviana 417-005 R2 o o o o M esculenta subsp. Peruviana 417-005 R3 30 o 41.6 o M carthaginensis 413-013 Rl o o M carthaginensis 413-013 R2 56.25 o 31 .25 o M carthaginensis 413-013 R3 o o lO o M carthaginensis 17f 417-00 l R1 o o M carthaginensis 17f 417-001 R2 - 23 7.7 M carthaginensis 17f 417-00 l R3 14.28 o 7.14 o M esculenta subsp.jlabe/lifolia. 444-002 R1 - 33.3 11.1 M esculenta subsp. jlabel/ifo/ia. 444-002 R2 100 60 100 lOO M esculenta subsp. Flabellifolia 444-002 R3 84.6 53 .8 o o M esculenta subsp. F/abellifolia 433-002 R1 76.9 61.5 53.8 46.1 M esculenta subsp. Flabellifolia 433-002 R2 70 50 80 50 - M esculenta subsp. Flabellifolia 433-002 R3 50 10 250 Figure 1: Plants of M. esculenta subsp.jlabellifolia 444-002 recovered form liquid nitrogen. Conclusions Preliminary results showed that plants from wild cassava species could be recovered after freezing wíth a 0-60% survival. In vitro shoot tips of wild material had different aspect compared to cassava. lt may be necessary to modify pre- and post-freezíng conditions. Simple MS with activated charcoal medium is a key factor to recover shoots after freezing. Different dehydration periods could improve recovery rates after freezing of wild material. M esculenta subsp. peruviana and jlabellifolia showed better response than M carthaginensis. Future actívities The cryo-methodology needs to be validated with other wild material. Keep fine-tuning conditions to improve recovery rates of wild material after freezing. Compare cryopreservation method with conventional medium-term conservation (at 5°C or - 20°C) helps to determine the best conservation alternatives. References Mario et al 1990 Escobar et al 2000 BRU Annual Reports Manrique NC. 2000 251 2.2.10 lemporary Immersion System (RITA) for Anther Culture of Rice E. Tabares1, G. Delgado\ R. Escobar', Z. Lentini1 1SB-2· 2IP-4 , Introduction Plant in vitro culture using temporary immersion offers all the advantages of a liquid medium system (automation, large scale production, easy changes of medium, filter sterilization, easy cleaning) without any of its drawbacks (reduced gas exchange, vitrification ). The RITA device has the additional advantages of low cost and automation. And it is suitable for both industrial production and research use. This system has proved its efficiency for somatic embryogenesis of banana (Alvarat 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); 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 & Alavarad, 1998). RITA system improves plant cell nutrition and gas exchange, and enhances the quantity and quality of micro-propagated tissue. Last year we reported the use of RITA for the induction of embryogenic callus derived rice mature zygotic embryos. Callus induction was noted 8 to 15 days earlier in the RITA system respect to petri plates. A larger number of embryos showed callus formation, and between 2 to 4 fold increase in the number of embryogenic callus was seen in the automated system. One of the major bottlenecks for generating doubled haploids from rice anther culture (AC) is that the response is highly dependant on the genotype. Japonica types are generally highly responsive in contrast to indica rice, which shows low response and severa! genotypes are recalcitrant. The low response of indicas restrain the use of AC in breeding for the tropics. Following is reported preliminary experiments directed to induce embryogenic callus and plant regeneration from rice AC using the automated RITA system. Materials and Methods For preliminary experiments, a japonica genotype (CT 6241-4-1-15-1 ) and an indica variety (Palmar) were used. Plants were grown in the field, panicles harvested, and anthers cultured according to Lentini et al., 1995. Anthers were either culture in liquid medium contained in baby food jars (control) or in RITA vessels. Seven hundred and fifty anthers were culture per RITA vessel. Immersion system was set up for cycles of 6 hr immersion. Induced callus from each culture system, was then transfer onto solid plant regeneration medium according to Lentini et al. 1995. Results and Discussion Callus induction was noted 5 days earlier in the RITA system respect to the control. A significant larger number of embryogenic callus was noted in RITA for both genotypes, and between 3 to 5 fo1d increase in the number of regenerated plants was seen in the automated system (Figure 1). Experiments are in progress to determine the reproducibility of these results with larger number of indica genotypes, and to optimize the system for recalcitrant genotypes important for the breeding program. 252 250 CT 6241 u 200 +--- - "' g 150 -+-- -- ~ 100 +---- u a: 50 o CaUus per 100 Plants per 100 anthers anthers Palmar Callus per l OO Plants per 100 anthers anthers mcontrol1 ~Control2 .RITA Figure 1.- Callus induction and plant regeneration of japonica and indica rice in RITA system References Alvarad, D.; Cote, F.; Teisson, C. 1993. Comparison ofmethods ofliquid medium culture for anana 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 immersion technique. ASIC, Kyoto, ol 11: 514-519C. Cabasson, D.; Alvarad, D.; Dambier, D.; Ollitrault, P.; Teisson, C. 1997. Improvement ofitrus somatic embryo development by temporary immersion. Plant Cell Tiss. Org. Cult. 1-5 Escalant, J. V.; Teisson, C.; Cote, F. 1994. Amplified somatic embryogenesis from maJe flowers riploid 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. CIA T Annual Report Project SB-02 p 84-86 - Etiene, H.; Lartaud, M.; Michaux-Ferriére, N.; Carron, M.P.; Berthouly, M.; Teisson, C. 1997. lmprovement of somatic embryogenesis in Hevea Brasi/iensis (Müll. Arg) using the Temporary immersion technique. In Vitro Cell Dev. Biol. 33:81-87 Lorenzo, J.C.; Gonzalez, B.; Escalona, M.; Teisson, C.; Espinosa, P.; Borroto, C. 1998. ugarcane shoot formation in an improved temporary immersion system. Plant Cell Tiss. rg. Cult. 54: 197-200 Teisson. C. & Al varad, D. 1998. In vitro propagation of potato microtubers in Ji quid medium sing temporary immersion. Conf. Potato Seed Production by Tissue Culture, Brussels, ost 822, European Commission. 253 2.2.11 Development of selection systems for the generation of transgenic rice according to current food biosafety requirements E. Tabares1, G. Delgado2, Z. Lentini1 1SB-2 Project; 2IP-4 Project lntroduction Most selection systems commonly used for the generation of transgenic rice using either biolistic or Agrobacterium mediated transformation, relies on the use of the hpt gene conferring antibiotic resistance to hygromycin. Although the selection using hygromycin resistance gene has been highly efficient for the production of transgenic rice, and hpt gene has been de-regulated for generating transgenic food for human consumption, it has not yet been approved for animal. As a response to public perception concems, recent intemational food biosafety recommendations suggest to use non-antibiotic selection markers genes. From the year 2005 it is most likely that food or feed crop containing antibiotic resistance will not be approved for commercialization. Recent developments using selection systems different from either antibiotic or herbicide resistance suggest that the phosphomannose isomerase (pmi) is an efficient selection gene for production of transgenic plants from maize, sugar beet, and cassava (Wang, et al. 2000; Negrotto et al., 2000). When non-transgenic cell tissue are cultured on mannose, after uptake mannose is phosphorylated by a hexokinase, yielding mannose-6-phosphate which accumulates in plant cells and causes severe growth inhibition. Mannose isomerase gene converts mannose-6-phosphate to fructose-6-phosphate, thus giving plant cells the capacity of metabolizing mannose as a carbon source. Last year we reported the optimization of a positive selection protocol using mannose containing medium, for callus induction from mature zygotic embryos Agroinfected with a construct containing the pmi gene. Callus showed a significant reduction in growth when cultured on mannose containing medium respect to sucrose. An inhibition of 59%, 64%, 95% and 98% were noted when callus were cultured on mannose 1%, 2%, 3%, and 6%, respectively. The leve! of this inhibition was even higher than that observed on medium without carbon source which shows a 87% of growth ínhíbitíon respect to 3% sucrose. The inhíbition of growth was due to the effect of mannose and not due to an increase of osmotic potential, since callus on 3% sucrose + 1.5% mannitol or 6% sucrose + 3% mannitol develop alike to those on 3% sucrose. Based on these results in the case of Cica 8 a concentration of 3% mannose is considered sub lethal, and 6% mannose a lethal dose. Following is described the progress made regenerating plants of rice from mannose resistant callus. Materials and Methods Mature zygotic embryos of rice indica varieties Cica 8 and Palmar were co-cultivated with Agrobacterium tumefaciens strain LBA 4404 containing a constuct containing the phosphomannose isomerase (pmi) gene and the uid-intron gene encoding the ¡3-glucoronidase for gus expression. After agroinfection , embryos were cultured on callus induction medium containing acetozyringone 1 OO¡.LM and cefotaxime 250 m gil for bacteria elimination. A stepwise selection for callus induction was conducted first on medium containing 3% mannose (261 mOsmollkg ), as the only carbon source, followed by a subcultured on índuced callus on medium with 6% mannose. Mannose resistant callus were then transferred onto plant regeneration solid medium containing 2% mannose as based line complemented with either 0.5 %, 1.0%, 1.5% or 2.0% sucrose. 254 Results and Discussion For Palmar variety, preliminary experiments indicated that although plant differentiation was noted on medium containing only mannose without sucrose, shoots did not elongate when mannose was used as the only carbon source (data not shown). Therefore, various concentrations of sucrose were added besides the mannose to aid the plant regeneration process. Initially, a total of 16 different combinations were tested which included mannose from 0.5% to 6% in combination with sucrose from 0% to 3%. Results suggested that in order to have only transgenic plant regeneration in Palmar, a mínimum of mannose 2% was needed in the plant regeneration medium (data not shown). Palmar showed about 50% plant regeneration on mannose 2% with either sucrose 1.5% or 2.0% (Table 1). All the regenerated plants showed gus expression (Table 1 ). In contrast, Cica 8 did not regenerate at all when mannose was included in the medium, although medium was complemented up to 3% sucrose (Table 1). About fifty percent plant regeneration was attained from Cica 8 mannose resistant callus only when 3% sucrose was used as the sol e carbon so urce, however of these plants only 5% showed gus expression. Southern blot analysis of these plants is in progress. Results obtained so far, suggest that all gus expressing plants had simple insertions of both pmi and iud-intron genes in the genome. References Joersbo M, Donaldson I,Kreiberg J,Petersen SG.Brunstedt J.Okkels FT(I998) Analysis ofmannose selection used for transformation ofsugar beet.Mol Breeeding 4: 111-117. Negrotto D., D. Jolley, S. Beer, A.R. Wenck, and G. Hansen. 2000. The use ofphosphomannose-isomerase as a selectable marker to recover transgenic maize plants (Zea mays L.) via Agrobacterium transformation. Plant Cell Resorts 19: 798-803 . Wang A.A., R.A. Evans, P.R. Altendorf, J.A. Hanten, M.C. Doyle, and J.L. Rosichan. 2000. A Mannose selection system for production of fertile transgenic maize plants from protoplasts. Plant Cell Reports 19: 654-660. 255 Table 1. Effect fo mannose : sucrose ratio on plant regeneration of mannose resistant callus 1 iud-intron gene (%) expression (%) (%) Callus analyzed Plants Regenerated Plants Plants (%) Genotype Treatment Mannose Sucrose regenerated plants tested Gus+ Gus + Palmar 1 2.0 0.5 4 o o 2 2.0 1.0 20 o o 3 2.0 1.5 9 5 56 5 5 lOO 4 2.0 2.0 4 2 50 l 1 lOO Control 2.0 2.0 20 o o Cica 8 1 0.0 3.0 35 20 57 20 1 5 2 2.0 0.5 6 o o 3 2.0 1.0 5 o o 4 2.0 1.5 6 o o Control 2.0 2.0 5 o o Putative transgenic callus derived on selection medium containing 3% mannose (sub-lethal) followed by 6% mannose (lethal) 256 2.2.12 A first initiative to support cassava seed production-for the industry R. H. Escobar', M.C. Arzuza2, L. Muñoz\ C. Restrepo2 and J. Tohme1 1 SB 1, 2Com Products Andina. Secciona) Barranquilla. Introduction In the Northem Coast of Colombia, cassava has been considered the most important altemative crop since it is the only that yields acceptable under marginal conditions and with minimal inputs. There is an intemational starch industry, Com Products, that has a branch Iocated in Malambo (Barranquilla, Colombia), formerly known as Inyucal. It is considered one of the largest cassava consumers for industrial purposes in Colombia. However, Com Products has encountered a chronic problem in the supply of raw material, fresh cassava roots for milling. To continually operate the cassava milling plant, it would be desirable to provide up to 100 ton per day of fresh roots. Under the actual conditions, only 0.5 ton/day are ground, and in sorne days no material is available to take to the mili (M.C. Arzuza personal communication). In vitro propagation systems could be a quick and safe manner to put enough, healthy planting material in farmer's fields to support Com Products' demand. This project is the first initiative to set up cassava seed production programs aimed to satisfy industrial needs. Methodology Based on CIA T' s experiences on handling cassava in vitro (Roca 1984; Escobar et al 2000, 200 1; Annual Report 2000), we performed a feasibility analysis based on the final number of plants needed per hectare. This allowed the estimation of propagation rates, lab requirements, manpower, screenhouse facilities among other inputs to start the project. After establishing the scale and lab requirements, we trained staff from Com Products. During the initial steps we formed a discussion group to diagram the lab to be established in Malambo. The lab included four essential areas: sterilization, propagation, washing and growth areas. Additionally, a screenhouse was also established. CIA T supported the technical and logistical scheme, while Com Products provided financia) support. Com Products signed an MT A to access cassava germplasm MTai 8 and MVen 25. They received plants that were given to us later to initiate propagatiorr schemes at CIA T while they were building the laboratory facilities in Barranquilla. Results and discussion A 75,2 m2 laboratory (including the four critica) areas and a small office), and a 72 m2 screenhouse were built (Figure 1). One stafffrom Com Products (Maria C. Arzuza) was trained in tissue culture during one week. She then trained other partners in Malambo. They signed the MTA #028 to receive in vitro plants from the Genetic Resources Unit (5 plants per clone; April 23-1999). At BRU, using conventional propagation schemes, we increased the planting material and divided it in two sets: 1) a frrst set of 1336 in vitro plants (682 plants of MTai8 and 654 plants of MVen25; July 16- 2001 ), was sent to Malambo to initiate propagation activities, 2) a second set was maintained at BRU as a back up under minimal growing conditions (35 Magenta boxes of MTai 8, and 12 Magenta boxes for MVen25; each magenta contained 20 plants). 257 Com Product staff initiated propagation with material (1,336 plants) received from BRU. We estimated, based on a 1:3 propagation rate, that Com Products would produce around 4,000 plants. They are currently propagating 25-40 containers/day/person, and maintaining 346 containers ofMTai 8 (1384 plants) and 534 ofMVen 25 (2136 plants), for a total of3,400 plants (88% of the expected number of plants ). We maintain communicatíon through e-mail, fax, phone and technical field visits to support ímprovements in staff skills and lab and screenhouse conditions in an attempt to keep losses below 3%. This will allow a more accurate planning of propagation schemes. Although Mven25 was one of the clones selected initially for propagation, its poor adaptation to soil conditions in the area, and the early stake sprouting makes it unattractive to keep it in the propagation scheme. For this reason Com Products is selecting other clones with better response and high dry matter content to propagate them. B Figures: (lA) In vitro growth area in the laboratory and (lB) Screenbouse facilities in Malambo- Barranquilla. Future activities • Adjust screenhouse conditions to recover higher percentages of transplanted plants. • Design propagation plans to maintain back ups in vitro to feed production lines, improve propagation rates and reduce costs. • Introduce other clones into propagation schemes. References Annual Report 2000. Project SB-02 pp 185-187 Escobar R.H., Muiioz L, Tolune J y W.M. Roca. Estado actual de la micropropagacion de la yuca. Seminario Internacional Programa Colombiano de Biotecnología Agrícola, Cartagena, Colombia, febrero 21-23,2001 Escobar R.H., Muiioz L y W.M. Roca. Cassava micropropagation for rapid seed production using temporary inmersion bioreactors. Poster at the symposium "From germplasm bank to farmer fields, role of CIA T biotechnology in research and training in Latin America", December 4 of 2000, CIA T, Cali, Colombia. Roca, W. M. (1984) Cassava. In: Handbook ofplant cell culture, Vol. 2. Crop species. (Sharp, W.R. Evans, D.A.; Ammirato, P.V.;and Y amada, Y.) p(269-301) MacMillan Publishing Co., New York. 258 2.2.13 Development of methodologies for in vitro m~ltiplication, plant regeneration, and genetic transformation of naranjilla (lulo) V. Segovia; Z. Lentini SB-2 Project 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 return for the farmers. Naranjilla (So/anum quitoense) is among these fruits. This species is native from Colombia and Ecuador, and it is normally cultivated between 700 and 2000 meters above sea leve!. Sorne of the main attributes of this fruit includes its high leve! 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. The high leve! of trait segregation restrains its multiplication through seeds. Rapid multiplication of quality planting materials is of paramount importan ce. 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 clones free of pathogens. The expected results include the mass multiplication of elite clones that then can be distributed to farmers. Since breeding for this species is almost non-existing, paralelly to the in vitro propagation effort, it will be important to develop plant regeneration and transformation systems to aid the development of germplasm. Last year it was reported preliminary results on developing a plant regeneration protocol. This year it is presented the advancement in establishing a system for maintenance of the in vitro germplasm collection, and the progress made identifying factors to increase the plant regeneration efficiency from elite naranjilla materials. Materials and Methods High quality and elite clones provided by the Andean Fruit Center (Centro Frutícola Andino- CEF A) and Corpoica La Selva are used. This collection includes naranjilla with or without thorns commonly grown by farmers. Various media were tested for in vitro maintenance of the clones. Statistical experimental design was used to determine the optimal plant tissue and medium for an increase efficíency in plant regeneration. Regenerated plants were taken to the greenhouse and the field to evaluate plant growth and development to maturity. Plants initially grown in vitro and then in soit, were taken back to the in vitro system to establish a protocol for renewing the in vitro collection. Results and Discussion In Vitro Propagation Last year it was reported that plant develops healthier and faster in vitro when cultured onto ~ MS medium supplemented with ANA 0.02 mg/1, BAP 0.04 mg/1, and GA3 0.05 mgll, and agar 4.5 g/1, in contrast to the regular micro-propagation medium used by CEF A or Corpoica La Selva (SB2 Annual Report 2000). However, after 1 O months of sub-cu1turing naranjilla in vitro plants in this medium, most of the plants showed white stripes on leaves, thick stems, and slow development. Reason why other media were tested again. For this purpose, it was evaluated media commonly used for long term maintenance of in vitro germplasm of other Solaneace 259 species. Of the medium tested, a medium contammg MS basal salts and vitamins, and supplemented with calcium pantothenic acid 2.5 mg/1 and gelrite 3.5 g/1 (Hussey and Stacey, 1981) used for potato, also showed to be the optimal for naranjilla (medium A). On this medium, in vitro shoots start showing root proliferation two weeks earlier respect to medium !h MS medium. Fully expanded new leaves were present at one month after subculture on medium A, whereas it took about 6 weeks on !h MS medium (Figure 1). Moreover, new Jeaves developed on medium A were completely green although the starting materials had leaves with white stripes. The number of leaves with white stripes increased as plants were sub-cultured on !h MS medium. By sub-culturing plants on medium A contained in jars rather than in test tubes, it is possible to obtain a large number of explants to conduct series of experiments for optimizing plant regeneration with various replicates every month. Medium A also seems to be a more appropriate for a long term in vitro maintenance of the germplasm collection since plants are healthier, develop faster, and differentiate normal green looking leaves. Figure 1./n vitro development ofnaranjilla plants in Yl MS medium (left) and medium A (right) one mothe after subculture Plant Regeneration A randomized block design of four replicates each of 15 experimental units was used to determine the best medium composition and explant to induce a direct plant regeneration in naranjilla. A non-parametric chi-square analysis indicated that petioles from the first and second node showed 17 times and about four times more plant regeneration that the corresponding leaves from thomy and non-thomy clones respectively. (Figure 2 and 3). A significant higher response was also noted on medium originally develop for plant regeneration on tomato (Ultzen et al, 1995), consisting on MS salts, B5 vitamins, supplemented with sucrose 1 O g/1, glucose 1 O g/1, gelrite 1.5 g/1, zeatine 2 mg/1 and IAA 0.02 mg/1 (Figure 2A). On this medium petioles from thomy genotypes showed three fold increase in plant regeneration respect to a medium reported for naranjila ( Hendrix et al., 1987) composed by MS salts and vitamins and supplemented with sucrose 30 g/1, agar 7 g/1, IAA 0.01 mg/1, kinetin 5 mg/1, or with a modification consisting on gelrite 2 g/1 and BAP 2 mg/1 (modification suggessted by Dr Richard Litz, Univeristy of Florida, laboratory which Hendrix work was conducted)(Figure 2B). Non-thorny genotypes did not regenerated any plant on mediurn developed by Hendrix (Figure 2B). 260 o/e Re spo nse A % Re sp 1 on Petiole • Thorny l rJ Non- Thorny 1 r.l Petiole • Learl Figure 2.- Plant regeneration from petiole or leaf explants of genotypes with or without thorns using medium previously develop for tomato (A) or reported by Hendrb: for naranjilla (B). Figure 3.- Multiple shoot formation from petiole of naranjilla Plant Evaluations in the Greenhouse and the Field A methodology was establish to transfer in vitro material (from the in vitro germplasm collection and regenerated plants) to the greenhouse and there after, to the field. The first regenerated plants evaluated shown a normal plant growth and development to maturity in the greenhouse and in the field. Fruit formation is being evaluated in the field only, since temperatures in the greenhouse at CIA T headquarters is to high for naranjilla fructification. The field selected is located at 1, 700 m.s.n.m. and with a mean temperature of 22 C, ideal for induction of fruit formation of naranjilla. The field plot is located in a farm (La Casona) at Dapa about 20 min from Cali, where there is a naranjilla production by farmers. Future plans Establish a cyclic culture from greenhouse to in vitro, to completely renew the existing germplasm in vitro collection and propagate it in medium A Evaluate other factors affecting plant regeneration response to increase it at about 50% 261 B Develop a genetic transformation protocol Complete evaluation of regenerated plants in the field and compare the growth and development with in vitro propagated and seed derived materials References Hendrix R. , R. Litz, and B. Kirchoff. 1987. In vitro organogenesis and plant regeneration from leaves of Solanum candidum, S. quitoense (naranjilla) and S. sessilijlorum. Plant Cell Tissue and Organ Culture. 11 : 67-73. Hussey G. and N.J. Stacey. 1984. Factors affecting the formation ofin vitro tubers ofpotato (So/anum tuberosum 1.). Ann.Bot. 53: 565-578. U1tzen T., J. Gie1en, F. V enema, A. Westerbroek, P. Haan, M. Tan, A. Schram, M. Grinsven, and R. Goldbach. 1995. Resistance to tomato spotted wilt virus in transgenic tomato hybrids. Euphytica. 85: 159-168. 2.2.14 Cryopreservation of Friable Embryogenic Callus (FEC) lines LG Santos1, R Escobar\ P Chavarriaga1, and WM Roca 1"2 1SB-2 Project / CIP-Peru Introduction Friable Embryonic Callus lines were established at CIA T for sorne commercial Latin American cassava cultivars (Lopez 2000). In vitro-maintained cell suspensions are genetically unstable, besides, keeping them is labor intensive and costly (Reinhoud et al 2000). Cryopreservation of FEC may provide a means of ensuring genetic stability of cell lines, and could be a source of "fresh" tissue useful for genetic transformation. Methodology To establish a cryopreservation protocol we are using cultivars MCol 2215, CM 3306-4, and keeping the cultivar TMS60444 as a control. Different cryopreservation methods were tested: Classic, vitrification, encapsulation-dehydration, encapsulation-vitrification, desiccation and desiccation-vitrification. Different temperatures of loading, PVS3 dilutions, and pretreatment of FEC were also tested. A programmed freezing protocols (1 °C/min) using a container with isopropanol was tested. Results and discussion Desiccation of FEC before freezing seemed to give tissues a good chance to recover. FEC grown on media with high agar content (6-9%) and growth regulators, desiccated during different times (10-30 days), are showing promising results after freezing. Frozen tissue was recovered on GD2- 50Pi solid medium, grown under low light intensity and transferred to liquid medium (Figure 1). Fine tuning re-growth conditions (temperature, light intensity and humidity) will probably increase the recover of tissues. We are testing GRU's growth room conditions to compare with those we have at BRU. Initial observations indicate that there may be strong differences between both growth rooms, with the one from GRU giving better results. 262 • Protocols involving vitrification did not seem to work well for the recovery of FEC after freezing. Toxic effects of cryo-protectant solutions like PVS2 (Sakay et al 2000) were observed even with no frozen tissues. PVS3, another vitrification solution showed better results than PVS2. We will however continue testing protocols that mix vitrification and dessication to find out which one gives better recovery rates. Figure l. Regrowth (arrows) of FEC after freezing (A) and control witbout freezíng (B) for cultivar TMS60444. Conclusions • lt is possible to recover cell growth after freezing ofFEC. • Modification of growth conditions could improve post-freezing response. • Desiccation and desiccation~vitrification could be used as potential cryopreservation methods. Media with high agar content could be used as a desiccation treatment. • PVS2 shows a detrimental effect on cassava cell suspensions. PVS3 has shown beneficia! effect on tissues. Futures activities • Adjust conditions to improve cell recovery after freezing. • Test best conditions on other FEC lines from Latín-American cultivars. • Recover plants from frozen cells. References Lopez2000 Reinhoud P. Van Iren F. and Kijne J.W. 2000. Cryopreservation ofundifferentiated plant cell. In: Cryopreservation of tropical plant gennplasm:Current research progress and applications. Engelman F. and H. Takagi (edt). JIRCAS-Tsukuba, Japan!IPGRI. Sakay A. 2000. Development of cryopreservation techniques. In: Cryopreservation of tropical plant germplasm:Current research progress and applications. Engelman F. and H. Takagi (edt). JlRCAS- Tsukuba, Japan!IPGRI. 263 2.2.15 In vitro propagation through micrografting of selecte~ clones of soursop (Annona muricata L.): Optimization of the technique and field evaluation of the agronomic performance of propagated trees Juan Jairo Ruiz3•8, Adriana Alzate\ Nelson Royero2•8, Silvio Cadena3"8, Adriana Nuñez3, Francisco Arboleda3"\ Jorge Cabra., William Roca7, Alvaro Mejía-Jiménez1 1SB-2 Project; 2Corporación BIOTEC; 3Project supported by Pronatta; 4Project supported by Fundación Banco de la República; sindependent consultant; 7Centro Internacional de la Papa, CIP, Perú; 8Universidad Nacional de Palmira Introduction The soursop (or guanábano in Spanish, Annona muricata L) is a fruit tree native to the tropical Americas. Its white pulp is used for the production of juices, yogurts, ice creams and desserts. One of the most important problems facing the soursop growers is the lack of disease-free planting materials from elite selections. Between 1997 and 1999, we developed a novel methodology for the in vitro clona! propagation of elite trees through micrografting (Royero et al., 1998). This methodology allows a rapid clonal multiplication of elite varieties and the production of disease-free planting materials. The ftrst trees derived through in vitro propagation were planted at CIA T and in farms belonging to experienced soursop growers located in Huila and Valle for field-testing betw~en January 1999 and January 2000. In 2001 ,our efforts were concentrated on the: Evaluation of the agronomic performance of the micropropagated trees in the field; Optimization of the propagation methodology; Adaptation ofthe developed in vitro propagation methodology to new promising soursop clones; Initiation of investigations on the use of rootstocks of different soursop selections or related annonaceus species; and lnitiation on investigations on the management ofthe micropropagated plants in the greenhouse. Methodology The in vitro propagation methodology of soursop through micrografting has been described in previous reports (Royero et al. 1998 and 1999) Results Optimization and adaptation of the micrografting methodology to new selected genotypes The in vitro propagation methodology of soursop developed jointly by Corporación BIOTEC and CIA T, consists of the in vitro propagation of buds isolated from selected soursop trees or varieties, and their micrografting onto rootstocks produced from seedlings germinated in vitro. Buds for micrografting can be induced from in vitro cultured stem fragments, isolated from greenhouse growing plants or from other previously produced micrografts cultured in vitro ( cyclic micrografting). Overall efficiency of propagation depends on ( 1) the multiplication rate of buds in vitro, (2) the efficiency of seed germination in vitro, (3) the efficiency of scion/rootstock union, and (4) further development in vitro and in the greenhouse of the micrografted plants. Each of these steps is being optimized independently in order to improve the overall propagation efficiency. 264 In general, the same methodology of propagation, developed initially with the commercial cultivar "Eiita" (Ríos Castaño and Reyes, 1996), could be adapted with few modifications to other three clones (named "Cristina", "Rosa" and "Francia") selected from large commercial plantations located in the Huila region ofColombia. High efficiencies of scion/rootstock union are already being achieved with the actual methodology (between 100 and 65% depending on genotype combinations and so urce of buds, figure 1 ). However with sorne combinations, further development of the micrografted bud is still low (between 33 and 74%). The improvement ofbud development in vitro can thus be identified as the most critica) step for increasing the overall efficiency of the propagation methodology of soursop through micrografting. The highest levels of bud development in vitro are being achieved when rootstocks of the genotype Cristina, and buds coming from previously produced micrografted plants, cultured in vitro, are used. Already new combinations of selected clones micrografted over different rootstocks have been planted at CIA T and sent to farms at Huila, for agronomic evaluation: 100 ... e Cll E 90 Q, o Cll 80 > Cll Q 'CI 70 :::1 11:1 'CI e 60 • e o e :50 ~ .w u -40 o ... 111 ... o 30 o a: ...... e 20 o u (/) .... 10 o M o C/C C/E C/R E/C E/C E/E E/R E/R R/C R/E R/E R/R SIC ele est ele ele est ele ele est 83t ele .st ele ele Scion/Rootstock Comblnatlon and Source or Buds Flgl. Micrografting success (sdon-rootstock union) and furtber development In vitro of micrografted plants of dlfferent combinations of soursop selections (E- Elita, C= Cristina, R• Rosa, F"' Francia) and rootstocks. The buds used for producing the micrografts in the combinations marked with "tic" or "est" were isolated from micrografted plants produced previously and cultured in vitro, or from axillary buds induced in vitro from isolated stem sections of greenhouse growing plants, respectively. 265 Eva/uation of the agronomic performance of soursop p/ants propagated through in vitro micrografting. The soursop plants produced through the in vitro micrografting methodology are being tested in the field in order to assess the usefulness of the technique for the production of planting materials of this tropical fruit species. As a part of a M.Sc. thesis (Juan Ruiz, Universidad Nacional de Colombia), the agronomic performance of El ita/El ita micrografted plants is being evaluated in five different locations of the Valle and Huila states. Micrografted plants were planted in the field between January 1999 and January 2000, after 10 months of hardening in the greenhouse. In CIAT (where the highest growth rates are being achieved), the trees at 13 months after planting in the field had become vigorous and healthy with an average height of 2 m. Since the trees have to be pruned after reaching this height in order to initiate tree formation, other parameters in addition to tree height should be taken into account, in order to ha ve a reliable estímate of tree growth. Perimeter of the stem 1 O cm over micrografting point was measured in replacement (fig. 2). After 13 months in the field, an -average of 5 reproductive structures (flower buds, flowers and developing fruits) were counted in each tree. First fruits were harvested 17 months after planting at CIA T. This is a rather fast development, and a first confirmation of the usefulness of the in vitro micrografting technique as a method for producing healthy and vigorous growing soursop planting material. o 2 • 6 8 10 12 1. 16 18 20 22 2• 26 28 Heootu .. n ... ,..,.t.., 111 t ... ''-"" -+-El ~tillo (Yollt) __ y.._¡¡ (Yol lt) -o.-Lo Esn.do (Yollt) -w-.CIAT -li-C«Titos (Hulla) -o-Son Fronclsco (Hullo Fig. 2 Growth of micrografted trees Elita/Eiita in five locations of the Colombian states Valle and Huila measured as the perimeter ofthe stem 10 cm over the mic:rografting point. 266 Use of rootstocks of other Annona species for the production of more vigorous, disease resistant and widely adapted soursop trees. The use of rootstocks of different genotypes or species from that of the se ion for the production of vigorous, widely adapted or disease resistant plants is a common practice in fruit tree propagation. In soursop this avenue has been largely under-exploited. We are testing the compatibility and agronomic performance of combinations of scions of selected soursop clones Elita, Rosa and Cristina micrografted in vitro over rootstocks of A. montana and A. glabra L., two annonaceous species originally from the Chocó rainforest of Colombia. These two species do not show symptoms of anthracnose infections, the most important disease of soursop, even under conditions favorable for the development of this disease (high temperatures and humidity). Grafting soursop over A. glabra or A. montana may also improve resistance to drought, high humidity or soils with bad drainage (Escobar and Sánchez, 1992). Compared to the A. muricata/A. muricata genotype combinations (for example, Elita/Cristina, fig. 3), micrografted plants produced over A. montana as rootstocks showed similar efficiencies of scion/rootstock union. However, further development efficiencies in vitro and later in the greenhouse, were much lower (13% greenhouse development for Cristina/A. montana micrografted plants; fig. 3) for sorne combinations. lt is note worthy though that viable plants were obtained in all genotype combinations, indicating that the micrografted tissues of these two species are compatible. The próduced plants were planted at CIA T or sent to farms at Huila State for agronomic evaluation . ... e • E ~= 80 ~ g 70 . ..: Q: e·~ 60 ~(.1) :§ ~ ~ ~ .. u 11: o- _. 'a •A ! e ""' o• :e 2A ........ -c- o > u 1/l .... o 20 M 10 o [llta/A.II*It... R>osa/A...-.taftl Cristila/A.II*It... [llta/Cristila Se 1 on/Rootstoclc Comb 1 nat 1 on a w Solon/Rootstock ~Ion •• 8ud O.Vtlopnont - •• Plont s.r.lvalln Fi¡. 3. Efficiency of scionlrootstock union, further development in vilro and plant survival in the greenhouse of dlfferent soursop clones (Cristina, E lita and Rosa) micrografted over rootstocks of A. montana. For comparison the data obtained from the combination Elita/Cristina (A. murlcataJA. muricllla) are presented. 267 Greenhouse management of in vitro propagated soursop plants One important advantage of the in vitro propagated plants over plants propagated through traditional methods in nurseries, is the possibility to produce plants free of viral and other infectious diseases. This healthy state should be maíntained during the greenhouse stage in order to prevent the recontamination of the plants and the spread of diseases through the planting material. As a part of a M. Se. thesis (Silvio Cadena, Universidad Nacional de Palmira), the management of the micrografted plants under greenhouse conditions is being standardized. The effect of four substrate combinations (mixtures of "cahaza", "carbonilla" , both by-products of the sugar industry- Pinus bark chips and CIA T -soil), and three strains of mycorrhizal endosymbiontic fungi: Glomus deserticola, Gigaspora margarita and Gigaspora rosae on the development ofthe micrografted plants is being investigated. Cooclusions The in vitro micrografting propagation methodology developed initially with the clone "Elita" could be -applied to other different soursop clones. The plants produced by this way develop healthy and vigorous growing trees, which initiate flowering after 13 months in the field under CIA T conditions. Efficiency of development of micrografted plantlets in vitro should be improved in order to apply the technology to the massive propagation of soursop clones. Future plans To continue with the evaluation in the field ofthe propagated soursop plants To further optimize the in vitro growing conditions ofthe micrografted plants, in order to increase development efficiency of micrografted plants To scale up the production of micrografted plants. The objective for 2002 is the production of at least 3000 plants To standardize the methodologies of acclimatization and handling of the micrografted plants in the greenhouse To continue with the evaluation of the compatibility between clones of soursop and rootstocks of different Annonaceous species, micrografted in vitro To initiate field evaluation of novel selected soursop clones and include more farmers in this evaluation References Escobar, W. and L. Sánchez. Guanábano. Fruticultura Colombiana, ed. A. Morales. Santafé de Bogotá: Instituto Colombiano Agropecuario-ICA, 1992. George, E.F. Plant Propagation by Tissue Culture. Part 1 The Technology. Second (revised) edition ed., Edington, Wilts: Exegenetics Ltd, 1993. Nelson Royero, Alvaro Mejia, Gladys Perdomo, Jorge Cabra, William Roca (1999). Development and standardization of an in vitro clona) propagation method of soursop (Annona muricata /.). Annual Report CIAT Ríos-Castaí\o D. and C.E. Reyes. Guanábano Elita. Agricultura Tropical33 , 3 (1996):97-101. Royero N., Muíloz L., Mejía A., Cabra J., Roca W. (1998). Development and standardization of an in vitro clona/ propagation method of Soursop (Annona muricata L.). Annual Report CIAT. 268 2.2.16 Developing cryopreservation alternatives for tropical fruits useful in National Programs: The case ofTree Tomato ( Cyphomandra betacea). Montoya, J.E.1, Escobar, RH1., and Debouck O 1•2· 1 SB-2 Project ; 2SB-1 Project Introduction Based on our experience in cryopreservation of cassava (Escobar et al 1997, 2000), and keeping in mind that CIA T is adopting tropical fruits as part of its mandate crops, we selected Tree Tomato as a case study for cryopreservation. Our objective is design long-term conservation altematives, comparable or better than standard conservation rnethods, which can be transferred to National Programs in the region. Tree tornato is important for small-scale farmers of cold, tropical zones (average annual temperature: l6°C and 22°C; 1600-3000 m.a.s.l., and annual precipitation between 1300-1600 mm), representing an altemative incorne source. Methods We collected ripen fruits of three materials (yellow, !amarillo and comun) available in local markets of Palrnira and Cali. They were selected on the basis of color, appearance, brightness, phytosanitary state arnong other characteristics. Seeds were extracted, disinfected and dried on open petri dishes in a flow charnber for one hour. Then seeds were cryopreserved at least for one hour and thawed by warming them at 37°C for one minute. Seeds were then grown on germination rnedium for 40-50 days, and shoots transferred to rooting media before taken to the greenhouse. We run five replicas with 20 seeds for each oftwo treatments (frozen and not frozen) . Treatments were then evaluated based on germination percentages. Results and Discussion No significant differences were observed between frozen and non-frozen seeds for /amarillo and comun, while for yellow we found differences at 40-day evaluation (Table 1.) . Table l. Seed germination (o/o) after freezing Tree Tomato materials. (***, significaot differeoces; os .. ooo-significaot) Species Amarillo Tamarillo Común Frozen 37.4 % Frozen 86% Frozen 94 % Treatments Non-frozen 97.5 % Non-frozen 90% Non-frozen 95 % Probability (0,001) ••• ns ns We evaluated seed viability after freezing for Amarillo using the Tetrazoliurn Chloride test (TZ) (viable tissues stain red). The results are summarized in Table 2. 269 Table2: TZ test with non-genninated seed after freezing. Repetition total seeds Viable seeds Viability (%) R1 18 15 83.33% R2 17 15 88.24% R3 16 11 68.75% R4 18 14 77.77% R5 18 15 83.33% (a) (b) (e) Figure 1: Plants recovered after freezen steps growing in the greenhouse. (a) Tamarillo. (b) Amarillo and (e) Comun. Conclusions There are not significant differences in germination of frozen and non-frozen seeds for Tamarillo and Común. Amarillo showed germination differences between treatments seeds, possibly due to fast desiccation rather than loss of embryo viability. Future Activities Test slow desiccation methods for Amarillo Compare cryopreservation with short-to-medium term conservation (5 to -20°C) Transfer to National Germplasm Conservation Systems References Elias, S. and Garay, A. ( 1999). Tetrazolium test (TZ)- A fast, reliable test for seed viability determination. Seed laboratory, February. Issue ofthe crop & soil news/notes. Vol. 13, No. 2. p. 1-3. http://www.css.orst.edu/newsnates/9902/seed-lab.html Escobar, R.H. ; Debouck, D. and Roca, W.M. (2000). Development of cassava cryopreservation. En: Cryopreservation oftropical plant germplasm, current research progress and application. Tsukuba: Japan Intemational Research Center for Agricultura! Sciences, Roma: lntemational Plant Genetics Resources Institute, Engelmann, F. and Takagi, H. Editores. p. 222- 226. Standwood, P.C. and BASS, L.N. (1981). Seed Germplasm Preservation Using Liquid Nitrogen.En: Seed Sci. & Technol., Vol. 9. p. 423-437. Withers, L.A. (1993). New technologies for the conservation of plant genetic resources. USA: Crop Science L .. p 429- 435. 270 Activity 2.3 ldentification of points of genetic intervention and mechanisms of plant-stress interactíon Main Achievements • Characterizing genes of the carotene pathway in cassava was initiated using consensus sequences to amplify the genes -Lycopene Cyclase (BLyc), -Carotene desaturase (CDes), Phytoene synthase (PSyn) and Phytoene desaturase (PDes). • The genetic variation in minerals concentrations of 411 cassava genotypes was evaluated. A substantial genetic variation that can be exploited to improve cassava micronutrient density was determined. Genotypes with higher con~entrations of Fe, Zn and pro-vitamin A were identified • The genetic variation of carotene content in leaves and roots of 682 cassava accessions was determined. A significant correlation (0.84) was observed for carotene content and color intensity ofthe roots. Carotene concentration in roots ranged from 0.13 to 0.92 mg/1 00 g FW, with a mean of 0.23 and a standard deviation of 0.1 O. Carotenes concentrate much more in the leaves than in the roots, illustrating, once again, the excellent nutritive value of cassava leaves. There was no correlation (0.02) between carotene concentration in the leaves and roots. • The Identification of target points for the control of post-harvest physiological deterioration in cassava was pursued. Two classes of secondary metabolites, hydroxycoumarins and flavan-3-ols, were identified in deteriorated cassava root extracts. Those metabolites proved to be biologically active, as antioxidants and antimicrobials ( 4, 5). The dominant compounds were scopolin, scopoletin and (+)-gallocatechin. 271 2.3.1 Characterizing genes of the ca roten e pathway in cassava L.l. Mancilla, A.F. Salcedo, J. Beeching, L. Chavez and J. Tohme CIAT Introduction Vitamin A deficiency is an important public health problem, especially in tropical countries. Cassava is one of the most important sources of energy in tropical countries, although it lacks the carotene content necessary, in the roots, to supply mínima! carotene intake requirements. It would be desirable to breed cassava for increased caro ten e content in roots, taking advantage of the great genetic diversity that exists for this trait within cassava germplasm. One way to do it is by conventional breeding, which is the longest way. A second, faster approach is through genetic transformation, introducing and expressing genes to elevate carotene content in roots. Characterizing cassava genes of the -carotene pathway to understand their regulation in roots will help in the identification of root specific promoters that could be coupled to bacteria!- or plant-derived genes known to enhance carotene content in other species (Y e et. al., 2000). Thus, the main objective ofthis work is to characterize cassava genes ofthe carotene pathway. Methodology We used consensus sequences from GeneBank to design primers for PCR amplification of the genes -Lycopene Cyclase (BLyc), -Carotene desaturase (CDes), Phytoene synthase (PSyn) and Phytoene desaturase (PDes). Genomic DNA ofhigh (Mper297) and low (CM523-7) carotene content varieties was extracted and used for PCR amplification. Amplified bands were cloned into pGEM-T Easy vector. Insert sizes were confirmed with M13 primers. Additionally, RT-PCR was also made from mRNA of roots from both varieties. Results and Discussion Consensus primers were successful at amplifying all targeted genes. With the exception of Psyn, it was impossible to obtain a unique band. The optimal annealing conditions for PCR amplification were 52 °C for genes BLyc, PDes, CDes and 50 °C for Psyn (Figures 1,2). PCR product sizes from DNA and mRNA agreed with expect sizes based on consensus sequences. BLyc M CDes PDes M PSyn - . -- . . . -- - - _,....., _ .. -· ~- ~ .. . .... . . ~· -· -- ~- .. - -· ~ Figure l. PCR amplification of carotene genes from genomic DNA of cassava. 272 mRNA samples from CM523-7 and MPer 297 genotypes were RT-PCR amplified . Severa! bands appeared. Genes BLyc, PDes and CDes showed expect size fragments. PSyn gene did not amplify. Bands were observed in both high and low carotene content genotypes. Different RT- PCR programs with other primer cornbinations, and annealing temperatures, will be performed for PSyn amplification. MPer297 MPer297 CM523-7 Figure 2. RT -PCR of cassava roots mRNA with specific primers for carotene genes. l. BLyc , 2. Pdes, 3. Cdes, 4. Psyn, B. Blank, M. fPst l. Blyc CDes PDes C+ r 1:3 1:4 2 3:1 3:3' r 2:1 ''A '1:2 1:4 1:3 2:4 2:3' 2.1 0.793K Figure 3. Evaluation of carotene genes amplified by PCR with primers of pUC13/Ml3. Tbe control clone (C+) has an insert size of 542 bp. Currently we have cloned 50 percent of the bands into pGEM-T vectors. Inserts from carotene clones ha ve been confirmed by PCR using M 13 and gene-specific primers. Results are shown in Figure 3. Cooclusioos aod oogoiog Work. Severa! fragments were amplified by PCR with primers designed from consensus sequences of carotenoid metabolism genes from cassava genomic DNA of two varieties having contrasting carotene levels in roots. Sorne of the arnplified fragments have been cloned into pGEM T easy vector for sequencing and data base comparisons. RT-PCR ofmRNA from roots ofthese cassava genotypes was also performed with successful amplification of bands. We will clone bands from 273 RT-PCR into the same vector for sequencing. Additionally, genomic and cONA libraries will be constructed in GEMll and pSPORT-1 vectors, respectively, to fish out genomic and cONA clones. References Bartley G.E., Scolnik P. A., & Gluliano G. 1994 Molecular biology of carotenoid biosynthesis in plants Annual Review Plant Physiology. Plant Molecular Biology. by Annual Reviews 45 :287-301. Chavez A. L., Bedoya J. M., Sanchez T. Iglesias. C. Cevallos & Roca W. 2000 Iron, carotene, and ascorbic acid in cassava roots and leaves Food and Nutrition Bulletin vol21 , no 4. 410 -413 Cunnngharn F.X., Jr & Grantt E. 1998 Genes and enzymes of carotenoid biosynthesis in plants. Annual Review Plant Physiology. Plant Molecular Biology. by Annual Reviews 49:557-83. Gluliano G. Aqilani R. & Dharmapuri S. 2000 Metabolic engineering ofplant carotenoids. Trends in plant science. October, Vol5 No.10: 406-409. Y e X., et. al. (2000) Engineering the provitamin A (b-carotene) biosynthetic pathway into (carotenoid- free) rice endosperm. Science 287:303-305 2.3.2 Genetic variation in total carotenes and minerals of cassava genotypes L. Chávez,1 ;H.Ceballos,2 A. Bolaños, 2 F. Calle/ T. Sánchez,2 M.C. Duque1 and J. Tohme1 1SB-02 Project 2IP-3 Project; Introduction Deficiencies of vitamin A, iron and zinc are widespread in sub-Saharan Africa and in many tropical areas where the diets of poor human populations are mainly plant-based and the intake of animal-derived products are low. As cassava is a staple food in regions where there are severe micronutrient deficiencies, the crop could be used as a vehicle to deliver vitamins and minerals in higher concentrations. In the field of human and animal nutrition there is an increasing amount of evidence of a synergistic effect between vitamins and certain minerals. Prelimínary results suggest that Fe and Zn contents in the diet increase vitamin A absorption and vice versa. Thus in the study of micronutrient availability from cassava roots and leaves, it is also important to measure mineral contents. Objectives 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 bioavailability of micronutrients and vitamins. The objective of this research was to evaluate the extent of genetic variation of mineral concentrations in 41 1 genotypes from the world cassava germplasm bank held at CIA T. 274 Methods Mineral concentration measurement. Leaves and roots were collected, dried, ground to powder and sent to the Analytical Laboratory of the University of Adelaide (sampling was the same as that for evaluating carotene content; see "Evaluation of genetic diversity for total carotenes content in cassava leaves and roots". There the samples were analyzed by inductively coupled plasma atomic emission spectrometry. Care was taken in the processing to avoid contamination from soil, which has higher mineral concentrations than those of plant tissues. Carotene concentration measurements. The extraction protocol for leaves and roots is described in "Evaluation of genetic diversity for total carotenes content in cassava leaves and roots". Postharvest Physiological Deterioration (PPD) measurements. PPD was measured 7 days after harvest on genotypes whose concentration of carotenes in roots was measured simultaneously (see "Evaluation of genetic diversity for total carotenes content in cassava leaves and roots". Results Trace mineral concentrations {mglkg) Roots averaged 14.88 ofFe, 8.15 ofZn and 803.11 ofCa on a dry matter basis (Table 1). Mineral concentrations in the leaves were much higher, averaging 281.61 of Fe, 47.53 of Zn and 13760 of Ca. These leaf concentration figures are much higher than those in most staple foods . Although the leaves, which can be eaten as a vegetable, have a high water content and· low mineral density, they supply high levels of minerals per calorie. Table l. Simple descriptive statistics for mineral concentrations (mg/kg) of 411 genotypes of cassava Mineral Lea ves Roots M in. Max. Mean M in. Max. Mean Iron 120.63 950.00 281.61 5.99 75.92 14.88 Manganese 18.67 200.00 61.39 0.47 5.00 1.47 Boron 4.02 31.16 14.04 1.14 3.44 1.95 Copper 2.81 12.36 7.39 0.49 40.31 6.78 Zinc 15. 14 150.47 47.53 2.63 37.52 8.15 Calcium 6300 32000 13760 303 2500 803.11 Magnesium 2600 11300 4710 520 2400 - 1015.11 Sodium 10.27 113 29.47 18.56 1230 135.5 Potassium 8700 23000 15148 5700 25000 12381 Phosphorus 2300 7600 3901 980 3200 1720.07 Sulfur 2400 5200 3319 123 550 280.7 Aluminum 59.5 880 211 4.42 43.95 8.77 Correlations among minerals in roots and leaves. There was a high positive correlation between the mineral content in the roots and leaves for Mn (0.508) and K (0.407). The accumulation of the other minerals in roots and leaves seemed to be independent. 275 10l . g¡ . . . .. . CDrr. •.Q.2131 .. . ID . .. . . . 70 .. ,· ··: . . .,.,,. . .,.m • . ·¿.. •• • • ¡i~Sl \ . . -· # •• • •• .,. • ... 4J • 1 f: 1. •• • • • • '•\. ~~. . .. 3) ·' .:..; ........ ~ .. . ~~·· . 4l . ,'· ~t;)~·~r • 10 ,.... ~~ ¿r' . " . \ .. o .... ...... .,."~ . o SlD m:D 1S:X:O aiXD 2tiD :mx ao 10.0 20.0 :no 4lO filO mo 10.0 m K(tnfi'IOI/g~ Figure l. Relationship between PPD and K concentration in cassava roots. Figure 2. Relationship between PPD and Fe concentration in cassava roots. Fe C"V' 1CXIg ~ Correlations among minerals and carotene contents. There was a very weak relationship between total carotene content and Zn and Ca in the roots ( correlation coefficients of 0.005 and 0.1792, respectively). There was no correlation for Fe. With respect to the leaves, the correlations were significant for Mn (0.150), Na ( -0.148) and Ca (0.146). Correlation among minerals in roots and PPD. In general the correlations between PPD and mineral concentrations in the roots were low. The higher correlation coefficients found were negative for K ( -0.2936, Fíg. 1) and Fe ( -0.1854, Fig. 2). Thus it would seem that there may be an inverse relationship between these minerals and the PPD process. Cluster analysis. In order to identifying elite genotypes for breeding programs, this analysis constructs groups ( clusters) that ha ve interesting characteristics in common. Of the 15 clusters that were determined, the genotypes that had the highest concentrations of minerals among them were M Bra 11 and CM 6068-3. SM 722-13 hada low K concentration anda high PPD. M Col 2436, M Col 2439, M Col 2459 and M Bra 206 hada low PPD anda higher carotene content. Table 2 summarizes these data. These results demonstrate that there is a substantial genetic variation that can be ~xploited to improve cassava micronutrient density. New varieties with higher concentrations of Fe, Zn and pro-vitamin A can be identified to exploit the synergy that may operate in absorption, interna! transport and function among these micronutrients. Future Activities Search for a group of cassava cultivars with a higher micronutrient content to help in the breeding approach to improve their micronutrient content Evaluate the expression and stability of minerals in elite genotypes under different environmental conditions 276 · Table 2. Genotypes with common characteristics identified from the cluster analysis for cassava roots. Genotype Characteristic MBra 11 High Fe CM 6068-3 High Zn HiiZh Na, K, Cu, P, S and Al SM 722-13 LowK Hi~h PPO M Col2436 High carotene M Col2439 HighK M Col2459 HighB M Bra206 LowPPO CM 5460-5, M Bra 49 LowK . M Bra 88, M Bra 89 LowB M Bra 90, M Bra 103 High PPO M Bra 104, M Bra 194 M Col 480, M Col 84 7 M Col 2456, M Ecu 6 M Col 2389, M Cub 35 M Bra 122, M Bra 465 High carotene M Bra 468, M Col 1995 High Ca M Col 2272, M Col 2363 High Mg M Col2412, M Col2528 M Bra 214, M Col2289 References Chávez, A.L.; Bedoya, J.M.; Sánchez, T.; Iglesias, C.; Ceballos, H.; Roca, W. 2000. lron, carotene, and ascorbic acid in cassava roots and leaves. Food and Nutrition Bulletin 21(4):410-413. Zarcinas, B.A.; Cartwright, B.; Spouncer, L.R. 1987. Nitric acid digestion and multi-element analysis of plant material by inductively coupled plasma spectrometry. Cornmon Soil Sci Plant Anal 18: 131-146. 2.3.3 Evaluation of genetic diversity for total carotene content cassava leaves and roots L. Chavez,1 H. Ceballos2, A. Bolaños, 2 F. Calle,2 T. Sanche:i and J. Tohme1 1SB-02 Project; 21P-3 Project; Introduction Cassava provides a large proportion of the daily intake of energy and other nutrients including vitamins for poor populations in many areas of sub-Saharan Africa. The cassava crop-given its flexibility with respect to planting and harvesting times and its tolerance to poor soils, pest and disease problems-plays an important role in food security as a stable food base in areas prone to drought. Root crops are usually considered primarily sources of low-cost energy, but not important sources of other nutrients. In the case of cassava, however, CIA T has found a wide range in the content of total carotenoids after screening its germplasm collection (CIA T 1999, 2000). Developing 277 cassava cultivars with high available carotene content could significantly improve the health and nutritional status ofthe poor, especially women and children. Postharvest physiological deterioration (PPD) is a constraint that limits the marketability of the fresh cassava roots and increases the costs of postharvest handling. Consequently, the roots must either be consumed or processed shortly after harvesting. In this study we evaluate the hypothesis that a high carotene content could help reduce PPD through its antioxidant capacity. Objectives The overall objective of this project is to improve the nutritional status of people living in marginal environments of the tropics, by selectíng and promoting cassava genotypes with high and good bioavailability of micronutrients and vítamíns.The specific objective of this research is to determine the extent of genetic variation of carotene content in 682 accessions from the cassava world germplasm bank held at CIA T. Metbods Harvesting and sampling. A selection was made of 682 cassava genotypes to represent the wide variability in carotene content and PPD susceptibility from the CIA T germplasm collection (6000 genotypes). Three plants per cultivar were harvested at 9-12 months of maturity. Carotene concentration measurements. The extraction procedure outlined by Safo-Katanga et al. (1984) was modified by extra"cting root parenchyma with petroleum ether. The extraction protocol for leaves had to be modified due to the presence of tannins and chlorophylls. The modified protocol included several extractions with petroleum ether at 35-65°C and washing with methanol in order to minimize interference from the other pigments. A 5-g sample was taken from randomly selected root or lea ves, 10-11 months after planting. The quantification was done by ultraviolet spectrophotometry, using a Shimadzu UV-VIS l60A recording spectrophotometer. UV detection was done at 1 = 455 nm for root extracts and 1 = 490 nm for leaf extracts. PPD measurements. Five commercially sized roots (mínimum length 18 cm) were randomly chosen and analyzed using the method of Wheatley et al. (1985) with one modification: The prepared roots were stored under ambient conditions for 7 instead of 3 days. The proximal and distal root ends were cut off, and the distal end was covered with Clingfilm. After 7_ days, seven 2-cm thick transversal slices were cut along the root, starting from the proximal end. A score of 1-1 O was assigned to each si ice, corresponding to the percent of the cut surface showing discoloration ( l = 1 0%, 2=20%, etc). The mean PPD score for each root was calculated. Results Analysis of carotene content. Carotene concentration in leaf tissue ranged from 18.71 to 96.2 mg/100 g FW, with a mean of 50.32 mg/100 g FW anda standard deviation of 10.43 (Table 1). These values are similar to those found in a group of 500 accessions evaluated the previous year (CIA T, 2000). The carotene distribution showed a symmetrical tendency (skewness = 0.2). Carotene concentration in roots ranged from 0.13 to 0.92 mg/1 00 g FW, with a mean of 0.23 and a standard deviation of 0.10 (Table 1). This distribution showed a strong skewness with long right tails (skewness = 2.83). A significant correlation (0.84) was observed for carotene content and color intensity of the roots. Carotenes concentrate much more in the lea ves than in the roots, illustrating, once again, the excellent nutritive value of cassava leaves. There was no correlation (0.02) between carotene concentration in the leaves and roots. 278 Table l. Carotene concentration in lea ves and roots of 682 cassava accessions from CIA T's germplasm bank collection. Data from Leaves Data from Roots Ran_g_e _{_ mg/1 00 FW) Frequency Range (mg/100 FW) Frequency 18.71-26.46 6 0.13-0.2 1 437 26.47-34.22 38 0.22-0.30 166 34.23-41.98 94 0.31-0.39 29 41.99-49.74 197 0.40-0.48 19 49.75-57.5 184 0.49-0.57 18 57.6-65.35 115 0.58-0.66 9 65.36-73.11 35 0.67-0.75 2 73.12-80.87 10 0.76-0.84 2 80.88-88.63 2 0.85-0.93 1 >88.64 1 >0.93 o Minimum 18.71 Mínimum 0.13 Maximum 96.2 Maximum 0.92 Median 57.45 Median 0.525 Skewness 0.2 Skewness 2.83 Mean 50.32 Mean 0.23 so 10.43 so 0.10 Correlations among vitamin content and PPD. The correlation between PPD and carotene content in cassava roots was -0.07 (Fig_ 1). This relationship is similar to the one reported previously (- 0.13). Although a high correlation between carotene content and PPD was not observed, it was noted previously (Chávez el al., 2000) that at carotene concentrations >50 mg carotene/1 00 FW in the roots, the PPD did not exceed 20%, suggesting a threshold effect (Table 2). Nevertheless, the results illustrated in Figure 1 clearly contradict previous findings. lt has been founthat PPD studies are affected by large experimental errors arising from environmental variations at the time the evaluations are carried out. Because of logistical limitations, gennplasm bank accessions must be harvested gradually over time; therefore the PPD evaluations are carri.ed out at different times and environmental factors (particularly temperature ). Beca use of this situation CIA T built a large chamber where temperature and relative humidity can be controlled. In the future all PPD evaluations will be perfonned under unifonn environmental conditions. ' .. •• •• J 10 1 .. .,. •• •• •• .. 0 . 10 0 . 20 0 . :10 0 . 4 0 o 1 0 0 . 1 0 o 70 0 . 1 0 0 . 1 0 1.0 0 corote noo lll g / 1 00g FW Figure l. Relationship between PPD and carotene content in cassava roots. 279 Table 2. Cassava genotypes with high carotene content and low PPD. Genotype PPD(%) Carotene mg/1 OOg FW M Col2086 5.7 0.51 CM 3199- 1 19.5 0.51 MCol676 19.0 0.51 M Col2068 5.7 0.53 M Bra487 20.0 0.55 M Col2498 4.3 0.57 M Col2099 9.5 0.57 CM 5655- 1 4.3 0.58 M Bra465 10.9 0.63 M Coi24IO 8.1 0.65 M Col2459 4.3 0.66 M Col2412 7.9 0.70 M Bra206 6.7 0.72 M Col2436 8.3 0.83 M Col2439 11.4 0.92 Future Activities • Continue screening the cassava landraces ( only those with nonwhite roots) from CIA T' s germplasm collection, exploring the genotypes with higher carotene contents. • Identify a group of clones with the highest carotene content to help in the plant breeding approach for improving the micronutrient content of cassava. • Evaluate the expression and stability of high concentrations of carotene in elite genotypes under different environmental conditions. • Determine the proportion of carotenoids (given that only total carotenoids have been identified) with pro-vitamin A activity present in cassava roots and leaves References Chávez, A.L.; Bedoya, J.M. ; Iglesias C.; Ceballos, H.; Roca, W. 1999. Exploring the genetic potential to improve micronutrients content of cassava. Improving human nutrition through agriculture. Los Baños, Philippines. Chávez, A.L.; Bedoya, J.M.; Sánchez, T.; Iglesias, C.; Ceballos, H.; Roca, W. 2000. Iron, carotene, and ascorbic acid in cassava roots and Ieaves. Food and Nutrit ion Bulletin 21{4): 410-413 CIAT (Centro Internacional de Agricultura Tropical). 1999 - 2000. Annual report. Cali, CO. Graham, R. ; Senadhira, D.; Beebe, S.; Iglesias, C.; Monasterio, l. 1999. Breeding for micronutrient density in edible portions ofstaple food crops: conventional approaches. Field Crops Res 60:57-80. Safo-Katanga, 0 .; Aboagye, P.; Amartey, S.A.; Olaham, J.H. 1984. Studies on the content of yellow- pigmented cassava. In: Terry, E.R. et al. (eds.). Tropical roots crops production and uses in Africa. IDRC, Ottawa, Canada. p. 103-104. Wheatley, C.; Lozano, C.; Gomez, G. 1985. Postharvest deterioration ofcassava roots. In: Cock, J.H.; Reyes, J.A. (eds.). Cassava: Research, production and utilization. UNDP-CIAT, Cali, CO. p. 655-671. 280 . 2.3.4 Enabling Genomics Tools for Understanding and Exploiting the Genetic Potential of Cassava Starch Cortés Diego Fernando, Jorge Veronique1, Verdier Valerie1 & Joe Tohme 1 Université de Perpignan, France. 2IRD Institute pour la Recherche et le Developpement, France. Introduction Expressed sequence tags (EST's) as candidate loci as quantitative traits, has been suggested as a way of increasing the accuracy of mapping complex traits. The EST's simplifies and directs genomics sequencing and isolation and cloning genes of agronomical interest genes. Nonnally the generation of EST's has been by sequencing of random cONA clones from libraries obtained from different tissues at various stages of development. Constructing cONA libraries from tissues and developmental stages are keys for study certain traits. Starch is certainly the principal carbohydrate storage fonn of CG mandate crops. Sorne of the key steps of the starch biosynthesis and deposition are well understood from the work in grain crops and root crop, such as cassava. A genome-wide gene expression approach will provide insights in the metabolism of starch in different crops. As one step to achieve this goal, a cONA library from high and low starch content cultivars has been constructed. Materials and Methods Source of plant tissue for the construction of the cONA library were storage roots from plants 1 O months of age. MPer 183 (low starch contend) and CM523-7 (high starch contend) were the cassava cultivars used in this study. Fresh roots from each cultivars were harvested and immediately storage at -80°C. Root tissue was ground to fine powder using liquid nitrogen and 3 g used for RNA extraction with lithium chloride and cleaning steps with phenol:chloroform. Only total RNA samples with high purity (A260/A280 ratio 1.8 -2.0) and good concentration were selected for mRNA isolation using magnetic poly A DYNAbeads according to the manufacture. The cONA synthesis and clonini was done using the Stratagene cONA Synthesis Kit, ZAP-cDNA ~ Synthesis Kit and ZAP-cDNA Gigapack~ m Gold Cloning Kit according to the manufacture. Results Prior to cloning, low molecular weight (less than 500 bp) cDNA was removed by size fractionation and the remaining fraction of cONA was divided in two portions for each genotype, the frrs portion contained fragments between SOObp and 1500bp and a second portion with fragments longer than 1500bp. The cONA synthesis was primed with oligo (dt) primer which contaíns a Not 1 restriction site and an adaptor on the other site which contains the apropiate EcoR 1 site, allowing that the libraries were directionally cloned. Alllibraries were obtained with a mínimum titer of 108 pfulml. The amplified libraries have titer of 108 to 1010 pfulml. This quantity is sufficient for severa! thousands screening. An insert size screening after cloning and the average obtained was 1 000 bp. Both cONA librarles were transferred to IRD (Francia) for the differential subtraction and sequencing the clones differentialty expressed, besides a copy of these libraries are still in CIA T for future activities 28 1 Future Activities • Identification of clones differentially expressed between the cultivars with high and low starch content. • Generation of thousands of EST' s sequen ces. • Establishment of starch related sequences obtained from public domain and EST's generated from the project. • Comparative analysis between cassava EST and with available dbEST's from public accessible databases. 2.3.5 Functional Geno mies Tools of Post-Harvest Physiological Deterioration in Cassava Cortés Diego Fernando, Gomez-Vazquez Rocio\ Beeching John1 & Joe Tohme 10epartment ofBiology and Biochemistry, Bath University, U.K. Introduction The developing of genomics and bioinformatics tools is increasing our knowledge of plant genome structure, organization and gene function. Novel technologies such an Expressed Sequences Tags (EST's) and cONA micro arrays are proving rapid ways to identify genes and to link sequence information to biological function. The post-harvest physiological deterioration (PPD) is a major constrain to the development of cassava for producers, processors and costumers alike. Extending the shelf-life of cassava to one or two weeks is perceived as a goal that would benefit many, particularly the sustainable livelihoods of small-scale rural farmers and would contribute towards poverty alleviation. The objective of this project is to identify the full set of major genes in volved in the response by exploiting the powerful high throughput analysis of cONA microarrays. This will enhance understanding ofthe problem and also provide the tools (clones) that could serve as components of gene constructs to modulate PPD. On this report we presentan important step to make this goal achievable, like is the development of a cONA library from different time-points during the PPD process in cassava roots. Materials and Methods Cassava roots from the cultivar CM2177 -2, the mal e parent of the F 1 cross used in generation the molecular genetic map of cassava, was used as source of plant tissue for the RNA extraction in the cONA library construction. Roots from the cultivar CM2177-2 were harvested from plants with 1 O months of age. The RNA was isolated form the fooling time-points after harvest: O, 3, 6, 12, 24, 48 and 96 hours. Poly (A)+ RNA was purified using magnetic poly A DYNAbeads according to the manufacture. 282 The cONA synthesis and cloning was done using the Stratagene cONA Synthesis Kit, ZAP~ cONA~ Synthesis Kit and ZAP~cDNA ~ Gigapack~ III Gold Cloning Kit according to the manufacture. Results Two directionally cloned cONA libraries in Lambda ZAP 11 were obtained: "Early" from pooled O, 3, 6 & 12 hour time points, and "Late", from 24, 48 & 96 hours. Based on size fractionation, low molecular weight (less than 500 bp) cONA was removed prior to cloning and the remaining fraction of cONA was divided in two portions according the fragment's size, the firs portian contained fragments between 500bp and 1500bp and a second portian with fragments longer than 1500bp. The two "Early and Late" original libraries were obtained with a ti ter of 108 pfu/ml. The amplified libraries ha ve ti ter of l. 7 x 109 and 7 x 108 pfulml respectively. This quantity is sufficient for severa! thousands screaning. Both cONA libraries were transferred to Dr. John Beeching's Laboratory in Bath University (United Kindong) for a specific gene screening and random clones sequencing A copy of these libraries are still in CIA T for future activities · Previous research showed that the key signalling events that trigger post-harvest physiological deterioration (PPD) in cassava occur during the earliest stages and that the lack of adequate wound repair in the detached root permits the spread of the deterioration response. The experimental work was to address these aspects. The 1ibraries were screened with PCR-generated subtraction probes enriched for PPO-related sequences. In addition we randomly se1ected clones from both librarles. The subtraction probes hybridised to 30 different clones, which were sequenced and characterised. Surprisingly 20 of these were for members of an extensin gene family, which fell into four classes. Extensins are proteins that are insolubilised in the plant cell wall by H20 2 during normal development and in response to wounding; they play a role in cell wall strengthening. This predominance of extensins could reflect their PPO-specificity or be an artefact of the PCR method used to produce the subtraction probes. Nonetheless, these results do demonstrate the activity of a large gene family of extensins during the early stages of PPD. Approximately 70 random clones from both librarles were also sequenced and characterized. The total of 100 clones sequenced included members of the following classes: cell wall strengthening, signa! transduction, stres_s responses, senescence, defence, metabolism, transcription, translation, and the frrst evidence of the activity of a Mutator-like transposable element in cassava. This and previous work had suggested that sorne aspects of wound repair ( cell wall strengthening) occurred during PPD. Therefore, we used immunological methods to detect extensin accumulation during PPD. These data confmned our supposition, showing an accumulation of extensin in the root, particularly associated with the vascular tissue; an accumulation that paralleled the development of PPO symptoms. However, the response occurred too late to heal the wound sites and thereby inhibit PPO. Future Activities cONA microarray spotting. 5,000 clones from each library will select after mass excision, PCR amplified and spotted onto replicate glass slides using the SPBIO spotting robot at CIA T. Interrogating of microarrays. Cy3 and Cy5 labeled cONA derived from poly(A)+ of time points of interest will be used as probe to hybridize cONA slides. 283 ldentification of clones of interest. Computer analysis ofthe microarray hybridization output will be use to identify clones that change in abundance during the time course ofPPD. 2.3.6 ldentifying target points for the control of post-harvest physiological deterioration in cassava M. X. Rodríguez1, H. Buschmann2, E. Okogbenin1, J. Tohme1 and J. R. Beeching2 1SB-02 Project; 2University ofBath, U.K. Introduction The major limitation for the development of cassava from a marginal farrners crop to an urban important crop is the rapid post-harvest deterioration starting 48 hours after harvesting, which renders the roots unpalatable and unmarketable for consumption and industrial utilization (6). The post-harvest deterioration process in cassava commences with a physiological deterioration (PPD), characterized by vascular streaking (blue-black discoloration of the xylem parenchyma followed by a general discoloration of the storage parenchyma). Subsequently, a microbial deterioration occurs. This secondary deterioration is a decay caused by the invasion of micro- organisms throughout wounds caused during root harvesting, handling and transport (2, 6). PPD is an active process involving the increase and de novo synthesis of proteins and changes in gene expression, which resembles wound responses in other more studied plants. Nevertheless, in cassava the wound repair and the resultant down-modulation of the signals are inadequate, which leads to continuous cascades of wound responses that spread throughout the cassava root (observed as PPD) (1, 2). The aim of this project is the identification of biochemical markers for post harvest physiological deterioration, studying the occurrence of secondary metabolites and enzymatic activities that may play a determinant role during PPD process. The identification of these PPD markers will generate the context necessary for the development of screening methods to explÓit the existing genetic diversity with respect to PPD. These screening methods will also help to identify cultivars that combine agronomic performance and reduced post-harvest deterioration. Methodology As previously reported, the strategy is based on the evaluation of cassava cultivars with contrasting responses to PPD, but following a deterioration time course of four days in order to avoid possible overlapping of defense responses caused by microbial deterioration. The biochemical tests are the quantification of coumarins (scopolin, scopoletin and esculin) and enzyme activities for Polyphenol oxydase (PPO), Peroxidase (POX), and Scopoletin-peroxidase. Taking advantage CIA T is generating a cassava genetic map and the parentals mapping cross ha ve different responses to PPD, a representative percentage of the population was sampled to try to determine the role of major genes influencing the PPD response. Comparing the distribution of the PPD response presented by the family K between the two different ecosystems, CIA T Palmira and CIA T Quilichao ( evaluation performed in 1998), CIA T -Palmira was chosen as sampling site because the deterioration score presented a normal distribution. Genotypes from the extremes ( 15 individuals) and the middle (five individuals) ofthe normal distribution were selected. From the three different groups the genotypes showing the minimal standard deviation between replications were chosen. 284 Results As reported before, two classes of secondary metabolites, hydroxycoumarins and flavan-3-ols, were identified in deteriorated cassava root extracts. Those metabolites proved to be biologically active, as antioxidants and antimicrobials ( 4, 5). The dominant compounds were scopolin, scopoletin and (+)-gallocatechin. Scopolin and scopoletin accumulate two to three days after harvest, whilst (+)-gallocatechin accumulates after four to six days. Based on this accumulation time, it was assumed that the coumarins are related to PPD while th_e flavan-3-ols are more related to microbial deterioration (3, 5). The analysis of al! root extracts chromatographic profiles showed two peaks, in the area of the non -polar compounds, which occurs with the onset of PPD. The structure elucidation of those candidates for PPD markers has not been successful. Besides that, the more prominent peak was quantified in terms of scopoletin. Figure 1 shows a major accumulation of the PPD peak in the cultivars wjth high response towards PPD, which may suggest this metabolite a good candidate for biochemical marker. "PPOrnutW" In FamllyK ~ 2.5 2.5 - 2 d ~= .ll • 2 i!'ii e .;1.s f: ~ '! 1.5 E ·; Cl"' .... l--u...PI'O --· -PI'O ·· ·•· ·· H¡11PI'O o.." o! 1 e& e; o " . -o.s .. ... · . · .· ____ .-- ____ PI'O ~ -: 1i 1 .. ! 0.5 .. ···,.···· _.,..- ¡ --LMPI'O ............ ..:<·~---- ···•··· Hg!PI'O o~-=~~~~~~----~~==~--~¡ 3 -...&.doy o 3 o Figure 1. "PPD marker" quantification in Scopoletin equivalents of different cassava cultivars anda percentage of Familiy K population with contrasting responses to PPD. Enzyme activities in crease from the initial stage of the deterioration. Only POX presents a small decrease in activity by the end of the time course period, and the differences between cultivars with contrasting PPD responses are not as clear as with the other enzymes. Previously was mentioned that peroxidases, in cassava roots, catalyse a reaction between scopoletin and H20 2 resulting in a black precipitate. The Iocalisation of POX activity around the vascular parenchyma suggests a correlation between PPD and oxidation of hydroxicoumarins. As well, the increase of scopoletin-peroxidase could clarify the decrease of scopoletin at the lasts time course days (Figure 2). The Family K was tested in detail for PPD damage. The entire population was grown, at the same time, in two different agro-ecological sites, Palmira and Quilichao, and during two following years, 1998 and 1999. Figure 3 shows the PPD response frequency for both planting locations and periods. Besides conclude PPD response is hereditable, is extensively affected by the environrnent. Even considerable differences in climate and soil factors exists between Palmira and Quilichao, it was not determined a specific environmeñtal factor that could explain the genotype by environment (GxE) interaction. 285 The PPD scoring method used at CIA T, for more than 20 years, has been considered subjective and non reliable, therefore this project is looking for a biochemical analytical method which grants an objective and accurate measure of the PPD response. All tests are subject to large variability between roots ofthe same plant and between those of genetically identical plants. For that reason, it was assumed that the enormous variation, which indeed complicates the collection of statistically significant data, is more related to environmental factors than the specific analytical method used. While a rapid and economic biochemical assay can be developed, the traditional visual method should be used. eo ~50 g o 1<40 -ª ~30 e fl e o u e20 j o Q. ~ 10 Scopoletin accumulation and Scopoletin.Peroxidase activlty comparison for Famlly K genotypea -sCPLowPPO - ..... - SCP Medium PPD .... . o o o .o ... o.... .. .... . .. .... , • / f¡ .•. ¿..o SCP Hi"' PPO •· ··. o • • ,- ~SCP.POXL- PPD • • 0:;,~;.~0. -- SCP.POXMedium PPO ·' / ··Ir· SCP.POXHi"' PPO ;-------- --,:~~~--- ,,,,''/ .;,·,. .· ,." ... ,,.' _ .. - 0.04 0.035 o.o3 S' t>l 0.025 ~ ~ (~ 3 ~ o.o2 a ;_ .. Q. 0.015 ~ 5 ~~ 0.01 ! 00005 0~----------+-----------~----------~----------~-----------+ o o 3 time cou,... day Figure 2. Comparison between scopoletin (SCP) accumulation ·and scopoletin-peroxidase (SCP- POX) activity during a time course of four days, between a selected percentage of Family K population with different responses to PPD. PPDPALMRA PPO QJILJCHAO Figure 3. Comparison of distributions of the PPD response amongst the Family K mapping population in two contrasting agro-ecologies in Colombia in 1998 and 1999. 286 On going activities Present and future activities are the detailed analysis of all data compiled during the three years of research References Beeching, J. R., H. Buschmann, R. Gómez-Vásquez, Y. Han, C. Iglesias, H. Li, K. Reilly, and M. X. Rodríguez. 1999. An abiotic stress response in cassava: post-harvest physiological deterioration, p. 205-210./n: M. F. Smallwood, C. M. Calvert, and D. J. Bowles (eds.), Plant Responses to Environmental Stress. BIOS, Oxford. Beeching, J. R., Y. Han, R. Gómez-Vásquez, R. C. Day, and R. M. Cooper. 1998. Wound and defense responses in cassava as related to post-harvest physiological deterioration. Recent Advances in Phytochemístry 32:231-248. Buschmann, H., M. X. Rodríguez, J. Tohme, and J. R. Beechíng. 2000. Accumulation of hydroxycoumarins during post-harvest deterioration oftuberous roots of cassava (Manihot escu/enta Crantz). Annals ofBotany 86:1153-1160. Buschmann, H., K. Reílly, M. X. Rodríguez, J. Tohme, and J. R. Beeching. 2000. Hydrogen peroxide and flavan-3-ols in storage roots of cassava (Manihot esculenta Crantz) during postharvest deterioration. Journal of Agricultural and Food Chemistry 48:5522-5529. Rodríguez, M X, Buschmann, H, Tohme, J, and Beeching, J R. Production of anti-microbial compounds in cassava (Manihot esculenta Crantz) root during post-harvest physiological deterioration. Carvalho, L J C B, Thro, A M, and Vilarinhos, A D. Cassava Biotechnology. IVth Intemational Scientific Meeting Cassava Biotechnology Network. pp. 320-328. 2000. Brasilia, Embrapa. Wenham, J. E. 1995. Post-harvest Deterioration ofCassava. A Biotechnological Perspective. FAO, Rome. 287 \ Output 3. Collaboration with public and private partners enhanced Activity 3.1 New collaborative arrangements and organization of workshops and training courses Main Achievements • During the period of Oct 2000-2001 a total of more than 70 people (researchers, joumalists, visitors and others) received training with SB-2 Project staff. • A second workshop on Biotechnology and GMOs biosafety was given by CIAT to Colombianjournalists. • Two training courses were held at CIA T to train and upgrade SB-2 staff knowledge. One training was on molecular approaches for disease resistance, and modulating gene expression in transgenic plants was conducted at CIA T in collaboration with Eric Lam and Nilgun Tumer from the Biotechnology Center for Agriculture and the Environrnent Rutgers University, New Jersey, USA. Total of 43 participants including 37 CIA T Support Staff and 6 Principal Staff attended the course. • The-second course was on the use of micro array and the use of a novel technology. Diversity Array technology (DaRT). The course was given by Damian Jacouq from CAMB lA Australia and was attended by sorne 20 assistants from SB-2. • CIA T obtained approval of a project from BMZ, Germany, on " An Jntegrated Approach for Genetic Jmprovement of Aluminum Resistance of Crops on Low-fertility A cid Soils" This proposal addresses a major strategic research issue through research partnerships linking CIAT, a regional network in Africa, two systemwide programs and an advanced research organization in Germany. The proposal will use a multidisciplinary approach to integrate the scientific capacity of both CIA T and the University of Hannover to alleviate a major soil constraint to agricultura! productivity on Iow-fertility acid soils of Africa and Latín America. • A collaborative project between Yale University and CIAT on insertion mutagenisis in rice using Ac-DS system was approved for funded by the USDA, opening the way to strengthen CIAT capacity in functional genomics. • A collaborative project between CIA T and the University of Costa Rica on Breeding, biosafety and deployment of RHBV resistan! transgenic rice was approved by the Rockefeller Foundation. • With funding from the Rockefeller Foundation, SB-2 organized a legume genomics meeting between the CG working on legumes and US Universities resulting in the preparation of a common research agenda and a collaborative proposal. • Staff of SB-2 organized the CG planning workshop for the biofortification proposal. • A database for bean microsatellites was established. • Al1 updated version of FloraMap was released in 200 l. Sorne 200 registered users from severa! countries ha ve obtained a copy. • Two assistants from the genome lab received full fellowships to attend visit a technology lab in Canada and to get trained in the development of fulllength cONA libraries. • In the period of Oct 2000-Sept 2001, SB-02 projects members published 40 Scientific papers in referred journal and books; abstracts and posters in conference. At he same time staff members gave keynotes and plenary presentations at Intemational and regional meetings. • In the period of Oct 2000-Sept 2001, projects members increased the contacts with prívate sector at the regional and intemational leve! to establish collaborative projects on technology transfer and to obtain freedom to operate for key technologies. Same kind of activities were established with regional NGOs in order to transfer CIA T technologies. • The Cassava Biotech Network was reestablished with funding from DGIS and IDRC and with a focus for Latín America. • In the same period 5 proposals were approved and I 1 more submined • A total of 24 organizations contributed to funding projects in SB-2. 288 3.1.1 Highligbts From The Cassava Biotechnology Network's Activities For 2001 Chusa Ginés SB-2 Project Introduction The CBN initiated activities under the project The Cassava Biotechnology Network in Latin America: Strategies for Integrating Sma/1-Sca/e End-Users in Research Agenda-Setting, Testing and Evaluation, jointly funded by DGIS and IDRC. The Coordination Office is based in Quito, Ecuador, a step taken in the process of decentralization of the Network to be eventually hosted with a Latin American organization. The Coordinating Unit - now composed of Dr. Chusa Ginés, the Coordinator, and Verónica Mera, Social Scientist- established contact and built the collaborative basis with the potential pilot sites' organizations in Brazil, Colombia, Cuba, and Ecuador. The development of proposals and funding of small projects took place. The pilot sites "reconnect" with activities funded by the previous phase of the CBN. As well, the projects in Brazil and Colombia buíld on the collaboration with the PRGA, to take advantage of the synergistic opportunities for development of participatory methodologies. This usually involves capacity building on participatory and gender-sensitive methodologies and better integration of multidisciplinary teams. In the case of Brazil, the use of biotechnology will support the successful Participatory Plant Breeding Program of EMBRAP A Mandioca e Fruticultura. The focus is to stimulate local research on anchorage and adaptation ofPRGA- Jínked biotechnology tools in key pilot sites. The situation in Ecuador presents both a challenge and an opportunity. There is no organization with the specific mandate of doing R&D in cassava, and activities are fragmented. Therefore the CBN initiated a multi-stakeholder analysis, lead discussions with a variety of organizations involved with cassava activities in different regions and constituted a working group to carry out a diagnostic study on the Status of Production and Use of Cassava. This work has started in sorne depth in the province of Manabí. The results will be integrated into a GIS platform to improve targeting of activities and to increase the accuracy of analysis of crop performance. The outputs of GIS-based approach could include targeting and spatial analysis of potential 1 limitations of cassava production; analysis to identify possible genotype x environment interactions; definition of spatially variable risks, especially associated with future climate change. As well, socio- economic variables will be integrated. This study will form the basis for developing and integrated user-needs analysis, and ultimately, a more in/depth analysis on stakeholder involvement in priority setting and evaluation. Presentations on the project were made at the following international meetings: Programa de Biotecnología Agrícola, in Cartagena, Colombia, Feb. 200 l . Possibilities for collaboration were explored with this DGIS-funded program with the intention to develop a joint research proposal to seek other funding and conduct a comparative analysis of the different methods of microprogation RedBio meeting in Goiania, Brazil, June 2001- CBN co-organized a session on Cassava Biotechnology with presentations by CBN partners. 289 CBN V meeting in S t. Louis, Missouri, 4-9 November 200 l. This meeting served to invigorate the network and initiate the creation of a new vision for CBN. The program included sessions on participatory research and technology development. International Society for Tropical Root Crops-Africa Branch (ISTRC-AB). 8th Symposium held at UTA, Ibadan, Nigeria, 12-16 November 2001. This meeting served to establish an exchange with other cassava researchers in Africa and start the formulation of a collaborative program for the CBN Network in Africa. Sociedad Latinoamericana de Raíces y Tubérculos (SLART) meeting in Lima, Perú 28-30 November 2001. The objective was to strengthen the links with liked-minded organizations and projects in Latín America. Overall, the first year of the project was focused on establishing a sol id base for a decentralized regional network, and stimulating interdisciplinary collaborations for the development of a conceptual-framework on participatory biotechnology. 3.1.2 Collaboration with Public and private sectors • A CG planning meeting at CIA T was organized by SB-2 to prepare a global pro posa! on: "Harnessing agricultura! technology to improve the health ofthe poor: "biofortified" crops to combat micronutrient malnutrition. The proposed project seeks to bring the full potential of agricultura! science to bear on the persistent problem of micronutrient malnutrition by integrating breeding, nutrition, nutritional genomics, participatory and communities leve! research. CIA T and the International Food Policy Research Institute (IFPRI) will co- coordinate this inter-disciplinary effort among plant scientists, human nutritionists, and social scientists. CIA T will coordinate the breeding and biotechnology components while IFPRI will coordinate the nutrition and policy parts. • To reinforce this global project SB-2 staff made a series of visit to Michigan State University, Cornell, IFRPI, Yale University, USAID and University of Frieburg to discuss the science and collaboration and establish the framework for the implementation of the project. • SB-2 staff attended the Cereal genomics meeting organized at CIMMYT by Dr. Bob Zeigler from Kansas State University. The meeting was between CG and US researchers and resulted in the formulation of a research proposal. • An intensive course on molecular approaches for disease resistance, and modulating gene expression in transgenic plants was conducted at CIA T in collaboration with Eric Lam and Nilgun Tumer from the Biotechnology Center for Agriculture and the Environment Rutgers University, New Jersey, USA. • With funding from the Rockefeller Foundation, SB-2 staff organized a legume genomics meeting between the CG working on legumes and US Universities resulting in the preparation of a common research agenda anda collaborative proposal. Sorne 30 researchers attended from four CG centers (CIA T, !CARDA, IRCISA T, liT A) and severa! US universities. Washington, D.C., USA, Aug 19-24, 2001. 290 • SB-2 Staff prepared with researchers from the Danforth Center a framework proposal for a Global Cassava biotechnology initiative. The initiative also involved liT A, EMBRAP A and FAO. SB-2 staffvisited the Danforth Center, Washington University sequencing centers, and F AO to prepare the proposal. Other meetings and interactions included: • SB-2 staffvisited CIRAD and IRD to discuss ajoint T-DNA project, Sept, 2001 • SB-2 staff visited University of Freiburg to explore possibilities of strategic alliance with CIAT. Cassava breeding through Biotechnology. Germany. September 9-11, 2001 • Clemson, South Carolina, USA, Aug 26-30, 2001, to visit collaborators at Clemson Unversity Genomics lnstitute and work on bioinformatics related to bean microsatellite development • SB-2 staff Visited FEDEARROZ- Sa1daña with Paola Ruíz and Juan José V ásquez July 17 -20, 2001 • SB-2 staff visited CIA T whiteflies fie1d at !bagué - Tolima with Dra. Jeanne Jacobs, New Zeland. July 18, 2001 • SB-2 staff trip to Venezuela. Visited National Programs to link Projects of lnvestigation between both countries. July 7-9, 2001. • SB-2 staff visited the Co1ombian Maize's Industry as . part of a technology transfer agreement. June 16-17, 2001. • SB-2 staff visited COLCIENCIAS- Bogotá. Genomic's Project. June 3-6, 2001 • SB-2 staff PBA Technical Meeting. Santa Marta. May 3-6, 2001 • SB-2 staff visited NOV ARTIS, San Diego to discuss collaboration on rice genomics. April 14-22, 2001. • SB-2 staffvisited INIFAP, Mexico. April22-29, 2001. • Davis, California, USA, to visit collaborators at UC-Davis in the lab of P. Gepts. March 8- 12,2001 • Santa Fe, New Mexico, USA, to design software for CG bioinformatics program. Jan 14-17, 2001 • SB-2 staff nominated as a member of National Biotechnology Counci1 of Colciencias, Colombia. • Collaborators in ESALQ-Piracicaba and EMBRAPA 1 Arroz e Feijao - Goiania 291 • SB-2 staff member of Laboratory for training Latín American Trainees in Biosafety and related technology. Nominated as a member ofthe United Nations University • SB-2 staff member of a review Panel to analyze the current Biosafety Regulation of GMOs in Colombia. {sponsors: Ministry of Environment and Agriculture; Colombian National Biosafety Council). • SB-2 Staff aided the Colombian Association of Scientific Journalist with the public dissemination of a study funded by Colciencias on news cover of GMOs in Colombia • SB-2 staff assisted the University of Costa Rica (UCR) on logistic to set up evaluations of RHBV resistant transgenic rice plants in Costa Rica according to biosafety recommendations • SB-2 staff assisted the Colombian National University (UNAL) in Palmira, Colombia, and the Institute of Advanced Studies (IDEA) in Caracas, Venezuela on file application and construction of biosafety greenhouse facility complying with National biosafety regulations • SB-2 Staff visited the Ministry of Science and Technology in Caracas, Venezuela, as a follow up initiative of Venezuela becoming a CIA T donor to fund research on rice biotechnology • SB-2 staff and stafffrom University ofCosta Rica initiated contacts with Val Giddings , Bio (USA), to get advice on how to proceed to get a non-interference approval from the prívate sector holding corresponding patents for the deployment to Latín American farmers of the RHBV transgenic resistant rice invention • SB-2 staff initiated contact with the International Food Biotechnology Committee of ILSI to get advice on how to proceed to initiate food biosafety analysis of RHBV transgenic resistant rice 3.1.3 Interoational Scientific Meetings • SB-2 staff participated in severa! international meetings and gave keynote and plenary presentations at: • Tohme, J. Opening presentation at the Euro conference on "Molecular genetics of Model legumes: impact for legume biology and breeding", organized by the Mx Planck Institute of Molecular Plant Physiology, Golm, Germany. Sept 15-19, 2001 • Tohme, J. Plenary presentation at the German Colombian workshop on Biotechnology research, University ofHanover, Sept 12-14, 2001 • Chavarriaga P., speaker InternationaJ Symposium on Biotechnology. Plant Biotechnology at CIA T: from genomics, through tissue culture to farmers . Nairobi, Kenia. September 2-5, 2001 • Z. Lentini. lnvited speaker: GMO derived food less safe than others?. IX Congress of the Colombian Association ofNutritionists. November 1- 2, 2001. Bogotá, Colombia. 292 • Lentini, Z. Invited speaker: Role of Biotechnology Developing Improved Germplasm for The Tropics. First International Congress on Challenges and Opportunities of GMOs for Agriculture and Agroindustry. Cartagena, Colombia. October 11-12, 200 l . • Z. Lentini, Invited speaker: Gene Flow Analysis for Assessing the Safety of GMOs in the Neo-Tropics: The case of beans and rice. lnternational Congress on GMOs: Real risks or chimeras? Brasilia, Brazil. September 3-6, 2001 • SB-2 staffparticipated in ng livelihoods ofthe resource poor through biotechnology. London. August 30- September 10, 2001 • Beebe, S Annual meeting of the Costa Rican national bean program (PIIT A). Oral Presentation: Razas de frijol común y sus implicaciones para mejoramiento genético. August 8-9, 2001. • Beebe, S. Annual meeting of the Costa Rican national bean program (PITT A). Oral Presentation: Reflexiones sobres retos y oportunidades en el mejoramiento de frijol en los próximos años August 8-9, 200 l . • Beebe, S. Jones, P Applications of GIS to tropical agriculture in CIA T. Oral presentation at the Conference on GIS and Biotechnology, VPI, Blacksburg, V A. 200 l. • Beebe, S.; Terán, H.; Quintero, C.; Pedraza, F.; Tohme, J Annual meeting of the Costa Rican national bean program (Pm A). Oral Presentation: Selección asistida por marcadores: consideraciones para su aplicación práctica. August 8-9, 2001. • Beeching JR. Strategies for the Modulation of Post-harvest Physiological Deterioration in Cassava for the Benefit of Poor People. Report commissioned by DFID's Crop Post-Harvest Programme. 2001 • Tohme J. Moderator and presentation at the panel on biotechnology at the V Encuentro Nacional para la Productividad y la Competitividad organized by the ministerio de come.rcio exterior, Medellin, July 9-11 , 2001 • Tabares E., L.Fory, L.Duque, F.Angel, G .Delgado, and Z. Lentini REDBIO. Rice Genetic Transformation efficiency using particle bombardment or mediated by Agrobacterium tumesfaciens. Goiania, Brazil. June 4-8, 2001 • Blair, M.W. Tohme, J. Beebe, S REDBIO. Bean genetic resources and genomics research at CIA T. IV Latín American meeting on Plant Biotechnology, Goianía, Brazil. June 4-8, 2001 • Florez C., R. Escobar, M.Duque and Z. Len ti ni REDBIO Optimization of RITA system for un automated mass production of embryogeni callus of Brachiaria species. Goiania, Brazil. June 4-8, 2001 . • Lentini Z. -key note speaker REDBIO. Gene Technology: Expanding Genetic Diversity and Adding Value to Rice. Brazil. June, 2001 293 • Lentini Z. Key note speaker. REDBIO. Biotechnology for Farmers and Consumers: Role of Biotechnology in Research and Delivering lmproved Germplasm for Latín America. Goiania, Brazil. June 4-8, 2001 • Martinez César P.and J.Tohrne. 200l.Progress in the genetic improvement supported by molecular markers.Paper presented in :REDBIO 200 l.N Latín-American Meeting on Plant Biotechnology. June 4-8, 200 l.Goiania,Goias.Brazil. • Mora A., L. Fory, Iván Lozano, E. Tabares, L. Calvert, and Z. Lentini REDBIO. Transgenic Rice with Hypersensitive Resistance to Rice (0 . sativa) Hoja Blanca Virus (RHBV) in the Field. Goiania, Brazil, June 4-8, 2001 • Lentini Z. - Plenary speaker Biosafety in Field Trials with Plants Modified with Gene Technology. VII Congress of the Colombian Society of Plant Breeding and Crop Production. !bagué, Colombia. May 23-25, 200 l. • Beebe, S.Virginia Polytechnic Institute, Blacksburg, VA, May 17-19, 2001, to attend conference on GIS and biotechnology. • Beeching JR. Molecular analysis of post-harvest physiologícal deterioration in cassava. BioVeg 2001 : Intemational Workshop on Plant Biotechnology. Ciego de Avila, Cuba, April 16-20, 2001 • Beeching JR. Post-harvest physiological deterioration in cassava: molecular and biochemical insights. INIVIT, Santa Clara, Cuba, April, 2001 • Gomez-Vasquez R. Defence responses in cassava suspension ce lis treated with elicitors. BioVeg: Intemational Workshop on Plant Biotechnology. Ciego de Avila, Cuba, April, 16- 20, 2001. • Martínez, C. Symposium on Quantitative Genetics and Plant breeding in the 21st Century. Lousiana State University, March 26-28.2001. • Tohme, J . Pleanry presentation on Biotechnology and agriculture productivity N Encuentro de Competitividad y Productividad. Pereira, February 2001 • Tohme, J. Plenary presentation at a meeting organized by El Espectador and Ministerio de comercio for the sector industrial on biotechnology, 2001 • Beeching JR. Deterioracion fisiologica post-cosecha en yuca. Colombian Biotechnology Programme in Agriculture - Intemational Seminar. Cartagena, Colombia. February 21 - 23,2001 • Beeching, J.; Tohme, J.; Escobar, R. Chavarriaga, P.; Verdier, V- Invited speakers Intemational Program of Agricultura! Biotechnology. Cartagena. January 20-24, 2001 • Chavarriaga P. invited speaker Conference on Genetic Engineering and Food for the World. USA, January 19-21, 2001 294 • Beeching JR. Post-harvest physiological deterioration in cassava: molecular and biochemical insights. llT A, Ibadan, Nigeria, November, 2000 • Buschmann H, Tohme J, Beeching JR. Biochemistry of post-harvest deterioration in cassava root tubers. German Society for Tropical Sciences. University of Hohenheim, Stuttgart. October 11-12, 2000 • Beeching JR. Post-harvest physiological deterioration in cassava: molecular and biochemical insights. Central Food Technological Research lnstitute, Mysore, India, October, 2000 • Beeching JR. Post-harvest physiological deterioration in cassava: molecular and biochemical insights. Central Tuber Crops Research lnstitute, Trivandrum, India, September, 2000 • Beeching JR. Post-harvest physiological deterioration in cassava. lntemational Society for Tropical Root Crops Symposium. Tsukuba, Japan. September 11-15, 2000 • Tohme, J . Plenary presentation on the VIll congreso Colombiano de farmacología y terapeutica, frrst Simposium International sobre biodiversidad como fuente de nuevos medicamentos., 19 August, 2001, Cali. • Blair, Tohme, J. and Beebe, S PCCMCA. Bean genetic resources and genomics research at CIAT. Costa Rica Aprill-6, 2000 • Muñoz, C. ; Blair, MW.; Roca, W. and Tohme, J. PCCMCA. Introgresión de genes de frijol tepari a frijol común por retrocruzas congruentes. Costa Rica April 1-6, 2000. 3.1.4 Workshops, training and Conferences • SB-2 staff trained in construction of cONA Libraries. Yale University. New Haven, Connecticut, USA. October 21 to November 18, 2001 • SB-2 staff trained in Métodos para cuantificar la variación genética y las relaciones filogenéticas en Bancos de Germoplasma. Costa Rica. November 5-7, 2001 • SB-2 staff attended: Conocimiento, construcción y análisis de técnicas de DNA Microarrays, October 13, 200 l. North Carolina S tate University • Biosafety in Developing and Using Insect Resistant Transgenic Plants Enthomology Forum. National University of Colombia. Palmira. October 11, 200 l. (Z.Lentini). • Workshop on Agriculture Biosafety for Colombian Journalists. CIA T. September 29-30, 2001 • Propagación a bajo costo en Santa Ana, Cauca." Perico Negro, Puerto Tejada, Septiembre 20 de 2001. • SB-2 staff invited speaker in Fundamentos de Biología molecular para Fitopatologos". ASCOLFI. September 7, 2001 295 • SB-2 staff invited speaker in Analysis ofMolecular Data. CIAT August 8-10 y 17, 2001 • SB-2 staff assistance to Seminar de Cryo' intercentre.Leuven, Amsterdam. June 30- July 8,2001 • SB-2 stafftrained in VIRTEK. Toronto, July 7-16, 2001 • SB-2 Staff attended Workshop on Cryopreservation of vegetatively propagated tropical crops. INIBAP, 2-6 july 2001 • SB-2 staff attended Course on Desarrollo de metodologías parttctpativas en Biotecnología. In: Curso sobremétodos y técnicas de participación de productores en la Investigación en CIAT. IPRA. June 26-29,2001 • sa:2 staff attended Statistic Course, North Carolina State University, 29 de May 29 - June 19, 2001 • SB-2 staff visited JDIAF(Instituto Dominicano de Investigaciones Agrícolas y Forestales).One month training in Plant Breeding. • SB-2 staff trained in Summer Institute in Statistical Genetics NCSU May 30-June 16, 2001 • SB-2 staffparticipated in Microarrays Course. CIAT May 20-27, 2001 • SB-2 staff attended a workshop on Phaseolus genomics, Mexico, March 2-7, 2001, • SB-2 staff participated in IV. Encuentro sobre Competitividad y Producción. Pereira, Colombia. February 15-16,2001. (J.Tohme invited speaker) • SB-2 staff trained a research scientist from Cenicaña on tissue and genetic transformation of sugar cane. 2001 • Course on the use of micro array and the use of a novel technology. Diversity Array technology (DaRT). The course was given by Damian Jacouq from CAMBIA Australia and was attended by sorne 20 assistants from SB-2. Mayo 20-27, 2001 • Course on molecular approaches for disease resistance, and modulating gene expression in transgenic plants was conducted at CIA T in collaboration with Eric Lam and Nilgun Tumer from the Biotechnology Center for Agriculture and the Environment Rutgers University, New Jersey, USA. Total of 43 participants including 37 CIAT Support Staff and 6 Principal Staff attended the course. Graduate students (current) • Ivan Ochoa- Pennsylvania State University, USA- conducting laboratory and field studies to understand the inheritance and mechanisms of low phosphorous tolerance in common bean and the role of adventitious rooting in adaptation to low phosphorous stress (collaboration with J. Lynch) 296 • Andrea Frei - 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) • Osear Checa - Universidad Nacional - Palmira, Colombia - studying the inheritance of climbing ability in common bean and the importance of genotype x environment interaction in this trait. • Juan F. Femandez; Molecular markers and population genetics of Quercus -PhD Program, Tropical Ecology, University ofMissouri in Saint Louis -UMSL, USA. • Hemando Ramírez; Tomato transformation for insect resistance -PhD program, Agronomic Sciences, Universidad Nacional de Colombia. • Gerardo Gallego; Gene cloning of rice disease resistance genes - PhD program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Eliana Gaitán; Molecular markers and diversity of patm trees - PhD program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Roosevelt Escobar; Genotypic stability of cryopreserved cassava plants- MSc Program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Nelson Royero; Molecular markers and diversity of Anonna spp - MSc Program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Fabio Escobar; Molecular markers to certify seeds of rice - MSc Program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Edgar Barrera; Molecular markers for ACMD resistance- MSc Program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Juan J. Ruiz; Field evaluation of in vitro propagated Annona - MSc Program, Agronomic Sciences, Universidad Nacional de Colombia, Palmira, Colombia. • Eyvar Andrés Bolaños Vidal. Caracterización de la diversidad genética en cuanto a contenidos de caroteno de raíces y hojas de 682 genotipos de yuca. • Galindo, L. M. 200 J. Aislamiento y caracterización de las secuencias L TR de retrotransposones del grupo TYI-copia en Phaseolus vulgaris. Dpto. de Biología, Universidad Nacional de Colombia, Bogotá. (Laureada). • Martinez A.K. 2.001. Obtención de microsatélites en la palma de chontaduro Bactris gasipaes (Palmae). Dpto. de Biología, Universidad Nacional. Bogotá. • Oiga Ximena Giralda. 2001. Universidad del Valle. Construcción del mapa genético de Brachiaria utilizando microsatélites y AFLP's. Dpto. de Biología. Fac. De Ciencias. 297 • Eliana González. 2001. Univalle. Diversidad genética de 3 poblaciones de colombo balanus excelsa (fagacia) especie endémica de los Andes Colombianos • Claudia Patricia Florez. Ph.D. Thesis. Universidad Nacional. Sede Palmira .. Development of Brachiaria Genetic Transformation mediated by Agrobacterium tumefacien. Sponsor: Colciencias. Currently on leave at Dr. German Spangerberg's Laboratory, Plant Biotechnology Centre, Agriculture Victoria, La Trobe University, Bundoora, Victoria, Australia, establishing a CIAT- La Trobe Univeristy collaboration Undergraduate students (current) • Sergio Prieto- Universidad Nacional • Juan José Vásquez. Universidad de los Andes, Bogotá • Paola Ruíz. Universidad Javeriana, Bogotá • Gloria Iriarte, Universidad de Tolima. • Caroliña Castaño, Universidad de los Andes, Bogotá • Andrés Felipe Salcedo. Universidad del Valle • Hector Fabio Buendía, Universidad de Tolima. • Carolina Ramirez Rodríguez, Universidad del Tolima. • Carolina Astudillo, Universidad del Valle. • Wilfredo Pantoja, Universidad del Valle. • Luis Guillermo Santos, Universidad Nacional-Palmira. • Andrés Bolaños, Universidad Nacional-Palmira. • Juan Esteban Montoya, Universidad Nacional - Medellín • Morgan Echeverry, Universidad del Valle 3.1.5 Visiting Research • Peter Wenzl, Center for the Application of Molecular Bio1ogy to Intemational Agriculture, Cambia. (Oct 2001 - Dec. 2001) • Rocío Gómez. Universidad de Bath (Oct. 2000-Aug.2001) • Adebola Raji - Nigeria (Sept 1- Nov. 30, 2001) • Maritza Berti, Universidad Santiago de Chile, Chile (Sept- Nov, 2001) • Martha Isabel Moreno, UNIV ALLE, Colombia. (Sept, 200 l - Dec. 2001) • Alexandra Narváez, IRD, Ecuador ( August- Sept, 2001) • Erika Alexandra Arnao, CONICIT- DANAC, Venezuela. (August, 2001 -Sep. 2001) • Enmanuel Okogbenin, University oflbadan, Nigeria ((July 1998- Nov.2001) • Juan Diego Palacio. Instituto von Humboldt, Colombia (July 1998- Dec 2001) 298 • María Paola Rangel, Centro Nacional de Investigación de Caña de Azucar, Colombia (May 99- Dec. 2001) • Yvonne Lokko- Ghana (April 4, - Sept 5, 2001) • Alba de las Mercedes Alvarez, OlEA, Cuba (Jan 2001 - Jun.2001) • María Ximena Rodríguez, University of Bath (Jan. 1999 - Dec. 2001) • Carlos Marío Hemández, Fundación para la Investigación y el Desarrollo Agrícola - FIDAR. Colombia ( Feb. 1999- Dec 2001) • Inés Sánchez Mosquera, Corporación Colombiana de Investigación Agropecuaria, Colombia (Feb.1995- Dec. 2001) A total ofmore than 70 people (research,joumalist, visitors) received trainning with SB-2 Project staff in different areas and /or for courses and workshops. Their backgrpunds were diverse, having from BS topos-doctoral degrees. Activity 3.2 Assembling databases, genetic stocks, maps probes and related information 3.2.1 Molecular genetics database constructed for a microsatellite parental survey of common bean germplasm M.W. Blair1, A.F. Guerrero2, F. Rojas3 1SB-2; 2IP-1 and 3Unidad de Sistemas de Información Introduction We are developing two databases to store the information about microsatellites tested on common beans. The frrst, uses Oracle (Develo~r 2000) software that we initially tested for storing, handling and presenting images within a relational database about RAPOs that we constructed last year (Annual report 2000). The software has a more user-friendly interface and the capacity to be loaded onto the web. As a relational database, Oracle has the advantage of being an efficient program for organizing and managing data that has multiple layers of relational structure and which is based on a series of data tables. Oracle is also the standard program for databasing the information from the breeding programs at CIA T. The second database we are using to store microsatellite data is Beangenes (http://beangenes.cws.ndsu.nodak.ed!!L), which is the AceDB genome database for beans which was established by the USDA - Plant Genome program to specialize in the genetic information relating to the crop. We hope that these databases will be the basis for collecting genotypic information on common bean and a dynamic analysis tool allowing researchers to ask such questions as: how many polymorphisms can 1 expect when comparing two varieties that might be potential parents? ; and which polymorphisms distinguish one variety from another? Results and Discussion 299 A total of 186 pbotographs of microsatellites surveys were scanned, annotated and loaded into both databases. Size estimates based on the molecular weight of each allele were also databased. This represents the diversity data for a total of95 microsatellite loci. Oracle Database: The gel images were loaded with the Oracle graphics development tool and the estimated band sizes were loaded using the Oracle worksheet development tool. The database provides a set of tools for asking new questions. The three principal components are tools to generate worksheets, reports and graphic images. The program has a web-compatible format that uses windows and buttons to allow for interactive searches and queries. The finished database has three main windows for "gel", "accession", "molecular weight" and "locus" as shown in Figure l. Each ofthese has a datasheet format with columns and entries. The "gel" and "locus" windows allow the user to call up parental surveys for individual microsatellites and to compare the molecular weight of the alleles found in all the genotypes in that survey. These window can be used to compare two microsatellites at a time. Embedded windows are used to call additional items such as the gel images as shown in Figure 2. The "accession" window can be used to compare the markers present in two genotypes. The "molecular weight" window allows a user to search for microsatellites of a certain size. Activities are realized either through the menu bar or action buttons. A console line indicates the status and location of the user. The first version of this database was written in Spanish. AceDE Database: A new class called "microsatellites" was created within Beangenes to contain the information about these markers. This was a central point for Iinking the images of the microsatellite parental surveys, which were added toan existing class, called "image". Beangenes is no longer being updated regularly at North Dakota Sate University, so it will be important for CIA T to incorporate and curate its own data for it to be included in the database. All the AceDB genome databases are being mirrored by the Demeter site at Cornell - USDA bioinformatics facility. These genome databases as well as Beangenes are written and curated in English. Future plans: are to load additional molecular marker data into the present databases. The database could accommodate data from other studies of genetic diversity using the microsatellites, as well as any additional parental surveys that will be conducted. All of the microsatellties used in the parental survey are also being placed on the bean genetic map at CIA T and therefore the chromosomal location of the markers will be the next important data to put in the database. lt will be very important to link the map information with the genetic diversity data accumulated for all microsatellites. A comparative mapping tool would allow researchers to compare the position of the markers in the CIA T population to mapping results from other laboratories (notably with the University of Florida and the University of California - Davis). Web access will be a priority, once the databases are released. AceDB is already internet compatible, through the program Web-Ace developed at the Sanger center. For the Oracle database, we hope to place the database on the internet using Microsoft Interdev or Web-DB. With either system, the database can be accessed from any type of computer, via common web- browsers such as Netscape or Internet Explorer. The information stored in these databases could be amenable to linkages with ICIS {lnternational Crop lnformation System http://www.cgiar.org/icis), which is the database system for managing and integrating genetic resource, crop improvement and crop management information of the CG-system. IPHIS (International Phaseolus lnformation System- http://www.ciat.cgiar.org/icisO is the version ofthe ICIS data base that has been developed at CIA T to hold bean-breeding data. Molecular data is foreseen to be an important part of these databases in the future. To realize the maximum potential of a molecular marker database, it should also be linked with other existing databases that contain germplasm data on Phaseo/us including SINGER (http://singer.cgiar.org), the 300 principal database on genetic resources held in the CGIAR system and GRIN (http://www.ars- grin.gov/npgsD, a comparable germplasm database ofthe USDA. Figure 1. Components of tbe main window Figure 2. el-image window linked to accession and band data. 301 wincinws columns 1 data lunCen• Data R.eleue 2 . ~·· ..... . •. ·-~~;..¡·~ 1.0.: .~ ..... ..~<¡tf,L:.!.,..: Se.rch: -1 fi!ili::::J In Ch .. l' Hep_Dat.. c .... Cene_Pf"'dwct. Petholocw Speeiea Re F'e f"..-.ee KewS.t Ot.her _C.f"'lllltJh~ ·-C.ne .. Chn s..,ene. Vl r.l .. Pa thocen , .... Jo..-~1 Modo! loe u• AU ele Probo Clono Authof" Colleecue tulthe r Deec Coll Col2 Hlc,.o .. tellt.e 91151 ::J:,riMfJu Hiero Table lono i&i lene 1 2 J • 5 6 7 8 ' 10 11 12 13 14 15 16 17 18 -1ewer t~~alu ... . r e 11360 e 11350 e 21657 ' 21078 ' 21242 e 14519 e 4825 e 19833 BAT 477 0011 364 e 3513 BAT 881 e 21212 e 24404 RAD-CER e 243'10 OOR 3'0 ' 19892 Culth• r 0011 476 5El 1309 11\A.K CBil ($) Bll.K CBB tR • BULK BQ-N