ANNUAL REPORT 1998 
PROJECT SB-02 
ASSESSING AND UTILIZING 
AGROBIODIVERSITY THROUGH 
BIOTECHNOLOGY 
-ciAT 
For Interna! Use Only 
lf~~ ®K,cu~'a ''~~at lntematlonal Center for Tropical Agrlcutlure 
PROJECT SB-2: ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH 
BIOTECHNOLOGY 
PROJECT OVERVIEW 
The Challenge: 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 modero 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 forrages, 
are grown in developing countries. Our research on the mandated crops expands to other crops of current or 
potential economic importance in Latín Arnerica. 
Objective: To apply modero biotechnology to identify and utilize genetic diversity for broadening the genetic 
base and increasing the productivity of mandated and non-mandated crops. 
Outputs: Improved characterization of genetic diversity of wild and cultivated species and associated 
organisms. Genes and gene combinations accessed and utilized. Collaboration with public and prívate sector 
partners enhanced. 
Initiation: 1997 
Duration: Five years 
Milestones: 
1998.Molecular linkage maps and DNA-based markers available for assessing diversity and tagging useful 
traits of Phaseolus, Manihot, and associated organisms. Transgenic resistant plants generated with rice; 
populations generated with useful traits from wild Oryza spp. Collaborative activities with CIA T partners 
implemented. 
1999.DNA-based markers available for other species in CIA T mandated agroecosystems. Modero methods 
developed for the rapid propagation of cassava and other species in CIA T mandated agroecosystems. Trangenic 
resistance to insect pests available in cassava. Cryopreservation technique for cassava. 
2000.QTL, from wild germplasm identified and mapped involved in yield and quality of rice and beans. 
Collaborative activities with CIA T partners, including privated sector implemented. 
2001.Gene transfer utilized for broadening the genetic base and germplasm enhancement of rice, cassava, 
beans, and Brachiaria. Collaborative activities with CIA T partners implemented. 
Users. CIA T and NARS partners (public and prívate) in volved in crop genetic improvement and 
agrobiodiversity conservation, AROs from DCs and LDCs. 
Collaborators: IARCs (IPGRI: systemwide program, CIP and llT A: root-tuber crops research; ISNAR: 
training, policies. NARS (CORPOICA, ICA, EMBRAPA, INlAs,). AROS of DCs and LDCs. Biodiversity 
institutions (A. von Humboldt, INBIO, SINCHI, Smithsonian). Corporations and prívate organizations (Corp. 
BIOTEC, BRL, Novartis). 
CGIAR system linkages: Saving Biodiversity (30%); Enhancement and Breeding (60%); Training (10%). 
CIAT project Iinkages: Inputs to SB-2: Germplasm accessions from gene bank project. Segregant 
populations from crop productivity projects. Characterized insect and pathogen strains and populations from 
crop protection projects. GIS services from land use project. Outputs from SB-2: Genetic and molecular 
information on gene pools, and populations for gene bank, and crop productivity projects. lnformation and 
material on identified genes and gene combinations for productivity and crop protection projects. Methods and 
techniques of propagation and conservation for gene bank and productivity projects. Interspecific hybrids and 
transgenic stocks for crop productivity and crop protection (IPM) projects. 
WORK BREAKDOWN STRUCTURE 
PROJECT SB~02: ASSESSING AND UTILIZING AGROBIODIVERSITY THROUGH BIOTECHNOLOGY 
PROJECT GOAL 
To contribute to increased productivity and to the conservation of agrobiodiversity in tropical countries 
PROJECT PURPOSE 
To apply modero biotechnology to identify and utilize genetic diversity for broadening the genetic base and increasing the productivity 
of mandated and non-mandated crops 
1 
OUTPUT l. Genomes ofwild 
and cultivated species and 
associated organisms 
characterized. 
- Molecular characterization 
of genetic diversity 
- Identification and mappmg 
of useful genes and gene 
pools 
- Development of molecular-
genetic techniques for 
assessing genetic diversity. 
1 
OUTPUT 2. Genes and gene 
combinations made available for 
broadening crop genetic bases. 
- Utilization of novel genes and gene 
combinations by means of cellular 
and molecular genes transfer 
techniques. 
- Identification of points for genetic 
intervention in plant/stress 
interactions. 
- Development of cellular and 
molecular techniques for genome 
modification. 
¡¡ 
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. 
NARRATIVE SUMMARY 
GOAL 
To contribute to increased 
productivity and to the conservation 
of agrobiodiversity in tropical 
countries. 
PURPOSE 
To apply modem biotechnology to 
identify and utilize genetic diversity 
for broadening the genetic base and 
increasing productivity of mandated 
and non-mandated crops. 
OUTPUT l. 
Genomes ofwild and cultivated spp 
and associated organisms 
characterized. 
OUTPUT2. 
Genes and gene combinations made 
available for broadening crop genetic 
bases 
OUTPUT 3. 
Collaboration with public and prívate 
sector partners enhanced 
PROJECT LOG-FRAME (1998) 
MEASURADLE JNDICATORS 
CIA T partners using improved 
germplasm and more effective 
breeding and conservation methods 
Characterized gene pools, improved 
genotypes and useful genes are 
available for crop improvement. 
MEANS OF VERIFICATJON 
Center's NARS and other 
publications, statistics. 
More efficient breeding techniques, 
publications. 
Conservation strategies based on Reports, publications, databa~es 
molecular knowledge of diversity; 
integration of genome analysis with 
agroecological data; projects using 
molecular markers; incorporation of 
new molecular and bioinformatic 
techniques. 
A vailability of rice fines with Reports, publications, germplasm. 
transgenic and wide crossing, 
improved traits through, respectively; 
cassava and Brachiaria tissue 
cultures expressing transgenes; 
engineered transformation cassettes 
available; 
NARS scientist participates in 
training courses and adapt sorne 
biotechnologies; partners use CIA T 
assembled strains and constructs, 
databases and maps; new partners 
become involved in CIA T work. 
¡¡¡ 
Courses organized; material 
distributed to partners; new work 
agreements; additional project 
proposals submitted; additional 
funding; reports, publications. 
IMPORTANT ASSUMPTIONS 
Continuation of donor support; 
successful partnership with private 
sector; good regulatory framework. 
Availability of phenotypic data for 
CIAT crops; establishment of 
effective partnerships; up-to-date 
equipment available at CIA T. 
Adequate support continues; private 
sector involvement. 
Adequate support; equipment and 
operational support. 
PROJECT RESEARCH HIGHLIGHTS 1998 
The research highlights for the period Oct. 1997- Sept 98 are summarized by outputs: 
l. Genetic diversity characterized. Microsatellites (SSRs) and expressed sequenced tags (ESTs) 
were added to CIA T's repertoir of molecular genetic markers. We have used these and other 
markers to describe diversity at molecular level, determine population structure and genetic 
distances and phylogenetic relationships, in mandated crop germplasm and extended to non-
mandated crops in cooperation with CORPOICA, A.V. Humboldt and the SINCHI Institutes. 
Genetic diversity of the cassava bacterial blight pathogen in Colombia was linked to the 
ecoregion and microgeographic origin of germplasm. Biological control organisms against the 
cassava green mite were also subject of genetic characterization this year. Assessment of 
diversity at DNA leve! provides information on potential new sources of variability for 
broadening crop genetic base, and for linking diversity in-situ with ex-situ collections. This year 
we have also used molecular markers for mapping crop genomic regions associated with 
resistance genes against cassava bacterial blight, for apomixis in Brachiaria, and for BGMV and 
P uptake in common beans. DNA sequencing allowed us to identify a putative Antracnose 
resistance gene in beans andan R-gene for rice blast. Finally, through cooperation with ARis 
we have constructed the first bacterial artificial chromosome (BAC) gene library for cassava. 
This is a step to the cloning of large genomic segments for future manipulation and usse in crop 
improvement. 
2. Crop plant genomes modified. We have obtained transgressive segregation for higher yield 
(15-25%) in interspecific backcross progenies of rice with wild Oryza spp. Screening these 
progenies with molecular markers, allowed to identify and map chromosomal regions (QTLs) 
from O. barthii and O. rufipogon associated to yield and yield components. Evaluation of 
RHBV resistant transgenic rice resulted in up to 65% higher yield than non-transgenic control; 
furthermore, we have obtained indications that the transgene can be transferred by crossing to 
other rice vars. In preparing for the genetic engineering of cassava, we ha ve screened a range of 
Latín American cassava cvs for their response to somatic embryogenesis. This year we have also 
made genetic constructs harboring a Bt gene agains the cassava stem borer. As a step towards an 
understanding of pathogen-host interactions, a CBB pathogenicity gene has been cloned and its 
protein characterized, and a highly sensitive and specific, PCR-based, detection method was 
developed. Due to their unique tolerance to Al toxicity, low N and low P, Brachiaria spp 
constitute an important source of stress-related genes. In collaboration with an ARI we have 
isolated two secondary metabolites which are biosynthesized in Brachiaria roots under stress; its 
anti-fungal role will be investigated. Long-term preservation of tissue and cells is basic to 
integral genome modification strategies; we have moved forward in the developing of a 
simplified, low cost cryoconservation technique for cassava shoot tips by an encapsulation-
dehydration method. As above, CIA T technological capacities have been extended to 
micropropagationlmicrografting of a fruit tree and genetic transforrilation of a horticultura! crop 
in collaboration with NARS. 
3. Collaboration with CIA T partners enhanced. In the period covered by this report, project SB-
02 staff has organized 6 workshops and courses dealing with biotechnology and biodiversity 
topics and planned a second phase ofCBN. The BRU has assembled a collection ofnear lOO E. 
coli strains, containing plasmid constructs for genetic transformation and over 100 A. 
tumefaciens strains. In the period Oct 1997 - Sept 1998, the project staff has produced 19 
refereed publications, 14 non-refereed and 13 project concept notes and proposals. Ten donor 
organizations have contributed to project SB-02 complementary outputs/activities in 1998. 
iv 
PROJEC T SB-02: ANNUAL REPORT 1998 
CONTENTS 
OUTPUT l. GENOMES OF WILD AND CUL TIVATED SPECIES AND 
ASSOCIATED ORGANISMS CHARACTERIZED 
ACTIVlTY 1.1 Molecular Characterization of genetic diversity 1 
Summary of Achievements 1 
1.1.1 Phylogenetic analysis in Phaseolus 1 
1.1 .2 Microsatellite variability in Brazilian cassava landraces 3 
1.1.3 Measurement of the genetic diversity in Xanthomonas axonopodis pv. Manihotis whithin 5 
different fields in Colombia. 
1.1.4 AFLP fmgerprinting: an efficient technique for detecting genetic variation of Xanthomonas 7 
axonopodis pv Manihotis 
1.1.5 Virulence variability of a Xanthomonas axonopodis pv Manihotis Colombian populations. 8 
1.1.6 Molecular characterization of selected species of Phytoseiidae lO 
1.1.7 Molecular characterization ofthe entomopathogenic fungus, Neozygites sp, pathogen of 14 
phytophagous mites. 
1.1.8 Studies on the Viral Pathogens found in Mononyche/lus tanajoa and M caribbeanae 15 
1.1 .9 Characterization and genetic variability analysis of Passijlora L. spp from the Andean 18 
region, using molecular markers. 
1.1.1 O Biochemical and molecular characterization of the Musa L. Colombian Collection 19 
ACTIVlTY 1.2. Identification and mapping of use fui genes and gene pools 21 
Summary of Achievements 21 
1.2.1 Mapping of irnportant agronomic traits in bean 22 
1.2.2 Gene Tagging with PCR markers for bean breeding: adaptation to low fertility conditions 24 
1.2.3 Identification ofDisease Resistance Gene Analogs (RGA): ldentification ofNBS type 26 
putative resistance gene in common bean 
1.2.4 Gene Tagging of Resistance to the African Cassava Mosaic Disease (ACMD). 29 
1.2.5 Marker-Assisted genetic analysis of Earliness and Root quality traits in Cassava 31 
1.2.6 Mapping of genetic resistance of Cassava to bacteria} blight disease (Xanthomonas 33 
axonopodis pv manihotis) 
1.2.7 Identification of genomic regions responsible for the determination ofwhitefly resistance in 35 
cassava 
1.2.8 Molecular markers of genes involved in physiological post-harvest deterioration (ppd) of 37 
cassava roots. 
1.2.9 Identification ofNBS and Protein Kinase type putative resistance genes to Rice Blast 42 
1.2. 10 Genetic mapping and gine mapping of the apomixis gene in Brachiaria. 44 
ACTIVITY 1.3 Development of molecular genetic techniques for assessing genetic diversity 46 
Summary of Achievements 46 
1.3.1 Development of cassava microsatellite markers and its use for assessing genetic diversity at 46 
genus level 
1.3.2 Construction of a Cassava Bacteria} Artificial Chromosome (BAC) Library: Towards 48 
Cloning ofDisease and Pest Resistance Genes 
1.3.3 Expressed Sequence Tags (ESTs) for the Genetic Map ofCassava 49 
1.3.4 Lowering the cost of reliable biotechnologies for national cassava programs: Simple 51 
Sequence Repeat (SSR) to Sature and Facilitate Use ofthe Cassava Molecular Genetic 
Map. 
1.3.5 lsolation and characterization ofmicrosatellites in Brachiaria sp. 53 
V 
OUTPUT 2. GENES AND GENE COMBINATIONS MADE AVAILABLE FOR 
BROADENING CROP GENETIC BASES 
55 
ACTIVITY 2.1 Utilization of novel genes and gene combinations by means of cellular and 55 
molecular gene transfer techniques 
Summary of Achievements 55 
2.1.1 Gene transfer between Phaseolus spp through interspecific hybridization. 55 
2.1.2 Use of P. coccineus and P. polyanthus to improve Common Bean. 57 
2.1.3 Development of genetic transformation methods for common bean. 60 
2.1.4 Improving somatic embryogenesis of a range of Latín American cassava cultivars. 63 
2.1.5 Construction and testing of plasmid containing a l3t-gene for cassava transformation 66 
2.1 .6 Cbaracterization of the PthB pro te in of Xanthomonas axonopodis pv manihotis 68 
2.1.7 Detection of Xanthomonas axonopodis pv. Manihotis in cassava true seeds by Nested- 70 
Polymerase Chain Reaction assay (N-PCR) 
2.1.8 Cryopreservation of cassava shoot tips: Encapsulation-dehydration technique 71 
2.1.9 Genes from Wild Rice Contribute to Yield lncrease in Cultivated Rice 74 
2.1.10 Control ofRHBV (rice hoja blanca virus) through nuclear protein mediated cross protection in 78 
2.1.11 
2.l.l2 
2.l.l3 
2. 1.14 
transgenic rice. 
Agrobacterium mediated genetic transformation ofrice genotypes adapted to irrigated, upland, 
and hillside agroecosystems. 
Advances in Brachiaria genetic transformation mediated by Agrobacterium tumefaciens 
incorporation ofresistance to fruitworm (Neoleucinodes elegantalis) and budworm 
(Scrobipalpuloides absoluta) in the tomato variety UNAPAL Arreboles by genetic 
transformation. 
Development and standardization of an in vitro clona! propagation method of Soursop 
(Annona muricata L.) 
82 
85 
87 
91 
ACTIVITY 2.2. ldentification of points for genetic intervention in plants/stress interactions 93 
Summary of Achievements 93 
2.2.1 Exploring the genetic potential and stability ofVitamin content in cassava 94 
2.2.2 Mechanism of acid soil adaptation in Brachiaria cultivars 97 
2.2.3 Validation of analitycal methods to quantify Vitamins A andE in Jessenia bataua Palm Oil 1 O l 
OUTPUT3 COLLABORATION WITH PUBLIC AND PRIVA TE SECTOR PARTNERS 
ENHANCED 
104 
ACTIVITY 3.1 Organization of Networks, Conferences/Workshops and Training Courses 104 
Summary of Achievements 104 
3.1.1 The Cassava Biotechnology Network 104 
3. 1.2 Training couses/workshos on agrobiodiversity and biotechnology organized in cooperation 107 
ACTIVITY 3.2 Assembling of data bases, genetic stocks, maps probes and related information 108 
Summary of Acbievements 108 
3.2.1 Genetic stocks 108 
3.2.2 E. Coli andA. tumefaciens collections 108 
ACTIVITY 3.3 Publications, project proposal development, and contribution to IRP and 109 
biosafety management. 
Summary of Achievements 109 
3.3.1 Publications by SB-02 staff in the period Oct. 97 - Sept. 98 110 
3.3.l.l Refereed Publications 110 
3.3.1.2 Other Publications 111 
3.3.1.3 Concept Notes and Project Proposal prepared 114 
DONORS CONTRIBUTING TO PROJECT SB-02 COMPLEMENTARY FUNDING IN 1997-98 115 
INTERNALLY COMMISIONED EXTERNAL REVIEW (ICER) OF PROJECT SB-02 116 
PROJECT SB-02 ST AFF LIST (1998) 117 
vi 
PROJECT SB-02: ASSESSING AND UTILIZING AGROBIODIVERSITY 
THROUGH BIOTECHNOLOGY 
ANNUAL REPORT 1998 
OUTPUT l. GENOMES OF WILD AND CULTIVATED SPECIES AND 
ASSOCIATED ORGANISMS CHARACTERIZED 
ACTIVITY 1.1 Molecular characterization of genetic diversity. 
SUMMARY OF ACIDEVEMENTS 
• In describing the phylologeny of the genus Phaseolus, 20 species have been grouped 
using interna! regions of ribosomal DNA. 
• As a collaboration with the ESALQ, Univ. of Sao Paulo, on the structure of Brazilian 
cassava landraces using AFLPs, alleles that are unique to Brazilian germplasm have 
identified. 
• In collaboration with CORPOICA, 52 Passif/ora spp have been characterized and 
patterns of genetic distances described. Genome grouping, was carried out with the 
Colombian Musas collection for duplicate identification. 
• AFLPs identified genetic groups within the cassava bacterial blight pathogenic 
population of Colombia, at ecoregional and microgeographic level. Further, AFLPs 
showed great discriminatory power to distinguish genetic differences at even the 
intrapathova level. 
• AFLPs where also used to characterize biological control agents and 
entomopathogens of the cassava mi te. 
1.1.1. Phylogenetic analysis in Pltaseolus. 
Background 
Molecular data plays an essential role in the reconstruction of evolutionary relationships 
among many organisms. For assessing molecular phylogenetic relationships among 
plants, the nuclear genes coding for the 18S and 25S ribosomal RNA components of the 
cytosolic ribosomes have been used extensively at family and higher taxa levels (Hamby 
and Zimmer, 1991). To study the phylogeny among closely related genera or species, 
however, a more rapidly evolving gene is needed for the analysis. Interna! transcribed 
spacer (ITS) regions have been shown to be relatively conservative in length and can be 
used as a source of nuclear DNA characters for phylogenetic reconstruction in plants 
(Baldwin, 1992). Both ITS regions (ITS 1 and ITS2) are part of the rRNA genes (rDNA), 
which are organized in a tandemly repeated manner. Each rDNA repeat unit consists of 
the genes coding for the 18S, 5.8S, and 25S rRNA, which are separated from the next 
repeat unit by the intergenic spacer region (IGS). The coding regions are interrupted by 
two interna! transcribed spacers: ITS 1 and ITS2. 
We have initiated a molecular phylogenetic study of the different section of the 
Phaseolus genus starting with the group of species related to the common bean 
(Phaseolus vulgaris). The second group to be anaiyzed is ofthe Lima bean. A third target 
for this project is a total phylogeny for the genus including aH species available at CIA T. 
Methodology 
Total DNAs were isolated from fresh or frozen leaves following the protocol of Afanador 
et al. (Afanador, L. pers comunication). PCR amplifications of the ITS sequences were 
done using a total volume of SOfll, containing 2.5mM MgC12, 0.1uM each primer, 1mM 
dNTPs mixture and lOX Taq polymerase buffer and 1 unit Taq DNA polymerase was 
added to each reaction. PCRs were performed in a MJ research thermocycler. PCR 
products were purified using Wizard PCR Preps (Promega Corp.) or by elution in a low 
melting point agarose. Fragments were sequenced on a ABI Prism 377 (Perkin-Elmer) 
using dye terminators and AmpliTaq FS (Perkin-Elmer). The sequences obtained were 
aligned using the Sel-Al program and phylogenetic analysis will carried using PAUP 
version 4. 
Results 
Twenty different species of the genus Phaseo/us have been amplified using primers from 
the ITS regions and complete sequences were obtained. A preliminary phylogenetic tree 
was constrcuted using macroptilium sp as an outgroup. P. lunatus and P. vulgaris 
complex were found to be in separated groups, as expected. P. vulgaris was closer to P. 
coccineus. The P. lunatus complex was found to be closer to P. augusti and P. 
pachyrrhyzoides than to P. vulgaris . These results are in agreement with previous 
analysis conducted at CIA T using the AFLP methodology (A. L. Caicedo, BRU annual 
report 1997) 
On going activities 
• Analysis of the data for the P. vulgaris complex and P. lunatus complex and 
associated wild Phaseolus species 
• Comparison with different species. 
• Sequencing of ITS from a wide range of Phaseolus species 
• Construction of phylogenetic tree for all species of the genus Phaseolus 
References 
Afanador L.K., Haley S.D. and Kelly J.D. Adoption of a "mini-prp"DNA extraction 
method for RAPD marker analysis in Common Bean (Phaso/us vulgaris L.). Department 
of Crop ad Soil Sciences. Michigan State University. 
Baldwing, B.G. (1992). Phylogenetic utility of the interna! transcribed spacers of nuclear 
ribosomal DNA in plants: An example from the Compositae. Mol. Phylogenet. Evo l. 1:3-
16. 
2 
Hamby, R. K., and Zimmer, E.A. ( 1992). Ribosomal RNA as a phylogenetic tool in plant 
systematics. In "Molecular Systematics ofPlants" (P.S. Soltis, D.E. Soltis, and J.J. Doyle, 
Eds.). pp 50-91, Chapman and Hall, London 
Collaborators: E. Gaitán, O. Toro, D. Debouck and J. Tohme 
1.1.2. Microsatellite variability in Brazilian cassava landraces. 
Introduction 
Cassava is one of the most important cultivated species among traditional farmers in 
Brazil, especially those from local indigenous communities in Amazonian and sub-
Amazonian regions. Cassava flour constitutes the basis of these comrnunities' diets. In 
these regions, farmers grow in small gardens dozens of cassava landraces (Cury, 1998). 
This type of agricultura! system conserves and maintains a substantial genetic variability 
in situ, in a way that is still not well understood. The expansion of modem comrnercial 
agriculture that affects directly and indirectly these traditional systems, often results in 
the genetic erosion of existing landraces (folk varieties, ethnovarieties). 
To understand better the genetic structure of locallandraces a project was initiated by the 
Genetic Department, ESALQ-USP, Piracicaba-SP, Brazil. One component of the project 
involves the analysis at CIAT of cassava microsatellites to evaluate the genetic variability 
of 55 cassava landraces from 13 'gardens' , obtained from 8 comrnunities in 3 Brazilian 
regions: Solimoes river, Negro river (amazonia), and Ribeira valley (Atlantic jungle, 
southem cost ofSao Paulo State). 
Material and methods 
DNA from lyophilized leafs of 55 Brazilian folk varieties was extracted by the method 
described in Dellaporta (1983). The tissue was obtained from the cassava germplasm 
collection of the Genetic Department, ESALQ-USP, Piracicaba-SP, Brazil. Twelve 
cassava microsatellites (Chavarriaga-aguirre, in press) were amplified with fluorescent 
primers purchased from Perkin-Elmer or Research Genetic. The fragments were 
separated in an ABI PRISM ™ 377 DNA Sequencer. The protocols used to amplify and 
separate the microsatellites were described in CHA V ARRIAGA-AGUIRRE (in press). 
DNA from the accession MCol-22 from the CIAT core collection was used as a standard 
in all gels to facilitate the comparison of results . Microsatellite data from the core of the 
core collection on CIAT were obtained by Roa, A.C. (BRU, personal communication). 
The results were analyzed with the GeneScan TM and Genopyper software. 
3 
Results 
The data of this study were also compared to 38 accessions that constitute the core of the 
cassava core collection present at CIAT, using the same set of microsatellites. The 
number of alleles found is displayed in the Table l. We found 14 alleles that are not 
shared by the 2 groups of plants. Seven are exclusives of the Brazilian landraces and the 
other ones are exclusive of the core of the core collection of CIA T. 
On going activities 
l . Statistical analysis of the microsatellite data to determine the genetic structure of the 
Brazilian collection 
2. Compare the diversity ofthe Brazilian collection wit h the core collection 
3. Deterriline the change if any on the core collection 
Table l. Microsatellites primers analyzed in cassava and wild Manilwt species. 
Microsatellite Forward primer/ reverse primer• Repeat 
sequence .. 
Size range (bp) No. of a lleles< % ofpolymorphism 
na me 
GAGG5 TAATGTCATCGTCGGCTTCG (GGGA)l(GCGA) 113-127 8 86 
GCTGATAGCACAGAACACAG (GGGA)(GAh 
GA 12 GATTCCTCTAGCAGTTAAGC (GA)n 133-171 17 lOO 
CGATGATGCTCTTCGGAGGG 
GA 13 TTCCCTCGCTAGAACTTGTC (GA)2GC(GA)a 134-140 4 57 
CTATTTGACCGTCTTCGCCG 
GA 16 GTACATCACCACCAACGGGC (GA)u 98-134 15 71 
AGAGCGGTGGGGCGAAGAGC 
GA21 GGCTTCATCATGGAAAAACC (GA)1 AA(GA)n 104-J 18 8 86 
CAATGCTTTACGGAAGAGCC 
GA 126 AGTGGAAATAAGCCATGTGATG (GA)a(GT),G(GT) 178-403 20 86 
CCCATAATTGATGCCAGGTT 
GA 13 1 TTCCAGAAAGACTTCCGTTCA (GA),. 75-1 28 21 100 
CTCAACTACT GCACTGCACTC 
GA 134 ACAATGTCCCAATTGGAGGA (GA),2GTA(GA)2 309-338 10 71 
ACCATGGATAGAGCTCACCG 
GA 136 CGTTGATAAAGTGGAAAGAGCA (GA)u 148-171 11 100 
ACTCCACTCCCGATGCTCGC 
GA 140 TTCAAGGAAGCCTTCAGCTC (GA) .. 148-172 JO 57 
GAGCCACATCTACTCGACACC 
• Chavarriaga et al., 1998. 
b Sequence ofthe SSR loci isolated from the cassava cultivar M COL 22. 
e Null alleles are not included. 
4 
References 
Chavarriaga-Aguirre,P; Maya, M.M.; Bonierbale,M. W.; Kresovich,S.; Fregene, M.A.; 
Tohrne,J.; Kochert,G. 1998. Microsatellites in cassava (Manihot esculenta Crantz): 
discovery, inheritance and variability. Trends in Genetics (in press). 
CURY,R. Distribuicao da diversidade genetica e associacao de caracteres em 
etnovariedades de mandioca (Manihot esculenta Cranz) provenientes da agricultura 
tradicional do Brasil. PhD Tesis, ESALQ- USP, Brasil.1998. 
Dellaporta, S.L.; Wood, J.; Hicks, J.B. A plant DNA minipreparation: version Il. Plant 
Molecular Biology Reporter 14: 19-21 . 
Collaborators: Muhlen, G.S1. ; Roa, A.C.; Tohrne, J. 
l . ESALQ, Universidade de Sao Paulo, Brazil 
1.1.3 Measurement of the genetic diversity in Xanthomonas axonopodis pv. 
Manihotis within different fields in Colombia. 
Introduction 
Cassava Bacteria! Blight (CBB) caused by Xanthomonas axonopodis pv. manihotis 
(Xam), is a major constraint to cassava cultivation worldwide. The movement of Xam by 
the exchange of contaminated planting material has important implications for the control 
of the disease. Thus it is important to understand pathogen migration and how it 
influences population genetic structure. A study of the distribution of pathogen diversity 
among ecological zones and sites in Colombia was conducted in 1995-1996. This study 
revealed the genetic variability of Xam in Colombia and population sub-structuring at the 
level of ecological zones (Restrepo and Verdier, 1997). 
The objective of the present study was to continue the analysis of Xam population 
structure in Colombia by examining the variation and microgeographic distribution of the 
pathogen in different fields located in four ecological zones. 
Materials and Methods 
A Xam collection was sampled in 1997 from eight experimental fields and analyzed by 
RFLP, rep-PCR and by inoculation on a cassava cultivar. The genetic diversity for all 
Xam isolates was estimated by Nei's diversity index (HT). A genetic differentiation 
coefficient (GsT) was calculated as GsT = (HT - Hs)IHT where Ht is the total genetic 
diversity of Xam in Colombia and Hs is the average genetic diversity in each ECZ or each 
field. 
5 
Results and ConcJusions 
In Colombia, the Xam population shows a high degree of genetic diversity (table 1 ). 19 
distinct haplotypes were found among the 244 Colombian isolates collected in 1997. Ten 
new haplotypes, that were not detected in a previous study, were characterized. This may 
reflect the potential of change of the Xam populations. Our results allowed us to assess 
the role of host selection in structuring pathogen population. Pathogen diversity was 
highly correlated to the number of cultivars planted. 
Seven haplotypes out of the 8 described in the ECZ2 were detected at site E. All the 
haplotypes present in ECZ 1 and ECZ5 could be detected in one si te, si te G and B 
respectively. For future studies to evaluate pathogen diversity in Colombia, we suggest to 
collect extensively within one field representative of each ecological zone. Further 
studies in single fields ( e.g, si te F) will give insights into the effects of cultivation of a 
host genotype on pathogen diversity. Based on the multiple correspondence analysis, the 
244 isolates were grouped into six clusters. Three clusters contained isolates collected in 
only one ecozone, clusters 1, 2 and 6 (ECZ 5, 2 and 1 respectively). 
ERIC and REP primers yielded two and three fingerprint types respectively. The rep-PCR 
was less discriminative than RFLP, showing less that 1% of polymorphism among the 
analyzed strains. 
It was possible to detect pathogen rnigration with the use of the RFLP analyses. While the 
haplotypic distribution is specific to each ecological zone (ECZ), within an ECZ the same 
haplotype strains were present in different sites. This supports the evidence that 
movement of the pathogen within ECZ results from the transmission of the strains 
through contarninated germplasm. 
References 
Restrepo, S. & V. Verdier. 1997. Appl Enviran Microbio/63, 4427-4434. 
Table 1: Haplotypic diversity (Hs) of Xam population within each field, each ecozone and for the Xam collection in 
Colombia (HT). 
Ecz• Locality Field Cassava cultivar planted Genetic diversity index 
Hs (field) Hs (ECZ) 
5 Cajibio A Traditional o 0.53 
Mondomo B lmproved 0.66 
2 Matazul e Improved 0.62 
Carimagua D Improved 0.64 0.68 
Villavicencio E Improved 0.68 
2-5 Santander Quilichao p Improved 0.46 0.46 
Caracoli F CG1141-l o 
0.52 
Pivijay G lmproved 0.61 
• ECZ = edaphoclimatic zone, ECZ 1: subhumid tropics, ECZ 2: acid-soil savannas, ECZ 5: high-altitud tropics and ECZ 2-
5: which shares ECZ 2 and ECZ 5 characteristics 
6 
Collaborators: C. Vélez, S. Restrepo and V. Verdier1 
l. ORSTQRM,CIAT 
1.1.4 AFLP fingerprinting: an efficient technique for detecting genetic variation of 
Xa11tlzomonas a.t:o11opodis pv manilwtis 
Introduction 
Xanthomonas axonopodis pv. manihotis (Xam) is the causal agent of cassava bacteria! blight 
(CBB), a particularly destructive disease of cassava in South America and Africa.genetic 
diversity of Xam has been mostly characterized by RFLP analyses. The analysis of a Colombian 
population of Xam, collected from different ecological zones, showed the existence of at least 
one clonal population, which was found in the high-altitude tropics ( ecozone 5) (Restrepo and 
Verdier, 1997). To verify the genetic homogeneity ofthis population, we developed a molecular 
technique that can detect smaller sequence variations than can the RFLP technique. 
Amplified restríction fragment length polymorphism (AFLP) is a recently developed technique 
for the fingerprinting ofDNAs of any origin or complexity. The high resolution power of AFLP 
was demonstrated in characterizing bacteria! strains at the subgeneric level. Until now, AFLP 
has not been applied to population studies ofplant-pathogenic bacteria at the pathovar Ievel. We 
previously established the conditions for applying the technique to the study of Xam. Here, we 
report the use of the AFLP technique in assessing the genetic diversity of Xam at the 
infrapathovar level. The efficiency of RFLP and AFLP analysis for measuring genetic diversity 
in Xam was compared. 
Methods 
Bacterial strains, DNA isolation and RFLP analysis. Xam isolates were collected in different 
sites in three ecological zones. Forty-six field isolates collected in 1995 and 1996 and one 
reference strain, CIAT 1121 were analysed. The RFLP-pthB haplotype of37 ofthe 47 Xam 
strains were previously characterized (Restrepo and Verdier, 1997). 
AFLP reactions. AFLP markers were assayed as previously described with modifications (Vos 
et al., 1995). The second PCR reaction was performed with the touch-down PCR thermal 
profile, with the initial annealing temperature of 60°C and subsequent reduction by 1° to 55°C 
over five cycles (Fig.1). The primers used were those selected in a previous study (CIAT, 
annual report, 97). The amplified products were separated on a 6% polyacrylamide denaturing 
gel. 
Data analysis. RFLP or AFLP markers from eight primer combinations were scored as either 
present (1) or absent (0). We used NTSYS-PC to calculate a similarity matrix, todo the cluster 
analysis and to construct the dendrogram. A correlation index was calculated between the 
similarity matrices resulting from AFLP and RFLP to determine the complementarity or 
redundancy of results. 
7 
Results and Conclusion 
AFLP is an extremely useful and reliable technique for detecting polymorphisms in bacteria! 
populations. Analysis of 47 Xam strains presented a total of 322 AFLP bands for the eight 
primer combinations. Between 28 and 64 bands per strain were obtained, ranging from 40 bp to 
350 bp. 173 bands were polymorphic and considered for the cluster analysis. AFLP allowed the 
discrimination of closely related strains collected in ECZ5. Ten clusters were obtained and 
ECZ5 strains were grouped in two separate clusters. The cluster analysis performed for the 
RFLP markers resulted in 8 clusters and ECZ5 strains were grouped in one cluster. 
The results obtained by AFLP and RFLP are correlated but not redundant. Matrices of genetic 
similarity estimates, based on both methods were correlated (r = 0.79). The AFLP data not only 
supported but also extended the RFLPipthB analysis by revealing the existence of subgroups 
among the ECZ5 Xam population. By having a superior discriminative power in differentiating 
highly related strains belonging to the same pathovar, AFLP analysis is a valuable altemative in 
Xam population studies. 
AFLP results support the hypothesis that strains from ECZ5 form a genetically and evolutively 
separate group. Strong evidence exists to suggest that the pathogen migrates between and within 
ECZs in Colombia, but because of the low similarity observed between ECZ5 strains and strains 
collected in other ECZs, the origin ofthe ECZ5 strains remains unclear. 
AFLP markers can also be used to identify particular races or pathotypes and for monitoring the 
dynamics of AFLP haplotypes within each region. In addition to population studies, AFLP 
DNA fingerprinting may facilitate the identification of polymorphisms linked to virulence 
factors and contribute to the understanding ofplant-bacteria interactions at the molecular level. 
References 
Restrepo, S. & Verdier, V. (1997). Appl Enviran Microbio/63, 4427-4434 
Vos, P., Hogers, R., Bleeker, M., Reijans, M., Van de Lee, T., Homes, M., Frijters, A. , Pot, J., 
Peleman, J., K.luper, M. & Zabeau, M. (1995). Nucleic Acids Res 23, 4407-4414. 
Collaborators: 
Silvia Restrepo, Myriam Duque, Joe Tohme and Valérie Verdier 1 
1 ORSTOM - CIA T 
1.1.5 Virulence variability of a Xanthomonas axonopodis pv manihotis Colombian 
populations. 
Introduction 
Cassava bacteria! blight (CBB) caused by Xanthomonas axonopodis pv. manihotis (Xam) is a 
particularly destructive disease in Colombia. The most appropriate and realistic approach for 
controlling the disease is to identify and deploy host-plant resistance. However, effective 
8 
breeding for resistance depends on information about the pathogen diversity and the spatial 
distribution of the diversity. Virulence variation of Xam has been reported by different authors 
and their results suggest that besides variation in virulence, sorne pathogenic specialization 
exists. Our main objectives are to determine the possible existence of pathotypes among the 
Xam population and to evaluate the usefulness of cassava cultivars as host differentials. 
Materials and Methods 
Cassava cultivars and bacterial strains. Based on a previous study of the gennplasm 
diversity (Sánchez et al., 1998), we selected a set of 17 cultivars as potential CBB differential 
hosts. A set of 26 Xam isolates representing the 26 haplotypes described in Colombia in 1996 
was used for inoculation. Plants were arranged in the greenhouse according to a randomized 
block design with five replications per isolate x cultivar combin'ition. The experiment was 
conducted twice. Inoculation techniques and disease rating were as previously described 
(Restrepo and Verdier, 1997; Verdiér et al., 1994). 
Data analysis. Inoculation datas were analyzed by a non-parametric analysis of variance, the 
Kruskal-Wallis analysis. The area under the disease progress curve (AUDPC) was calculated 
for each interaction and data transformed to log (AUDPC) and then analyzed. Data from the 
AUDPC analysis and the leaf inoculations were analyzed by an analysis of variance using the 
General Linear Models procedure (SAS). A Duncan's multiple range test was used to compare 
the entry means (significance was declared for p < 0.05). The leaf reaction was assigned to one 
of five pathogenicity classes (A to E), which were determined by a descriptive method 
(univariate procedure ofSAS) (SAS). The analysis was done for each cultivar separately. 
Results 
Leaf and stem inoculations for detecting strain X genotype interactions .. Lesion sizes obtained 
after the leaf inoculation varied widely among cultivars. By the stem and the leaf inoculation 
methods, it was possible to detect a variation in the reaction of the isolates when inoculated on 
the cultivars. However, using the leaf inoculation method the genotype X isolate interactions 
were not statistically significant (p = 0.1884). Data of the stem puncture of the 26 isolates on 
the same 4 cultivars showed that the interactions were highly significant (p < 0.001). 
Isolate X cultivar interactions. The interactions between Xam isolates and cassava genotypes 
were highly significant (p<0.01). No significant differences were found between the two 
repetitions of the experiment. Each Xam strain showed different reaction after the inoculation 
with the 17 cultivars. Reactions between isolates and cultivars adapted to the ECZ of origin of 
the isolates were studied and specific pathotypes were defined for each ECZ. Among ECZ2 
isolates, 8 different pathotypes were defined. Six pathotypes were found among ECZI isolates. 
Cultivars adapted to ECZ5 were all susceptible to ECZ5 isolates therefore only one pathotype 
was described. The ANOVA analysis of the areas under the disease progress curve 
demonstrated that significant differences between cultivars, isolates and among the interactions 
(isolate x cultivar) (p<O. 001) exist. 
9 
Discussion 
Two inoculation methods were assessed for their usefulness in detecting cultivar x isolate _ 
interactions. Multiplication and spread of Xam were monitored in cassava leaves in susceptible 
and resistant cultivars as expressed by the leaf area lesion. Our results confirmed previous 
results that have shown that bacteria! spread in the mesophyll is identical in resistant or 
susceptible cultivars. Cassava bacteria! blight is a foliar and vascular disease and plant 
resistance mechanisms occurred at a vascular leve!. Stem inoculation is the most appropriate 
method for detecting isolate x genotype interactions and must be recommended for such 
studies. The lesion sizes obtained after the leaf inoculation may be correlated with the leaf 
structure rather than with the resistance as expressed in the vascular stem. 
The results of the Kruskal-Wallis analysis of the stem inoculations and results of the ANOV A 
analysis of the areas under the disease progress curve showed the possibilíty of sorne varietal-
specificity. Resistant cultivars react differently to different highly virulent isolates, indicating 
that different resistant factors may be involved in the expression of resistance. Isolates were 
grouped into pathotypes based on their reactions on the cultivars adapted to the ECZ where 
they were collected. 
Our results provide a basis for CBB control strategy in specific ECZ utilizi..'lg a selective 
deployment ofECZ-adapted cassava resistant cultivars. In the case of Xam in Colombia, sorne 
genotypes with good level of resistance could be proposed for further breeding. MBRA 685 
and MNGA-2 presenta good leve! ofresistance to the majority ofECZ2 isolates, CM6438-14 
and CM523-7 are resistant to ECZI isolates. Resistance to ECZS isolates has to be found. The 
durability of these sources of resistance is likely to be affected by the capacity of the bacteriurn 
to undergo mutation for pathotype change in the field and by the changes in the pathogen 
population due to pathogen's migrations. 
References 
Restrepo, S. & Verdier, V. (1997). Appl Environ Microbio/63, 4427-4434 
Sánchez, G., Restrepo, S., Duque M. C., Fregene, M., Bonierbale, M., and Verdier, V. 
Genome, submitted. 
Verdier, V., Boher, B., Maraite, H., and Geiger, J. P. 1994. Appl. Environ. Microbio!. 
60:4478-4486. 
Collaborators: S. Restrepo, G. Sánchez, M.Duque, and V. Verdier1 
l. ORSTOM- CIAT 
1.1.6 Molecular Characterization of Selected Species of Phytoseiidae 
The complex of phytoseiidae found on cassava in the Neotropics has its greater genetic 
diversity in Colombia where 40 species have been sent to Africa for the biological control 
of M tanajoa. Three of these have become established, Neoseiulus idaeus, 
Typholodromalus manihoti and T aripo. It has been found that isolates of N. idaeus do 
not show any variability at the isoenzymatic level, perhaps oving to its wide distribution 
in different ecosystems. T manihoti is the predominant species in rnost of the cassava 
growing areas of South America. In isoenzymatic studies done at CIA T show that T 
10 
manihoti has considerable diversity in its banding pattem. Because of this hybridization 
studies have been carried out between populations of this species to determine if 
reproductive differences exist or if there is an isolation of this group. 
Materials and Methods 
In the present research molecular studies are using AFLP's is being done in populations 
of the different species found in hot semiarid and high altitude regions of the neotropics 
that correspond to similar regions of Africa (the Sahara and the East African plateau. The 
objective of this work is to identify their phylogenic relationships and contribute to the 
selection and release ofthe most promising species for biological control of M tanajoa in 
Africa. 
Results and Discussion 
Eight phytoseiid species, including numerous populations collected in severa! countries 
were analyzed with AFLP's (Table 1). Due to the problems that have occurred in the 
reproducibility of banding pattems of AFLP's with arthropods in general, special 
extraction techniques have been developed, which have given good results. The 
methodology of Vos et al, 1996, with Operen Technologiest adaptors and Primers was 
used. Non degenerated DNA, at good concentrations, between 10 and 25ng/ul, was 
obtained. 
Ofthe species indicated in (Table 1) diverse combinations have been analyzed and at this 
time results have been obtained for the combinations PEIC/PMIC for N. idaeus and T. 
manihoti. Using a NTSYS program, with UPGMA methods of classification realized 
with a matrix of similarity calculated with the DICE index, the following dendograms 
were obtained (Fig. 1 and 2). 
Populations of N. idaeus showed a very high similarity index of 0.96 in the case of 
populations coming from different geographic zones such as Brazil and Venezuela and up 
to 1.00 in the case of populations form the same geographic zone. These similarity 
indexes, being so high, show no significant difference and we can say they are 
monomorphic populations. 
Populations of T. manihoti display indexes of 0.2 to O. 7, tending to be more similar when 
from the same geographic zones, as in the case of Villanueva (Colombia) and Y aracuy 
(Venezuela) of 0.7. These populations come from the north coast of both countries, 
geographically close and with similar climates. It can also be concluded that the 
populations of T. manihoti analyzed show a high polymorphism. The analysis of the 
remaining populations is continuing. 
Collaborators: A.Bohorquez, J. Tohme, A. Bellotti 
11 
Table l. Phytoseiidae species anaJyzed with AFLP's. 
Species 
T. manihoti 
T. aripo 
T. tenuiscutus 
N. ideus 
G. annectens 
G. helveo/us 
Euseiusho 
N. californicus 
Country 
Colombia 
Venezuela 
Ecuador 
Brazil 
Colombia 
Brazil 
Benin 
Colombia 
Ecuador 
Colombia 
Ecuador 
Brazil 
Venezuela 
Ecuador 
Colombia 
Ecuador 
Ecuador 
Ecuador 
12 
Locality 
Cajibío, Cauca 
Chinchiná, Caldas 
Barbosa, Antioquia 
Armenia, Quindío 
Copacabana, Antioquia 
Sta. Rosa Cabal, Risaralda 
Bijagual, Santander 
Bucaramanga, Santander 
Villanueva, Guajira 
Pivijay, Magdalena 
Marin, Yaracuy 
.Calderon, Manabf 
Cruz das Almas, Bahia 
Palmira, Valle 
Pivijay, Magdalena 
Cruz das Almas, Bahía 
Ab-Calavi, Benin Station 
Los Córdobas, Córdoba 
Chone, Manabí 
Cantagallo, Portocayo, Manabí 
Portoviejo, Manabí 
Guayaquil, Guayas 
Armenia, Quindío 
Fonseca, Guajira 
Carretalito, Guajira 
La Paz, Cesar 
Danzarín, Manabí 
Petrolina, Pernambuco 
Crató, Ceará 
Capim Grosso, Bahía 
Piritiba, Bahía 
Carora, Lara 
Crucita, Portoviejo, Manabí 
Calderón, Manabí 
Charapoto, Manabi 
Rocafuerte, Danzarín, Manabí 
Fonseca, Guajira 
Santa Ana, Manabí 
El Rodeo, Portoviejo, Manabí 
Portocayo, Cantagallo, Manabí 
Santa Ana, La Teodomira, Manabí 
Santa Ana, Santa Ana, Manabf 
Rocafuerte, Danazarín, Manabí 
Puertocayo,Cantagallo, Manabí 
Portoviejo, Manabí 
Chone, Manabí 
Macha la 
0.16 0.32 0.48 0.64 0.00 
.--------------- 13- ~llfria, O'atbl:a, Chlorrbia 
'--------------- 4&- Q!ibío, OltJa.i. Chbrbia 
~----------------------------- 47- Quzch<; A!Jm;, &00, 13m.til 
45- ViUin.levd, l.aG~gin~ Colon1Jia 
46- Mtin, Yaa:uy, Vln2l.lda 
Figure l. Similarity index of five populations of T. manihoti with the PElC combination. 
o .9 6 o .9 7 o. 9 8 o. 9 9 1. 0 o 
ls . Pi rilib a. B a b (a . 8 raa i l ~------------------------------------------------------------------~ 7 · C aviru Gru, ~u . B•~(• ..--------------------ti a · P •tr• 1 i u . P u u ru bu e o . 8 ra 111 
--4 6 · Cru 6. C eari. B ruil 
9 · C u o ra , lan, V oae.~ucla 
Figure 2. Similarity index of five populations of N. idaeus with the PElC/PMlC combination. 
13 
1.1.7 Molecular characterization of the entomopathogenic fungus, Neozygites sp, 
pathogen of phytophagous mites. 
Neozygites cf. jloridana, a fungal pathogen (Zygomycetes: Entomophthorales), cause 
irregular or periodic mortalities on mite populations in Colombia and the NE Brazil. The 
pathogen has been found on cassava mites throughout many cassava growing regions of 
the neotropics. This fungus shows considerable promise for biotogical control of the 
CGM and is being studied in Brazil and Africa, as well as at CIAT. 
Due to taxonomic problems with this type of fungus, it has been difficult to determine 
exact species identification. There is interest to introduce the neotropical strain of the 
fungus into Africa to control CGM, but before this can be done, it is necessary to 
determine certain characteristics of the fungus, such as the existence of different strains 
or species. 
Material and Methods 
Molecular techniques are presently being used to determine taxonomic identification of 
entomopathogenic fungi, such as AFLP's. However dueto the difficulties in rearing this 
fungus in vitro, it is difficult to obtain sufficient quantities of DNA. We have improved 
on the method developed by Luis Leite and Donald Roberts (Boyce Thompson Institute, 
lthaca, New York) (see 1997 Annual Report). Work by Delalibera 
(CNPMFIEMBRAPA, Cruz das Almas, Brazil) has also contributed significantly to the in 
vitro culture of this fungus. 
DNA extraction was done by modifying the methodology of DNA MiniCT AB 
(developed by Zeller at Purdue University) developed for the Pyricularia fungus. The 
extraction ofDNA is made from fungal isolates that have been growing in a liquid media, 
TNMFH, for 23 days. This is then lyophilized to reduce humidíty and facilitate 
maceratioq. Fu AFLP's, the methodology developed by Vos et. al. (1995), with sorne 
modifications was used. Two kits were tried, the Small Genome of Gibco BRL 
Technologists, and the kit that contains the primers and adaptors of Operon 
Technologists. 
Results and Discussion 
lt was possible to obtain (Zeller technique) DNA of good quality and concentrations of 5 
to 150 ng, depending on the quantity of tissue obtained from the in vitro culture. Strains 
from severa! different sources were used in the analysis (Table 1). In sorne cases (Mt 
Benin) we were unable to obtain sufficient DNA to observe bands in the gels, perhaps 
dueto the insufficient amount ofhifal bodies produced in the in vitro culture. 
14 
Tab1e l. Strains of Neozygites used for DNA a na1ysis collected from severa! sources. 
Mite Host 
Tetranychus urticae 
T. urticae 
T. urticae 
Mononychellus tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
M tanajoa 
Locality where collected 
CIA T, Palmira, Valle (Colombia) 
CIAT, Palmira, Valle (Colombia) 
A vakpa, Benin 
CIAT, Palmira, Valle (Colombia) 
CIAT, Palmira, Valle (Colombia) 
CIAT, Palmira, Valle (Colombia) 
Media Luna, Magdalena (Colombia) 
Santander de Quilichao, Cauca (Colombia) 
A vakpa, Benin 
Cruz das Almas, Bahía (Brazil) 
Cruz das Almas, Babia (Brazil) 
Caruaru, Pernambuco (Brazil) 
Piritiba, Bahia (Brazil) 
N ame 
TuCIAT 1 
TuCIAT2 
TuBenin 
MtCIAT 1 
MtCIAT2 
MtCIAT 3 
MtML 
MtSQ 
MtBenin 
MtCDA 1 
MtCDA 2 
MtCar 
MtPir 
With the small Genome kit it was possible to obtain good banding patterns, although the 
numbers are small (10 to 20) and insufficient to make a true characterization. With the 
Operon kit, it was possible . to evaluate more primer combinations (Table 2). The 
combinations with the greatest numbers of bands were + 1/+ 3 and the analysis has 
focused on these types of combinations. At this time we only have partial 
characterization that indicate marked differences between the fungal strains. However 
these analysis need to be repeated to confirm these results and the statistical analysis is 
being done. 
Table 2. Strains of Neozygites and primer combination for AFLP analysis. 
No. Strains of Neozygites sp. Type of 
Combination Combination No. of Bands 
6 +2 1 +3 PE1AAIPM1A 32 
8 +2 /+3 PE1AAIPM1B 44 
8 +2 1 +3 PEIAAIPM1C 39 
5 +2/+3 PE1AAIPM1E 24 
5 +2 /+3 PE1AAIPMIF 40 
3 + 1 1 + 3 PE1AIPM1B 44 
3 + l/+3 PE1AIPM1C 40 
3 + 1 1 + 3 PE1AIPM1D 34 
3 + 1 1 + 3 PE1AIPM1F 57 
3 + 1 1 + 3 PEIAIPMIG 70 
3 + 1 /+ 3 PE1AIPMIH 64 
3 + 11+ 3 PE1AIPM11 67 
Collabora tors: A.Bohorquez, J. Tohme, A. Bellotti 
1.1.8 Studies on the Viral Pathogens found in Mononychellus tanajoa and M. 
caribbeanae. 
Previously we have reported at CIA T, epizootics causing high mortality in field and 
greenhouse populations of M tanajoa and M caribbeanae. The pathogen, analyzed 
under the election microscope appears to be of viral origin. The particles are 
15 
approximately 60 nm in size. Extractions of DNA have been made from the infected 
mite populations for molecular analysis. In agarose gels that displayed complete DNA, 
the presence of six different sized molecular bands was observed. This confirmed the 
presence viral particles. It has also been observed that these viral epizootics occur 
normally during wet or humid periods. 
Reports in the literature on viral diseases of mites are few and little is known about its 
biology, mode of infection, symptoms, transmission, mite stages attacked, etc. In 
addition it is not known if it will attack beneficia! mite predators. The present study is 
aimed at understanding and describing the signs and symptoms, oviposition of diseased 
females, mortality, viral presence all mite stages and transoval transmission. 
Material and Methods 
Four cassava leaves containing diseased M tanajoa mites, and four leaves with diseased 
M caribbeanae mites were used. From each leaften repetitions of each mite stage (eggs, 
larvae, nimphs, and adults), that showed symptoms of viral ínfectíon were removed. 
Infested leaves with healthy mites were used as a control. Observations with the election 
mícroscope were used to insure that mites were infected with the virus. 
The biology, mortality and oviposition of mite stages were observed for 12 days. Each 
infected development stage was placed on a clean cassava leaf disc in a petri dish with 
moist tissue and in a 23 to 25°C chamber. 
Viral disease symptoms are expressed by a clear amber discoloration of mite eggs, larvae 
and nymphs, and dark amber to brown colored adults. Upon death, hemocyle can be 
observed coming out of the anal pore resulting in a completely dry cadaver. Mortality is 
probably produced by the virus as virus particles have been found in the cadavers, of all 
stages of both species of mites. Mortality is highest in eggs and adults which could be 
due to the delicacy of the eggs and the age of the adults and the time of ínfectíon of the 
virus (Fig. 1). After 12 days, individuals in the control treatments showed no evidence of 
viral infection and those examined under the electron microscope showed no virus 
particles. 
Oviposition by díseased females was greatly reduced when compared to healthy females 
(Fig. 2). When the F2 generation was evaluated mortality for all the stages of M 
caribbeanae was 25.3% and for M tanajoa 29.8% due to the virus. These results 
indicate that the virus is tranovil transmission of the virus. The F2 generation of the 
control mites did not show these symptoms. 
Preliminary results with this virus disease indicate a good potential for biological control 
of cassava mites. Considerable research, however, still needs to be accomplished before 
this disease can be of practica! use in an IPM program for control of cassava green mi te. 
Collaborators: A.Bohorquez, J. Tohme, A. Bellotti 
16 
25 
• ! 20 
e: 
• e 1s 
• 11. 
10 
5 
M. tanajoa M. caribbeanae 
Stages 
Figure l. Percent Mortality of for Stages of Mononychel/us tanajoa and M caribbeanae due to 
Virus Infection. 
100 E!l M. tanajoa 
Q) 80 C') 
ca 60 ~ S: 
Q) 40 (,) 
... 20 Q) 
a. o 
Eggs Larva e Ninfs Adults 
Stages 
Figure 2. Oviposition of M. tanajoa and M caribbeanae females infected with a virus disease. 
17 
1.1.9 Characterization and genetic variability analysis of Passijlora L. sp. from the 
Andean region, using molecular markers. 
Introduction 
The characterization and evaluation of wild and cultivated population of Passiflora is a 
priority for Andean countries. 
In this study genetic characterization of variability evaluation with RAPD markers and 
mutational analysis of chloroplast DNA variation have been used to explore the evolution 
and the relationship of species complex in Passiflora. 
The results by genomic and molecular characterization derived from Passiflora analysis 
will provide the bases to characterize different species, accesions, and origins, traits in 
intra and interspecific crosses for further improvement with multiplication of selected 
genotypes and conservation of the Passiflora as a valuable genetic resource. This work 
will contribute to the recent national efforts to collect, characterize and evaluate the 
Passiflora fruit germplasm. 
Material and Methods 
The collections of Passiflora a were established at the Experimental Station La Selva, 
Rionegro (Antioquia) of CORPOICA. DNA Extraction was described by Dellaporta et 
al. (1983). The RAPD markers were carried out as described by Williams et al. (1990). 
For the RFLP, five cloned fragrnents of cpDNA from Mung bean (4) were used as probes 
for detecting variation in the Passiflora chloroplast genome. DNA labeling and 
hybridization procedures were carried out as described by Angel et al. (1993). Three Psti 
chloroplast probes from Petunia hybrida were also used. RFLP and RAPD band patterns 
were scored as a 1/0 data matrix. The dendrograrns were constructed by the clustering 
method of UPGMA. 
Results and Discussion 
The resuls with RAPD markers carried out on 52 accessions representing 14 species of 
the genus Passiflora L. showed polymorphism, indicating high leve! of polymorphism 
intra species in P. /igularis given good bases for the improvement of this species. The 
RAPD markers also indicated a great genetic flow ínter species specially between P. 
india and P. mollissima. This result of a great value for carring out taxonornic 
classification studies, shows similarity coefficients ranged from 0.929 to 0.075 showing a 
diverse genepool in the genus. 
The studies with 71 more accessions shows a large intraespecific variation in P. ligularis 
confirming previous studies published in Euphytica 101 :341-347. 1998. Further estudies 
will be carried out with new accessions. The results with RFLP markers demostrated 
profound cpDNA diversity among the species evaluated. Intraspecific variatiation was 
observed in four of seven species. 
18 
References 
l. Escobar, L.K. 1988. Flora de Colombia. Ins. Cien. Nat., Mus. Hist. Nat., Fac. Cien., 
Univ. Nacional. p.l4. 
2. Dellaporta, S.L., Wood, J. and Hicks, J.,B., 1983. A plant DNA minipreparation: 
version II. Plant Mol.Biol. Rep. 1:19-21. 
3. Williams, J.G.K., A.R. Rubelik, K.J. Livak, J.A. Rafalski & S.V. Tingey, 1990. DNA 
polymorphism amplified by arbitrary primers as useful as genetic markers Nucleic Acid 
Res 18:6531-6535. 
4.Palmer, J.D. and Zamir, D. 1982. Chloroplast DNA evolution and phylogenetic 
relationships in Lycopersicon. Proc .. Natl. Acad. Sci. USA, 79: 5006-5010. 
5. Angel F., Arias, D.I. , Tohrne, J., Iglesias, C., and Roca C., 1993 
Biotechnology 31:103-113. 
Joumal of 
Collaborators: l. Sánchez1, F. Angel2, D. Fajardo, Maria Femanda Castillo, J. Tohrne, 
and W. Roca. 
l . CORPOICA-CIAT; 2. Current address: CENICAÑA, Cali. 
1.1.10 Biochemical and molecular characterization of the Musa L. Colombian 
Collection. 
Introduction 
The Colombian Musa Collection has 130 accesions but very few of them have been 
characterized at a morphologic and agronomic level. The genetic studies which we 
propase, based on a molecular characterization, will allow a better understanding of 
Musa's geno me and genetic diversity of the CCM. 
This proyect also allow the identification of duplicated accessions of Colombian Musa 
Collection (CCM), thus contributing to reduce the costs of mantaining this collection in 
vitro and in field, by allowing its appropriate management. 
Material and Methods 
The agronomic characterization will be done to the field collection which is kept at the 
Agrado, Armenia - Quindío, 131 O m.o.s.l. 
DNA was isolated, using the method of Doy le, 1990. 
19 
The molecular biology methods that will be used are: AFLP (Amplified fragment length 
plymorphism) described by VOS, P. et al. 1995. and Random Amplified Polymorphic 
DNA (RAPD), described by Williams, et al.1990. 
Result and Discussion 
Pattem differences between the accessions were evident in eight of the 52 tested RAPD 
primers. The dendrogram with RAPD markers diferentiated the plantain groups (AAB 
and ABB) from the acuminata.(AA and AAA). 
From 32 primer combinations of AFLP tested, the combinations E-AAG+M-CAT and 
E-AAG+M-CTT were selected. The number of bands scored per gel was of 166 which 
an average of 30% were polymorphic. similarity matrix based on the similarity 
coefficient of DICE (1945), was produced. The greatest similarity (100%) was found 
between of the cavendish, gros michel genotypes, larger differences were found between 
poupoulou genotypes. A dendrogram based on average linkage cluster analysis was 
generated from the AFLP data The dendrogram shows clustering of the genotypes 
according to their genome origin . 
AFLP analysis was found to be higly effective in distinguish genotypes from genome 
groupe. The percent similarity values between the 128 Musa genotypes, using one primer 
combination, ranged from 0,4 to 100%. In the previous study isozyme electrophoresis 
has been shown to have potential for discriminating between cultivars of Musa L. and 
RAPD variation was also found to be effective in distinguishing among the groups 
acuminata and balbisiana. On the other hand AFLP markers reveal differences within 
each group. Mientras que los AFLP muestran las diferencias existestes al interior de cada 
grupo. The AFLP markers data show that, genetically, the plantain genotypes are more 
closely related to those of the bananas. In the present study, with a few exceptions, the 
128 genotypes clustered together according to genome type. 
Future Activities 
Characterization of Musa clones using microsatelites primers, given by the CIRAD -Iabs. 
References 
Dice, L. R. 1945. Measures of the amount of ecologic association between species. 
Ecology. 26: 297-302. 
Doyle, J.J. , J.L. Doyle LE. 1990. Isolation of plant DNA from fresh tissue. Focus. 
12: 1:13-15. 
F AO, 1991 . Policy Developments and Prospects for Coordinated Intemational Action 
on Bananas. F AO, Rome. 
Horry JP. Chimiotaxonomie et organization génétique dand le genre Musa. Parts 1 ,2,3 . 
Fruits 44:455-474, 509-520, 573-578. 
20 
Rowe Pr ., 1981. Breeding an intractable crop . Bananas in: Rackie KO. Lyman JM 
(eds) Genetic engineering for crop improvement. Working Papers. The Rockeffeller 
Foundation., pp 66-84. 
Scandalios JG. 1969. Genetic control of multiple molecular forms of enzymes in plants: a 
review. Biochem. Genet. 3:37-79. 
Simmonds NW . 1962. The evolution of the bananas. Longman. London. 
Vos, P. , Rene, R., Marjo, B., Martín, R., Theo Van De Lee, Miranda, H., Adrie, F., 
Jerina,P., Martín, K., and Marc, Z., 1995. AFLP; a new technique for DNA 
fingerprinting. NAR. 23: 21: 4407-4414c. 
Williarns JGK, Kubelik AR., Livak KJ., Rafalski JA., Tingey SV. 1990. DNA 
polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic 
Acids Res. 18:6531-6535. 
Collaborators: 
Inés Sánchez 1, Duverney Gaviria1, Gerardo Gallego, Jorge Alberto Valencia2, Mario 
Lobo3, Joe Tohme, WilliarnRoca 
1
. CORPOICA- CIAT; 2 • CORPOICA- El Agrado, Armenia; 3 • CORPOICA- Río 
Negro. 
ACTIVITY 1.2 ldentification and mapping of useful genes and gene pools 
SUMMARY OF ACHIEVEMENTS 
• QTLs for P uptake by roots has been mapped; and major and rninor resistance genes 
to BGMV, bean anthracnose and bean angular leaf spot have also been mapped; and 
QTL for BNF in presence of low P was tagged. 
• Using the strategy of degenerate primers, we have identified sequences of a putative 
anthracnose resistance gene in beans and of a Pyricularia R-gene in rice. 
• A genomic region involved in general defense against cassava bacteria! blight was 
localized using the cassava molecular map. But initial work found no QTLs 
associated with marker genes involved in cassava post harvest deterioration. 
• A region ofthe Brachiaria genome was associated to the apornixis gene and linked to 
an RFLP marker at 3 cM from the gene, to an AFLP at 11 cM and to a SCAR marker 
at 3 cM from the gene. All these markers correspond to the rice chromosome l. 
21 
1.2.1 Mapping of important agronomic traits in bean 
Introduction 
Last year we reported on the tagging and mapping of genes in a population of RILs 
derived from the cross of DOR 364 x G19833. QTL for two traits were mapped 
previously: root architectural differences that contribute to phosphorus (P) uptake in 
bean; and resistance to BGMV. Since that time we have obtained additional data for 
BGMV reaction, for other traits related to P absorption, and for other diseases. 
Materials and Methods 
In addition to data on root growth and P uptake in the field in Darién, Colombia, this year 
we obtained from colleagues at Penn State University, data on basal (or crown) root 
development, on root hair development, on leaf acid phosphatase activity, and on H+ ion 
exudation. At CIAT the RILs were evaluated for resistance to three different isolates of 
the anthracnose pathogen, and one isolate of the angular leaf spot (ALS) pathogen. All 
these isolates were characterized previously as Mesoamerican in reaction, as were used to 
test for the presence ofresistance genes derived from the Andean parent, G19833. Days 
to flower, seed size and grain yield were also registered. Phenotypic data were analyzed 
with the QGene program to identify likely regions in the genome for QTL. 
Results 
It was possible to identify QTL for all the traits for which the analysis was performed. 
Thus, this is perhaps the most agronomic traits of common bean for which genes have 
been mapped on any single map to date. The mapping of QTL for the severa! traits 
related to P uptake permitted a far better understanding of the mechanisms that contribute 
to P acquisition in bean. Similarly, the mapping of resistance genes for anthracnose and 
ALS revealed aspects of the genetics of resistance. Both major and minor genes were 
identified. The biological dimensions of these results are discussed in sections 1.1 and 
3.2 of Project IP-1. For purposes of the present report, we shall refer to an overview of 
the bean map that is emerging from these activities. 
For the traits analyzed, at least 28 important regions of the genome were identified. 
These were distributed across all eleven linkage groups of the common bean genome, and 
are represented in Fig.l was observed that QTL for traits related to P absorption tended 
to cluster together. This was observed in almost all linkage groups. Sorne of these traits 
may in fact have a common physiological basis, although not necessarily. Likewise, 
sorne coincidence of resistance genes for anthracnose and ALS was observed ( eg, around 
the region marked by the RAPD AA193D in linkage group B03). This could occur if 
these genes have a common evolutionary origin. 
However, clustering of genes for apparently distinct traits was also observed. For 
example, in linkage group B03, the RAPD P076G was associated with P uptake and root 
architecture, with anthracnose resistance, with flowering date and with seed size. In 
linkage group B 11, the RFLP BngOO 1 was associated with P uptake and root architecture, 
22 
with anthracnose and BGMV resistance, and with seed size. Many other examples were 
observed. The significance of this for the evolution of the crop genome is not clear. 
However, apart from the genetics of any single trait or possible applications in marker-
assisted selection - the clustering of genes for very different types of traits is significant 
from the standpoint of breeding. lf useful genes are linked in repulsion, it will be 
difficult to recombine such genes in breeding lines, in which case breeders may need to 
prioritize among closely linked genes and choose those that are indispensable for their 
objectives. Breeders may wish to develop strategies to utilize those genes that are not 
linked when different genetic strategies are available ( eg, altemative genes for a disease 
resistance), or to make a specíal effort to detect genetic recombinants. This issue is 
especially important when breeding objectives call for the improvement of a larger 
number of traits. In any case, a systematic strategy requires the continued mapping of 
genes. 
Collaborators: S. Beebe, F. Pedraza, J. Tohme 
MAPPING OF QTL FOR AGRONOMIC TRAITS 
·: .H 
P, IH _.OH .• 
ll.~r. \ 
-
¡,\,l 
-
?.1<1 . ....... ./ 
''"·" ::::::::. 
-
23 
" 
.•.. 
f'.RI . \m. 
B( •. S.>; 
~ '"'·"·' / ~ 
1.2.2 GeneTagging with PCR Markers for Bean Breeding: Adaptation to Low 
Fertility Conditions 
Introduction 
Phosphorus (P) deficiency is thought to be a primary limitation on Biological Nitrogen 
Fixation (BNF) of cornrnon bean in tropical soils. In 1997 it was reported in Project IP-1 
that bean accessions had been identified for the tolerance of BNF to low soil P. A 
breeding line, BA T 477 was found to be relatively tolerant to low P with regards to SNF. 
It has also demonstrated excellent N fixation capacity in unstressed conditions in two 
contrasting soil types and in drought conditions. A germplasm accession , 03513 that 
was previously identified as tolerant to low P in grain yield was also found to be 
relatively tolerant in BNF. In contrast, DOR 364 was found to be sensitive to low P with 
regards to BNF. The work with the sources of low P tolerant BNF was extended to the 
search for the genes responsible through the use of molecular markers. 
Materials and Methods 
RILs were developed of crosses of DOR 364 x BAT 477 (DxB)and DOR 364 x 03513 
(DxO). These were evaluated in the F5 and F6 generations in the case of DxB, and F4 
and F5 generations for DxO. A greenhouse sand.culture system was used employing 3 
treatments: 1) Low P+BNF, in which plants were inoculated with Rhizobium tropici 
strain CIAT 899; 2) Low P without Rhizobiurn, to estímate the contribution of seed N to 
plant growth; 3) High P with Rhizobium, to estímate BNF capacity in the presence of 
adequate P supply. Data taken on all treatrnents included aerial, root and total DW. %N 
and %P analyses were performed on the Low P+BNF treatrnent in both the F5 and F6 
generations, using a bulk sarnple of ground tissue from the four repetitions. Shoot'root 
ratio, Total N and Total P were calculated from primary data. 
After an initial screening of more than 300 primers to identify those that produced 
polymorphisms among the parents, primers were tested on genotypes in.each population 
that presented extreme values in biomass accurnulation and/or total N. Eventually 42 and 
32 primers were evaluated on all RILs of DxB and DxO, producing 82 and 49 RAPD, 
respectively. A total of 12 linkage groups were thus formed in the population DxB, 
identified as DB1 to DB12, with 9 RAPD unlinked. These represented about 615 cM 
which is approximately half of the total length of the bean genome (Vallejos et al, 1991: 
Nodari et al, 1993; Beebe et al, 1998). N in e linkage groups were formed in the 
population DxO covering 563 cM. 
Results: Severa! genomic regions in both populations were identified as containing QTL 
for different traits. In the popu1ation DxB, the most important linkage group, DB5, was 
composed of only three RAPD covering 5.1 cM. Within this group the marker N201 
explained 16% of D W and 1 0% of Total N variability in the Low P + BNF treatrnent in 
the F5 generation. In the F6 generation the magnitude of its effect dropped to about 5%, 
but was significant nonetheless. This linkage group was also the only one in the DxB 
cross that presented a consistent effect for %P. In linkage group D B 1 O the RAPD U 1403 
hada significant effect on DW (13%) and also Total N (9%) in F5, while expressing a 
24 
1 
much reduced effect in F6. Markers in groups DB13 andan unlinked marker 01002 also 
presented relatively stable effects. The possible agronomic significance of these QTL is 
discussed in the IP-1 report. Field trials are underway to determine the contribution of 
the QTL to field performance. 
In the DxG population, QTL were identified in the Low P + BNF treatment that were 
expressed in both generations for: root DW; shoot DW; root/shoot ratio; %N; %P; and 
NIP ratio. Subsequently the RILs of DxG were planted in the low P field in Darién and 
were evaluated for yield. Three promising QTL were identified for yield at Iow P, but 
only one of these corresponded to a QTL recognized in the greenhouse system, where it 
expressed an effect on root/shoot ratio. While it is hopeful that yield QTL can be 
identified, these yield QTL remain to be confirmed with additional yield trials. It remains 
to be seen what significance exists in the field for the QTL that were marked in the 
greenhouse in the DxG cross. G3513 expresses other unique traits related toN transport, 
therefore the RILs will be used for studying those traits and the partial genetic map that 
has been developed will possibly be useful for those traits as well. 
R' 1 
Linkage Size DW %N %P Total N N/P DW 
Group 
081 
085 
0810 
DBlJ 
UL 
Adj R' 
(bp) high p 
F5 F6 F5 F6 F5 F6 F5 F6 F5 F6 F6 
Marker RSq RSq RSq RSq RSq RSq RSq RSq RSq RSq 
01601 B 6.1 B 0.8 o 0.3 o 0.2 B 0.4 o 0.9 B 6.02 B 0.79 o 1.1 B 0.6 B 
H1201 B 7.9 B 1.9 o 0.2 o 0.3 B 0.1 o 1.5 B 8.33 B 1.77 o 0.4 B 1 B 
P701 B 8.8 B 3.7 o 2.8 o 0.4 o o o 2.1 B 5.74 B 3.58 o 0.6 B 0.7 B 
X303 B 3.1 B 0.5 B 0.1 B 1.9 B 1.5 B o B 4 R 3.47 o 1 B 2.5 o 
N201 B 16 B 4.7 o 5.3 o 0.1 D 5.8 D 3.7 B 10.4 B 3.28 B 2.3 B 3.8 B 
Ul80l B 11 B 5.2 o 3.5 B o D 6.3 D 4 B 8.93 B 4.89 B 3.9 B 7.3 B 
02001 B 12 B 5.1 o 3.4 o o D 6.3 D 5 B 9.2 B 4.39 B 4 B 7.8 B 
M901 B 6.3 B o o 6.5 o 5 o 0.1 o 3 B 2.29 o 0.98 o 0.6 o 0.1 B 
U1304 B 6.6 B 0.6 o 5.3 o 6 o 0. 1 o 6.7 B 2.84 D 0.26 o 0.6 B 0.2 B 
Ul303 B 7.5 B 1.1 o 5.3 o 5.2 o 0.4 o 6.3 B 3.52 o 0.01 o 0.1 B 0.3 B 
X903 B 10 B 3.4 o 2.2 o 3.5 o 0.3 o 5.4 B 8.03 B 0.85 B 0.2 B 0.6 B 
U1403 B 13 B 3.1 o 3.7 o 3.2 o 0.3 o 6.3 B 9.08 B 0.77 o o B 1 B 
U1402 B 13 B 3.1 o 3.7 o 3.2 o 0.3 o 6.3 B 9.08 B 0.77 D o B 1 B 
M902 B 12 B 0.8 o 3.6 o 2.6 o 0.3 o 2.8 B 8.21 o o o o o o B 
Q1701 o 1.4 o 5.2 B 0.4 B 0.1 o 0.2 B 1 o 1.14 o 4.22 B 0.9 D 0.8 o 
Q1001 D 5.6 D 5.3 B o B o B 0.2 B 0.6 D 6.33 D 5.08 o 0.1 o 0.3 D 
X901 o 6 D 3.6 o 0.3 o 1.3 o o B 0.3 o 10.5 D 7.97 o 0.7 o 4 D 
G1002 D 3.7 D 8.9 o o o o B 0.7 B 3.4 D 4.95 D 11.9 o 0.7 o 5. 1 B 
Conclusions 
Severa! QTL were identified that were expressed in two generations in a greenhouse test 
of SNF under low P, although the level of expression varied widely. So me of these were 
also expressed at high P. At least one should be considered for conversion toa SCAR for 
use in marker assisted selection. 
Collaborators: S. Beebe, F. Pedraza, A. Velázco, J. Lynch1, G. Ribet, J. Tohme. 
l. Pennsylvania State University. 
25 
RSq 
0.6 
0.2 
0.9 
o 
6.8 
4.5 
3.4 
3.7 
2.5 
3 
6.2 
5.3 
5.3 
5.9 
0.8 
6.2 
3.8 
0.5 
1.2.3 ldentification of Disease Resistance Gene Analogs (RGA): ldentification of 
NBS type putative resistance gene in common bean. 
Introduction 
In the past few years severa! resistance genes have been isolated and cloned different 
plant species, using either map-based cloning or transposon-tagging. Molecular 
characterization of these genes uncovered common sequence motífs, even though they 
confer resistance to a wide spectrum of pathogens, i.e., viruses, bacteria or fungi. The 
majority of cloned R-genes are characterized by the presence of an N-terminal nucleotide 
bínding sites (NBS) and a C-terrninal stretch of leucine-rich repeats (LRR). The presence 
of these conserved domains allowed the grouping of these genes into severa! classes, and 
established their possíble functíon in the defense response as part of the sígnal 
transduction pathway (Baker et al., 1997). Proteins similar to serine-threonine kinases are 
another group which has been suggested to interfere in protein phosphorylation, one of 
the common mechanisms of protein control (Bent, 1996). Other families ínclude 
transmembrane receptors either with long extracytoplasrnic LRR domains, like Cf-2 and 
Cf-9 in tomato, or with a LRR transmembrane region as seen in the product of HS 1 pro·! 
from sugar beet. There are also transrnembrane receptors with extracellular LRR 
domains and also an intracellular serine/threonine kinase, like the one encoded by Xa21 
in rice (Baker et al., 1997). 
PCR-based cloning with degenerate primers rnade possible the identification of 
Resistance Gene Analogues (RGAs) in rice, A. thaliana, and lettuce (Leister et al., 1998; 
Aarts et al., 1998; Shen et al., 1998). Their sequences have shown high homology with R-
genes previously reported, and they rnapped within or near to disease resistance loci. 
We have initiated a project to use this strategy to identify potential R-genes in rice and 
Phaseolus vulgaris L. 
Common bean (Phaseolus vulgaris L.) production is severely affected by severa! diseases 
such as anthracnose, angular leaf spot and Bean Golden Mosaic Virus (BGMV). 
Anthracnose is caused by the fungi Colletotricum lindemuthianum (Sacc et Magn) and is 
the principal pest in Africa and Latin America. Genetic analyses of the disease, and race-
cultivar specificity, strongly suggest a gene-for-gene interaction (Adam-Blondon et al., 
1994), where the resistance is determined by the interaction of dominant resistant genes 
(R) in the plant and the corresponding avirulence genes (A vr) in the pathogen. 
Materials and Methods 
DNA was extracted from leaftissues oftwo genotypes, G19833 and DOR364. G19833 is 
highly resistant to anthracnose and sorne strains of angular leaf spot but susceptible to 
BGMV, whereas DOR364 is susceptible to angular leaf spot and anthracnose, but is 
resistant to BGMV. Data on 87 RIL-F9 plants from the cross of G19833 by DOR364 
screened wíth severa! Andean and mesoamerican isolates of anthracnose, angular leaf 
spot and BGMV are available. A RAPD, RFLP and AFLP map is also available (BRU 
annual report, 1997). 
26 
PCR reactions were performed using degenerate primer combination I (s 1 y as 1) and 11 
(s2 y as2) (Table 1). These primers were designed based on Leister et al. (1996). The 
PCR products were separated by electrophoresis in a 1.2% low melting point agarose 
(LMP -GIBCO-BRL). Each band was eluted and purified with the PCR-Preps kit 
(Promega). PCR-purified products cloned into the pGEM-T vector system (Promega) and 
transformed into E.coli DH5a cells by electroporation following GIBCO-BRL 
instructions. One hundred clones from each band were randomly picked and grouped 
based on restriction digestion pattems using 4-base-pair cutting enzymes. At least three 
clones from each group were used as probes. They were hybridized to blots containing 
parental DNA digested with a set of five restriction enzymes (EcoRI, EcoRV, Xbai, 
Hindiii y Drai). RFLP segregation was evaluated on the 87 RIL-F9 plants. Polymorphic 
probes and at least one clone from each group were sequenced usíng the Dye Terminator 
Cycle Sequencing Kit and an Applied Biosystems Prism 377 DNA sequencer (Perkin-
Elmer). 
Table l. Primers used for PCR based on Leister et a l. (1996). 
N-terminal 
Peptide sequence G G V G K T T 
Primer s l GGT GGG GTT GGG AAG ACA ACG 
Primers2 GGI GGI GTI GGI AAI ACI AC 
e-terminal 
Peptide sequence G L p L A L 
Primer asl CAA CGC TAG TGG CAA TCC 
Primer as2 lA A IGC lAG IGG IAA ICC 
Results 
With primer combination I it was possible to amplify a single band, RGAl.l, of about 
1200 bp. By using primer combination II we detected three bands: RGA2.1 (1200bp), 
RGA2.2 (800bp) and RGA2.3 (500bp). The restriction pattem of the RGAl.l was the 
same for 100 clones, indicating homogeneity of PCR products (i.e., a single group was 
formed). The restriction pattem of bands RGA2.1 , RGA2.2 and RGA2.3 was 
heterogeneous allowing the creation of groups 9, 10 and 7 respectively. 
We have sequenced 35 clones from different groups of each band. A GenBank search, 
using the BLASTX algorithrn, of sequences from bands RGA2.1 , RGA2.2 and RGA1.1 
did not show any significant homology with R-genes or RGAs reported in this database. 
Eighteen clones of band RGA2.3 were sequenced and grouped, according to sequence 
homology, into three classes: 1) Class I contained five clones identical to each other (98% 
sequence identity); 2) Class II included clone RGA2.3.33.alone, although it had 73% 
sequence identity to Class 1; and 3) Class III formed by clone RGA2.3.78, with 47% 
sequence identity to Class l. A fourth class grouped clones that did not have homology to 
the aforementioned classes, nor homology to any R-gene or RGAs. The majority of them 
27 
did not have continuous open reading frames and their hybridization patterns suggested 
multiple-copy sequences. 
Clones of Classes 1 and 11 showed significant sequence similarity not only to resistance 
genes but also to like-NBS RGAs. However, clone RGA2.3.33 did not have a full-length 
open reading frame, suggesting that it is a pseudogene. A Gene:3ank search of clone 
RGA2.3.78 revealed high homology toRGAs ofthe kind NBS-LRR. 
Several clones of the first class, and the clone RGA2.3.33, were mapped and were placed 
at the distal regían of linkage group B 11 of the integrated new bean map developed by 
Gepts's lab a t UC Davis. A highly significant QTL for anthracnose resistance (LR140) is 
mapped at this regían (Pedraza, F. Personal communication). These clones may be part of 
the anthracnose R-gene of common bean. 
Sequence of bean NBS gene fragment have been deposited in the GenBank database 
(accesion no. AF084026) (Figure 1). 
On going activities will concentrate on: 
4. Mapping RGA2.3.78r 
5. Using different degenerate primers to amplify new RGAs 
6. Sequencing more clones, 
7. Generating cDNA libraries to isolate the candidate R-gene. 
Figure l. Sequen ce of bean NBS gene fragment GenBank database accession no. 
AF084026 
1 acgactctcg ctcaacatgt attcaatgac ctgagggtgg atgaggctaa atttgatgtt 
61 aaagtttggg tttgtgtttc agatgaattt gatgttttca agatatctag agcaattctt 
121 gaggcagtta ctaaatcagc cgatgatagt agagatctgg agatggtcca tagaggaatg 
181 aaagaagaat tgacgggaaa gaaatttctt cttgttttgg atgacgtttg gaacgaaaac 
241 caacctaaat gggaggaagt gcagaagccc cttgttttag gagtccaagg gagtaagatt 
301 cttgtgacca cacgtagtaa ggaagttgct tctaccatgc gttcagaaga atactcccta 
3 61 caacaattac aagaagatga ttgttggaag ttgtttgcta aacatgcatt tcgaggtgat 
421 tgtactcaac taaacccaga gtgcaagaag attggaatga agattgttaa gaaatgtaaa 
/translation="TTLAQHVFNDLRVDEAKFDVKVWVCVSDEFDVFKISRAILEA VTKS 
ADDSRDLEMVHRGMKEELTGKKFLLVLDDVWNENQPKWEEVQKPLVLGVQGS 
KILV 
TTRSKEV ASTMRSEEYSLQQLQEDDCWKLF AKHAFRGDCTQLNPECKKIGMK.IV 
KKCK" 
Reference: 
Aarts, M.G.M., Hekkert, B.L. , Holub, E.B., Beynon, J.L., Stiekema, W.J., Pereira, A. 
1998 Identification of R-gene homologous DNA fragments genetically linked to disease 
resistance loci in Arabidopsis thaliana. MPMI 11 : 251-258. 
28 
Adam-Blondon, A.F., Sevignac, M., Banneerot, H., Dron, M. 1994. SCAR, RAPO and 
RFLP markers linked to a dominant gene (Are) conferring resistance to anthracnose in 
common bean. Theor. Appl. Genet. 88: 865-870. 
Baker, B., Zambryski, P., Staskawicz, B., Dinesh-Kurnar, S.P. 1997. Signaling in Plant-
Microbe Interactions. Science 276: 726-733. 
Bent, A.F., Kunkel, B.N., Dahlbeck, D., Brown, K.L., Schrnmidt, R., Giraudat, J., Leung, 
J ., Staskawicz, B.J. 1994. RPS2 of Arabidopsis thaliana: a Leucine-Rich Repeat class of 
plant disease resistan ce genes. Science 265: 1856-1860. 
Lawrence, G.J., Finnegan, E.J., Ayliffe, M.A., Ellis, J.G. 1995. The L6 gene for flax rust 
resistance is related to the Arabidopsis bacteria! resistance gene RPS2 and the tobacco 
viral resistan ce gene N. Plant Cell 7: 1195-1206. 
Leister, D., Ballvora, A., Salamini, F., Gebhardt, Ch. 1996. A PCR-based approach for 
isolating pathogen resistance genes from potato with potential for wide application in 
plants. Nature genetics 14: 421-430. 
Leister, D., Kurth, J., Laurie, D.A., Yano, M., Sasaki, T., Devos, K., Graner, A., Schulze-
Lefert, P. 1998. Proc. Natl. Acad. Sci. USA 95: 370-375. 
Shen, K.A., Meyers, B.C., Islam-Faridi, M.N., Chin, D.B., Stelly, D.M., Michelmore, 
R.W. 1998 Resistance gene candidates identified by PCR with degenerate 
oligonucleotides primers map to clusters of resistance genes in lettuce. MPMI 11 :815-
823. 
Whitham, S., Dinesh-Kurnar, S.P., Choi, D., Hehl, R., Corr, C., Baker, B. 1994. The 
product of the tobacco mosaíc virus resistance gene N: similarity to Toll and the 
lnterleukin-1 receptor. Cell 78: 11 O 1-1115. 
Collaborators: López,C., Gallego,G., Gaitán,E., Pedraza,F. and Tohme, J. 
1.2.4. Gene Tagging of Resistan ce to the African Cassava Mosaic Disease (ACMD). 
Introduction 
The Ugandan epidemic of the African cassava mosaic disease, caused by a recombinant 
strain of the African cassava mosaic virus (ACMV), reinforces the need for new 
approaches to resistance breeding as a preventive measure for new situations which may 
arise elsewhere in East Africa and sub-Saharan Africa, as the virus continues to evolve. 
Identifying and pyramiding disease resistance genes will provide stable resistance against 
a broad spectrum of the viral strains. Pyramiding genes, known only by indistinguishable 
phenotypes, requires molecular marker-aided genetic analysis. Map-based cloning of 
resistance genes, using the existing BAC library and tagged genes, will facilitate even 
more efficient movement of genes around cassava genepools, and is the ultimate goal. 
Strategies for gene tagging of disease resistance genes in out breeding crops have been 
29 
described and the require a population segregating for disease resistance and a molecular 
genetic map. 
A half-sib backcross mapping population, generated at CIA T by crossing, in both 
directions, a set of 5 F 1 plants, chosen for their profuse flowering, to their maternal 
parent, TMS30572, an ACMD resistant improved cassava variety from West Africa, is an 
ideal segregating population. The paternal parent of the F 1 progeny, CM2177-2, an 
improved cassava variety adapted to the Colombian Caribbean coast, is susceptible. The 
back-cross population was established in vitro from immature seeds, 40 days after 
pollination, to considerably reduce the long period required for establishing cassava 
mapping populations. An existing embryo culture protocol for immature cassava seeds 
was modified for in vitro gerrnination of the immature seeds. A total of 276 plants were 
transferred to to the International Institute of Tropical Agriculture (liT A), lbadan for 
post-flask management, field establishment and disease resistance scoring. Gene tagging 
of ACMD resistance required that this half-síb BC 1 population is genotyped with frame 
work markers from the genetic map of cassava and phenotyped for disease resistance in 
mutli-locational, replicated trials. 
Materials and Methods 
Two hundred and thirty plants showing sufficient growth were micropropagated at liT A, 
from the initial in-vitro cultures, by explanting and transferred to the green house in 
preparation for transfer to the field Ocober 1997. The green house material was 
established in the field in Abuja, Nigeria, a disease- free area for multiplication of 
vegetative material from the half-sib family January 1998. An ACMD disease resistance 
trial, with 3 replicates, 6 genotypes per replicate, in 3 high disease pressure sites, in a 
randomized complete block design was established early September 1998. Disease 
resistance will be scored at 3 and 5 months after planting. 
Controlled screening, in the green house, of ACMD disease resistance is also desired, to 
maximise virus infection efficiency, and thus obtain true disease resistance phenotypes 
from each genotype of the BC 1 cross. A full-length, infectious DNA clone of the West 
African strain from Nigeria was recently obtained by abutting primer PCR (AbP-PCR) at 
the John Innes center (JIC) (Briddon et al. 1993). The infectious clone was used to infect 
cassava plants in the green house without the whitefly vector, using a hand held particle 
gun. Twenty four stakes each of the parental genotypes of the original F 1 mapping 
population were shipped to the JIC and bombarded with the infectious viral clone. 
Transferring the genetic map of cassava to the half-sib backcross population has also 
begun at CIAT. One hundred and fifty RFLPs, 5 microsatellites, and 3 isoenzyme frame 
work markers, from the female- and male-derived genetic map of cassava will be used to 
genotype the population, and a genetic map constructed using the MAPMAKER linkage 
analysis program. 
Results and Discussion 
Preliminary screening of ACMD resistance with a virus DNA clone from ACMV-
Nigeria, conducted at the John Innes center, on twelve plants established from stakes in 
30 
the green house, from each of the parents of the F 1 Map population reveal the female, 
parent, TMS30572, is indeed resistant, while the male parent is susceptible. TMS30572 
showed no symptoms after bombardment with the infectious viral particle after severa! 
weeks after inoculation, while symptoms developed in 1 O plants out of twelve in the 
susceptible, male, parent, CM2177-2, after two weeks. This results shows the 
appropriateness of these populations for gene tagging studies of diesease resistance. The 
half sib backcross population is expected to pro vide increased variation in the phenotypic 
expression of disease resistance thereby ma"<imixing the chance that underlying 
segregating QTLs can be detected. 
DNA was extracted from 240 genotypes from the BC 1 half-sib cross, they correspond to 
genotypes in the field in Nigeria and also in the field here at CIAT. Southern blots for 
survey of the BC1 parental genotypes, with five restriction enzymes are being prepared, 
to screen the selected markers for polymorphisms. Progeny filters with the relevant filters 
will then be prepared. 
Further activities 
• Phenotype the BC 1 half-sib cross for disease resistance both in the field in Nigerian, 
and the screen house at JIC. 
• Genotype the BC 1 with frame work markers from the genetic map of cassava. 
References 
Briddon R. W., Prescott A. G., Lunness P., Charnberlin L. C. L., and Markham P .. 1993. 
Rapid production of full Iength, infectious geminivirus clones by abutting primer 
PCR (Abp-PCR). Journal ofVirological Methods 43 : 7-20. 
Collaborators: J. Guitierrez, M. Fregene, C. Iglesias, J. Tohme 
1.2.5 Marker-Assisted Genetic Analysis of Earliness and Root quality traits in 
Cassava 
Introduction 
A quantitative trait loci (QTL) tria! to identify region of the cassava genome controlling 
time of initiation and rate of storage root thickening, post-harvest deterioration (PHD), 
starch content, culinary quality, and salient morphological traits in cassava with 
molecular markers from the cassava map, was established earlier in the year. Genetic 
analysis of genes controlling these important traits is expected to lead to more efficient 
schemes for simultaneous development of early, better keeping, and high starch cassava 
varieties. In order to take powerful technologies for marker-assisted cassava breeding out 
of lnternational centers to national programs in regions where the most pressing needs for 
cassava improvement exists, a young Afrícan scientist was involved in the QTL 
experiments through the award of a Ph.D. studentship, funded by the Rockefeller 
foundation. 
31 
Materials and Methods 
The F l mapping population, from which the cassava map was developed, was also used 
as the QTL mapping population; the F 1 cross has been genotyped with o ver 200 markers, 
including known starch biosynthesis genes, cyanogenesis genes, and expressed sequence 
tags. The field experiment was a triple partially balanced lattice design with 20 plant per 
genotype/plot per replication, with three replications, in two sites, CIAT, Palmira and 
CIAT, Quilichao. The Quilíchao experiment was planted on the 15th of January 1998, 
while the CIAT plot was established the 29th of January, both plots were irrigated prior to 
planting. To test for early genotypes, storage roots from three plants from the border 
row of genotype plots, were harvested at 7 months, or early August, and dry matter 
content of the roots determined. Leaf morphology was also be measured at 6 months. 
The rest of the traits wil be harvested at maturity. Heritability of the dífferent traits will 
be calculated from environmental, genotypic and total variance derived from the 
ANOV A of results. Single point. by regression and t-test with conditioning will be 
carried out on the mean of phenotypic scores from the three replication by the QTL 
analysis software package Q-gene and PGRI. 
Results and Discussion 
Storage roots was harvested from three border plants from each plot at 7 months. Fresh 
roots from all three plants were weighed in air and water and the specific gravity of roots 
determined, which was later used to calculate dry matter. Phenotype values were 
averaged over replication for each site and analyzed by single point analysis using 
regression, with markers from the female-derived map being the independent variable, 
with the Q-gene package. Table 1 shows significant QTLs (P<O.O 1) found only on 
chomosome D for weight of storage roots at 7 months for the data from the Quilichao 
trial. Two regions on Chromosome D explains more than 60% of phenotypic varíance. 
Heritability estimates, calculated from the ANOVA, (SAS Institute) remains to be 
completed, and should yield estimates of genotypic variance explained by the markers. 
No significant QTLs were found with the CIAT, Palmira data. It was however noted that 
the CIA T Palmira trial suffered from multiple problems, including: flooding in March, 
extensive pesticide damage late March, residual maize pesticide "atrazin" effect in May, 
and sporadic pest and disease problems. The Quilichao trial remained relatively clean. 
The experiment will be repeated the following year and in controlled trials in the green 
house to confirm the identified. The leaf morphology data is currently being analyzed. 
Further activities 
• Scoring the QTL tria! in November for dry matter, starch content, post-harvest 
deterioration, culinary quality and important plant morphological characteristics such 
as plant height, height of first branching, stem color etc. 
• Analyze earliness data with an aim of setting up controlled green house experiments 
with 40 genotyopes each from the extremes to measure components of earliness such 
as, increase in girth of secondary roots, time of frrst starch accumulation, or 
differentiation of cambia, rate of starch accumulation, etc. 
32 
Table l. Markers of linkage group D of the genetic map of cassava showing significant linkage to a 
QTL effect for storage root dry matter weight at 7 months. QTL analysis used only data from the 
Quilichao tria!. 
Marker Chrom F RSq p AA SE N A a SE 
Ai18b NgD 15.72 0.1733 0.0002 521.72 32.46 39 695.5 29.35 
GY219 NgD 14.45 0.1598 0.0003 530.49 33.07 41 696.76 27.86 
GY42 NgD 14.43 0.1632 0.0003 530.8 33.25 41 701 .23 28.97 
rGY167 NgD 14.37 0.1608 0.0003 536.93 31 .7 44 706.36 29.55 
rGY180 NgD 14.24 0.1578 0.0003 532.95 32.6 42 698.5 28.23 
GY181 NgD 14.01 0.1556 0.0004 535.42 31 .93 43 700.2 29 
GY50 NgD 13.96 0.1569 0.0004 533.93 32.57 42 700.29 29.14 
rGOT-2 NgD 13.81 0.1521 0.0004 537.47 30.49 47 704.25 30.81 
GY125 NgD 13.8 0.1554 0.0004 536.26 32.11 43 702.24 29.55 
GY222 NgD 12.18 0.1447 . 0.0008 532.27 33.4 41 692.3 29.82 
GY179 NgD 10.51 0.1229 0.0018 539.05 34.14 40 685.76 29.16 
019 NgD 8.97 0.1068 0.0037 554.42 30.85 40 684.22 30.26 
AD4e NgD 8.89 0.1086 0.0039 550.79 31 .97 38 682.32 30.34 
K11b NgD 7.52 0.097 0.0077 536.29 35.21 38 669.35 32.88 
Collaborators: M. Fregene, E. Okogbenin1, J. Bedoya, F. Calle, C. Iglesias, and J Tohme 
l. University oflbadan, Nigeria. 
1.2.6 Mapping of genetic resistance of Cassava to bacteria} blight disease 
(Xanthomonas axonopodis pv manihotis) 
Introduction 
The most common biotic constraint to cassava (Manihot esculenta Crantz) production 
worldwide, cassava bacteria! blight (CBB), is invoked by a vascular bacteria! pathogen 
Xanthomonas axonopodis pv. manihotis (Xam) . Resistance has been introgressed in 
improved varieties through intra and interspecific crosses (M esculenta x M glaziovii). 
However, genetic of resistance to CBB have never been extensively studied until now. 
Although Xam populations genetic studies have been recently extended, there is no gene 
for gene relation clearly demonstrated. The difference between resistant and susceptible 
varieties invoked a variation in the speed of vascular reaction to Xam colonization, and 
suggest that resistance is quantitative. The objective of the present study was to locate 
cassava genomic regions responsible of resistance to Xam in order to facilitate their 
manipulation in breeding programs. 
33 
Material and Methods 
Material. The plant material used in this study was the F1 (MNGA 2x CM2177-2) 
composed of 150 individuals including 90 individuals described previously (Fregene et 
al., 1997). Five strains of Xam belonging to different haplotypes and collected from 
different ecozones have been selected. 
Evaluation of resistance. In greenhouse, nine plants per genotype and per strain were 
evaluated at seven, 15 and 30 days after inoculation by puncture following a scale 
previously described. Bacterial population densities in the stem were scored at seven and 
15 days. The average of the log 10 value of Xam population was used in QTL mapping. In 
the field, the 150 F 1 i.p.dividuals were evaluated for resistance to CBB during 2 cyles 
(1997 and 1998) in completely randomized experiment at CIAT center ' La Libertad' in 
Llanos. Each plant was evaluated at four and 7 months using a scale from 1 (no 
symptom) to 5 ( dieback of the whole plant). The disease rating of each plant was used 
separately for the QTL analysis. 
Mapping of QTL. The framework linkage map used is the female map and contains 142 
markers including RFLP, RAPD, isoenzymes and microsatellites as described previously 
(Fregene et al. , 1997). Molecular data for each individual were used to QTL single 
marker analysis. Statistical analysis of qualitative data was performed with SAS for 
resistance measured on an ordinal scale (greenhouse and field), and with 
QTLcartographer computer package for bacteria! population densities in stem 
Results and Conclusions. 
For greenhouse evaluations, the x2 test for homogeneity showed significant differences 
between resistance rating to the five different strains. For bacteria! quantification in stem, 
the x2 test showed that high levels of population are associated with higher notes and low 
levels with lower notes at 7 and 15 .da.i .. Test for association between each of these 
populations and symptoms at 30 d.a.i. showed significant association only in few cases. 
Spearman correlation analysis showed that there is no significant correlation between 
field disease rating at four and 7 month. There was no correlation between greenhouse 
and field evaluation. Among possible explanations, one can be that strains inoculated in 
greenhouse are very different from those present in field, and another is that sorne 
individuals may have escaped the disease in field. A second cycle of evaluatíon ís under 
progress. 
A total of five linkage groups and 26 markers were involved in resistance based on 
greenhouse evaluation to the five strains of Xam. Three linkage groups were common to 
almost two strains with several common markers (C, D, and L linkage groups), 16 
markers being in common for resistance to CI0-84 and ORSTX-27. For the other strains, 
regions of cassava genome involved in resistance to Xam are different. This result 
indicates that there is probably an interaction between pathogen and plant i. e. factors of 
resistance are specific for each strain. We know that the difference between each strain 
used in this study at molecular level is based on a probe corresponding to pathogenicity 
34 
gene. These factors of pathogenitcity may interact with resistance factors in a quantitative 
mode. Five markers on five different linkage groups are significantly associated with 
log10 of bacterial population in stem at 7 and 15 d.a.i. Nine linkage groups and 17 
markers were involved in field resistance at 7 month and none at 4 month. Three linkage 
groups (D, K and X) and seven markers were cornmon to greenhouse and field evaluation 
based in disease rating at 30 d.a.i .. Despite the lack of correlation between field and 
greenhouse disease rating, linkage analysis showed that linkage group D was tightly 
linked to resistance in both cases. Resistance factors of this linkage group must be 
involved to general defense mechanisms against CBB. These resistance factors may have 
come from M glaziovii. Two evidences for this hypothesis are (i) the numerous markers 
present on group D reveals low levels of recombination like found in interspecific 
crosses; (ii) an hypothesis is that resistance to CBB has been inherited from M glaziovii; 
it will be interesting to identify and map markers specific of M glaziovii that are present 
in TMS 30572. 
A second population of 270 individuals was developed in CIAT backcrossing five F 1 
individuals with female parent TMS 30372. Increased size of this population will allow 
us to map resistance to CBB more precísely. Obtention of homozygous indivíduals for 
sorne genome segments will give us the possibility to identify resistance loci with 
recessive effects. 
Reference 
Fregene MA;..Angel F, Gómez R, Rodríguez F, Roca W, Tohrne J, Bonierbale M (1997). 
Theor Appl venet 95:431-441 
Coll~bqrators: V. Jorge1 , M. Fregene, M.C. Duque, M. Bonierbale2, J. Tohrne, and V. 
Verdter 
1.0RSTOM-CIA T 
2.Current address: CIP 
1.2. 7. ldentification of genomic regio os responsible for the determina ti o o of whitefly 
resistance in cassava 
Introduction 
Whiteflies are a major pest of many agricultura! crops in most areas of the world. Yield 
losses are estimated in the hundreds of rnillions of dollars. Most whitefly species cause 
crop losses through direct feeding, while sorne are very efficient vectors of several 
econornically-important plant viruses. There are almost 1200 species with a wide range of 
hosts like legumes, fruit trees and ornamentals where this insect causes big econornic 
losses (Butler and Henneberry 1994, Brown et al. 1995). 
A single whitefly species Bemisia tabaci (this is actually a complex of biotypes and may 
include the specíes B. argentifolia), is the vector of Africa Cassava Mosaic, Bean Golden 
Mosaic, Bean Dwarf Mosaic and at least 30 other gemíniviruses of important food crops. 
The species Aleurothrachellus socialis and Trialeurodes vaporariorum cause 
considerable yield loss on cassava due to their direct feeding damage. The main 
symptoms in the plant are: total chlorosis, the apical leaves tum curly, the basal leaves 
35 
tum yellow and dry, and it stops the developing of the plant. The adult insects are found 
preferentially in the apical zones of the plant, where they extracting large quantities the 
sap of the conductive vessels, causing a considerable damage by loss of vigor, with low 
yield in the production. The honeydew which excrete, as a result of the copious sap 
intake, serves as a substrate for sooty mold fungus, which can also damage hosts by 
blocking photosynthesis. 
The entomology section has reported the existence of different sources of resistance to 
whítefly (CIAT, 1995). The most resistant genotypes are: Mbra-12 and MEcu-72. We 
have initiated a project to map the resistance genes of MEcu-72 using RFLP markers 
from the Cassava map, RAPDs and AFLP. 
Materials and methods 
Two mapping populations were generated from the crosses of CG489-34, a resistant 
clone and two susceptible genotypes Mcol-2026 and Mcol-1505. CG489-34 is resistant 
clones that combine the both so urce of resistance Mbra-12 and MEcu-72. A third 
mapping population was generated from the cross of MBra-12 by Mcol-2026. The 
population size ranges from 108 to 135 
Based on field testing, contrasting individuals (highly resistant and susceptible clones) 
were selected for each population. DNA was extracted from the different individuals and 
susceptible and resistant groups were formed and bulked with the respective parental 
genotypes. Bulk segregant analysis was conducted using RAPD primers. 
Results 
Sixty two RAPDs markers have been screened with the resistant and susceptible bulks of 
the three populations (table 1). One primer, OP.P3, showed a clear polymorphism 
between the susceptible and resistant group; presence of a strong band in the resistant 
bulk and absence in the susceptible bulk. This marker is being evaluated on the whole 
population of each cross to confirm its association with the resistance. Parental 
genotypes from each family were also evaluated with RFLP markers from the cassava 
map (Fregene et al, 1997). Seventy-two probes were found to be polymorphic 
On going activities 
l. Isolation and sequencing of the polymorphic band from OP.P3 and development of a 
Scar for efficient and rapid screening. 
2. Screening of the bulk DNA AFLP markers 
3. Development of a framework map using RFLP, microsatellites from the CIA T 
cassava map, RAPDs and AFLP markers for the linkage analysis. 
References 
-Brown, J. K., D. R. Frplich, and R. C. Rosell. 1995. The sweetpotato or silverleaf 
36 
whiteflies: biotypes of Bemisia tabaci or a species complex. Annu. Rev. Entorno!. 40: 
511-534. 
-Butler, G. D., Jr. and T . J. Henneberry. 1994. Bemisia and Trialeurodes (Hemiptera: 
aleyrodidae), pp.325-352. In G.A. Matthews and J. P. Turmstall, Insect pests of cotton. 
CAB, Wallingford, U.K. 
-CIAT, 1995. Cassava program annual report. 54p. 
-M. Fregene, F. Angel, R. Gomez, F. Rodriguez, P. Chavarriaga, W. Roca, J. Tohme, M. 
Bonierbale. 1997. A molecular genetic map of cassava (Manihot esculenta Crantz). Thor. 
Appl. Genet. 95:431-441. 
Collaborators: E.Barrera Sabogal, A.Bohórquez Chaux, C. lgh:sias, T. Bellotti and J. 
Tohme. 
1.2.8 molecular markers of genes involved in physiological post-harvest 
deterioration (ppd) of cassava roots. 
Introduction 
The short period of storage of the roots of cassava affects the marketing of the product. 
The process of deterioration begins 2 or 3 days after the harvest, followed by a microbial 
deterioration between the 5 to 7 next days. The physiologic deterioration is observed as a 
discoloration of the vascular tissues and the stored parenquima, accompanied by changes 
typical in the response to lesions in other plants systems (Beeching et al., 1994a; 
Bee~hing et al., 1994b ). 
The mechanism response of plants to lesions and attacks of pathogens include multiple 
biochemical responses, which are common to a wide range of plants. This include: 
increase in the synthesis of sorne proteins as phenylalanine ammonia-lyase (PAL), an 
associate enzyme with the synthesis of phenol, peroxidases and phenoloxidases, key 
enzymes in the biosynthesis of phenols and poliphenoles; changes in the composition of 
the membranes lipids and production of etilen lithic enzymes like b, 1-3, glucanase and 
Chitinase (Beeching et al. 1994a; Beeching et al. 1994b; Rickard, J.E. 1985). 
The correlation among phenotypic values of deterioration and specific neutrals molecular 
markers will allow indirect selection in breeciing programs forthese traits. Recently a 
molecular linkage map of cassava was developed at CIA T on a F 1 population using 
RFLP's, RAPD's and isoenzimes markers. The mapping population is the result of a cross 
between two lines elite of cassava, MNga 2 (TMS 30572) and CM 2177-2 (ICA 
Cebucan) (Fregene et al., 1994). 
The objectives of this investigation, are to transfer the existent knowledge and the 
molecular tools developed in other plants systems to cassava, and the mapping of the 
genes invovled in the physiologic post-harvest deterioration. 
37 
Methodology 
Plant material. The mapping population of cassava consists 150 F 1 plants, from a cross 
between "TMS 30572" "CM2177-2" (the male parent). This population was developed to 
construct a genetic map of cassava (Manihot esculenta Crantz) by Fregene et al, (I 997). 
DNA Extraction. DNA was obtained from young leaves (3-4 gr. from fresh weight) The 
tissue was collected from the plants maintained at CIAT at -80°C. The obtained DNA 
using Dellaporta et al. 1983 protocol was dissolved in T10E1 for incubation to 4°C over 
night and then quantified by fluorometry (TKO 100 Hoefer). 
RFLP's. Digestion with restriction enzymes (EcoRI, EcoRV, Haelll, Hindlll and Pstl) 
was made using 1 O ng of DNA,. The digestions were carried out following standard 
protocols provided by the manufacturer (Biolabs New England). The digestion products 
was runned into a gel agar 0.8%, using TBE 0.5X buffer solution :'or 16-18 hours. They 
were transferred to a membrane of nylon Hybond N+ (Amersham) following the method 
of Southem blotting alkaline (Southem, E. M. 1975). Then the filter were prehybridized 
for four hours and later hybridized over night to 65°C with the probes marked with a32P 
dNT. 
EvaJuation PPD. The phenotypic evaluation of the population to PPD susceptibility was 
carried out in two different environments: I . CIAT Palmira and 2. CIAT Station at 
Villavicencio, during the first semester of 1998. In addition an evaluation carried out by 
Fregene in 1997 at CIAT, was also included as a different environment. These 
evaluations are based on a subjective method, based on the works ofWheatley, C. (1982). 
Analysis of Information. The analysis of variance was carried out in order to determine 
the range of the components of variance for PPD, genotypic [G], environmental [E] and 
interaction genotype- environment [GxE]. 
The phenotypic correlation of PPD was analyzed employing the statistical programs of 
analysis linkage: Q-Gene, MapMarker and QTL Cartographer (Lander, et al., 1989; 
Nelson, J.C., 1995; Basten, C.J., 1997). 
Generation of the Map Linkage. F our of the seven clones ha ve been included in the 
map previously developed by Fregene (1997), GLUl and MEPX segregating for the 
female parent (NgA-2), PAL1 and MEPAL segregating for the male parent (CM2177-2). 
The linkage map was generated using the program MAPMAKER/ EXP 3.0 (Lander et al., 
1987; Lincoln et al., 1992). The RFLP's markers were grouped using a LOD 4.0 
(probability of error is of 1 in 1 04) . 
38 
Mapping QTLs. The QGENE program (Nelson, 1994) was used to obtained the linear 
regression of the markers from the map over the trait (PPD) and to identify through 
simple interval the regions or QTLs associate markers and the variability of the 
characteristic explained (R2). The mapping for compound interval was carried out using 
QTL Cartographer (Basten, C.J. , 1997). 
Results 
Obtaining the Genes. The genes of Pheylalanine ammonia-lyase (P AL 1 ), Ca talase 
(CATl), Acc Oxidise (ACC01), Hydroxyproline- Rich Glycoproteins (HRGPl), and b-
glucanase (GLUl) were given by J.R. Beeching (University of Bath, United Kingdom). 
This were cloned in the site of restriction for Eco RI of plasmid pUC18. L.F. Pereira 
(University of Guelph, Canada) gave the genes of Peroxidase (MEPX) and another 
Phenylalanine ammonia-lyase (MEPALl). This cloned in the site ofrestriction for Eco RI 
of the plasmid pGEM 10. These were amplified by reaction PCR using primers M13, 
obtaining amplified fragments of the expect sizes. 
RFLP's markers. With the availability of seven candidates genes in order to include in 
the genetic map of cassava, was carried out hybridization in order to determine 
polymorphism generated restriction enzymes of between the parents, which have a high 
probability of segregating in the individuals of the pro gen y. 
For each one of the genes polymorphism was detected with the following restriction 
enzymes: Eco RI for GLUl , Taq 1 and Hae 111 for PALl and ACCOl , Eco RI and Taq 1 
for MEPX, Taq 1 and Hind III for HRGPl , Hind III and Taq 1 for MEPAL. CATl , is the 
only gene that was monomorphic with the tested enzymes. Only the 23% of the enzymes 
were polymorphic. The 46% of the polymorphism founded were with Taq 1 and 18% for 
Eco RI, Hae 111 and Hind III respectively. 
The evaluation of the clones in the progeny was carried out with the enzymes that 
generated polymorphism in the parents. Except GLU1 and MEPX, the clones were 
multiple bands, that could correspond to severa! loci of the same gene or severa! 
sequences of the same gene family. In the progeny not all of the polymorphism was as 
evident as which how it in the figure 4, and the difficulty genotyping the individuals 
conceming severa! markers was something frequent, running the risk of identifying the 
genotype ofthe individuals conceming to one or severa! clones of erroneous way. 
Phenotypic Evaluation. The phenotypic evaluation was carried out in the two planned 
environments. It was found that the values for PPD, doesn't have a normal distribution. 
The observed distribution is asymmetric on the left, presenting a high frequency low 
values of PPD between 0- 5%. A similar distribution was obtained by M. Fregene. The 
frequencies of individuals by any interval of val u es of PPD were different among the two 
environments. The highest grade of asymmetry was in environment 1 (Villavicencio ), 
prevailed the low value deterioration. The data were adjusted to a normal distribution 
through statistical transformation (In PPD). 
The data already in a normal distribution was used for the analysis of variance in Q-
GENE and QTL-Cartographer. 
39 
PD's QTL. Table 1 summarizes the QTLs found in the analysis of the segregates marker 
in the map ofthe female parent only. 
Table l. Summary of QTLs associated with markets, R2 and value additive. Analysis for 
parent female (NGA-2). 
Martin Villavicencio CIAT 
Grupo de QTL R'' a QTL R" a QTL R. a 
Liaamiento 
p A RGY20 0.25 0.94 RGY208 0.25 -1.143 RGY20 0.25 -1.006 
8 8 
p GY213 0.14 0.70 
RGYI9 0.185 -0.786 
D B 1 
GYI97 0.14 -0.673 
rGYI47 0.14 0.82 
GYI97 0.17 -0.81 
e RGYI7 0.23 0.915 
4 
GYI96 0.17 0.769 
F GY203 0.156 -0.852 
GY122 0.160 -0.849 
GY186 0.214 -1.000 
G rAMI8 0.18 0.783 
J ADIC 0.149 -0.859 
GY34 0.246 -1.079 
K GY119 0.180 0.921 
rCDYIO 0.332 1.059 
6 
N GYI48 0.189 0.746 
rGY74 0.207 1.042 
Q F19A 0.23 0.883 
GY183 0.19 0.832 
X CBB1 O .lO 0.649 
Discusión 
The apparent low level of polymorphism (23%) between the parent lines of the mapped 
population could be attributed the high percentage of relationship between the cassava 
genotypes. 
The linkage map is of great use in the development of markers for indirect selection, the 
management of germplasm and the genetic study of complex traits. The inclusion of 
markers in the genetic map of cassava in order to select either resistant or susceptible 
genotype to the PPD has been identified as an important objective. The effort for the 
development of these markers involves the study of the inheritance genes implied in the 
defense process of the plant, which is supposed of quantitative inheritance because the 
high contribution of the environment in the variability expression of the characteristic. 
The next step includes the evaluation of the PPD in several environments in order to 
40 
reduce the as much as possible the component of environmental variance and elucidate 
the genetic component. The RFLP's markers are useful for this analysis. 
The restriction enzymes Taq I, EcoRI, Hae III, and Hind 111 generated polymorphism for 
genes involved in the PPD, that it is segregates in the progeny. 
The results of the phenotypic evaluations have been confusing. At CIAT, like at 
Villavicencio, a high frequency of values of very low PPD was observed, even 1 O days 
after the harvest. This was n6t expected for a progeny, in which the parents have 
intermediate values ofPPD. However the mapping and QTLs analysis were carried out. 
The markers of PALl, MEPX, GLUl and MEPAL present high distances in 
centimorgans are from the nearest markers. This indicates either that markers are wrongly 
mapped or that saturation ofthe map of cassava is low. 
No QTL showed association with the markers of the genes involved in PPD. The QTLs 
were associated with effect between environments suggesting that the environmental 
variation is high. Only the group of linkage was a QTL that maintained its effect and 
association to a marker between the ambient. 
In order to resol ve this molecular and of field problems, a news analysis is being carried. 
In addition, all the genes will be mapped again using RFLP-based PCR and make other 
phenotypic evaluation will be made with three repetitions in every environment. 
References 
l . BeechingJR, DodgeAD, MooreKG, PhillipsHM, WenhamJE. (1994) physiological 
deterioration in cassava: possibilities for control. Tropical Science. 34: 335-343. 
2. Beeching JR, Dodge AD, Moore KG, Wenham JE. Physiological deterioration in An 
incomplete wound response? The Cassava Biotechnology Network: Proceedings of 
the second intemational scientific meeting: August 22, 1994-August 26, 1994; Bogor, 
Indonesia. Thro A-M, Roca W (Eds.) Cali, Colombia: CIAT; 1995: pp. pp. 729-736-
3. Dellaporta, S.L.; Wood, J.; and Hicks, J . R. 1983. A plant DNA minipreparation: 
version II. Plant Mol. Biol. Rep 1:19-21. 
4. Fregene M, Angel F, Gomez R, Rodríguez F, Maya M, Bonierbale M, Tohme J, 
Iglesias C, Roca W. A linkage map of cassava (Manihot escuienta Crantz) based on 
RFLP and RAPO markers. The Cassava Biotechnology Network: Proceedings of the 
second internacional scientific meeting: August 22, 1994-August 26, 1994; Bogor, 
Indonesia. Thro A-M, Roca W (Eds.) Cali, Colombia: CIAT; 1995: pp. 49-61. 
41 
5. Lander ES, Green P, Abrahamxon J, Bariow A, Daly MJ, Lincoin ES, Botstein D 
( 1989) Mapping Mendelian factors underiying quantitative traits using RFLP linkage 
maps of experimental and natural populations. Geno mies 1: 17 4-181. 
6. Nelson, JC (1 995) QGene. A computer package for the marker-based anaiysis of 
population. U ser Manual. 35pp. 
7. Rickard JE. (1985) Physiological deterioration of casssava roots. J. Sci. Food Agríe. 
36: lb7-176. 
8. Sothem. E.M. (1975). Detection of sequences among DNA fragments separted by gel 
electrophoresis. J. Mol. Biol. 98: 503-517. 
9 Wheatley C (1982) studies on ~assava (Manihot esculenta Crantz) root postharvest 
deterioration. PhD Thesis. Univ. ofLondon; London. 246 p. 
Collaborators: D.F. Cortés, M. Fregene, J. Tohme, C. Iglesias 
1.2.9 Identification of NBS and Protein Kinase type putative resistan ce genes to Rice 
Blast 
Introduction 
In rice we are specifically interested in fmding the resistance genes against Pyricularia 
grisea, one of the most devastating and variable fimgus that affect up to 85 countries 
where rice is cultivated. Breakdown of resistance is quite comrnon and most varieties 
One major exception is Oryzica Llanos 5, developed at CIA T. This cultivar has been 
grown experimentally and commercially for more than 8 years without any symptoms of 
a breakdown. Resistance remains stable because severa! resistance sources are combined 
in the lines, which were selected through a long and complex breeding scheme. Such 
characteristics make it ideal 1 to study blast resistance and to design breeding strategies 
incorporating so urce of resistance to the different blast lineages. We are working 1) on 
the dissection of resistan ce of Llanos 5 using gene tagging and 2) on the identification of 
RGA linked to resistance genes. 
Methods 
The same strategy described for the identification of RGA in bean is been followed for 
rice. Based on a consensus sequence in the protein domain, degenerate primers were 
designed (see Table No.l) and were used to amplify fragments analog to resistance genes. 
DNA templates were used from the varieties Fanny and Oryzica Llanos 5, susceptible 
and resistant to blast, respectively. 
After the PCR reaction the different bands were separated and cloned into E. coli. Groups 
of sequences for the different bands were established using restriction enzyme pattems 
42 
following the methods used by Leister et al. (1998). Then five to ten clones of each group 
were tested by RFLPs in the parentallines; and the polymorphic ones were hybridized to 
DNA from the progeny ofthe cross O. Llanos 5 x Fanny. Using the MapMaker program 
(Lander et al., 1987), these probes were placed on the RFLP map generated for the F7 
RILs population. The fragments linked to known resistance genes were sequenced using 
the automatic ABIPrism 377 DNA sequencer (Perkin-Elmer) and compared with the 
database of proteins (blastx) in GenBank. 
Results 
Sa far, we have sequenced 9 fragments from the 776 clones of the NBS library and 3 
from the 140 clones in the protein kinase library. All ofthem showed high percentage of 
homology with known resistance proteins like Oryza longistaminata receptor kinase, 
Oryza sativa receptor kinase-like protein or NBS-LRR type resistance protein [ Oryza 
saliva] . 
We have placed 11 of the identified resistance gene analogs (RGA) as new markers on 
the rice map derived from the cross of Llanos 5 and Fanny. Many of them (6) map to 
chromosome 11 where Xa 21 as well as Pi-1 and Pi-7 are located, suggesting that we 
might be getting close to resistance genes to Pyricularia in rice. Another one map to 
chromosome 6 and one more map to chromosome 7. 
On going activities: 
• fine map rice blast resistance genes in the O.Llanos x Fanny cross; 
• Sequence additional fragments and map them on the RIL populations 
• Initiate the isolation of the genes responsible for that resistance by screening rice 
BAC librarles with the different RGA obtained from degenerate primers. 
Table No. l 
Primer N ame Nucleotide Seguence Reference 
NBS ASl 5 ' -CAACGCT AGTGGCAA TCC-3' Leister et al., 1996 
NBS AS2 IAAIGCIAGIGGIAAICC .. 
NBS SI GGTGGGGTTGGGAAGACAAGG .. 
NBS S2 GGIGGIGTIGGIAAIACIAC .. 
NBS AS3 IAGIGCIAGIGGIAGICC .. 
LRR TCAAGCAACAA TTTGTCAGG ICAIA TICC Ronald, P. pers.com. 
PK PKI T AACAGCACA TTGCTTGA TTTIAIITCICGITG .. 
PK PK2 T AACAGCACA TTGCTTGA TTTIAIITCICAITG .. 
PK PK3 TAACAGCACA TTGCTTGA TTTIAIITCICTITG .. 
References 
Bent, Andrew F. 1996. Plant Disease Resistance Genes: Function Meets Structure. The 
PlantCel!Vol. 8,1757-1771. 
Leister, Dario et al. 1998. Rapid Reorganization of Resistance Gene Homologues in 
Cereal Genomes. Proceedings ofthe National Academy ofScience Vol.95, 370-375. 
43 
Leister, Dario et al. 1996. A PCR-Based Approach for Isolating Pathogen resistance 
Genes from Potato with Potential for Wide Application in Plants. Nature Genetics Vol. 
14, 421-429. 
Causse, M.A. et al. 1994. Saturated Molecular Map ofthe Rice Genome Based on an 
lnterspecific Backcross Population. Genetics Vol. 13 8, 1251-127 4. 
Collaborators: G. Gallego, M. Santaella, F. Correa and J. Tohme. 
1.2.10 Genetic mapping and fine mapping of the apomixis gene in Brachiaria 
Introduction 
The gene tagging of apomixis and construction of a Brachiaria map was pursued using 
the crosses B. ruziziensis x B. decuinbens andB. ruziziensis x B brizantha . Heterologous 
RFLP probes from the rice (Comell University), maize (Missouri University) and 
sorghum (Texas University) map were used. Other molecular markers used in this study 
were RAPDs, SCARs and AFLPs. A modified protocol for DNA extraction was also 
implemented in order to improve the efficiency of the mapping project. 
Materials And Methods 
A sexual tetraploid B. ruziziensis clone (CIAT 44-3) was used as a female parent in a 
cross with natural and apornictic tetraploid genotypes B decumbens and B ruziziesis B 
brizantha. 
DNA Extraction DNA was extracted using the protocol described by Carlos Colombo 
(personal communication) with modifications. 1g of tissue was dried at 48 oc for 20 
hours and ground to fine power using a modified piston constructed for grinding bean 
seeds; 15 ml of extraction buffer (O.lM Tris-Hcl pH8.0, 0.05M EDTA pH8.0, 0.7 M 
NaCl, 4% CTAB and 1% BMe) was added and incubated at 65 oc for 10 rnin; 15 ml of 
chloroform:isoamyl alcohol (24:1) was added and centrifuged at 3000 RPM for 30 min. 
The aqueous phase was transferred to a new tube and 8 ml of chloroform:isoamyl alcohol 
was added and centrifuged at 3000 RPM for 30 rnin, repeated twice. A volume of cold 
isopropanol was added to the supematant 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 concentration of 20 ug/ml. DNA 
was quantified on a DYNA QUANT 200 fluorometer (Hoffer Scientific Instruments, San 
Farancisco CA). Protocols for RFLP, RAPD, and AFLP markers in Brachiaria have been 
described previously (BRU Annual report pp 105-11 O 1997). 
Segregation of markers in the genetic model for single dose restriction fragment (SDRR) 
markers was determined by departure from the hypothesized 1:1 ratio by the Chi-square 
test. The data matrixes obtained for presence or absence of the bands were analyzed for 
44 
linkage to the apomixis gene with Mapmaker Macintosh using the Kosambi function and 
LOO score of 6.0 
Results 
Four sets of heterologous probes were screened on the parental genotypes and the 
mapping populations. Table 1 shows percentage of hybridization, polymorphism and 
mapped probes). Linkage analysis revealed a region of Brachiaria genome, 
corresponding to chromosome 1 in rice, bearing the apomixis gene linked to RFLP, 
AFLP and SCAR markers. The apomixis gene is linked to three rice RFLP markers (RZ 
276, RZ 995 and RZ 413), at a distance of approx. 3 cM from the apomixis gene. All 
three makers map near the centromere region in the rice map. The AFLP marker EeMf28 
was located at 11 cM and SCAR marker (SN14) at 6 cM. Attempts to fine map the 
apomixis gene failed so far. 
A second activity involves the screening of selected Brachiaria genotpyes with the 
SCAR SN 14 (MW = c.a. 1100 bp) to determine the efficiency of the marker as a too! for 
indirect selection of the reproductive mode ofF 1 plants or Brachiaria accessions. 
Table 1 Percentage hybridization, polymorphism and mapped probes 
So urce Hybridized probes % Polimorphic probes %Mapped probes 
Rice probes from the Japan 6 83 50 
map 
Rice cOna 83 82 41 
Oats cONA from the rice map 18 100 72 
Sorghum probes 13 23 8 
Maize probes 4 100 75 
On going activities 
• 
• 
• 
• 
Implementation of a modified bulk segregant analysis for fine mappmg of the 
apomixis gene 
Mapping of Brachiaria microsatellites obtained 
Development of a framework map using RFLP, RAPOs, AFLP and microsatellites 
markers 
Screening Brachiaria accessions and hybrids plants with the Scar SN 14 and compare 
with embryo sac analysis. 
Collaborators: J. Vargas, C. Roa, J. Miles and J. Tohme 
45 
ACTIVITY 1.3 Development of molecular genetic techniques for assessing 
genetic diversity 
SUMMARY OF ACHIEVEMENTS 
• We have embarked in an effort to generate microsatellite (SSR) markers for cassava 
and Brachiaria. Cassava SSR were highly polymorphic and have more allelic 
diversity with wild Manihot than cassava. Through a collaboration with the Swiss and 
Sweedish, the aim is to generate 500 SSRs to saturate the cassava molecular map. 
• For the first time we have constructed a Bacteria! Artificial Chromosome (BAC) 
library for cassava. The BAC library contains over 55,000 clones and has 5 x 
coverage ofthe genome; and thus 95% probability offunding a desider DNA clone in 
a gene cloning effort. 
1.3.1 Development of cassava microsatellite markers and its use for assessing genetic 
diversity at genus level 
Introduction 
Microsatellites are very polymorphic simple-sequence repeats, abundantly present in 
eukaryotic genomes. Their amenability for PCR arnplification and serniautomated 
genotyping makes them suítable markers for typing large gennplasm collections. Due to 
conservation of priming si tes, microsatellites may be arnplified across species and genera. 
We characterized 14 rnicrosatellites in cassava [1]. To evaluate the genetic diversity in 
Manihot, and the ability of cassava microsatellite primers to arnplify (homologous) loci at 
the genus level, we typed a genetically representative sarnple of cassava, as well as six 
wild Manihot species with ten-primer pairs [2]. The degree of relationship between 
cassava and its wild relatives was also assessed. Furthermore, 521 accessions of the 
cassava core collection were typed using four primer pairs [3] to characterize diversity 
and to identify redundancies. 
Materials and Methods 
Microsatellites were isolated from accession MCol22 using standard cloning protocols. 
Typing of accessions was performed using radioactive- and fluorescent-labeled primers. 
Multiplex PCR reactions were performed combining 2 up to 4 different fluorescent 
primers. Electrophoresis and allele detection was carried out on automatic DNA 
sequencers (models ABI Prism 373 and 377). Statistical analysís included heterozygosity 
estimation, allele nurnber and frequency, segregation analysis, similarity indexes 
calculation and dendrograrn construction employing the unweighted pair group method 
(UPGMA). 
46 
Results and Discussion 
Heterozygosity values in the cassava core collection ranged from 0.00 to 0.88 and the 
number of alleles detected varied from 1 to 15. Chi-Square tests (X2) showed that most 
microsatellite loci segregated following Mendelian ratios. Linkage was detected among 
few loci, and most loci showed two alleles per genotype. Typing the core collection with 
four microsatellite loci revealed between-country allele number and frequency variation, 
which agreed with between-country allele size variation at the same loci. Unique alleles 
were present in countries such as Brazil, Colombia and Guatemala. A low percentage 
(1 ,34%) of redundancy was found in the core collection through microsatellite, isozyme 
and AFLP analysis. Based on these findings we recommend 1) to use more 
microsatellites for typing cassava collections (í.e., the entire core), and 2) to employ 
DNA-based markers for future selections of cassava germplasm to ensure genetically 
representative, non-redundant samples. 
Cassava-derived microsatellite primers produced PCR-amplified products across the 
genus demonstrating their utility for cross-species amplification. However, a decline of 
amplification success was observed with increase of genetic distance. For example, in M. 
aescu/ifolia, M. carthaginensis and M. brachyloba no PCR product was detected for two 
loci. The presence of null alleles in these species also affected the level of observed 
heterozygosity. Previous work with AFLP markers showed that these three species are 
the most distant relatives of cassava, whereas M. escu/enta susbsp. flabellifolia and M 
esculenta subsp. peruviana are the wild forms most closely related to the crop. Such 
findings were confirmed with microsatellite analysis. 
Overall, a high level of polymorphism was observed (83 %) in the Manihot genus. The 
number of alleles varied greatly (from 4 to 21) among the loci evaluated, but in all cases 
more alleles (79 out of 124) were found in the wild taxa than in cassava. Considering 
only cassava and the wild subspecies, 53 additional alleles were present in the second 
group, suggesting a larger pool of alleles for the M escu/enta subspecies than for cassava. 
This result pointed out the primary gene pool of cassava as a potentially source of 
diversity that could be useful for the improvement ofthe crop. 
References 
l. Chavarriaga-Aguirre, P., Maya, M.M., Bonierbale, M.W., Kresovich, S., Fregene, M., 
Tohme, J., and Kochert, G: Microsatellites in Cassava (Manihot escu/enta Crantz): 
r)t.}-- discovery, inheritance and variability. Accepted for publication in Theoretical and 
Applied Genetics ( 1998). 
2. Roa, A.C., Chavarriaga-Aguirre, P., Duque, M.C., Maya, M.M., Bonierbale, M.W., 
Iglesias, C., and Tohme, J. Cross-species amplification of cassava (Manihot escu/enta 
O~ Crantz) microsatellites within the genus: allelic polymorphism and degree of relationship. Submitted to Genome (Sept. 1998) 
3. Chavarriaga-Aguirre, P., Maya, M.M., Tohme, J., Duque, M.C., Iglesias, C., 
Bonierbale, M .W., Kresovich, S., and Kochert, G: Using microsatellites, isozymes 
47 
and AFLPs to evaluate genetic diversity and redundancy in the cassava core 
collection and to assess the usefulness ofDNA-based markers to maintain 
germplasm collections. Submitted to Molecular Breeding (Sept-1998) 
Collaborators: P. Chavarriaga-Aguirre, A.C. Roa, M.M. Maya1, M.C. Duque, M. 
Bonierbale2, C. Iglesias and J. Tohrne. 
l. Purdue University; 2. CIP, Lima 
1.3.2. Construction of a Cassava Bacterial Artificial Chromosome (BAC) Library: 
Towards Cloning of Disease and Pest Resistance Genes. 
Introduction 
Disease and pest epidemics arises from time to time in major cassava growing regions 
with the attendant loss of income for resource poor farmers and erosion of valuable 
germplasm. Resistance breeding is the most effective way of combating intermittent, 
widespread disease and pest attacks. But, cassava's long growth cycle, heterozygous 
state, and vegetatively propagated nature considerably slows down resistance breeding. 
The advent of cloning resistance genes in plants, together with reproducible 
transformation protocols in cassava makes genetic transformation with homologous 
disease and pest resistance genes, a faster and more efficient way of taking on disease 
epidemics. 
Cloning genes known only by their phenotypes and pos1t1on relative to molecular 
markers on a genetic map, or map-based gene cloning, depends on the ability to generate 
large DNA fragments and ordering them into contigs, sparining the genomic regían 
carrying the gene(s) of interest. Two important criteria therefore are markers closely 
linked to genes of interest, and large DNA fragment libraries. The cassava genetic 
mapping project at CIAT has identified regions of the cassava genome controlling 
resistance to severa! strains ofthe cassava bacteria! blight (CBB). Similar experiments are 
ongoing to identify genes controlling resistance to the cassava mosaic disease (CMD). 
Materials And Methods 
Megabase size DNA was isolated from the cassava variety TMS3000l and embedded in 
agarose plugs as described by Zhang et al. (1995). TMS30001, developed at the 
Intemational Institute of Tropical Agriculture (liT A), shows extreme resistance to the 
cassava mosaic disease (CMD) and resistance to the cassava bacteria! blight (CBB). 
Large genomic DNA, in one-third of an agarose plug, was partially digested with Hind 
III, 1.5U for 20 min at 37°C, and DNA fragments of sizes 100-300kb size selected by 
pulse field gel electrophoresis (CHEF MAPPER, Bio-Rad Corp). Size selected DNA was 
ligated into the Hind III cloning si te of pBeloBAC 11 in a vector:insert ratio of 10: 1, using 
14U of ligase, in a fmal volume of 1 00~-tl. Twenty microliters of DH1 OB competent cells 
(GIBCO BRL) were transformed, with 2~-tl of the ligation reaction, by electroporation, 
and white colonies picked for DNA insert sizing. Colonies were grown for 14hr in LB+ 
30mg/ml Chlorampenicol and plasmid DNA isolated by the Autogen automatic plasmid 
48 
isolation robot (Kurabo Inc.). Plasmid DNA was digested with 1 GU of Not I to libera te 
inserts and separated on a 1% agarose gel by pulse field gel electrophoresis. The rest of 
the ligation was transforrned, plated out and picked with the Q-bot robot (Genetix PLC). 
Results And Discussions 
Twenty colonies carrying recombinant plasmids were randomly picked and inserts were 
sízed. DNA insert sizes ranged from 40 to 11 Okb, with an average of 80kb. Using the Q-
bot robot (Genetix PLC), 55296 clones were picked and transferred to 384-well plates 
contain1ng culture media suitable for storage of bacteria colonies at -80°C. A total of 144 
plates were required to store the library. The BAC library was also gridded onto 3 high 
density filters, each containing 18,432 clones per filter, for rapid screening of the library. 
The cassava genome has a size of 760 million base pairs. The cassava BAC library 
therefore has a 5X coverage of the genome and a 95% probability of finding any desired 
DNA clone. Screening the BAC library with cassava genomic DNA clones has so far 
revealed a satisfactory representation, up to 30 hits. The BAC library was constructed 
wíth the objective of map-based cloning of disease and pest resistance genes in cassava. 
A good commencement point of cloning resistance genes will be already identified 
markers tightly linked to CBB resistance genes. Fine mapping of these genomic regions 
has also been initiated, to be followed by assembling a contig of BAC clones traversing 
the region. 
Further activities 
• The cassava BAC library has 55,296 clones, of average size 80kb available for gene 
cloning, the next stage of this project will be to use the BAC library to construct 
contigs of regions already identified in cassava carrying resistance genes to CBB. 
The library is publicly available through CIAT. 
References 
Zhang H., Zhao X., Ding X., Paterson A., Wing R. 1995. Preparation of megabase-size 
DNA from plant nuclei. The Plant Joumal 7(1 ): 175-184. 
Collaborators: M. Fregene, G. Gallego, J. Lopez, R. Wing1, J. Tohrne 
l. Clemson University, USA 
1.3.3 Expressed Sequence Tags (ESTs) for the Genetic Map ofCassava. 
Introduction 
A project was initiated last year to saturate the molecular genetic map of cassava with 
EST markers. ESTs markers have been widely used in the human, mouse and severa! 
plant genome projects for the candidate loci approach to mapping polygenic traits, to 
organize YACs, BACs, or other large-insert DNA clones into physical maps, and for 
identification of DNA regulatory sequences responsible for gene expression. The EST 
49 
data base division of Gene Bank contains over 1.2 million sequences, derived from a 
diverse collection of organisms. Cassava EST markers, obtained from sequencing 
transcipt derived fragment (TDF) polymorphic between the parents of the F 1 mapping 
population for cassava, is expected to provide more informative markers for the map. 
Putative identities of ESTs so far identified confirms the usefulness of this approach in 
saturating the molecular genetic map of cassava, and is also giving added insight into the 
genome organization of a putative allotetraploid. 
Materials and Methods 
Total RNA was isolated from young leaves and roots of the two genotypes of the parents 
of the cassava mapping population TMS30572 y CM2177 -2 at 4 weeks after planting. 
Messenger RNA was isolated from total RNA by Dyna beads and cDNA was syntesized 
from the mRNA templates as descibed by Bachem et al. (1996). Using the modified 
AFLP method for cDNA templates (Bachem et al, 1996), a preamplification PCR without 
selective bases prirners and a subsequent amplification with two selective nucleotides 
primers, using all the possible 256 combinations, was done on the cDNA library. PCR 
reaction was electrophoresed on a 5% sequencing gel. Polyrnorphic AFLP bands 
correspond to TDFs differentially expressed in the parents excised from the dried gels, 
was eluted and reamplified. 
The eluted cDNA fragment was cloned into pGEMT (Promega Inc) and transformed into 
E. Coli strain, DH5 alpha . Plasmid DNA was isolated from overnight rniniprep cultures 
and TDFs were PCR labelled for use as probes in an RFLP screen of parental filters. 
Filters contained genomic DNA frorn the parental genotypes digested with five enzyrnes: 
EcoRI, Hindiii, Haeii, Asel, Taql. TDFs found polymorphic in the parental survey were 
hybridized to progeny filters prepared with the corresponding enzyme. Raw rnarker data 
was combined with previous segregation data from the frame work map of cassava and 
linkage analysis was by the MAPMAKER program. New markers are added to the 
existing framework map with a LOD thresh hold score of 3.0. TDFs were also 
sequenced, regardless of if they have been mapped or not using the automated ABI 377 
sequencer in the Biotech Research Unit, CIAT. The Basic Alignment Search Too! 
(BLAST ) version 2.0, of the National Center for Biotechnology Information found at 
http://www.ncbi.nlm.nih.gov/gorf/wblast2.cgi was used to search for homology to known 
genes published sequences of Gene Bank. 
Results and Discussion 
More than 500 TDF polymorphisrns were obtained from the AFLP screening using 256 
2-nucleotide selective primer combinations. Two hundred of these were eluted frorn the 
gel and PCR amplified. Forty eight ofthe TDFs were randornly chosen for rnapping, and 
cloned. F orty four were used in the parental survey. Leve! of RFLP polymorphism in the 
parents was 55, and six of the polymorphic TDFs were mapped to the existing frame 
work of cassava, four to the female-derived map and two to the male-derived map. The 
others have to be confirmed in the hibridization with the progeny. Forty four ofthe TDFs 
were sequenced and homology to known genes in revealed mostly housekeeping genes. 
They include Chlorophyll binding proteins, in Lycopersicum esculentum and Brassica 
napus, a transcription factor from Nicotiana tabaccum PRL-2 genes from Arabidopsis 
50 
thaliana , andan AFLP fragment from Dioscorea tocoro. These sequences, in the process 
of submission to the EST data base of the GenBank. 
Complementary DNA derived from cONA/ AFLPs show much higher RFLP 
polymorphism compared to random full length cDNAs found during genetic mappin gof 
cassava, 55% to 20%. The teclmique is therefore a powerful way to map many more 
cDNAs, using the highly informative RFLP technique, towards a transcript map of 
cassava. Preliminary QTL mapping with mapped TDFs reveals the power of mapped 
cDNAs to uncover genes involved in complex trait, a chlorophyl binding protein mapped 
el ose to a QTL for dry matter content and explained 10% of phenotypic variance. 
However, the short length of the TDFs, 200-400bp considerably reduces sensitivity of 
Southern hybridization resulting in rather weak signals. Other marker techniques, such 
as single nucleotide polymorphisms (SNP) might provide a way around the RFLP bottle 
neck encountered in mapping TDFs. Results thus far have shown that it is possible to 
develop a transcrip map of cassava, that might simplify genetic analysis of agronomic 
traits of interest. 
Further activities 
• Primers are being designed for sequences TDFs to implement the SNP mappmg 
approach. 
• More TDFs arwill be cloned, sequenced and mapped and their putative identities 
determined 
Collaborators: M.C. Suarez, M. Fregene, A. Berna!, J. Guitierrez and J. Tohme 
1.3.4 Lowering the Cost of Reliable Biotechnologies for National Cassava Programs: 
Simple Sequence Repeat (SSR) to Saturate and Facilitate Use of the Cassava 
Molecular Genetic Map. 
Introduction 
The applicability of the majority of molecular markers, currently RFLP markers, available 
on the molecular genetic map of cassava, in marker-assisted studies of genetic diversity, 
complex traits and marker-assisted improvement of cassava has been questioned by 
collaborators in the NARs. Considering the low technology status of most cassava 
breeding programs, RFLP markers, though highly informative, are beyond the capacity of 
all centers in Africa and Latín America and most centers in Asia. Recent advances in the 
identification and development of simple sequence repeat (SSR) markers, combined with 
its speed and high levels of allelic diversity, PCR-based assay, and codominant nature 
make them the markers of choice for saturation of the cassava genetic map and for 
transfer of marker technology to collaborators in the National Programs. PCR technology 
requires a relatively small capital out-lay for its integration into day-to-day plant breeding 
and germplasm management operations. More SSR markers can be assayed for a large 
number of plants, in a given time, for a given lab attendant, for a given budget, compared 
51 
to RFLP markers. Besides SSR markers can be shared "online" with collaborators, since 
only information as to primer sequences are required. 
A project to generate a few hundred SSR markers for the molecular genetic map of 
cassava was developed, with support from the Swiss Development Cooperation (SDC) 
and the Swedish International Development Agency (SIDA). A cassava breeder from the 
National Root and Tuber Program in Nigeria was invited to participate in the entire 
process of SSR marker identification, development and mapping. 
Materials and Methods 
A DNA library from variety TMS30572, was enriched for SSR markers, using the 
affinity capture method for the following di-, tri, and tetra- nucleotides: (CT) s, (GT) s, 
(CAA) 6, (CAG) 6, (AA T) 6 , (ACG) 6, (GATA) 4, and (CAGA) 4 and cloned into pUC 18 
plasmid as described by Edwards et al. (1996) . The library was transformed into DH10B 
competent cells and total of 18,000 colonies from the enriched library was screened for 
SSR markers by hybridization with (CT) 20, (GT) 20, (CAA) 14, (CAG) 14, (AA T) 14 , 
(ACG) ¡4, (GATA) 1o, and (CAGA) 4 oligonucleotides and by anchored primer PCR. A 
second enriched library from variety CMC40 enriched for (CT) s, (GT) s, deve1oped by 
Keith Edwards, Bristol University, UK, and cloned in pN1 , was screened with the oligos 
(CT) 20, and (GT) 20 by dot blot screening. Plasmid DNA was isolated from putative 
clones by the QIAprep plasmid isolation kit, and l-3ul of plasmid preparation was 
sequenced on an ABI377 automated sequencer, using the Universal and reverse M13 
primers. Forward and reverse sequence strands were aligned using, either the GCG 
software, or the Basic Alignment Search Tool (BLAST) version 2.0 site of the National 
Centerfor Biotechnology Information found at 
http://www.ncbi.nlm.nih.gov/gorf/wblast2.cgi. Primer design was by the web based 
software Primer 3.0 found at http://waldo.wi.mit.edu/cgi-bin/primer/primer3.c!Zi. 
Results and Discussions 
A total of 120 putative clones were identified in the first library and sequenced. Of this, 
64 clones contained unique SSR markers; primers, 20-mers long, could be designed for 
40 clones from regions flanking the di-or tri-nucleotide repeats. Other clones had the 
SSR too close to the end of the fragment, or had very short, <~. di- or tri-nucleotide 
repeats. Results reveal a low efficiency of enrichment <1% enrichment. More than 1100 
putative SSR markers were identified from the second library, corresponding to an 
enrichment of 50%. Plasmid DNA was isolated from 550 clones and so far 50 of these 
clones have been sequenced, all clones contained the corresponding di-nucleotide repeat. 
Further activities 
• Sequencing of putative SSR clones from the second enrichment, and primer design is 
continuing; the target by the end of the year is 500 sequenced clones with 
corresponding primer pair. 
• Genetic mapping of the SSR markers is expected to begin at the beginning of next 
year. 
52 
Reference. 
Edwards KJ, Barker JHA, Daly A, Jones C., and Karp A 1996. Microsatellite libraries 
enriched for severa! microsatellite sequences in plants. Biotechniques vol 20, No. 5:758-
760 
Collaborators: M. Fregene, C. Mba1, J. Guitierrez, E. Gaitan, K. Edwards2 , J. Tohme 
l. National Root Crops Research Institute, Nigeria. 
2. Bristol University, U.K. 
1.3.5 Isolation and characterization of microsatellites in Brachiaria sp. 
Background 
Microsatellites are short DNA sequences of repeated nucleotide motifs between two and 
six bases, flanked by conserved sequences . They are codominant inherited genetic 
markers and have proven to be highly polymorphic. They can be analysed using PCR 
with specific primers flanking the repeat region, resulting in simple sequence length 
polymorphism (SSLPs). DNA microsatellites have been used for mapping, gene tagging, 
germplasm characterisation and evolutionary studies. Severa! methods have been 
developed to speed up the isolation of genomics fragments containing such simple 
repeats (J.M.H. Kijas, 1994, K.J. Edwards, 1995). We have initiated a project to assess 
the presence of microsatellites in Brachiaria decumbens and their potential as markers for 
fine mapping the apomixis gene and spitellbug resistance .. 
Methodology 
50 ug of genomic DNA were digested using 500 units ofTsp509-I (AATT) at 5°C during 
3 hours and ran on a 1.0% low melting point agarose gel. Fragments between 100 and 
600 bp were excised and purified using Phenol- Chloroform methodology. About Sug 
size- fractionated genomic DNA were ligated with SOuM of Tsp adaptors containing 
EcoRI and Tsp509-1 si tes. Enrichment for microsatellites was carried out using 1 OOng of 
the ligated, denatured DNA in 500ul of hybridisation buffer using filters with GA and 
CA,AT,GC oligonucleotides for 24 h at 37°C. After hybridisation, filters were washed 
during 1 O minutes in lxSSC at 65°C. Bound DNA was eluted in 200 ul sterile distilled 
water by boiling fo 5 min. Two microlites from the eluted DNA were amplified by PCR 
in 25 ul using 1 unit Taq DNA Polymerase. After the PCR enrichment step, amplification 
were repeated once more using all of DNA generated from the previous step. 
Approximately 250 ng from enriched and cleaned DNA were taken directly from the final 
amplification and ligated to 50 ng of pGem-T easy vector (Promega) during 24 hours at 
l5°C. Plasmids were cleaned and electroporated into DH5a cells and plated on LB-agar 
plates containing ampicilin, IPTG and X-Gal. Following incubation ovemight at 37°C, 
colonies were transferred to nitrocellulose filters and hybridised with each oligonucletide. 
53 
Positive colonies were prepared using the Promega Wizard Kit and sequenced in a 377 
Perkin- Elmer sequencer with dye terminators. 
Results 
1771 colonies were hybridised with GA oligonucleotide, 759 colonies with GC 
oligonucleotide, 1 O 12 colonies with CA oligonucleotide and 1518 colonies with A T 
oligonucleotide. 92 positive colonies were obtained of which 68 were GA, 6 were GC 
and 18 were CA. We not found positive colonies using AT oligonucleotide. Primers for 
the relevant SSR are being designed and will be used to screened the different mapping 
populations available. 
On-going activities 
l. Using enrichrnent methodologies for other motifs. 
2. Sequencing of additional GA and CA clones. 
3. Design ofprimers 
4. Estimation ofthe number alleles in Brachiaria sp 
5. Testing the usefulness of the microsatellites in gene tagging and genome 
mapping in Brachiaria sp. 
References 
J.M.H. Kijas, J.C.S. Fowler, C.A. Garbett and M.R. Thomas. 1994. Enrichrnent of 
Microsatellites fron Citrus Genome Using Biotinylated Oligonucleotide sequences bound 
to Strptavidin- Coated Magnetic Particles. Biotechniques 16(4):657-662 
K.J. Edwards, J.H.A Barker, A. Daly , C. Jones and A. Karp, 1995. Microsatellite 
Libraries Enriched fo severa} microsatellite sequences in Plants. Biotechniques 
20(5):758-760. 
Collaborators: R. Pineda, E.Gaitán, J. Miles and J. Tohrne 
54 
OUTPUT 2. GENES AND GENE COMBINATIONS MADE AVAILABLE FOR 
BROADENING CROP GENETIC BASES 
A CTIVITY 2.1 Utilization of novel genes and gene combinations by means of 
cellular and molecular gene transfer techniques 
SUMMARY OF ACHIEVEMENTS 
• Resistant interspecific progenies involving Phaseolus coccineus and P. polyanthus have been 
sent to Africa for field trials against the Bean Fly, a major pest of common bean in that 
continent. 
• We have screened 23 Latín American Cassava vars for response to somatic embryogenesis. 
A gene construct harboring the cry lA(b) Bt gene has been made in the BRU, anda highly 
virulent A. tumefaciens strain identified, for use in a transgenic approach agains the cassava 
stem borer. 
• A pathogenicity gene (Pth B) of cassava bacteria! blight (CBB) pathogen has been cloned. 
Characterization of the PthB protein has began to allow interaction studies with the host 
plant. We have also developed a PCR-based highly sensitive, specific, and simple technique 
for the detection of the CBB pathogen in sexual seed. 
• A simpler, lower cost, cryopreservation technique through encapsulation-dehydration of 
shoot tips has been tested with severa! cassava genotypes. 
• QTLs from O. rufipogon andO. barthii associated with yield increase (up to 25%) have been 
localized using microsatellite markers from the rice molecular map. 
• Transgenic rice resistant to RHBV displayed 45-65% higher yield potential than non-
transgenic CICA 8 control. Inheritance and expression ofthe transgen indicates that it can be 
utilized to complement the natural resistance source in different rice varieties. 
• Work has been initiated to adapt Agrobacterium- mediated transformation of rice and 
Brachiaria, with promising results. Common bean transformation appears more promissing 
with the biolistic method. 
• Extension of CIAT biotech capacities has included collaboration with the Biotec 
Corporation, Cali through the developing of a micrografting-micropropagation technique for 
the Soursop fruit tree. Similarly, in collaboration with the Agronomy Faculty, Univ. Nac. 
Palmira, a highly efficient plant regeneration protocol for tomate has been developed as a 
step towards a transgenic approach to the tomate fruitworm and budworm. 
2.1.1 Gene transfer between Phaseo/us spp througb interspecific hybridization 
Background 
Useful traits such as resistance to somepests (ej . Bruchids and leaf hoppers) and diseases (e.g. 
Ascochyta blight, bean golden mosaic, common bacteria! blight, web blight, and white mold) are 
not found, or are inadequately expressed or the genetic base is very narrow in common bean. 
There is need of introgression of useful genes from related genepools, and wild populations of 
common bean (primary gene pool) and from related spp ( of the secondary and tertiary gene 
pools). 
55 
Common bean breeding and selection methods using genes from distantly related races, gene 
pools, wild populations, and species often results intermediary products or parental stocks. 
(Singh, et al 1998). The stocks can be used as parents in subsequent crossing with elite lines to 
develop commercial cvs. Backcrossing is often used to transfer the necessary traits, like 
recurrent and congruity backcrossing. Introgression of useful genes from distant sources often 
requires plant-the-lant hybridizations, sometimes bridging-parents, or embryo rescue (Mejía-
Jimenez, et al, 1994). 
Material and Methods 
In the period late 1997 mid 1998, interspecific crosses were performed between the wild P. 
acutifolius P. parvifolius, P. parvifolius is highly, tolerant to CBB and presses good levels of 
resistance to leafhopper and bruchids (Singh, et al, 1998). 
Results and Discussion 
(i) Congruity backcrosses carried between a P. parvifolius genotype with ICA Pijao, 
required embryo rescue. Progenies were selfed for 2 and 3 generations to develop 
inbred-backcross lines. Seed of these material, will be multiplied for extensive 
evaluation. 
(ii) Hybrid plants also were generated (15-50%) from P. vulgaris x P. acutifolius crosses. 
Around 70-75% of the plantlets had hybrid constitution, as evidence through 
Dehydrogenase isozyme profiles. More congruity backcross are underway in order to 
allow increased recombination. 
(iii) From the original between common and tepary beans carried out since 1995, severa! 
hybrid lines, highly resistant to bacteria! blight, have been produced. These lines have 
moved further to regional trials, and others used as donor parents of CBB resistance. 
Use molecular markers, molecular linkage maps, and in situ DNA hybridization will be 
implemented to facilitate genes introgression. The availability of routine transformation 
protocols for P. vulgaris will make gene transfer from alíen source possible. 
P. acutifolius can be genetically transformed since wild it can be used as a bridge for alíen gene 
introgression into common bean through wide crossing with the transgenic tepary. 
References 
Singh, S.P., Debouck, D. and Roca, W. 1998. Interspecific hybridization between Phaseolus 
vulgaris L. and P. parvifolius Freytag. BIC Ann. Report, 41: 7-8 
Singh, S.P., Debouck, D. and Roca, W. 1998. Broadening the genetic base of common bean 
cultivars. BIC Ann. Reports 41 : 39-40 
Mejía-Jimenez, A., Muñoz, C. Jacobsen, H.J. , Roca, W.M., Singh, S.P. 1994. lnterspecific 
hybridization between common bean and tepary bean: Increased hybrid embryo growth, fertility, 
56 
and efficienty of hibridization through recurrent and congruity backcrossing. Theor. Appl. Genet. 
88: 324-331. 
Collaborators: S. Singh, D. Debouck, P.V. Herrera, W. Roca 
2.1.2. Use of P coccineus and P polyanthus to improve Common Bean 
Phaseolus coccineus and Phaseolus polyanthus express traits that are largely or totally absent 
fiom the primary gene pool of P. vulgaris, including resistance to Bean Stem Maggot (BSM) and 
ascochyta blight. P. coccineus also has a significant level of resistance to BGMV which could 
be of utility for common bean. 
Material and Methods 
Under a Belgian funded project, we are investigating the potential of these specíes through the 
use of a core collection of 140 accessions drawn from CIA T' s collection of sorne 1,000 en tries. 
To select the core we employed a GIS model that involved a clustering of environments to assure 
sampling a range of ecological origins. For seed increase we have made an effort to utilize seed 
that is as close to original seed as possible, that is to say, seed that has not been increased under 
open pollenation, although such seed is not always available. We have performed seed increases 
under controlled conditions and now have seed available of about 60% of the core to initiate 
evaluations. 
We have also given continuity to the former Belgian-sponsored project and evaluated the 
interspecific progeny that were generated previously. 
Results and Discussions 
Seed Production. lt has proven more difficult to produce seed of (P. coccineus) and (P. 
polyanthus) than we had expected. Seed production by different accessions varíes from more 
than 2,000 seed to none at all at the time of writing. However, about 80% of the accessions have 
produced enough seed to initiate evaluations in the coming. Accessions from the north and 
central parts of Mexico produced seed without problems in most cases. However, accessions 
from further south (starting in the state of Chiapas in Mexico and continuing through to Peru) 
were variable in seed production and often produced little seed (<200 seeds) as of yet. This 
pattern implies a problem of adaptation related to latitude. The seed production as per 
agroecological cluster will be investigated further. Although the difference in daylength in 
Colombia is small as seasons change, there still might be sorne slight response as days become 
slightly shorter in September. 
A nursery for the vegetative propagation of the accessions has been established, from which to 
draw cuttings and thereby to conserve sexual seed for the traits for which it is absolutely 
necessary. It is hoped to use vegetative cuttings for evaluations of foliar pathogens. Accessions 
have likewise displayed wide differences in the ability to reproduce from cuttings. 
Inheritance studies. For this purpose we have made the necessary crosses for BGMV, for BSM 
and for ascochyta. Populations are in different stages of development. 
57 
In the case of BGMV, we studied the reaction of a number of accessions in the glasshouse to 
mechanical (hand) inoculation with the virus. The PC accessions previously identified as 
resistant maintained their expected reaction, while PP accessions expressed a high degree of 
susceptibility. However, contrary to expectations, most of the PC accessions inoculated in the 
greenhouse expressed high resistance, although in the past we had observed that many PC are 
susceptible under field conditions. Two cross combinations are in F2 generation and another four 
are in Fl. Families in the F3 generation must now be produced for replicated field trials. 
In the case of Bean Fly, crosses have been successful. Two populations are in F2 , and another 
two are in F 1 (Table 1 ). We are waiting for feedback from Africa on the reaction of the parents 
which are putatively resistant and susceptible, based on past data. 
In the case of ascochyta blight, we could confirm the reaction of severa! selections with a PP or 
PC plant morphology but expressing s.usceptibility to ascochyta blight. Twelve such selections 
were prograrnmed in crosses with resistant PP and PC. The PP crosses have preven difficult but 
the crosses with resistant PC ha ve produced a good number ofF 1 seed. In the next year we will 
advance these populations and initiate inheritance studies. 
Interspecific crosses. Within the families selected for ascochyta resistance, 344 individual plant 
selections were taken and planted for reinoculation in the second semester. Another 44 
selections were made among progeny with a PC or PP morphology but presenting a susceptible 
reaction, to be used in the inheritance studies. All selections were inoculated three times in 
disease plots, while a separate lot was protected with fungicides to produce pathogen-free seed 
for export. In the disease plots, ascochyta pressure was even more severe than in the frrst 
semester, and 1 O lines were identified as most resistant although their resistance did not equal 
that of resistant PP or PC. 
One of the objectives is to take advantage of the genetic variability in PC and PP in the practica! 
improvement of bean varieties. F our of the resistant lines were prograrnmed in 8 simple crosses 
for crossing in multiple combinations with common bean cultivars (Table2). Crosses will 
involve cultivar parents of both dry bean and snap bean types. Previous experience with small-
scale producers of snap beans in the Andean zone had demonstrated that pesticide abuse is now a 
common phenomenon, and that most producers apply fungicides as often as once a week. 
Furthermore, ascochyta is the first disease to appear, and therefore is the key to setting off a 
vicious cycle of pesticide application. Therefore, the deployment of ascochyta resistance as 
posed in the present project would have a doubly beneficia] effect in snap beans by improving 
yields and by forestalling pesticide applications. 
Other crosses have been planned to combine sources oftraits with other common bean cultivars. 
Among the pro gen y evaluated in the field in 1997, a great number contained PP or PC parents 
that reputedly possess resistance or tolerance to Bean Fly. These progeny were selected for 
agronornic value, and seed of the most promising were produced for shipment to Tanzania for 
testing there with the insect (Table 3). 
58 
Sorne accessions of PC have been reported to be resistant to the white mold fungus (Sclerotinia 
sclerotiorum). Therefore, it is of interest to evaluate interspecific progeny for this disease also. 
Collaborators: S. Beebe, C. Cajiao, J .P. Baudoin1, P. Geerts1 
l. Univ. of Genbluux, Belgica. 
Table 1: Hybrid seeds for inheritance study of BGMV resistance. 
Obs. Female Species X Mal e Species 
1 035171 cace X 035337 PLAN 
2 035171 cace X 035347 PLAN 
3 035172 cocc X 035337 PLAN 
4 035172 cace X 035347 PLAN 
5 035171 cace X 035337 PLAN 
6 035172 cace X 035337 PLAN 
Gene ratio No. seeds 
n 
Fl 59 
Fl 101 
F1 41 
F1 35 
F2 964 
F2 1047 
Table 2: Crosses involving the best interspecific selections for ascochyta resistance, to create multiple crosses. 
Popayán, 1998a. 
Obs Female (*) Species X Mate(*) Species Purpose Fl Seed 
1 239-(8)A-FI 1 VULO X 346-F4 VULO ASC 27 
2 239-(8)A-F 11 VULG X ICTA Hunapu VULG ASC 53 
3 249-(8)A-F11 VULO X 346-F4 VULO ASC 40 
4 249-(8)A-F1 1 VULO X ICTA Hunapu VULO ASC 15 
5 282-(2)A-F12 VULO X 346-F4 VULO ASC 22 
6 282-(2)A-F12 VULO X ICTA Hunapu VULO ASC 39 
7 283-(6)A-F12 VULO X 346-F4 VULO ASC 9 
8 283-(6)A-F 12 VULO X ICTA Hunapu VULO ASC 45 
59 
Table 3: Interspecific crosses involving P. vulgaris and P. cocci11eus or P. polyalltflus. 
Obs Female Spp. X M ale Spp. Purpose Fl Seeds 
1 G4090 VULG x G35171 cocc BGMV 50 
2 G 4090 VULG x G35172 cocc BGMV 27 
.., 
.J RAB50 VULG x G35171 cocc BGMV 35 
4 RAB50 VULG x G35172 cocc BGMV 38 
5 TALAMANCA VULG X G35171 cocc BGMV 43 
6 TALAMANCA VULG x G35172 cocc BGMV 177 
7 G 2333 VULG x G35345 PLAN BSM 6 
8 G 2333 VULG x G35348 PLAN BSM 61 
9 G 2333 VULG x G35350 PLAN BSM 56 
10 G 685 VULG x G35345 PLAN BSM 7 
11 G 685 VULG x G35348 PLAN BSM 4 
12 G 685 VULG x G35350 PLAN BSM 5 
13 FEB 192 VULG X G35348 PLAN BSM 38 
14 FEB 192 VULG x G35350 PLAN BSM 18 
15 CAL 143 VULG x G35345 PLAN BSM 42 
16 CAL 143 VULG X G35348 PLAN BSM 19 
17 CAL 143 VULG x G35350 PLAN BSM 1 
2.1.3 Development of genetic transformation methods for common bean 
Introduction 
The genetic transformation of common bean has been reported by Russell et al. (1993) and 
Aragao et al. ( 1996) by partí ele bombardment of mature seed meristems. This methodology 
however has not yielded until now transgenic plants expressing agronomic important genes, 
possibly because of the low efficiency of transformation achieved and/or transgene expression 
problems. Using Agrobacterium as a vector the genetic transformation of a wild genotype of 
tepary bean have been possible (Dillen et al. 1997). Due to the difficulties in common bean 
transformation, the latter authors suggested the use of tepary bean as bridge to introgress 
transgenes to the common bean, by means of interspecific hybridization. 
The main obstacle impeding the development of efficient genetic transformation in common 
bean, has been the lack of protocols for efficient in vitro plant regeneration from somatic cells or 
tissues. 
In the past years, we have been able to develop a methodology for in vitro plant regeneration of 
common bean (Mejía-Jiménez et al. 1998), based in: The induction of an undifferentiated, long 
60 
term maintainable, highly morphogenic tissue composed solely of meristems (meristematic calli 
or m-calli) in agronomic important cultivars of common bean and other Phaseolus species. 
The regeneration of fertile plants through micrografting of differentiated buds from the m-calli. 
During 1996-1997 we have attempted to transform m-calli through particle bombardment, but 
problems possibly related to gene integration, germline transformation or gene silencing 
prevented the recovery of transgenic plants expressing correctly the transformed genes. 
During 1998 our efforts were concentrated in increasing the transformation efficiencies of m-
calli, and in the achievement of the genetic transformation of m-calli through Agrobacterium. 
Materials and Methods 
A meristematic callus line of the cultivar Bayo Madero has been used in the particle 
bombardment experiments. For Agrobacterium transformation, this and other genotypes of 
common and tepary bean have been used. 
Before bombardment or Agrobacterium inoculation the tissue were sonified in medium 
containing 2,4-D, in order to induce at the meristems wounding, cellular reorganization, actívate 
cell divisions and expose cell layers below the epidermis to the transfonning treatment. The 
bombarded plasmid was the pCambia 1304, which contains the GUS gene fused to the GFP gene 
anda Hygromycin resistance gene both under the control ofthe p35S prometer. The plasmid was 
bombarded in four different configurations: Closed circular, Iinearized, single circular DNA 
strand + (coding strand for the GUS gene) and single circular DNA strand - (complementary 
strand). Single strand DNA was produced by cloning the whole plasmid pCambia 1304 into the 
plasmid pBluescript SK + or -. GUS expression of the bombarded m-calli was evaluated 2 and 
15 days after bombardment. 
For the transformation with Agrobacterium sonified m-calli were cocultivated with ovemight 
cultures (OD 0.5) of binary (EHAl O 1 piB Guslntron and A281 piB Guslntron) and cointegrate 
LBA4404 pTOK233 hypervirulent Agrobacterium strains, which carry the GUS-Intron gene and 
the Bastan" or hygromycin resistance genes. Transient GUS expression were assayed after 3 days 
of coculture. Agrobacterium growth was controlled using ticarcillin, clavulonic acid, combined 
with vancomycin 
Two weeks after transformation, the m-calli were transferred to selective medium containing 
antibiotics (hygromycin or phosphinothricin). 
61 
Results 
Meristematic calli induction and long term maintenance. 
• The number of genotypes responding to the methodology of induction of m-calli could be 
expanded. This has been possible in six agronomic important genotypes of common bean, 5 
oftepary bean and one of P. polyanthus. 
• Meristematic calli lines of Bayo Madero older than three years retain its regeneration 
capacity. This shows a high genetic and physiological stability of this kind of cultures. 
· • Efficient plant regeneration from m-calli is performed now routinely 
Genetic transformation tlrrough particle homhardment 
• The highest levels of GUS-expression 15 days after bombardment were obtained by 
bombarding with a single strand + DNA plasmid (the coding strand for the GUS gene). No 
expression 2 or 15 days after bombardment was obtained by bombarding with its 
complementary single DNA strand! This shows a totally elirnination of the transient reporter 
gene expression by bombarding with the - strand. 
• From these experiments more than 50 putatively transgenic clones that are resistant to 
concentrations of up to 30 mg/1 hygromycin are still under selection 
Agrohaclerium transformation 
• Methodologies for large scale wounding and inoculation of m-calli with Agrobacterium and 
have been developed. 
• High levels of transient expression of the GUS Intron gene (up to 20 GUS expression 
spots/100mg tissue) in m-calli inoculated with Agrobacterium is routinely obtained. Factors 
important for achieving this were the inoculation medium, coculture temperature, the 
Phaseolus genotype, and the use ofhypervirulent Agrobacterium strains. 
• The m-calli of common bean are usually very susceptible to media changes and antibiotics 
presence in the media, making it difficult the controll of Agrobacterium growth after 
inocu!ation. A treatment for controlling Agrobacterium growth without killing the m-calli 
with the high do ses of antibiotics required has been developed. 
References: 
Aragao FJL, Barros LMG, Brasileiro ACM, Ribeiro SG, Smith JC, Sanford JC, Faria JC and 
Rech EL. Inheritance of foreign genes in transgenic bean (Phaseolus vulgaris L.) cotransformed 
via partid e bombardment. Theoretical and Applied Genetics 1996;93 : 142-150. 
Dillen W, De Clercq J, Goossens A, Van Montagu y M, Angenon G. Agrobacterium-mediated 
transformation of Phaseolus acutifolius A. Gray. Theoretical and Applied Genetics 1997; 94 (2): 
151-158. 
62 
Mejia-Jimenez, A., H. J. Jacobsen, et al. (1998). Development of an in vitro regeneration system 
in common bean suitable for genetic transformation. EUCARPIA International symposium on 
breeding of protein and oil crops, Pontevedra- Spain 1-4 april. 
lmproved regeneration method may speed common bean transformation ( communicated by 
Timothy Pratt) In ISB News report from July 1998 http://W\vw.nbiap.vt.edu/news/news.rpt.html 
1 1 
Collaborators: A. Mejía, F. Giraldo, L. F. Galindo, A. Hassa, H. J.Jacobsen and W. M. Roca 
1 
University of Hannover 
2.1.4 Improving soma tic embryogenesis of a range of Latín American cassava cultivars. 
Introduction 
Agrobacterium-mediated transformation offers a powerful tool to introduce genes of agronomic 
interest and thus contribute to expand the genetic base of cassava. To use this technology, 
efficient transformation-regeneration systems are required for cultivars of economic importance. 
We tested 23 cultivars from nine Latín American cassava~growing countries for their ability to 
produce somatic embryos. An economically important cultivar in northern Colombia and 
Venezuela (MCol2215), was further tested for its ability to produce in vitro-derived plantlets 
from somatic embryos (somatic embryo maturation and elongation). 
For somatic embryo induction, the CIAT's standard medium (CIAT,1995) was tested. Similarly, 
for somatic embryo elongation-maturation severa! media and growth conditions were tested. 
Treatments to evaluate embryo germination included the effects of light, desiccation and the 
presence of activated charcoal in different media. Criteria for evaluation of embryogenesis 
included embryo developmental stage and growth, and percentage of friable, non-embryogenic 
callus produced. 
Results and Discussion 
Table 1 show six cassava genotypes, including MCol2215 and the advanced line CM 2177-2, 
which had high-to-very-high embryogenic response. It was comparable to the embryogenic 
response shown by the model cultivars MCol22 and MCol1505. The other cultivars tested had 
medium-to-low embryogenic response, although they produced friable, potentially-embryogenic 
calli. 
Regarding induction of somatic embryos and their development into plants with MCol2215, in 
vitro-derived, imrnature apical leaves, and axillary buds were the best explants for embryo 
induction. The best gelling-agent/auxin combination for embryo induction was Gel rite and 2,4-
D. Somatic embryo proliferation responded better on medium containing different concentrations 
of2,4-D anda mixture ofGel rite-Agar. 
63 
Treatments did not show significant differences for the production of friable, non-embryogenic 
calli, neither for embryo growth. For embryo development the best treatment included MS 
medium plus NAA, and MS plus ABA and cytokinin (BAP) (Figure 1). Embryos in the stages of -
early heart-shape and trumpet developed better under light conditions (Figure 2). Desiccation 
favored embryo gennination (Figure 3). Currently we are evaluating elongation conditions. 
Table l. Somatic embryogenic response of Latin Americans cassava cultivars 
CULTIVAR RESPONSE CULTIVAR RESPONSE CULTIVAR RESPONSE 
1 MCOL 1505 ++++ 9 MBRA 893 - 17 MDOM 2 ++ 
2 MCOL2215 ++++ 10 MBRA 888 
-
18 MPAR 110 
-
3 MCOL 22 ++++ 11 MPAN 139 + 19 CM 523-7 + 
4 MBRA 12 ++++ 12 MMEX - 20 CM2177-2 ++++ 
59 
5 MCOL 1468 ++++ 13 MPAR 104 + 21 CM 3306-4 -
6 MPER 183 ++++ 14 MCOL1522 ++ 22 CG 1141 -1 -
7 MECU 72 ++ IS MBRA 894 
-
23 MCRGJ + 
8 MCUB 74 +++ 16 MBRA 900 
-
(-) Absence; (+) Low; (++) Medium; (+++) High; (++++) Very High 
Collaborators: Ladino,Y.J; Segovia, V; Chavarriaga, P; Roca, W. 
64 
90 
80 
70 
60 
50 
40 
30 
20 
10 
o 
T1 
100 
90 
80 
70 
60 
50 
40 
30 
20 
10 
T2 T3 T4 T5 T6 T7 T8 T9 
O Trumpet 
l:ll Early-heart-shape 
•Both 
Fig. 1 Effect of medium composition on growth and 
Development of cassava (cv. Mcol 2215) somatic 
embryos 
• T1 : MS • No hormone. 
• T2: MS + cytokinin. 
• T3: MS +ABA. 
• T4: MS +ABA· cytokinin . 
• TS & T6: MS + cytokinin . 
• T7: MS + activated charcoal. 
• TS: MS + auxin . 
• T9: MS + auxin- mannitol. 
o -+oic....t__.l~~!b....lb;,-
Trumpet heart-
shape 
• Light O Nolight 
Fig 2. Effect of light on cassava 
(cv. MCol 215) embryogenesis. 
65 
Trumpet Early-heat- 8oth 
shape 
~No desiccation • Desiccation 
Fig 3. Effect desiccation on cassava 
(cv. MCol 2215) embryogenesis. 
2.1.5 Construction and testing of plasmids containing a Bt-gene for cassava 
transformation 
Introduction 
The cassava stem borer ( Chilomina clarkei, Lepidoptera) causes important economic losses in 
cassava growing area of Northem Colombia and in Venezuela. The introduction of insect 
resistant genes (Bt-genes) through Agrobacterium-mediated transformation offers a powerful 
component of integrated control of the pest. We constructed several plasmids containing a Bt 
gene (cryiA(b); Figures 1 & 2) plus severa! selectable and scorable markers employing standard 
cloning techniques. These plasmids were introduced into different Agrobacterium strains and the 
bacteria tested for their ability to transfer the T-DNA into cassava embryogenic tissues. A 
supervirulent Agrobacterium strain (LBA4404-pTOK233) was also tested to study the effect of 
containing extra vir genes on the efficiency of transformation of cassava tissues. Cassava has 
been transformed with this strain [ 1]. Poten tia! expression of the crylA(b) gene, and the effect of 
extra vir genes on transformation efficiency were indirectly measured by the transient expression 
of the GUS-intron gene. 
Results and Discussion 
Initial results indicated that C58C1-pBIGCry was very efficient at transferring T-DNA and 
expressing the GUS-intron gene in cassava explants (Table 1). Between 23 and 50% of explants 
showed multiple infections (blue foci) and whole blue tissues after 3-4 days of coculture and 15 
to 30 min. of incubation in GUS buffer. With the strain LBA4404-pTOK233 the percentage of 
explants showing multiple infections increased to 66-87%. The overall efficiency of 
transformation was comparable between these two strains although the presence of extra vir 
genes enhanced significantly T -DNA transfer. Non-supervirulent strains, carrying no extra vir 
genes (i.e., LBA4404-pBIGCry) produced few-to-none multiple infections. Since somatic 
embryogenesis in cassava is of multicellular origin [2], regeneration of non-chimeral transgenic 
plants requires multiple infections, i.e., multiple, adjacent cells must be transformed 
simultaneously on the same explant. As we show here, supervirulent strains like C58C 1, or 
strains carrying extra vir genes like LBA4404-pTOK233 produced multiple infections, which, 
with proper selection methods, can potentially developed into whole transgenic plants. We are 
currently in the process of selecting transgenic cassava tissues and constructing plasmids that 
carry extra vir genes to enhance T -DNA transfer efficiency. 
Table l. Efficiency of transient expression of the GUS-intron 
gene in cassava embryogenic tissues with different strain-plasmid 
combinations 
Plasmids: 
pBIGCry 
pKC2301 
pTOK.233* 
LBA4404 
++ 
+ 
++++ 
C58Cl 
++++ 
n.d. 
n.d. 
(++++) Between 20 and more than 80% multiple infections produced 
(+), (++) Below 10% multiple infections 
(n.d.) Not done 
66 
Figures 1 and 2. Plasmids pBIGCry and pKC2301 containing the crylA(b) gene. 
pBIGCry(15.4 kb) 
E E H E LB 
gus int 
H B B 
pKC2301 (14.62 kb) 
E H LB 
35S 11 gus int JSST Q 
H : Hind ill; E : Eco R 1; B:Bam H I;X:Xba 1 
References 
l. Li HQ, Sautter C, Potrykus 1, Puonti-Kaerlas. J. Nature Biotechnology 14:736-740 1996. 
2. Stamp JA. Annals of Botany 59:451-459 (1987) 
Collaborators: P. Chavarriaga, V. Segovia and L.I.Mancilla, W. Roca 
67 
2.1.6 Characterization of the PthB protein of Xanthomonas axonopodis pv. manilzotis. 
Introduction 
Phytopathogenic bacteria generally have limited host ranges, often to members of a single plant 
species or genus. Microbial genes involved in plant-microbe interactions may be functionally 
classified into four broad categories: those involved in parasitism, pathogenicity, host range, and 
avírulence (1). A farnily of aviru1ence (avr) and pathogenicity (pth) genes has been identified in the 
genus Xanthomonas, including X campestris pvs. vesicatoria (avrBs3) and malvacearum, X oryzae 
pv. oryzae, X citri (reviewed by Leach and White, 1996) and recently Xanthomonas axonopodis pv. 
manihotis named pthB (Verdier et al., 1996). The avrBs3-like avirulence genes sequenced to date are 
94 to 98% identical with each other. A virulence genes which are present in bacteria! pathogen 
interact with matching resistance genes in the host to triggering a cascade of responses in the plant. 
The majar plant defense mechanisms include production of anti-microbial compounds and 
enzymes, cell-wall fortification, production of reactive oxigen species, and the hypersensitive 
reaction (HR), characterized by the rapid necrosis of the plant at the site of infection and the 
accumulation of phytoalexins. 
Recently, it has been shown that bacteria! avirulence proteins functions as elicitors that are perceived 
within the plant cell. Our purpose was to express and purify the PthB protein for producing 
polyclonal antibodies. Interaction of the PthB protein with the host plant will be further studied by 
immunoelectronmicroscopy. 
Methods. 
Bacteria/ strains and plasmids. Strains and plasmids used in this study were Escherichia coli DH5 
a, X axonopodis pv. manihotis CI046 and CFBP1851 and vectors pGEX series 5X-l, 5X-2 and 
5X-3 (Pharmacia). The pthB gene of Xam was previously cloned and purified using standart 
methods (BamHI fragment =: 2.9kb ). The BamHI fragment of pthB was cloned in pGEX vectors 
previously digested with BamHI and dephosphorylated. Strain Al-28, A3-15 (pGEX 1 and 3 
respectively with the insert pthB) were grown on LB supplemented with ampicillin at lOOng/ml. 
Express ion of the pthB gene and purification of the protein PthBIGST 
Strain A3-15 was grown in LB+Arnp and induced with IPTG l.OmM. Bacteria! suspension were 
sonificated and run in a SDS-PAGE gel, stained with comassie blue R-250 for detecting the PthB 
protein The protein was ecised from the gel·and purified by an amonium carbonate 0.5%, SDS 
0.1% and mercaptoetanol 2% treatment, then dyalised for 48h to eliminate salts residues. 
Quantification of total protein was performed using the Bradfor method and using bovine serum 
albumine (0.5mg/rn1) as standard protein. For immunization of white rabbit, the protein (0.3mg) 
was injected intramuscularly three times at 1 O days interval. Specificity of the serum was evaluated 
by ELISA and westem blot. Antibodies for immunoelectron microscopy were purified from the 
serum by an amonium sulfate and dyalisis procedures. 
To determine the activity of protein, leaves of the susceptible cassava cultivar MCOL 1522 were 
inoculated with 0.03mg ofprotein; bacteria! suspensions of X axonopodis (CIO 46 and CIO 1851). 
IPTG induced strain of E. Coli A3-15 was used as positive control. Sterile water, polyacrylamide 
gel and non induced A3-15 strain were used as negative control. 
68 
IPTG induced strain of E. Coli A3-15 was used as positive controL Sterile water, polyacrylamide 
gel and non induced A3-15 strain were used as negative control. 
Results and discussion 
Cloning the pthB gene in pGEX vectors was successful with the pGEX 5X-1 , 5X-2 and 5X-3 
series, the insert orientation was then evaluated by restriction digests with EcoRI and Xhoi 
respectively. Strain A3-15 (pGEX5-3 - BamHI fragrnent) was selected for further studies. The 
protein fusion PthB/GST was induced by addíng IPTG (1 OOmM) to the cultured cells of strain A3-
15 (O.D 600= 0.6-0.8). The protein was then separated by SDS-PAGE gels. SDS-polyacrylarnide 
gels were stained and showed a band of 123KDa which corresponds to the protein fusion 
pthB/GST. Negative control (pGEX 5X-1 alone) showed one band between 30-26 kDa 
corresponding to the glutation transferase protein (GST). The protein was excised from SDS-
polycarylamide gels and 1.3mg of purified protein was obtained. 
Specificity of the serum was tested by ELISA. Using a dilution of 11250 of the protein (0.1mg) 
Elisa tests were positive with a O.D 490 was > 2.0. Positive reactions were observed using dilutions 
of the serum (1/10.000). A rabbit serum inoculated with freund 's adjuvant was used as negative 
control, (O.D = 0.2). Different samples were used as antigen control: strains of Xam CIO 46 and 
CIO 1851 (O.D > 1.5), strains A3-15induced (O.D >0.8) and not induced (O.D<0.5), extracts of 
cassava leaves MCOL 1522 (O.D <0.2). Western blots were used to confirm the reactivity of the 
serum (111000 dilution) versus protein fusion PthB/GST. Our results demonstrated the high 
specifity of the PthB serum against the recombinant protein. Reactivit:' of the polyclonal antibody 
to PthB protein will be further evaluated. 
Cassava leaves inoculated with 0.03 mg of protein show a yellow/brown spot after 48 hours post-
inoculation. Leaf angular spots appear 5 days after inoculation with the Xam strain CIO 46. Induced 
strain A3-15 caused light brown spot while inoculations with sterile water, acrylamide gels, and 
non induced strain A3~ 15 did not induce any symptoms. Serial dilutions of the protein extracts 
were made and 6 days after inoculation light leafbrown spots were visible. 
We will futher investigate the location ofthe PthB protein from Xam in inoculated plant tissues and 
bacterial cultures. Knowledge of the location of pthB gene products in bacterial-plant interactions 
could provide new insight regarding A VR protein function in the elicitation of resistance. 
References. 
Leach, J.E., and., White, F. 1996. Bacteria! avirulence genes. Ann. Rev. Phytopathol. 34: 153-179. 
Verdier,V., Cuny,G., Assigbétsé, K., Boucher, C. 1996. Characterization of a pathogenicity gene 
pthB in Xantomonas axonopodis pv. manihotis In: Abst. of the 8th International Congress on 
Molecular Plant Micro be Interactions, 14-19 july. Knoxville. 
CoUaborators: Valérie Verdier 1, Monica Chávez , Lida Mancilla. 
l. ORSTOM-CIAT 
69 
2.1. 7 Detection of Xantlwmonas a.mnopodis pv. manilwtis in cassava true seeds by Nested-
Polymerase Chain Reaction assay(N-PCR) 
lntroduction 
Cassava Bacteria! Blight (CBB) caused by Xanthomonas a.x:onopodis pv. manihotis (Xam is a 
particularly destructive disease in South America and Africa The success of a cassava seed 
certification program will depend on the availability of reliable tests to detect the pathogen in 
true seeds. Current methods to identify and detect Xam rely on isolating the bacterium and 
conducting irnmunoassays of tissue. Enzyme-linked irnmunobsorbent assays to detect this 
pathogen are not entirely specific. Recently, a PCR assay has been developed for detection of 
CBB in cassava lea ves and stems (V erdier et al., 1998). In cassava seeds, the bacterium was not 
detected by current PCR assays. 
The nested polymerase chain reaction (N- PCR), is a PCR based assay for which a second round 
of amplification is developed by using primers internal to the first ampiification product. For the 
present study, we developed a pair of internal primers and established the conditions of a N-PCR 
for seed detection. Our objective is to develop a rapid and more sensitive method for detecting 
Xanthomonas a.:r:onopodis pv. manihotis in true cassava seeds. 
Materials and Methods 
PCR Conditions. PCR reactions were as previously described (Verdier et al., 1998). 20ng of 
DNA were used for the amplification. For the experiments performed with cultured cells, the 
bacterium was suspended and serially diluted in distilled sterile water; plated on LPG and the 
colonies developed were counted. Each dilution was boiled for 1 O min prior to PCR and 1 O ~1 of 
each dilution was used as source of DNA template for the PCR reaction. Negative control 
reactions (no DNA or bacterium) were run in all the experiments. 1 O ~1 of each PCR product 
were then transferred to the N-PCR reaction mix with the interna! primers which were designed 
with the software Cprimer. Amplification profile was as previousiy described. The amplification 
products were analyzed by agarose electrophoresis, stained with etidium brornide. 
Detection of Xanthomonas axonopodis pv. manihotis in artificially infected seeds. Cassava 
healthy seeds were inoculated as follows: an inoculum of 108 cfu!ml ( optical density at 600nm= 
0.1) was prepared from a fresh culture. Three groups of ten seeds each were washed and 
scarified. The seeds were then inoculated by absorption in vacuum for 45 min. The inoculated 
seeds were washed twice by shaking in distilled sterile water, then each seed was macerated 
individually. Each sample was serially diluted in distílled water, and plated on LPG, the colonies 
that developed were counted. 10~1 of each dilution were used as the source of DNA template and 
incubated at 95 oc for 10 mín. One ¡..tl proteinase K (10 ¡..tg/ml) was added and samples were 
incubated at 55 oc for 12 min. Finally, the samples were incubated at 20 oc for 3 min during the 
addition of Taq polymerase. PCR pro file was as described above. For the N-PCR, 1 O ~1 of each 
PCR product were used as the source ofDNA template. 
70 
Detection of Xantltomonas axonopodis pv. maniltotis in seed samples collected in infected 
fields. Cassava fruits collected in infected fields were dried and the seeds were extracted. 
Samples of ten seeds were washed in distilled sterile water, serially diluted and plated on LPG. 
lO ¡..¡.1 of each dilution were used as DNA template for PCR after being treated as indicated 
above. l 0¡..¡.1 of each PCR product were used for the N-PCR. 
Results and Conclusions 
We have developed a N-PCR based assay for the detection of Xanthomonas axonopodis pv. 
manihotis, in true sexual seeds. The primers X-V2 and X-K2 directed the amplification of an 
509bp nucleotide product interna! to the 898bp product amplified with the first round PCR. The 
N-PCR assay improves the sensitivity of our previous single stage PCR assay. The N-PCR 
method allowed us to detect 1.4 cfu!reaction in cultured cells, 1 to 2 c:fu/reaction in artificially 
infected seeds and 1.2. 102 cfu/reaction in naturally infected seeds. The PCR and N-PCR 
procedures are simple and fast methods to perform. With its specificity and sensitivity the 
combined PCR assays described here have a potential as a reliable procedure for detecting and 
identifying the CBB pathogen in seeds. 
References 
Verdier V., Mosquera G. and Assigbétsé K. 1998. Plant Disease, 82:79-83. 
Collaborators: S. Ojeda and V. Verdier1 
l . ORSTOM- CIAT 
2.1.8. Cryopreservation of cassava shoot tips: Encapsulation-dehydration technique. 
Background 
Cassava shoot tips were successfully cryopreserved using an encapsulation/dehydration 
technique. The technique (Palacio, 1997) allows direct placing of shoots into liquid nitrogen, 
preventing the use of expensive programrnable freezing equipment and opening the possibility of 
large-scale conservation methodology at CIAT (Escobar et al. 1998). 
Materials and Methods 
T o optimize the factors controlling the rates of plant recovery after freezing, cytokinins, gelling 
agents mixtures and supplementation of beads were tested. 
Results and Discussion 
Results obtained in 1997-98 are shown in Table 1, 2 and 3. 
71 
Table l: Effect of growth regulators on viability and shoot recovery 
after freezing of cassava encapsulated-dehydrated shoot tips. 
Cultivar Growthregulator (mgll) 0/o % Shoot recovery 
Viability 
M Coll468 BKZ (0.33 each) 90 25 
Kl 71.4 4.8 
2iPKZ (0. 17 each) 90 45 
2iPKZ (0.33 each) 80 20 
2iPKZ (0.5 each) 90 20 
M Bra 507 BKZ (0.33 each) 100 35 
Kl 80 40 
2iPKZ (0.17 each) 95 65 
2iPKZ (0.33 each) 95 50 
2iPKZ (0.5 each) 90 10 
M Ven 232 BKZ (0.33 each) 84.2 o 
Kl 90 65 
2iPKZ (0.17 each) 88 .8 5.6 
2iPKZ (0.33 each) 89.5 10.5 
2iPKZ (0.5 each) 89.5 o 
M Col22 BKZ (0.33 each) 100 45 
Kl lOO 47.4 
2iPKZ (0.17 each) 95.4 72.8 
2iPKZ (0.33 each) 93 .7 68 
2iPKZ (0.5 each) 100 52.6 
B, K, Z and 2iP are all growth regulators 
Plant regeneration rates could be increased through the inclusion of Benzyladenine (B), Kinetine 
(K), Zeatin (Z) and 2iP. We previously found that the type and concentration of cytokinin could 
improve shoot response after freezing (BRU-Annual Report 1995). Kl and 2iPKZ (at 0.17-0.33 
mg/1 each) had the best effect on shoot recovery after freezing (Table 1 ). Two recalcitrant cvs, 
MVen 232 and MCol 1468, improved their response with this treatment. 
Viability after freezing is more consistent with the encapsulation-dehydration than with 
programmed freezing; the average viability value per cv. was more than 80%. This gives us the 
opportunity to recover more shoots per treatrnent. We have observed that when using beads with 
growth regulator response of frozen shoot improved. 
Table 2. Effect of agar brand and consistency of recovery medium on viability and 
plant recovery from frozen shoot tips. 
Cultivar Consistency of medium Agar relation 0/o % Shoot 
®Duchefa :®Phytagel Viability recovery 
M Col 1468 Solid (0.45%) 3:1 89.4 58 
Semisolid (0.35%) 3:1 88.9 50 
Control (0.35%) 1:0 100 56.2 
M Bra 507 Solid (0.45%) 3: 1 100 66.7 
Semisolid (0.35%) 3: 1 100 38.9 
Control (0.35%) 1:0 100 52.6 
72 
Sorne agars contain inhibitory substances which may prevent morphogenesis in certain cultures, 
rate of growth can be slow, toxic exudates from explants do not diffuse away quickly. 
Hyperhydracity could be avoided using mixtures of Gel-rite and Agar. We have found that a _ 
relation 3: 1 with the K 1 medium shows improves shoot recovery from frozen cassava shoot tips 
(Table 2). 
Table 3. Effect of bead supplementation on the response of cassava 
shoot tips after freezing in L.N. 
Cultivar Bead % % Shoot 
supplementation Viability recovery 
MCol 4E 83.3 50 
1468 
Kl 80 70 
BKZ (0.33 each) lOO 90 
Without 81.8 63.6 
MBra 507 4E 90 SO 
Kl 63.6 45.4 
BKZ (0.33 each) 54.5 9.1 
Without 66.6 41.7 
MVen 232 4E lOO 81.8 
Kl 90.9 90.9 
BKZ (0.33 each) 91.6 83.3 
Without 66.7 66.7 
MCol22 4E 91.7 75 
Kl 75 66.7 
BKZ (0.33 each) 75 50 
Without 66.7 50 
It seems that beads are not so permeable to media components at the beginning of culture; tissue 
could starve. Supplementation of beads with media components could support the initial growth 
till shoot emerges. We have found that medium K1 is more effective especially after freezing 
MVen 232 and MCol 1468 (table 3). 
Conclusion and future plans 
We found that adjusting the shoot recovery steps (media and conditions) the percentage of shoot 
recovery after freezing can be increased. Encapsulation/dehydration could be a simple way to 
introduce cassava collection liquid nitrogen. We will test this improved methodology on a sub 
set of the cassava core collection .. 
References 
BRU-Annual Report.l995 
Escobar R.H, Palacio J.D., Rangel M.P and W. M. Roca.l998. Crioconservación de ápices de 
yuca mediante encapsulación-deshidratación. in: III Latin american meeting on plant 
biotechnology REDBI0'98 La Habana-Cuba June 1-5 
Palacio J.D. 1997. Crioconservación de ápices de yuca (Manihot esculenta Cranz) utilizando la 
técnica de encapsulación-deshidratación. Tesis CIA T. 
Collaborators: R. H. Escobar, M. P. Rangel, W. M. Roca 
73 
2.1.9 Genes from Wild Rice Contribute to Yield Increase in Cultivated Rice 
Introduction 
Twenty one wild species and two cultivated species (O.sativa and O. glaberrima) represent 
ample genetic variability for rice breeding programs. It has been suggested that the Oryza wild 
species represent a potential so urce of new alleles for improving yield, quality, and stress 
resistance of cultivated rice. However, limited use of thís variabílity has taken place. Barriers 
still exist in effectively utilizing genes from wild species and molecular mapping techniques are 
needed to readily detect these new alleles in segregating populations. 
CIA T started in 1994 a collaborative project aimed at characterizing and utilizing wild rice 
species for the improvement of cultivated rice. We here report on progress made in the 
identification of QTLs associated with yield increase in O. ru.fipogon and O. barthii. 
Materials and Methods 
Two improved rice cultivars (BG90-2 and Lemont) were crossed to O. ru.fipogon and O. barthii, 
respectively (Table 1). Few plants (2-3) in each of the wild species were hybridized to severa! 
plants of each of the improved cultivars (recurrent parents). Single crosses were obtained and 
grown in the greenhouse at CIA T in 1994. Three F 1 hybrid plants were backcrossed to the 
improved cultivar; 153-1 Q8 BC1F 1 seeds were obtained per cr()ss combination. The resulting 
BC 1F 1 p1ants were transplanted (30x50 cm) and evaluated based on phenotype; negative 
phenotypic selection for undesirable agronomic traits (spreading plant type, excessive shattering, 
long awns, dark-color grains, high sterility, etc.) was used to narrow the selection down to the 
best (30-50) individuals. Each selected BC1 individual was back crossed again to the recurrent 
parent and approx. 30 BC2F1 seed were sown in wooden trays in the screenhouse and later on 
transplanted (30x40 cm) under irrigated conditions. A negative phenotypic selection was applied 
again and best individuals per cross were selected and harvested individually to generate BC2F2 
seed; 300 BC2F 1 plants were selected in the BG90-2/ O. ru.fipogon cross for field testing, whilst 
high sterility was found in the Lemont 1 O. barthii cross (Fig 2). Therefore, another BC to 
Lemont was done and 326 BC3F 1 plants were selected for field testing as BC3F2 families. 
l pi :anta cxh 
3 phutts Screcn HCHQC 
153 -Uplanu 
1200 ~ Ncpth·~ Phtnotypir 
1 Sd«tion 
l
l 300 fumilics ~~! Yidd Tri al- CIAT Palmira: 
Di~tA on lltniU 
Fi:!. l. Scheme uscd to dcvelnp BC:f1 familics with O. ruflpogou 
74 
"umh4.•r nfSe<'dl St'lt•t:ti on 
l pl .. tsCKtl 
.W lpluta 
(stcrillly) 
\ dÍ\ iÜt" ~ 
Seodhl& 
Scl'fttl8oUH 
191 JO plantJ Neptin Pkooripk 
(ncrility) SdtctlM 
J0.-50 Jttdllplut 4J pluts 
(lltriricyJ 
1.489 Jl6 tutUc Nq::ach-c PbHorypic: 
Plaatl Sdccdon 
ll6 Ca•ilia Yk:ld Trial: CIA T(Pal,.lra) 
D•t a • • ll1ni11 
Fig. 2. Schemc uscd to dcvclop RCJFz familics with O. bartllii 
Field trials 
The 300 BC2F2 families derived from the cross BG90-2/ O. rufipogon, and the 326 BC3F2 
families from the cross Lemont 1 O. barthii were planted in replicated yield trials in CIA T-
Palmira. Transplanting (20x30) was done anda completely randomized design with two reps., 2 
row-plot, 5 m. long was used. Data on 12 agronomic data, including plot yield/family were taken 
on 1 O randomly selected plants/plot. 
Based on yield potential and good agronomic traits, 38 BC2F2 families from the cross BG90-2 1 
O. rufipogon were selected and further evaluated for grain yield; a completely randomized 
design with four reps, 4 row-plot, and 5 m. long was used. 
Molecular characterization 
DNA of young leaves from the parentales genotypes, the BC2F2 and BC3F2 families was extract 
by the Dellaporta Method (McCouch et al. 1988) modified for PCR assay by CIAT 
Biotechnology Research Unit (unpublished data). Parental surveys filters containing O. 
rufipogon, O. barthii, BG90-2, and Lemont were prepared using five restriction enzymes 
(EcoRI, EcoRV, Hindlll, Xbal and Dral). Approximately 140 markers from the rice molecular 
framework linkage map were selected at 10-20 cM intervals throughout the genome. A set of 78 
mapped rice rnicrosatellite markers, developed at Comell University, was also used to 
complement the RFLPs in QTL analysis. 
Conclusions 
The distribution of grain yield (kg/ha) of 300 BC2F2 families (BG90-2/0. rufipogon) derived 
from plot yields of 40 plants (20 plants/row x 2 row) averaged over two replications. 
Transgressive segregation can be observed, with severallines (11 %) having between 5 and 25% 
higher yield than the recurrent parent BG90-2. Transgressive segregation for other yield 
components was also observed. Grain yield data taken on 38 BC2F3 families (Table lA) 
confirmed results obtained in the BC2F2 generation. 
75 
Based on the 88 RFLP and 14 microsatellites from the RF- Comell framework map screened on 
300 BC2F 2 fami1ies, putative linkages were identified with yield, anci yield components from 
replicated data available for the whole mapping population. Preliminary results using one way 
anova and t- test indicate associations between markers and yield on chromosome 2 similar to 
J.Xiao and S. McCouch results (Fig. 2). No linkage for yield was detected on chromosome 1 as 
reported by J. Xiao and S. McCouch. Other associations were also identified for yield and the 
various yield components (Fig. 3-4). These data from different groups working with diverse 
recurrent parents suggest that DNA introgressed from O. rufipogon can contribute positively to 
yield in elite rice cultivars. 
The distribution of grain yield (kg/ha.) of 326 BC3F2 families (Lemont 1 O. barthii) derived from 
plot yields of 40 plants (20 plants/row x 2 row) averaged over two replications. Transgressive 
segregation can be observed, with severa! lines having up to 30% higher yield than Lemont. 
Based on the 54 microsatellites from RF-
Comell framework map screened on 326 BC3F2 families derived from the cross Lemont 1 O. 
barthii putative linkages were identified. Preliminary results using one way anova and t-test 
indicate association between markers and yield on chromosome 2 similar to Xiao et al. and on 
chromosome 7. 
Table l. Grain Yield (Kglha) of sorne BC1F3 families from BG90-
2/0. rujipogon cross at CIA T. 1997 
Line/pedigree Yield (%) of BG90-2 
CTI3941- 11-M 7880 21 
CTI3958-1 2-M 7746 19 
CTI31946-1-M 7535 16 
CTI3976-7-M 75 19 16 
Ct13946-26-M 7359 13 
BG90-2 6496 o 
O. rufipogon 4998 -23 
• Mínimum signif difference. Dunnett's - 1582 
Grain Yield (Kg/ha) of sorne BC3F2 families from Lemont/0. Barthii cross at CIA T. 1997. 
Line/Eedigree Yield (%)de rendimiento 
CTI4949-28 6193.5 136.05 
CTI4937-24 5899.24 129.59 
CT14964-3 5629.59 123.66 
CTI4938-5 5620.73 123.47 
CT14937-10 5526.44 121.4 
CTJ4955-15 5514.25 121.13 
CTI4966-9 55 13.34 121.11 
CTI4937-5 5498.15 120.78 
CTI4946-69 5496.81 120.75 
CTI4937-26 5450.14 119.72 
Lemont 4552.39 100 
O. barthii 1065.72 23.41 
76 
ON-GOING ACTIVITIES 
l. Complete the characterization of agronomic and molecular data, and QTL analysis to 
determine the number of QTLs associated with yield increase across environments for 
both crosses. 
2. Determination of contribution for positive alleles of each of the parents. 
3. Development of Nils based on QTL analysis carrying specific QTLs for use in breeding 
programs. 
4. Start agronomic and molecular characterization of severa! other populations involving 
crosses with O. glaberrima to determine the presence of QTLs for yield in crease. 
Other Activities 
Agronomic evaluation of 346BC2F2 families from the cross Oryzica3/0.rufipogon under 
irrigated conditions at CIAT.Palmira. Grain yield data showed that several lines outyielded 
Oryzica 3. Molecular characterization needs to be done. 
Seed multiplication and agronomic evaluation of 350 doubled haploid lines from the cross 
Lemont/ O.barthii. These lines were developed from BC3Fl plants and evaluated under irrigated 
conditions at CIAT.Palmira. Bulked seed from each line was harvested for further evaluation in a 
replicated yield trial. 
Two hundred ninety six BC2F2 families from the cross Bg90-2/ O barthii are being evaluated in 
a replicated yield trial under irrigated conditions at CIAT. Palmira. Data on main agronomic 
traits are taken. 
The BC2Fl population from the cross Progresso/O.barthii was planted and evaluated at CIAT. 
Palmira. Very high sterility was observed and 320 fertile plants were harvested for further 
evaluation under upland conditions in a replicated yield trial. This population was also run 
through anther culture and the response in terms of callus production has been very good. 
The BC2 cross between BG90-2 and O.glaberrima was completed and the BC 1 population is 
under evaluation at CIAT.Palmira. The F2 seed will be shared with WARDA for evaluation in 
West Africa in 1999. 
One hundred advanced lines were received from WARDA's rice interspecific hybridization 
project (O.sativa/O.glaberrima) and evaluated under upland conditions at La Libertad 
Experiment Station, Villavicencio in collaboration with M.Chatel and Yolima Ospina. Data on 
main agronomic traits, disease reaction, tolerance to acidic soil conditions, and grain yield were 
taken; 15 lines were selected for further evaluation in 1999. 
77 
Seed of aprox.600 F3 lines from severa! interspecific crosses (BG90-210.rujipogon, Oryzica 
3/0.rufipogon, and Lemont!O.barthii) was provided to C.Bruzzone for field evaluation in Santa 
Rosa and Palmira. On the other hand, seed of 305 BC2F2 families from the cross BG90-
2/0.rujipogon was sent to W ARDA and evaluated in an observational nursery in lvory Coast 
.Many lines were selected by W ARDA' s breeders for further evaluation in severa! ecologies next 
year. 
Funding proposals and concept notes for Colciencias ,the European Community ,CIDA, and 
South Korea were developed and presented to potencial donors . 
Acknowledgements 
This research was supported in part by grants from the Rockefeller Foundation, USAID, 
with matching support from the International Center for Tropical Agriculture (CIAT). 
References 
Chen, X.; Temnykh, S.; Xu, Y.; Cho, Y. G. and McCouch, S. 1997. Development 
of a microsatellite framework map providing genome- wide coverage in rice 
( Oryza sativa L.). 
McCouch, S.; Kochert, G.; Yu, Z.; Wang, Z.; Khush, G.; Coffinan, W. and 
Tanksley, S. 1988. Molecular mapping of rice chromosomes. Theor. Appl. Genet. 
76: 815- 829. 
Xiao, J. ; Grandillo, S.; Ahn, S.; McCouch, S. and Tanksley, S. 1996a. Genes from 
wild rice improve yield. Nature. 384: 223- 224. 
Xiao, J. ; Li, J. ; Yuan, L. and Tanksley, S. 1996b. ldentification ofQTLs affecting 
traits of agronomic importance in a recombinant inbred population derived from a 
subspecific rice cross. Theor. Appl. Genet. 92: 230- 244. 
Collaborators: C. P. Martínez, J. López, A. Almeida, G. Gallego, J. Borrero, W. Roca, M. C. 
Duque and J. Tohrne. 
2.1.10. Control ofRHBV (rice hoja blanca virus) through nuclear protein mediated cross 
protection in transgenic rice. 
Rice hoja blanca virus (RHBV) is one of the major diseases afecting rice in tropical Americas. 
RHB V di se ase was first reported in 193 5 and sin ce then, mayor outbreaks of the disease had 
caused up to 80% of yield loss. Most popular varieties are resistant to the vector but are 
78 
susceptible to the virus. The breeding resistance is conferred by one or two genes, but plants 
carrying this source of resistance are susceptible at younger ages than 25-day-old. The 
uncertainty of epidemics induces farmers to spray insecticides to control sogata, the planthopper 
vector of RHBV. There is a need to incorporate additional sources of resistance into improved 
germplasm to ensure stable and durable resistance, since the resistance present in most varieties 
is from a single resistant source. The main goal of this project is to provide new source(s) of 
resistance to minimize the possibility of an outbreak of the disease by (i) transforming rice with 
novel gene(s) for RHBV resistance; and (ii) incorporating these genes into Latín American 
commercial varieties or into genotypes to be used as parents in breeding. Previous reports 
described the particle bombardment genetic transformation protocol optimized for indica 
Latinamerican genotypes, the preparation of gene constructs containing the RHBV nuclear (N) 
protein gene and the RHBV NS4 nonstructural protein antisense gene. Last year we reported the 
generation and selection of RHBV -N transgenic lines from the Colombian rice commercial 
variety Cica 8, showing stable RHBV resistance on T2 progeny plants. The RHBV -N 
transgenic lines A3-49-56, -60, and -1 O 1 showed a significant delay in the development of the 
disease and reduced severity ofthe symptoms (disease reaction from 0.1 to 0.3) in contrast to the 
non-transgenic control Cica 8 (disease reaction 0.8, when using a rating scale of Oto 1) which 
was highly susceptible at 23 days after the RHBV infection. These transgenic lines showed a 
yield potential of 46% to 64% higher than the non-transgenic Cica 8 control. 
Following is reported the study in progress directed to determine the inheritance and expression 
of the N-protection in other genetic backgrounds of interest for breeding. This study aims to 
elucidate if the N transgene could be used to complement the breeding resistance so urce already 
available to fully protect plants younger than 25 days of age. 
Materials and Methods 
lnheritance and expression of the RHBV nucleoprotein cross protection in different genetic 
backgrounds. 
In this work, T3 progeny plants were selected based on the resistance level and agronomic traits 
from the corresponding T2 progeny line (Table 1 ). Individual T3 pro gen y plants from the 
selected Cica 8 N-transgenic lines were frrst analyzed by nested-PCR to detect the plants 
carrying the N-transgene. Those plants were chosen as female parents and crossed with: 1) the 
breeding fixed line CT8008-3-12-3P-M-1P highly susceptible to RHBV; 2) the variety Oryzica 1 
with moderate resistance; and 3) the variety Fedearroz 50 (FDSO), highly resistant to RHBV. 
Controls consisted ofF 1 crosses between the non-transgenic Cica 8 and CT8008, Oryzica 1, and 
FDSO, respectively. Plants 1 0-day-old and 20-day-old of each F 1 cross were infected with 
RHBV under greenhouse conditions. Viruliferous nirnphs from a vector colony with at least 
85% of virulence were used. Five nimphs per plant were placed onto each plant contained 
within a plastic tube. Nimphs were allowed to feed on the plant for 5 days. Plants were 
evaluated weekly and up to 54 days after infection for the development of RHBV disease and 
plant vigor. Evaluations for performance of agronomic traits is still in progress. 
79 
Results 
Results showed that the non-transgenic F 1 s were significantly more susceptible than the resistant 
parent, suggessting that the natural resistance source is encoded by non-dominant gene(s). 
Crosses with the transgenic lines A3-49-60-4-5, A3-49-60-4-13, A3-49-60-12-3, and A3-49-101-
18-19 were significantly more resístant (about 40%) to RHBV than the corresponding FI non-
transgenic cross when using 10-day-old p.lants (Figure 1). A similar trend was obtained when 
plants were infected at 20-day-old. The higher level of resistance of the transgenic F 1 s was noted 
on the crosses with the susceptible, the intermediate resistant, and the highly resistant genotypes, 
and in sorne cases the resistance leve! was similar to the resistant parent (Figure 1 ). These results 
suggest that the protection conferred by the RHBV-N transgene is expressed independently of 
the genotype background, and that the transgene could be used to complement the natural 
resistance source. These crosses are currently being evaluated for its performance for agronomic 
traits. Future work will include the evaluation of the resistance segregation in F2 population to 
determine the inheritance and stability of this trait through selfmg, and the initiation of a marker 
assisted selection backcross breeding scheme for the introgression of the RHBV -N transgene into 
CT8008-3-12-3P-M-1P, Oryzica 1, Fedearroz 50, and Cica 8 parental genotypes. The selection 
process will be aided by MAS using the presence of the RHBV -N transgene as the trace marker. 
In order to implement this type of molecular selection, the correlation between the level of 
resistance and the presence of the transgene as well as the effect of the plant age on the level of 
the resistance conferred by the RHBV -N gene are currently being analyzed. 
Table 1.- Disease reaction and yield potential ofT2 progeny plants from RHBV-N Cica 8 transgenic lines 
selected as female parents. 
Line Disease reaction Grains/ plant 
A3-49 
56-17 0.04 1332 
60-4 0.09 728 
60-12 0.04 582 
101-5 0.00 1240 
101-18 0.03 67 
Cica 8 infected 0.51 (0.10) 70 (48) 
Cica 8 non-infected 0.00 1218 (343) 
Numbers in parentheses refer to the standard error. 
Collaborators: Z. Lentini, L. Calvert (IP4), C.Gamboa1 , L.F. Fory (IP4), I. Lozano (IP4), E. 
Tabares. 
l. Danac, Venezuela!FLAR. 
80 
e 
o 
-.;: 
Cj 
el: 
~ 
l. 
~ 
~ 
el: 
~ 
Cl) 
o 
e 
o 
·.;: 
Cj 
el: 
~ 
l. 
~ 
Cl) 
el: 
~ 
.~ 
Q 
100 
90 
80 
70 
60 
50 
4ü 
30 
20 
10 
o 
100 
90 
80 
70 
60 
50 
4ü 
30 
20 
10 
o 
100 
90 
80 
70 
60 
50 
40 
30 
20 
10 
o 
Cica 8 N-transgenic/ FD 50 
5 12 19 26 33 4ü 47 54 
Days after infection 
Cica 8 N-transgenic/ Oryzica 1 
5 12 19 26 33 4ü 47 54 
Days after infection 
Cica 8 N-transgenic/ Cf8008 
-}' ... ... 
/ 
JI 
CL n. . O 
F.~~ : ;_-w.~.: ... IAJ:: ¡¡¡, • (S.'~ 
..... r. ~ .• {)· . ...., 
//! . 
-q . 
• r.i ~
5 12 19 26 33 40 47 54 
Days after infection 
---cr 8008 
-lr-- Cica 8/FD 50 
--+-- FD 50 
· ··A··· 6)-4-5/ FD 50 
···~ ·- 60-4- 13/ FD50 
· • - ~ • • 60- 12-3/ FD 50 
· · ·"· · · 101-18-19/ FD 50 
• cr 8008 
A Cica 8/ Ory 1 
---+-- Oryzica 1 
···A· ·· 60-4-5/ Ory 1 
· • · ~- - · 60-4-13/ Ory 1 
... ~ .. 60- 12-3/ 0ry 1 
···0-- · 101-18-19/ 0ry 1 
--.- cr8oo8 
-A-- Cica 8/ CT8008 
·· · A·· · 60-4-5/ CT8008 
. . · ~- · · 60-4-13/ cr 80o8 
.. ·El·· · 60-12-3/ CT8008 
· · · "-- · 101 -1 8-19/ CT8008 
Figure 1.- Disease development on 1 0-day-old progeny plants from crosses 
between Cica 8 RHBV N-transgenic lines and three genotypes with different Jevels 
ofRHBV resistance. Fedearroz 50 (FD 50), highly resistant; Oryzica 1, 
interrnediate resistant; and CT 8008 highly susceptible. 
81 
2.1.11. Agrobacterium mediated genetic transformation of rice genotypes adapted to 
irrigated, upland, and hillside agroecosystems. 
Until recent years it was thought that monocots, such as rice, were recalcitrant to 
Agrobacterium-mediated transformation. Recent reports have shoum high transformation 
efficiency and demonstrate the potential of agroinfection in economically important cereals, 
such as maize, wheat, barley and rice. Transformation based on the use of Agrobacterium 
tumesfaciens is being preferred in a larger number of laboratories nowadays, because it appears 
to result in high transformation efficiencies and in more predictable segregation pattems of the 
transgenes when compared to the particle delivery system or electroporation. At CIAT there is 
interest to adopt this technology for transformation of local rice varieties such as indica rices 
(CICA 8, IR-72, INIAP 12) adapted to irrigated (flooded) conditions, and japonica rices (CT 
6241, CT 10069, and Lastisday-Fofifa) adapted to acid soils (savanna) and highland (hillside) 
environments. Research was initiated this year to review the protocols established for rice (Hiei 
et al., 1994; Aldemita,1996; and Toki., 1997) and introduced the necessary modifications for 
optimization using the selected breeding genotypes. 
Material and Methods 
Optimization of Agrobacterium mediated transformation six genotypes adapted to Latin 
American rice habitats 
The three protocols developed for rice (Hiei et al., 1994 and 1997; Aldemita,1996; and Toki , 
1997) were tested. Preliminary results showed differences between the three protocols 
established for rice transformation. Following are described the modifications introduced into 
Aldemita and Hodges (1996)'s protocol, which gave the highest response for the genotypes 
tested so far. Instead of using N6 as recommended by Aldemita and Hodges (1996), scutellum 
derived callus was induced and sub-cultured for O, 3 and 7 days on NBA mediurn (Li te al., 
1993) containing proline and NAA prior the co-cultivation (preculture) with the bacteria. The 
the callus was co-cultivated with A.tumefaciens strain LBA4404 (pTOK233) in NBA-AS 
medium containing 1 OOJ.1M acetosyringone for 3 days. Casamino acids and kinetin were omitted 
in the co-cultivation and se1ection media, and 20j.l1 1 OOJ.lM acetosyringone were added 2 hours 
prior the co-cultivation and onto the co-cu1tivated callus to reactivate further the vir genes. 
After co-cu1tivation, the agro-infected callus were washed with N6 salts (Chu et al., 1975) 
containing carbenicillin (250 mg/L), cefotaxime (100 mg/L), and hygromycin (50 rng/L) to kill 
the bacteria. The callus were then transferred onto the selection medium A [NBA containing 
carbenicillin (250 mg/L), cefotaxime (100 mg!L), and hygromycin (30 mg!L)] for three weeks. 
The healthy looking callus were sub-cu1tivated onto medium B [NBA containing carbenicillin 
(250 mg/L), cefotaxime (100 mg/L), and hygromycin (50 mg/L)] for other three weeks. 
Following the transgenic calli were first transferred onto a proliferation medium (LS with 0.5 
mg/1 2,4-D and 50 mg/L hygromycin) for 3 weeks and then to a regeneration medium (MS with 
NAA 1 mg/1, kinetin 4 mg/1). 
82 
Results and Discussion 
Preliminary observations indicated that embryogenic scutellum derived callus showed higher _ 
transient GUS expression than imrnature embryos (data not shown). The effect of the callus 
preculture time on GUS expression was evaluated. Embryogenic scutellum derived callus from 
Cica 8, CT10069, CT6241, INIAP 12, IR-72 and Latsihday-Fofifa were preculture forO, 3 and 7 
days on NBA medium prior co-cultivation, then co-cultivated with the bacteria for 3 days, and 
transient GUS expression was determined. Results indicated higher GUS transient expression 
on callus precultured for O or 3 days prior co-cultivation for indica rice (Cica 8 and INIAP 12), 
whereas no effect of preculture time was noted for japonica type (CT6241, CT10069, and 
Latsihday-Fofifa) (Table 1). Higher number of hygromycin resistant callus (Hygr) was 
recovered from Cica 8 (irrigated, indica) with three days of preculture, and from CT6241 
(lowland-upland,japonica) and CT10069 (highland-upland,japonica) with O days of preculture 
after selection with 30 mg/1 and 50 mg/1 of hygromycin, respective! y (Table 2). Between 60% 
and 100% ofthese Hygr callus showed stable GUS expresion (GUS+) at 60 days (2 months) and 
90 days (3 months) after the agro-infection (Table 2). Plant regeneration from these Hygr and 
GUS+ callus is in progress. The studies conducted so far suggest that the selection of the 
explant source and preculture of the target tissue are majorkey factors for Agrobacterium 
mediated transformation of rice. 
References: 
Aldemita R.R., T.K. Hodges T.K. (1996). Agrobacterium tumefaciens-mediated 
transformation of Japonica and Indica rice varieties. Planta 199:612-617. 
Hiei, Y., S. Ohta, T. Komari and T. Kumashiro. 1994. Efficient transformation ofrice 
(Oryza sativa L.) mediated by Agrobacterium and sequence analysis ofthe boundaries of 
the T-DNA. Plant J. 6:271-282 
Hiei, Y., T. Komari and T. Kubo. 1997. Transformation of rice mediated by 
Agrobacterium tumefaciens. Plant Mol. Biol. 35: 205-218. 
Li L., R. Qu, A. de Kochko, C. Fauquet, and R.N. Beachy. 1993. An improved rice 
transforamtion system using biolistic method. Plant Cell Rep. 12: 250-255. 
Toki, Seiichi. 1997. Rapid and efficient Agrobacterium-mediated transformation in rice. 
Plant Mol. Biol. Rep. 15(1) 
Collaborators: Luis Orlando Duque, E. Tabares, Z. Lentini 
83 
Table l.- Transient GUS expression on scutellum-derived callus after cocu ltivation with A. 
tumefacieus strain LBA4404 (pTOK133). 
~ICC Pre-culture · Callus · Callus Mean 
Genotype (days) Co- GUS+ (%) (Serror) 
cultivated 
IR -72 o 20 6 ¿5 l.T(0.7J o 39 o 
o 34 1 2.9 
" .) 33 4 12.1 NO 
1NIAP 12 g 6 5 83 .3 48.8 28 4 14.2 (24.4) 
3 28 22 78.5 
3 7 1 14.2 35.6 
3 7 1 14.2 ( 17.5) 
7 18 1 5 
7 6 o o 
2.5 ( 1.8) 
CICA 8 o 18 17 94.4 ID 
o 42 23 54.7 ( 13.4) 
o 25 9 36 
o 32 32 100 
3 67 62 92.5 
3 20 20 100 78.9 
3 22 8 36.3 (10.0 
3 8 3 37.5 
" .) 37 12114 85 .7 
3 14 14 100 
3 24 24 100 
7 12 o o 
7 8 3 37.5 
7 9 4 44.4 37.9 
7 34 7110 70 (12.5) 
CT624 1 g á6 8 50 65.1 11 27.5 ( 13.2) 
o 42 20 47.6 
o 24 24 100 
o 27 27 100 
3 4 1 39 95 .1 
3 25 24 96 74.9 (9.1) 
3 12 9 75 
3 18 10 55.5 
3 22 8 36.3 
3 12 11 91.6 
7 7 3 42.8 
7 18 10 55.5 ?3.3~) 7 12 11 91.6 11.9 
CT 10069 g 30 18 60 67.1 67 21 31.3 (12.7) 
o 36 16 44.4 
o 27 27 100 
o 24 24 100 
3 4 1 39 95.1 
3 26 25 96.1 55.3 
3 14 8 57.1 (16.4) 
3 12 1 8.3 
3 25 2110 20 
7 14 12 85 .7 
ND 
Lats lhday- .) l? ~ 20.0 23.8 (3.9) Fofifa 3 18.1 
3 15 5 33.3 
7 10 7 70.0 NO) 
84 
Rice 
genotype 
CICA8 
CT6241 
CTI0069 
Table 2.- Production of Hygromycin-resistant (HygR) calli derived from scutellum inoculated with A. 
tumefaciens strain LBA4404 (pTOK233). 
Pre- Selective medium GUS expression 
culture A Callus B Callus GUS~/ evaluated 
(days) Callus Hyg %Hyg Analyzed Hyg' %Hyg' %Hy,g in 60 (%) 90 
Analyzed callus callus callus callus Aand B days ous· days 
3 23 15 65 15 10 67 43 12/ 12 100 8/12 
7 24 8 33 8 6 75 25 0/15 o 0/ 12 
o 18 18 lOO 18 16 89 89 ND ND 17/ 17 
3 25 15 60 15 9 60 36 ND ND ND 
7 20 10 50 lO 6 60 30 ND ND ND 
o 24 24 lOO 24 22 92 92 ND ND 14/ 14 
3 13 9 69 9 4 44 31 ND ND ND 
7 8 3 37 3 1 33 12 ND ND ND 
2.1.12. Advances in Braclziaria genetic transformation mediated by Agrobacterium 
tumefaciens. 
Brachiaria species are important components of the pastures grown in the tropical lowlands of 
America, Asia, Africa, and Australia. B. decumbens cv Basilisk is one of the most extensively 
cultivated species, which has adaptation to acid soils, rapid growth, good soil coverage, and high 
nutrional value as a feed pasture. This species however, is highly susceptible to spittle bug 
(homoptera). Resistance to this pest is present in B. brizantha which does not outcross with B. 
decumbens. B. ruziziensis is used as a bridge between decumbens and brizantha species, thus 
recurrent backcross is needed to recover the agronomic characteristics from decumbens. Plant 
genetic transformation offers an expedite altemative to transfer genes between unrelated species. 
A protocol for genetic transformation of Brachiaria will be particularly useful to introduce 
resistance gene(s) for this homoptera pest, and to improve further the quality traits associated 
with the nutritional value of the pasture. 
Earlier work at CIAT's included the establishment of tissue culture methods for plant 
regeneration (Lenis, 1993 ), and genetic transformation by direct methods using partí ele 
bombardrnent (Galindo, 1997) of Brachiaria species. Agrobacterium tumefaciens is a high 
efficient vector for the transfer of allien genes into dicotyledonous. Recent advances in the 
development of protocols for monocotyledoneous species is allowing to use this technology for 
transforming various cereal species such as rice (Hiei et al., 1994; Rashid et al., 1996; Dong et 
al., 1996; Toki, 1997), maize (Ishida et al., 1996; Escudero et al., 1996), and barley (Tingay et 
al., 1997). Genetic transformation mediated by Agrobacterium seems to have certain advantages 
respect to direct methods, which include a higher transformation effficiency, the transfer of large 
segments of DNA with minimal rearrangement, and the integration of fewer numbers of copies 
ofthe transgene(s) into the plant genome. 
85 
(%) 
ous+ 
67 
o 
100 
ND 
ND 
100 
ND 
ND 
Materials and Methods 
Improvement and adaptation of callus culture and plant regeneration for genetic 
transformation of Braclliaria 
The present work is in its initial phase and is airned to establish a transformation protocol 
rnediated by A. tumesfaciens of B. decumbens. In this research, matured ernbryos and 
ernbryogenic scutellum derived callus of B. decumbens cv Basilisk accession 606 is used as 
target explants. The bacteria strain LBA 4404 carrying the hypervirulent plasmid pTOK 233 
(50.35 Kb, kindly provided by Dr. Toshihiko Kornari, Japan Tobacco Inc. , Japan) or other 
constructs provided by CAMBIA (Dr. Richard Jefferson, Australia) will be tested. The pTOK 
233 plasmid is a cointegrated systern with three chimeric genes for expression in plants, 
including hygrornycin resistance (hph), gus-intron (uid-intron) and neomycin resistance (npt /1) 
genes driven by the 35S prometer. Tlús plasmid has been used successfully in Agrobacterium 
rnediated transformation of various monocots species. As a fust step, the transformation 
protocols already established for rice, rnaize, barley and other grasses will be tested and rnodify 
accordinghly to optimize agroinfection for B. decumbens. Following is reported the preliminary 
work in progress .. 
Results and Discussion 
The success of plant genetic transformation highly depends on the type and physiological status 
of the explant used. Work suggest that in monocots, the target explant should be in an active 
cellular division stage to be amenable for allien gene introduction. Embryogenic callus derived 
frorn the scutellum has shown to be the rnost appropriate explant for transformation of cereals. 
The protocols for callus induction and plant regeneration of Brachiaria developed at CIA T 
(Lenis, 1992, and Galindo, 1997) include a very short callus phase followed by plant 
differentiation as soon as the embryogenic callus is being formed. This rapid plant formation 
difficults using this tissue as target for agroinfection. At present, different basal medium 
cornpositions are being tested to define the rnost appropriate medium sequence for callus 
induction, maintenance of embryogenic callus, and plant regeneration. Preliminary results 
suggest that rnatured embryos cultured on a modified N6 medium develop embryogenic callus 
that can be rnaintained in the callus phase by subculture on the same medium. The embryogenic 
structures differentiate into plants when transferred onto a modified MS medium. Different 
mediurn sequences are being tested to define the best protocol to maintain the target tissue in the 
optimal stage for agroinfection, to select the putative transgenic tissues, and regenerate the 
transgenic plants. 
Collaborators: C.P. Flores, Z. Lentini 
86 
References 
Aldemita R.R., T.K. Hodges T.K. (1996). Agrobacterium tumefaciens-mediated 
transformation of Japonica and Indica rice varieties. Planta 199:612-617. 
Dong, J., T. Weimin, W. Bucholz, and T. Hall. 1996. Agrobacterium-mediated 
transformation of Javanica rice. Molecular Breeding 2: 267-276. 
Escudero, J.; N. Gunther, M. Sclappi, and B. Hohn. 1996. T-DNA transfer in meristematic 
cells of maize provided intracellular Agrobacterium. The Plant Journal 1 O (2): 355-360. 
GALINDO, L. 1997. Transformación genética de la gramínea forragera Brachiaria spp., 
mediante la técnica de bombardeo de partículas. Tesis Ingeniería Agronómica, 
Universidad Nacional de Colombia. Sede Palmira. 130 pp. 
Hiei, Y.; S. Otha, T. Komari. , and T. Kumashiro. 1994. Efficient transformation of rice 
( Oriza saliva L.) mediated Agrobacterium and secuence analysis of boundaries of T-
DNA. Plant Journal 6:271-282. 
Ishida, Y., H. Saito, S. Otha, Y. Hiei, T. Komari, and T. Kumashiro . 1996. High 
efficiency transformation of maize (Zea mays L.) mediated by Agrobacterium 
tumefaciens. Nature Biotechnology 14: 745-750. 
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. pp 80. 
Rashid, H. , Y. Shuuji, K. Toriyama, K. Hinata. 1996. Transgenic plant production by 
Agrobacterium in Indica rice. Plant Cell Reports 15: 727-730. 
Tingay, S.; D. McEleroy, R. Kalla, S. Fieg, M. Wang, S. Thomton, and R. Brettell. 1997. 
Agrobacterium tumefaciens-mediated barley transformation. The Plant Journal11(6): 
369-1376. 
2.1.13 Incorporation of resistance to fruitworm (Neoleucinodes elegantalis) and budworm 
(Scrobipalpuloides absoluta) in the tomato variety UNAP AL Arreboles by genetic 
transformation 
Tomato (Lycopersicon esculentum Mill) 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 yielding 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 
87 
Campus, has as main objective the development of varieties with resistance or tolerance to sorne 
of these traits. In 1997, this program released the tomate variety UNAP AL Arreboles, which 
has severa! traits atractive to tomate growers such as fruit firmness and good adaptability 
specially to the Valle del Cauca region. But this variety is susceptible to the two major -
limitations to tomate in this region: the fruitworm (Neoleucinodes elegantalis) which damage 
the fruit even at early stages of development, and the budworm (Scrobipalpuloides absoluta), 
which eats the tomate buds and young leaves. lt had been dificult to breed tomate resistant to 
these two pests by standard breeding. The only sources of resistance genes is from wild tomate 
species which are incompatible with the cultivated tomate, and so far the attempts for an 
interspecific breeding program has not been successful (Lourencao et al., 1985). 
An attractive altemative to introduce resistance in tomate against the~e two pests is by genetic 
transformation. Nowadays, it is well known the progress attained in the incorporation by genetic 
transformation of crystal proteins genes from Bacillus thuringiensis (Bt) for resistance to 
Lepidoptera pests. At present, other noyel insecticida! proteíns genes for insect resistance, such 
as vegetative insecticida! proteins (ViPs ), proteinase inhibitors, chitinases, peroxidases, 
cholesterol oxidases, arnong others have been reported. The main objectíve of this work is to 
transform the tomate variety Arreboles with the Bt gene cryiA(b ), which had been used 
susscessfully to obtain resistance against Lepidoptera pests in various economical important 
crops (i.e. maize, cotton). 
A prerequisite for transfering genes into plants is the availability of efficient transformation and 
plant regeneration systems. Although the transfer of genes into tomate using Agrobacterium 
mediated transformation has been reported (McCormick et al., 1986; Fillatti et al., 1987; Ultzen 
et a1.,1995 etc.), the efficiency of plant regeneration and transformation is genotype dependent. 
Following is reported the results from the frrst phase of this project on the adoption and 
adaptation of an efficient plant regeneration system of tomate Arreboles. 
Material and Methods 
Comparison of three media sequence for callus inductíon and plant regeneration from the 
tomato variety Arreboles. 
Three protocols cornrnonly used for tomate were compared. These protocols included the media 
MI (Narvaez, 1993), M2 ( Fillatti et al., 1987), and M3 (Ultzen et al., 1995). The components 
of MI, M2 and M3 callus induction media are shown in Table l. Ml is the most simple while 
M3 is most complex of the three media. M3 consists of MS salts, B5 vitarnins, with the addition 
of 0.05 mg/1 biotin, 0.5 mg/1 folie acid and 2 mg/1 glycine. Sucrose is reduced from 30g/l to 
1 Og/1 and 1 Og/1 of glucose is added. The response following the three protocol medium 
sequence were compared by evaluating callus induction and plant regeneration using the 
following criteria: No. of callus per explant, No. of callus with shoots, No. of shoots per explant, 
and No. of plantlets per explant. Three replications with a total of 108 explants were evaluated 
per each medium. Data was analyzed using a randomized complete block design (SAS, 1988). 
88 
Results and Discussion 
Cotyledonary lea ves from 7-1 O day-old seedlings were used as starting materials. Callus 
formation from the cotíledonary leaf explants was noted two weeks after culture on the 
callus induction mediurn. One month later, callus were transferred onto the corresponding 
plant regeneration medium. Duncan's Multiple Range Test for each of the parameters 
analyzed indicated that the highest response for callus induction and plant regeneration is 
noted on M3 medium sequence (Table 2). An increase in response of about 2-fold and 4-
fold on callus induction and plant regeneration was noted on M3 media respect to Ml and 
M3 medium, respectively. The lowest response was obtained on M2 medium. M2 
contains 3 fold as much sucrose as M3, and does not contain glucose nor an auxin source. 
A total of 238 plantlets were regenerated from 278 initial explants on medium M3 giving 
a regeneration efficiency of 86%, whereas the plant regeneration efficiency for M 1 and 
M2 were of 73% and 34%, respectively. Based on these results M3 medium has been 
selected to conduct the genetic transformation studies with the variety Arreboles. The 
review of key factors affecting the efficiency of Agrobacterium mediated genetic 
transformation of tomato Arreboles is in progress. 
Collaborators: H. Ramírez, Z. Lentini 
References 
Fillatti, J. J., Kiser, J. , Rose, R., and Comai, L. 1987. Efficient transfer of a glyphosate 
tolerance gene into tomato using a binary Agrobacterium tumefaciens 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., Barnason, A., Horsch, K. and Fraley, R. 1986. 
Leaf disc transformation of cultivated tomato (l. esculntum) using Agrobacterium 
tumefaciens. Plant Cell Reports 5: 81-84. 
Narváez-V ásquez, J. 1991. Expr ession of proteinase inhibitor genes in transgenic plants: 
Effect on insect resistance, levels of accumulation in four plant species, and cellular 
compartmentalization. Ph.D. thesis. Washington State University, Pullman, USA. 
Rick, C. M. and Yoder, J. I. 1988. Classical and molecular genetics of tomato: Highlights 
and perspectivs. Annu. Rev. Genet. 22: 2821 300. 
SAS Institute, 1988. SAS User' s guide. SAS Institute, 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 spoted 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. 
89 
Table t. Medium composition for tomato callus inductio n 
Medium 
M I M2 M3 
MS salts 4.5 g / 1 4.5 g 11 4.5 g 1 1 
Mvo-lnositol 100 mg / 1 100 mg / 1 100 mg / 1 
Thiamine-HCI 10 mg/1 0.5 ma 1 1 0.5 mg / 1 
Nieotinie acid 1 me: / 1 5 mg / 1 5 me:/1 
Pyridoxine 1 me: / 1 0.5 mg / 1 0.5 m a 11 
Biotin 
-
0.05 mg / 1 0.05 mg / 1 
Folie aeid 
-
0.5 mg / 1 0.5 mg / 1 
Glvcine - 2 mg/1 2 me:/1 
Sucrose 30 e: 1 1 30 e: 1 1 10 g/1 
Glueose ·- - !Og/1 
BAP 2.5 mg / 1 
- -
IAA 1 mg 11 
-
0.02 mg 11 
Kinetin 
-
4 mg/ 1 4 mg: / 1 
PH 5.8 6.0 6.0 
Phytagel 1.8 a /1 1.8 e: 11 1.8 g /1 
Table 2. In vitro response of tomato Arreboles on medium Ml , M2, a nd M3 commonly used for 
tomato callus induction and plant regeneration. 
M I M2 
Total Total 
Means per 108 o/os Means per 108 
explants explants 
Callus per explant 1.90 b 206 32. 11 1.45 e 157 
Callus w/ shoots/Expl 1.12 b 121 29.42 0.62 e 67 
Shoots 1 Explant 1.53 b 165 27.33 0.78 e 84 
Plantlets 1 Explant 1.40 b 151 34.14 0.50 e 54 
Means followed by the same letter are not significantly different at p: 0.0 l . 
ll Fraetion from the total number of callus 
21 Fraetion from the total number of callus with shoots 
31 Fraetion from the total number of shoots 
4/ Fraction from the total number of plantlets 
51 Response pereentage relative to the other media 
90 
M3 
Total 
% Means per 108 
explants 
24.5 2.57 a 278 
16.3 2.07 a 223 
13.9 3.30 a 356 
12.2 2.20 a 238 
% 
43 .4 
54.3 
58.8 
53.7 
Callus 
Per Expl 
MI 1.90 b 
M2 1.45e 
M3 2.57 a• 
Table 3. Comparison of the three plant culture media on the base of the evaluation of 
four parameters 
Total Rclaúve Callus Total Relative Shoots Total Relative 
Callus Response With shoots Callus Response Per shoots Response 
No. To others w/shoots Toothers Explant No. Toothers 
Media(%) No. Media(%) Media(%) 
206 32.1 1.12 b 121 29.4 1.53 b 165 27.3 
157 24.5 0.62 e 67 16.3 0.78 e 84 13.9 
278 43.4 2.07 a 223 54.3 3.30 a 356 58.8 
• Means with the same letter are not significantly different 
11 Mean from total of 108 explants 
Plantlets 
Per 
Explant 
1.40 b 
0.50 e 
2.20a 
2.1.14 development and standardization of an in vitro clonal propagation method 
of Soursop (Annona muricata 1.) 
Introduction 
Total 
Plantlets 
No. 
151 
54 
238 
Annona muricata L. (soursop) is a tropical fruit tree originated from Colombia and/or 
Brazil (J. León, 1968). Its fruits have interesting properties for agwindustry (Toro, 1995) 
andas a source of antitumoral and insecticide metabolites (Cavé et al., 1993). 
Soursop production in Colombia is very inefficient due to the use of sexual seed for 
propagation. Genome heterogeneity of soursop is very high and the development of seed-
originated trees is therefore heterogeneous. 
Asexual propagation methods like grafting are also used but they contribute to the 
dissemination of diseases like the anthracnose produced by Colletotrichum gloesporioides. 
Clonal propagation methods that use pathogen-free, elite clones will improve the 
production of soursop in Colombia. 
Materials and Methods 
We developed a micrografting method for micropropagation using elongated axillary 
shoots of "Elita" clon (Ríos Castaño et al. , 1996) as scions and in vitro germinated "Elita" 
plantlets as rootstocks. The explants were surface sterilized in 1% NaOCl (v/v) and 
washed with sterile distilled water. Axillary shoots were elongated aseptically in Woody 
Plant Medium (WPM - Lloyd and McCown, 1980) containing low concentrations of 
benzylaminopurine (BAP). Seeds were germinated aseptically in B5 (Gamborg, 1968) 
medium containing low concentrations of gibberellic acid (GA3). 
9 1 
Relative 
Response 
To others 
Media(%) 
34.1 
12.2 
53.7 
Axyllary shoots and apexes were grafted onto hypocotyls of 10-15 days-old, in vitro 
germinated plantlets. Micrografted plants were grown in half strength WPM incubated at 
29 °C (under white fluorescent tubes, 3600 lux, with a 12-h photoperiod). After 30 days, 
plantlets were transferred to the greenhouse under semi-shade conditions. 
Results and Discussion. 
Table 1 sumrnarizes the steps required to obtain micrografted, pathogen-free soursop 
plants. 
Table l. Steps required to obtain micrografted, pathogen-free soursop plants 
Steps 
Mother plants* selection and preparation in the 
screen-house (trimming, fungicide and antibiotics 
sprying, hormone treatment); selection of best 
stems for in vitro planting. 
Stem ~lanting in vitro and elongation of axillary 
buds. 
Rootstock production from sexual seed (seed 
sterilization, in vitro germination and growth) @ 
In vitro micrografting. 
Hardening in greenhouse conditions. 
Time (days) 
60 
30 
30-45 
35-40 
180 
Total: 355 
Average success 
8 stems 1 2 year-old mother plant 
47% of stems produce optimum 
scions 
81% of seeds produce optimum 
rootstocks 
61% successfully graft 
66% of successful micrograftings 
continue developing in the 
greenhouse 
(*) A mother plant is a nursery-grown, vegetatively grafted soursop plant that serves as so urce of stems. 
(@) These two steps are performed simultaneously to synchronize the process. 
To reduce in vitro contamination below 10%, it was necessary to tr~at mother plants with 
fungicides and antibiotics in the screen-house. 
To get an average of 81% in vitro germination of seeds it was necessary to 1) dry the 
seeds after sterilization using an oven (15 days at 37°C) or air-drying (21 days); 2) scarify 
completely the seeds; and 3) supplement the gerrnination mediwn (B5) with GA3 (0.5 
mg/1). 
"V shaped" grafting, which has been first used for bean in vitro grafting (A. Mejía, 1995), 
seemed to be an efficient method to guarantee grafting survival. 
From each stem planted in vitro we could obtain up to 12 scions for in vitro grafting, 
which in creases the multiplication rate and shortens the whole process to lO months ( one 
month shorter than the time required for standard comrnercial production of vegetatively-
grafted plants). 
After 35-40 days rnicrografts were transferred to the greenhouse to a non-sterile soil 
mixture containing soil/sand/ rice husk/perlita.(I :2:1: 1). Survival index (66%) in the 
greenhouse was higher than that (35%) reported by Bejoy and Hariharan (1992) for 
micropropagated, non-micrografted A. muricata plantlets. 
92 
Currently we have 200 micrografted, two months old plants growing in the greenhouse 
that will be transferred to the field to begin agronomic evaluations. 
The entire in vitro micropropagation process is being improved towards its transfer to 
soursop growers. 
References 
Bejoy, M.; Hariharan, M. (1992). In vitro plantlet differentiation in Annona muricata. 
Plant Cell, Tissue and Organ Culture 31: 245-247. 
Cavé, A.; Cortes, D.; Figadere, B.; Hocquemiller, A.; Laprevote, 0 .; Laurens, A.; 
Leboeuf, C. (1993). Recent advances in the acetogenins of annonaceae. In: Phytochemical 
Potencial ofTropical Plants. Edited by K. R. Downum et al ... Plenum press. New York. 
pp 167-202. 
Gamborg, 0.; Miller, R. A.; Ojima, K. 1968. Exp. Cell Res. 50: 151. 
Leon, J. (1968). Anonáceas. En: Fundamentos Botánicos de los Cultivos Tropicales. 
IICA-OEA (Costa Rica). pp 467-473. 
Lloyd, G.; McCown, B., (1.980). Commercially feasible rnicropropagation of mountain 
laurel Kalmia latifolia by use of shoot tip culture. Comb. Proc. Int. Plant. Prop. Soc. 30: 
421-427. 
Mejía, A.; Gómez; A. M.; Jacobsen, H. J.; Wolter, F.; Heinz, E.; Roca, W. M. (1995). 
Development of common bean genetic transformation protocol. In: Biotechnology 
Research Unit. Annual Report. CIAT. pp 59-61. 
Ríos Castaño, D.; Barona, P.; Reyes, C. E. (1996). Guanábano Elita. Agricultura Tropical 
33 (3): 97-101. 
Toro, J. C. (1995). Informe de entrega del germoplasma de guanábana, Annona muricata 
L. pp. 15. ICA. 
Collaborators: N. Royero1, L. Muñoz, A. Mejía, J. Cabra1, W. Roca. 
1 Corporación BIOTEC, UNIV ALLE, Cali 
ACTIVITY 2.2 Identification of points for genetic intervention in plants/stress 
Interactions 
SUMMARY OF ACIDEVEMENTS 
• Al-tolerance of all Brachiaria spp is superior to that reported for Al-tolerant crop vars. 
B. decumbes is best adapted to low N- supply and has higher N-uptake efficiency. 
Brachiaria thus becomes an important source of genes for these traits. 
93 
• Two new secondary metabolities, biosynthesized via the shikimate pathway, have 
been isolated and chemically characterized from Brachiaria decumbens roots grown 
under N and P stress conditions. Inhibitory effects by the two compounds against 
fungal pathogens will be evaluated. 
• CIA T work in exploring the potential of cassava germplasm for vitamins and anti-
oxidants sources has been extended to the Colombian tropical oil palm 'mil pesos' as a 
collaborative extension with Univ. del Choco. Antioxidants Carotene and Tocopherol 
contentare similar or higher than sunflower, but lower than soybean. 
2.2.1 Exploring the Genetic Potential and Stability of Vitamin Content in Cassava 
Introduction 
Improving cassava's nutritional value can improve living conditions of poor farmers and 
urban consumers who use the crop as the main source of nutritional energy. Few have 
studied the concentration and variability of vitamins and minerals in cassava root 
parenchyma. This work intends to determine the potential improvement that can be 
reached through selection and recombination. The nutritional value also depends on the 
availability of nutrients to the human body once the product is processed and consumed. 
The objectives ofthis work are to: 
a) Characterize the genetic diversity within CIAT cassava core collection, (630 
genotypes) selected to represent the genetic diversity in the global cassava 
germplasm collection (approximately 5,500 genotypes) with respect to ascorbic acid 
(vitamin C) and carotenes (vitaminA) content. 
b) Screen elite genotypes for physiological post-harvest deterioration and stability of the 
vitamins after processing. 
e) Evaluate the potential of cassava lea ves as a so urce of vitamins for human nutrition. 
Materials and Methods 
Carotenes. The extraction procedure outlined by Safo-Katanga et al. (1984) was adjusted 
by extracting root parenchyma with petroleum ether. The extraction protocol for leaves had 
to be modified due to the presence of tanins and chlorophylls. A sample of 1 O g was taken 
out of the root or lea ves, at random 1 O to 11 months after planting. The quantification was 
done by ultraviolet spectrophotometry using a Shimadzu UV-VIS 160A recording 
spectrophotometer. UV detection was done atA.= 455nm for root extracts andA.= 490 nm 
for leave extracts. 
Ascorbic acid. The protocol for the determination of ascorbic acid by Fung and Luk (1985) 
was adjusted for cassava leaf and roots taking as base the procedure outlined and involved 
the following steps: 
94 
a) Homogenization of 1 g of fresh leaves or 6 g of fresh roots in a turrax with 20 mi of 
extraction buffer ( 3% phosphoric acid and 8% glacial acetic acid). 
b) Centrifugation for 5 min at 1 ooc and 3000 rpm. 
e) Separation of supematant and vortex of 1 mi of the extract with 2 mi of 1 0% hydrochloric 
acid. Reading was taken immediately with an UV-VIS spectrophotometer. UV detection was 
done at A. =245 nm. Quantification was done using a previously decomposed extract with 
1M sodium hydroxide solútion as blank. During the whole process, samples were protected 
from air in order to avoid oxidation. 
Results 
Ascorbic acid concentration in leaftissue ranged from 1.68 mg/100 g FW to 419.25 mg/100 
g FW. Concentration of ascorbic acid in leaves is higher than in the roots. Average ascorbic 
acid concentration in the leaves was ten times higher than the one observed in the roots. The 
data about ascorbic acid concentration in root parenchyma showed a broad distribution of 
concentrations from less than 1.0 mg/100 g FW to 39.52 mg/100 g FW of fresh roots. 
Vitamin content in the leaves was not correlated to those in the roots. 
Cassava is processed before comsumption using heat treatments which can affect the 
ascorbic acid and carotene content. At the moment we are measuring the effect of following 
treatrnents on the ascorbic acid content of 40 elite clones (genotypes with the highest 
concentration of ascorbic acid): solar drying of cassava flour, oven drying of cassava flour 
and cooking fresh roots 30 minutes. Results from this study are pending. Boiling reduced 
ascorbic acid concentration at least in 60-70% from fresh root content. In addition we are 
evaluating post-harvest root deterioration of selected genotypes in order to correlate this 
results with the carotene and ascorbic acid content. 
Another target for 1998 is to complete the evaluation of the carotene content in the 
cassava core collection. So far screening of the lea ves of the 600 accessions has revealed 
a wide range of variation in carotene content, 23.28 mg/100 g FW to 172.45 mg/ 100 g 
FW. This implies that 6 g of fresh leaves (2 g of dry leaf flour) of genotypes with the 
highest concentration of carotenes in the leaves will supply the daily requirement of 
vitaminA for an adult male (between 600-700 ug vitamin A/day). 
Once screening of the roots is completed, we hope to be able to identify germplasm with 
particularly high or low carotene concentration. 
References 
Safo-Katanga, 0., Aboagye,P., Amartey, S.A., and Olaham, J.H.(1984). Studies on the 
content of yellow pigmented cassava. In: Terry, E.R., et.al., Root Crops Production and 
Uses in Africa. IDRC, Ottawa, Canada, pp.l03-104. 
Ying-Sing Fung and Shiu-Fai Luk (1985). Determination of Ascorbic Acid in Soft Drinks 
and Fruit Juices. Partl. Background Correction for Direct Ultraviolet Spectrophotometry. 
Analyst, 11 O, pp. 201-204. 
Collaborators: C.lglesias, W. Roca, A.L. Chavez, J.M. Bedoya, F. Calle, T. Sanchez 
95 
Vitamin C content in the CIAT cassava core collection 
(lea ves) 
# of accessions 
mg Vitamin e /100 g FW 
96 
2.2.2 Mechanisms of Acid Soil Adaptation in Brachiaria Cultivars* 
Background 
Brachiaria cultivars are the most widely sown forage grasses used for livestock 
production in tropical savannas. An ongoing breeding program at CIA T seeks to combine 
favorable traits, such as adaptation to acid soils, resistance to spittlebug and forage quality, 
within new apomictic cultivars. Easy and quick methods are thus required to screen large 
progenies for acid soil adaptation. To develop physiologically based screening methods, 
we focused on the three major factors that probably contribute to the poor persistence of 
less adapted Brachiaria cultivars on low fertility acid soils: Al-toxicity, P-deficiency and 
N-deficiency. 
Interspecific differences in adaptation to acid soil stress 
Al-toxicity. Al toxicity has been long been identified as the major growth-lirniting factor 
in acid soils. The elongation of the primary root of seedlings in solutions containing only 
A¡3+, Ca2+, and Cr ions was measured after 3 days to evaluate whether interspecific 
differences in Al-tolerance exist among Brachiaria species. The results demonstrated that 
(i) Al-tolerance of all Brachiaria species is superior to that reported for Al-tolerant crop 
varieties, and that (ii) B. ruziziensis was clearly less Al-tolerant than B. decumbens and 
B. brizantha (Fig. 1; left half). At present, HPLC analyses of organic acids in roots and 
root exudates are underway. Preliminary results suggest that citric acid as well as other 
organic acids within roots rnight contribute to the high level of Al-tolerance in Brachiaria 
species. However, there seem to be no significant interspecific differences with respect to 
organic acid accumulation/exudation. This implies that B. decumbens must have 
additional Al-tolerance mechanisms, which are probably not related to chelation of Al3+ 
ions by organic acíds. 
Adaptation to P-deficienc~. P in acid soils is extremely immobile because of chemical 
fixation by A¡3+ and/or Fe + ions. The root systems of acid soil-adapted plants must thus 
explore large soil volumes to take up sufficient P. This can be accomplished by extensive 
root systems. Thin roots can reduce the amount of biomass needed to construct a root 
system of a given length. An experiment was set up to investigate aspects of root system 
morphology, such as root thickness, in relation to P-deficiency. Plants were grown in 
hydroponic culture with declining P-supply in the greenhouse. Upon harvest, root systems 
were stained and scanned with a flatbed scanner. The resulting images were then analyzed 
with WinRHIZO software. 
97 
100 
80 
~ ~ 60 
_J 
0::: 40 0::: 
20 
o 
'E 
g 0.24 
... 
a.> 
...... 
a.> 
~ 0.20 
:o 
...... 
o 
o 
O 8. fUl /Zie nSIS 
• B. decumbens 
v B. bnzantha 
ma.12a vanattes 
--- wheat 1sogemc hnes 
20 40 60 80 100 120 
e o.1s 
-o- B. ruzizienSJs 
_..._ B. decumbens 
-v- B. brizantha 
o 2 4 
phosphate (IJM) 
Figure l. Differential adaptation to Al-toxicity and P-deficiency. Left half: relative root length (RRL) of three 
Brachiaria species as compared to maize and wheat varieties of contrasting AHolerance. Error bars denote SE 
(n = 28-36). Al-sensitive (s) and tolcrant (t) maize varieties were Tuxpeí'lo and South American 3. Al-sensitive (s) 
and toleran! (t) wheat isogenic lines were ES3 and ED. Data of maize varieties and wheat isogenic lines were tnken 
from the literature. Right half: mean root diameter of three Brachiaria species grown in nutricnt solutions under 
decreasing P-supply. Error bars denote SE (n = 8). 
The results demonstrated that B. decumbens has the finest root system, independent of the 
level of P in the nutrient solution, and that B. ruziziensis produced the thickest roots 
(Fig. 1; right half). Based on geometric considerations it can be estimated that 
B. ruziziensis needs 50- 55% more biomass than B. decumbens to construct a root system 
of a given length. Under P-deficient conditions this is a significant advantage, given the 
importance of active foraging for soil P-reserves. 
98 
250 
C]afl.llizxrtsls 
- a ck:tntl!ns 200 B Hbriz.rha 
~ 150 
2: 
:¡¡ 100 
z 
50 
o 
-
N 
E 10 (.) 
::::. 
o 
E 8 
:l. 
....... 
>-(.) 6 
e: 
<V 
·n 4 
le 
<V 
-o- 8 . ruzizi~nsis ~ 2 _._ 8 . decumbens 
ro -'V- 8 . brizantha 
-a. o 
::::l 
1 
z o 100 200 300 400 500 600 
nitrogen (JJM) 
Figure 2. Differential adaptation to N-deficiency. Left half: comparison of the leve! of nitrogen at which 50 % of 
maximum growth was achieved (N50) . Values were calculated from plot of the concentration of N in the nutrient 
solution versus dry matter production, using non-linear curve-fitting (Marquardt-Levenberg algorithm) and a 
Michaelis-Menten-like function. Error bars indicate SD. Right half: N-uptake efficiency as calculated as the amount 
of nitrogen taken up per root surface area. Error bars denote SE (n = 8) 
Adaptation to N-dejiciency. Declining soil N-reserves might cause degradation of 
pastures over time. An experiment was conducted to evaluate whether interspecific 
differences in tolerance to N-deficiency exist. Plants were grown in nutrient solutions in 
the greenhouse under declining N-supply. Upon harvest, root systems were stained, 
scanned and analyzed in a similar manner as above. Significant variability among species 
was found for total dry matter production in response to N-supply (Fig. 2; left half). 
B. decumbens was the species which was best adapted to low N-supply. It was further 
shown that B. decumbens had a superior N-uptake efficiency, that is, it can take up the 
largest amount ofN per root surface area (Fig. 2; right half). 
Biochemical responses to acid soil str ess 
Aromatic metabolites. Secondary aromatic metabolites frequently accumulate under 
nutrient stress when root growth is stimulated at the expense of shoot growth. In sorne 
cases they have been demonstrated to be involved in fungal protection and signal 
exchange with soil microorganisms such as V AM and N2-fixing bacteria. Using reverse-
phase HPLC, two dominant aromatic metabolites were detected in roots of Brachiaria 
species. Their level increased under P- and N-deficient conditions. A purification protocol 
99 
was developed and their structure was elucidated by rneans of structure 1H-NMR, 
13C-NMR, COSY, CD and positive-ion FAB rnass spectroscopy. 
OCH3 
10' 
HO 
o 
OH 
10" 
Figure 3. Structure ofDFQA. FCQA lacks an -OCH1 group ofone ofthe two feruloyl moieties. 
The compounds were shown to be L-1,3-di-0-trans-feruloylquinic acid (DFQA) and 
L-1-0-trans-p-cournaroyl-3-0-trans-feruloylquinic acid or L-1-0-trans-feruloyl-3-0-
trans-p-cournaroylquinic acid (FCQA) (Fig. 3). They are new hydroxycinnarnic acid 
conjugates of quinic acid. Both the hydroxycinnarnoyl and the quinic acid portien are 
synthesized vi a the shik.irnate pathway. Shikirnate kinase, an enzyrne in the middle of the 
shikirnate pathway, is subject to control by energy charge. Under energetically 
unfavorable conditions, a larger portion of shikirnate rnolecules are di verted frorn the rnain 
trunk into secondary products such as quinic acid and its derivatives. This could explain 
their accurnulation in roots of Brachiaria species under P- and N-deficiency. 
Transgenic tobacco plants with suppressed levels of phenylpropanoid products showed 
increased disease susceptibility. This indicates that 5-0-caffeoylquinic acid ( chlorogenic 
acid) rnight be a chernical barrier against microbial attack, since it is the rnajor soluble 
phenylpropanoid in tobacco. In a similar way, DFQA and FCQA rnight accurnulate in 
roots of Brachiaria species as preformed protectants against attacks of soil-bom fungi and 
by this increase root lifespan. An longer root lifespan is expected to increase the arnount 
of nutrients taken up per unit of biornass invested in roots. Under nutrient deficient 
conditions, when biornass is costly to construct, this rnight improve the plant's overall 
performance. Experirnents are underway to evaluate whether these cornpounds inhibit 
growth of the pathogenic fungus Rhizoctonia solani or stimulate growth of V AM fungi in 
vitro. 
Collaborators: P. Wenzl, A.L. Chávez, N.L. Lasso, A. Hemández, G.M. Patiño, L.I. Mancilla, C .. A. 
Pineda1, M. Nai~, I.M. Rao, J.E. Mayer3 
l. National Accelerator Centre (NAR), South Africa; 2. Michigan State University, USA; 3.CAMBIA. 
Canberra, Australia. 
100 
2.2.3 Validation of analitycal methods to quantify Vitamins A and E in Jessenia 
bataua Palm Oil 
Introduction 
Colombia has a wide variety of ecosystems. The tropical humid forest is one of them; it is 
characterized by a wide diversity of flora in wich palm trees occupies a very important 
place. In the Colombian Pacific region, considered as a tropical forest, the inhabitants 
obtain a diversity of products derived from the palm tress. In the present study we have 
selected the "Mil Pesos" palm tree (Jessenia bataua) because of its socioeconomical, 
cultural and enviromental importance. One of the products derived from J bataua is an 
edible oil , extracted from the palm fruit and comparable to olive oil because of it's 
aroma, flavor and chemical composition. In spite of this, we analized the nutritional 
qualities based on the fact that the ·oil of Jbataua stored at room temperature conserves 
it's organoleptic quality for up to three years (Balick 1992). 
It has been preved that the presence of natural antioxidants such as carotenes and 
tocopherols give stability to the oils (Nargiz 1991). These substances have elicited great 
interest in present days because of the possible role they play in ·Jle prevention of root 
damage and many degenerative process. They are also considered micronutrients as well 
as vitamins that actas buffers and reductors, and with this characteristic, they may be used 
for industrial production . 
In this study we assess the methods reported in the literature for quantification of carotene 
and tocopherols of vegetable oils and we validate the two best analitical methods using 
UV-VIS spectrophotometry technique for caro ten es determination and HPLC (High 
Performance Liquid Chromatography) for tocopherols. We applied these methods for 
quantification of carotenes and tocopherols for various samples of oil J bataua and 
others vegetable oils. 
Methods 
Carotenes. Samples of oil, obtained from Jbataua palm fruits were extracted with a 
mixture of petroleum ether, acetone and water ( 1:1:1) at room temperature . This extract 
was saponified with methanol at a 1 :4 ratio of oil to alcohol, catalysed by 10% potassium 
hydroxide. The reaction mixture was stirred (12 hours) until all the triglycerides were 
converted to alkyl esters. The concentration of carotenes was determined by UV-VIS 
spectrophotometry, using a shimadzu UV-VIS 160A recording spectrophotometer. UV 
detection was done at A = 455 nm using oil previously decolored as blannk(2). 
Tocopherols. Aliquots of 1 g palm oil were dissolve in hexane. This mixture was 
homogenized by centrifugation and the tocopherols were analyzed by triplicates using 
normal-phase HPLC (AOCS 1992) using a Hewlett Packard 1 OSO chromatograph, 
column Lichrosorb SI 60, 10 J.!M (Merck), hexane:dioxane (94:6) as mobil phase, flow 
rate : 1.2 mllmin and fluorometric detection (A. ex= 295 nm and A em = 330 nm). 
101 
Results. 
Ten methods reported by literature to determine carotenes and five to characterize 
tocopherols were evaluated. The saponification method at room temperature permited a 
good extraction and characterization of carotenes. The quantity of the detected carotenes 
and tocopherols depends on factors as harvest, site, and extraction methods, and 
therefore may vary within samples. 
Complete conversion of the triglycerides into alkyl esters took place without destroying 
the carotenoids during the saponification reaction. This was monitored by the interna! 
standard method. There were not significant differences between concentrations of 
tocopherols and carotenes of J bataua oil and the others comercial edible oils. The data 
were evaluated as mean ± standard deviation of triplicate deterrninations. Statistical 
significance within sets of data was determined by one-way analysis of variance. The 
significance level was in all cases p.< 0.05. This study showed the carotenes vary in the 
following ranges: J bataua oil (0.601 ± 0.02 mg carotenes/100 g oil.), sunflower oil 
(0.184 ± 0.08 mg carotenes/ 100 g oil) and soya oil (1.036 ± 0.08 mg carotenes/100 g oil). 
The average values of a.-tocopherol content were: J bataua oil (16.573 ± 0.8 mg/100 g 
oil), sunflower oil (16. 777 ± 0.1 mg/1 00 g oil) and soya oil ( 38.969 ± 0.9 mg/1 00 g oil). 
Palm oil obtained from the fruits of J bataua showed to be as resistant to oxidation as the 
other edible oils because their unsaponificable componentes including tocopherols and 
carotenes. Since crude J bataua oil is not refined, the natural oxidants are partly 
preserved and these compounds are reported responsible for their higher stability to 
oxidation. 
The storage stability of P-carotene and a.-tocopherol in the J bataua oil were observed 
for a period of 5 months. The carotene were more stable at 1 O oc than at 28-30 oc. There 
was a slight decrease (1.76%) in the carotene content ofthe J bataua oil when stored in 
dark glass. However the bigger decrease (41.09%) in the carotene concentration of the oil 
when it was stored in clear glass could be due to greater exposure to light, leading to an 
increased oxidation and degradation of the carotenes. 
References 
Balick, J. M. y colaboradores (1992). Jessenia y Oenocarpus, palmeras aceiteras 
neotropicales dignas de ser domesticadas. Roma: F AO. 
Macrae, R. (1992). HPLC in food analysis. London:Academic pres, Inc., pp.1 87-205. 
AOCS (1992). Official method to determination of tocopherols and tocotrienols m 
vegetable oils and fats by HPLC (1992). Ce 8-89. 
NARGIZ, C. and K. Unal. (1991). J. Sci. Food Agric. 56:79, pp. 208-212. 
Collaborators: A.Rios1, W.Roca, F.Villota, A.L.Chavez 
!.Universidad del Chocó, 2.Universidad del Valle 
102 
Figura l . UV Spectra of: (!) P-carotene standard (50ppm). (2)J bataua 
saponifed oil. 
200Ar------r~--------------------------~ 
1.50 
1.00 
0.50 
r-, 
rl2\ 
f \ 
1 
1 
1-
2-
o. o L-L----.---------.--------.,---_:¡:=r-------r---__~ nm 
200 800 
( 100 nm/div) 
Figura 2. Comparison between the chromatograms of a mix of tocopherols 
and J bataua oil. 
Column: Lichrosorb SI ffi {250 X 4 mm),Flow: 1 .2 ml/min, roobi\ phase: 
Hexane:Dioxane (94:6),detection: Fluorescente ex. 295 nm, em. 330 nm. 
2 
5 10 15 
1.Aa!il2J. t:aaua 
2 a-Tarferd 
3 r-Tarff!ld 
4. li-Tax:lflfOI 
4 
1 
1 
11 
11 
11 
11 
11 
11 
11 
1 1 
1 1 
11 
11 
11 
1 1 
1 
1 
20 25 
103 
OUTPUT 3. COLLABORATION WITH PUBLIC AND PRIVATE SECTOR 
PARTNERSENHANCED 
ACTIVITY 3.1 Organization ofNetworks, Conferences/Workshops and 
Training Courses. 
SUMMARY OF ACHIEVEMENTS 
• DGIS accepted to funda second phase ofthe Cassava Biotechnology Network (CBN) 
to facilitate the stakeholder-participatory process and develop a regional structure for 
LAC. CBN has also been involved in the development of projects for a low-cost, 
artesanal, cassava tissue culture scheme, and for restoring national cassava collections 
in Ecuador. 
• In the period Oct. 1997- Sept 1998, Project SB-02 has organized/carried out 6 
Workshops and Training Courses in Biotechnology and Agrobiodiversity, including 
tissue culture of fruit tress, genetic transformation, biodiversity, QTL analysis and 
IPR. 
3.1.1 The Cassava Biotechnology Network 
The Cassava Biotechnology Network (CBN) is a global voluntary research network, 
founded at CIAT in 1988. Members include CG centers, advanced laboratories, national 
programs, and farmers organizations in 35 countries. 
CBN' s goal is to mobilize the contribution of biotechnology tools to enhancing cassava's 
value for: 
• food security; 
• poverty alleviation; 
• econornic development via farm in come and rural jobs; and 
• natural resource management (particularly through providing resources and incentive 
for small-farm conservation practices) 
CBN's principal objectives are to: 
l. Integrate priorities of small-scale farmers and processors in cassava biotechnology 
research 
2. Stimulate cassava biotechnology research on priority topics 
3. Foster exchange ofinformation on cassava biotechnology research, techniques, results, 
materials 
From July 1992 to June 1997, the Government ofthe Netherlands, through DGIS, funded 
a CBN Coordination project. This year, from July 1997 to Aug 1998, DGIS accepted to 
fund an extension (retroactive; approved ll Dec 1997) for a stakeholder-participatory 
104 
process to design a new regional structure for CBN. This added a fourth objective for the 
year, to ( 4) facilitate the stakeholder-participatory process to design a new regional 
structure. In 1999, DGIS anticipates funding a new three-year project, based at CIAT, 
for coordination and activities of the CBN Latín American regional network. 
Regionalization of the network is expected to permit: closer contacts with end-users; 
better integration of farrner participation; enhanced regional ownership and 
accomplishment; enhanced responsiveness and cost-effectiveness; and expanded funding 
base. It is possible that regionalization willlead to enhanced sustainability of the network. 
Sorne concerns have been raised conceming the lack of a global coordination facility after 
1998: individual national or laboratory interests may domínate in a region; competition 
for scarce collaborative opportunities and funds may exclude late comers; public 
information in the North, about cassava and its importance to millions of people, may be 
neglected. 
Outputs by objective: 
l. Integrate priorities of small-scale farmers and processors in cassava biotechnology 
research 
• Farmers' workshop in the Latín American regional stakeholders meeting (see 
appendix for the farrners' workshop recommendations to CBN) 
• Farmer-participatory research, planned with farmers (Ecuador) or on the basis of 
farrners' stated priorities (Colombia) 
(i) Implementation began September: Low-cost tissue culture of traditional cassava 
variety and varieties from participatory selection; extension to other local staple crops 
with similar propagation problems (bananas, plantain); redevelopment of indigenous 
root crops; possible local micro-business; economic opportunity. Colombia; FIDAR 
(NGO), CIAT, community associations; supplemental funding anticipated SWG-PRGA 
(ii) Implementation beginning October: Restore national cassava collection from in-
vitro back-up at CIAT; propagate and disseminate variety selected by 
farmers+national program; rescue, conserve (field + in-vitro ), and use local cassava 
genetic diversity; reconstruct production-processing-market chain; restore food 
security and economic opportunity. Coastal Ecuador; CIAT, INIAP, Uni. Tec. 
Manabí, UATAPPY farmers' cooperative; El Nino disaster relief; funded by USAID 
• Field day for CBN biotechnology members to visit small-scale cassava farmers in 
small groups, and plenary and working group sessions on farmer participatory 
research in CBN, organized in 1997-98; to be held at the CBN IV meeting 2-7 Nov 
1998, in Brazil 
105 
2. Stimulate cassava biotechnology research on priority topics 
New cassava biotechnology research projects developed and research begun (see above, 
Obj. 1): 
• Planting material : need for responsive seed systems for high quality and sufficient 
quantity of desired varieties: Proposal for farmer-participatory research in collaboration 
with an NGO, FIDAR (see above) 
• Conservation and use of cassava diversity; new and improved products: Participatory 
proposal design and implementation, with farmers in Ecuador for El Nino disaster relief 
(see above) 
3. Foster exchange of information on cassava biotechnology research, techniques, results, 
materials 
• Facilitated combined electronic mail listserver between CBN, SWP-PRGA Plant 
Breeding Group, and invited biotechnologists outside CBN, on selected issues related 
to feasibility, value, and organization to biotech-assisted farmer participatory research 
• CBN IV fourth international scientific meeting organized with EMBRAP A (to be 
held 2-7 Nov 1998 in Salvador, Brazil). 
4. Facilitate the stakeholder-participatory process to designa new regional structure 
• Latín American stakeholders regional meeting and report 
• Concepts developed for CBN-Asia and CBN-Africa, in collaboration with CIP and 
IITA 
• Global stakeholders meeting and report 
• CBN regionalization plan presented (donor briefing) CGIAR Mid Term Meeting, 
Brazil 
Training courses/Workshops on agrobíodíversity and biotechnology organized m 
cooperation with LDC and DC CIA T partners 
• CBN IV fourth intemational scientific meeting organized with EMBRAP A (to be 
held 2-7 Nov 1998 in Salvador, Brazil). 
106 
Collaborators: 
• CIAT: Ann Marie Thro, Willy Roca, Martín Fregene, Luis Alfredo Hemandez 
Principal partners 1997-1998: 
• Brazil: EMBRAP A: Luis JCB Carvalho, Marcio Porto, Maria Jose Sampaio 
• Ecuador: Uni. Tecnico Manabí, Heman Caballero; FLACSO, Susan Poats 
• Colombia: FIDAR (NGO), Jose Restrepo 
• CGIAR: CIP, Gordon Prain; liTA, Mpoko Bokanga and Robert Booth; SWG 
PRGA: Charlie Spillane, Louise Sperling 
• Advanced research organizations: Univ. of Bath: G. Henshaw; NRI: N. Poulter and 
A. Westby; AU Wageningen: R. Visser; ILTAB: C. Fauquet, N. Taylor 
• CBN regionalization process participants - Latin American and African regions : 
Over 40 partners from national programs, prívate industry, CG centers; farmers ' 
associations 
• CBN Scientific Advisory Committee; Regional Technical Committee; Local 
Organizing Committee for CBN IV meeting organization: over 20 persons from 
national programs, advanced research institutions, and CG centers 
3.1.2 Training courses/workshops on agrobiodiversity and biotechnology 
organized in cooperation with LDC and DC CIAT partners. 
• Workshop on Intemational Systems on Intellectual Property Rights Applied to 
Biotechnology and Biodiversity, 1998. Jorge A. Goldstein PhD., J.D. Attomey at 
Law, specialist IPR Director Steme, Kessler, Goldstein & Fox P.L.C. Washington, 
D.C. CIAT. Cali-Colombia. April27-28, 1998. 
• Course/Workshop on Tissue Culture of Perennial Tropical Fruits. 1998. Proffessor 
Richard Litz and Pamela Moon. University ofFlorida, Homestead, USA. CIAT. Cali, 
Colombi. August 24-28. 
• Training on rice genetic transformation and anther culture were given to two 
scientists from Venezuela (Ariadne Vegas, Fonaiap, and Ramiro de la Cruz, 
Universidad de Los Llanos); one scientist from Colombia (Fdiz Milena Quintero, 
Cenicafé); and one from Surinam (Jerry R Tjoe-Awie, ADRON). 
• Course on Biodiversity: genetic diversity and phylogeny. Support/organization with 
Institute, Washington DC 
• Workshop on QTL analysis for CIAT staff. 
• Workshop on Genetic Diversity with participation of CIP and CIMMYT scientists. 
Collaboration of Comell Univ. Institute of Genomic Research. 
107 
ACTIVITY 3.2 Assembling of data bases, genetic stocks, maps probes and related 
information. 
SUMMARY OF ACHIEVEMENTS 
• Project SB-02 has assembled a range of genetic stocks, which are product of its 
research, including anther culture-derived rice plants and lines, trangenic rice lines, 
transgenic cassava Iines and micrografted soursop plants. 
• The strain collection assembled in the BRU has increased to 101 E. Coli strains 
containing plasmids with genes of interest for genetic transformation; and 112 A. 
tumefaciens and 17 A. rhizogenes strains for use in genetic transformation. 
3.2.1 Genetic stocks. 
(i) The rice anther culture laboratory generated total of 17,673 plants from October 
1997 to September 1998. These lines includes material with cold tolerance, 
resistance to rice blast, sogata, rice hoja blanca virus, high yield, high milling 
grain quality. Resposible Z.Lentini: 
• 1, 711 anther culture derived plants for the wild QTL CIAT -USAID project. 
• 7,764 plants from anther culture for FLAR breeding program (Brazil, Colombia, 
Costa Rica, Uruguay, Panama, Paraguay, and Venezuela); 
• 3,758 somaclones for the Venezuelan National Plan ofRice leaded by Fundarroz 
with the participation ofFonaiap, Danac, and la Universidad de los Llanos. 
• 4,440 somaclones plants for Fedearroz, Colombia. 
• 815 doubled haploids lines from 1997 were in advanced trials of evaluations for 
disease send as part of the VIOFLAR for selection at Brazil, Paraguay and 
Uruguay. 
(ii) Transgenic rice lines carrying gus, hygromycin resistance, and resistance to rice 
hoja blanca virus. Responsible Z. Lentini. 
(iii) Transgenic cassava plants carrying bar gene (resistance to herbicide Basta). 
(iv) . Micropropagated and micrografted (annona muricata) plants: 200 in vitro. 
Collaboration with Corporación BIOTEC, Cali. 
3.2.2 E. coli andA. tumefaciens coJiections 
The BRU Agrobacterium tumedaciens collection contains 112 strains. Among the most 
important strains of A. tumefaciens are LBA4404, AT650, C58C1 and EHA101 , which 
are hypervirulent wild strain isolated from cassava. 
The Escherichia coli collection contains 1 O 1 strains. Among them, the strains DH5a, JM 
109, XL1-Blue and DH10B are often used in cloning. The collection contains many 
plasmids with different genes of interest for plant transformation like hygromicyn and 
108 
basta resistance genes, cry IA(b) (lepidoptera resistance), and metabolic enzyme such as 
rubisco. PEP carboxylase and starch branching enzyme. 
The viability of bacteria collections was evaluated this year and three copies per strain 
were stored at -80oC. 
Table l. Collection of E. coli strains. 
Eschericltia coli 
Wild strains 
Strains with plasmids 
Strains with genes of interest 
Total strains in the collection 
Table 2. Collection of A. tumejaciens strains. 
Agrobacterium tumejaciens 
Wild strains 
Strains with plasmids 
Strains with genes of interest 
Total strains in the collection 
Number 
30 
45 
26 
101 
Number 
32 
30 
51 
112* 
• The total number of A. tumefaciens strains may be smaller since few 
strains may have lost viability 
Table 3. Collection of A. rltizogenes strains. 
Agrobacterium rhizogenes 
Wild strains 
Strains with plasmids 
Total strains in the collection 
Number 
12 
5 
17 
Collaborators: L. Mancilla, P. Chavarriaga, V. Segovia, J. Ladino, Z. Lentini, W. Roca 
ACTIVITY 3.3 Publications, project proposal development, and contribution 
to IRP and biosafety management. 
SUMMARY OF ACHIEVEMENTS 
• In the period Oct. 1997~Sept 1998, project SB~02 staff has produced: 19 refereed 
publications, 14 other publications as book chapters and conference proceedings, and 
has supervised 15 degree Theses. 
109 
• The project staff has also generated/prepared 13 Project Concept Notes and Proposals 
for seeking financia! support. 
• The Team has contributed to at least 5 activities (workshops, conferences, 
consultancies, etc) to develop IPR and biosafety management at CIAT in Colombia 
and the LAC region. 
• In the period of this report, Project SB-02 has enjoyed the contributions of 1 O donors 
through complementary funding. 
• In Nov. 1997, Project SB-02 was the subject of a CIAT ICER. The review Panel' s 
recommendations ha ve been submitted to CIA T management. 
3.3.1 Publications by SB-02 staff in the period Oct. 1997- Sept. 1998. 
3.3.1.1 Refereed Publications 
Sanchez 1., Angel, F., Grum M., Tohme, J ., Lobo, M. and Roca, W. 1998. Variability of 
chloroplast DNA in the genus Passiflora L. In press Euphytica. 
Sanchez, l. Angel, F. Grum, M. Tohme, J. Lobo, M. and Roca, W. 1998. Characterization 
molecular, base sólida para el mejoramiento genético de Passifloraceae Juss. Fitotecnia 
Or Colombiana, Organo de la sociedad Colombiana de Fitomejoramiento y producción de 
cultivos. In press. 
Sanchez, I., Angel, F., Grum, M., Tohme, J., Lobo, M. and Roca, W. 1998. Estudios de 
la variabilidad genética de Passifloraceae - familia de plantas nativas promisorias. 
Revista Corpoica. In press. 
Fajardo D., Angel, F., Grum, M., Tohme. , Lobo, M., Roca, W., Sánchez, 1., 1998. 
Genetic variation analysis of the genus passiblora L. using RAPD markers. Euphytica 
101:341-347 
Restrepo S. and V.Verdier. 1997. Geographical differentiation of the population of 
Xanthomonas axonopodis pv manihotis in Colombia. Appl. Env. Microbial, 63 :4427-
4434. 
Verdier, V., Mosquera, G. and Assigbetse K. 1998. Detecting the Cassava Bacteria! 
Blight Pathogen, Xanthomonas axonopodis pv manihotis, using the polymerase chain 
reaction. Plant Disease, 82: 79-83. 
Verdier, V., Restrepo, S., Mosquera, G., Duque, M.C., Gerstl A. and Laberry R. 1998. 
~ The Xanthomonas axonopodis pv manihotis populations in Venezuela: its genetic and 
O pathogenic variation. Plant Pathology, 4 7: in Press 
Thro, A.M., and M. Fregene 1998. The Cassava Biotechnology Network: Impact of the 
network and scientific advances in cassava biotechnology. Tropical Agriculture 
tf.J\.. (Trinidad). In press. 
110 
¡ 
1 
Thro, A.M., N. Taylor, K., Raembarkers, R. Visser, J. Pounti-Kaerlas, C. Schopke, C. 
Iglesias, W. Roca, M.J. Sampaio, C. Fauquet, and l. Potrykus. 1998. Maintaíning the 
Cassava Biotechnology Network: An agenda to make a difference. Nature 
Biotechnology 16(5):428-430. 
~ Thro, A.M., W.Roca, G. Henry, C. Iglesias, and S.Y.C. Ng. 1998. Contributions of in 
- r itro biology to cassava, a small farrner's crop. African J. Crop Sci. Kampala (accepted). 
j 
Lentini, Z., Martinez, C.P. andE. Nossa. 1997. Introgression ofhigh response to anther 
culture into indica rice. lntemational Rice Research Notes (IRRN). 22(1 ): 18-20. 
Jones, P., G.; Beebe, S.; Tohme, J. Galwey, N.W. 1997. The use of geographical 
information systems in biodiversity exploration and conservation. 
Roa, A.C.; Maya, M.M.; Duque, M.C.; Tohme, J. , Allem, A.C. Bonierbale, M. 1997. 
AFLP analysis of relationship among cassava and other Manihot species. 
Fregene, M.; Angel, F. Gomez, R. Rodríguez, F. Chavarriaga, P. Roca, W. Tohme, J. 
Bonierbale, M. 1997. A molecular genetic map of cassava (Manihot esculenta Cranzt). 
Theor. Appl. Genet. 95:431-441. 
Escobar, R. Mafia, G., Roca, W. 1997. A methodology for recovering cassava plants 
from shoot tips maintained in liquid nitrogen. Plant Cell Reports 16: 474-476. 
Yeoh, H.H. Sanchez, T. and Iglesias, C. 1998. Large-scale screening of Cyanogenic 
Potential in Cassava Roots using the Enzyme-Base Dipsticks. J. Food Composition and 
Analysis 11: 2-1 O 
\ Fregene, M.A., Ospína, J.A. and Roca, W. 1998. Recovery of cassava (Manihot 
j 1}--.- esculenta Crantz) plants from inmature zygotic embryos. Plant cell and orgon culture (in 
\O press. 
'~ Thro, M.M., Roca, W. Henry, G. Iglesias, C. and Mg, SY.C. 1998. Contribution of in 
vitro biology to cassava, a small farmer ' s crop. African J. Crop Sci. (accepted) 
Chavarriaga-Aguirre, P. Maya, M.M. Bonierbale M. Kresovich, S., Fregene, M., Tohme, 
• Jkochert, G. 1998. Microsatellites in cassava (Manihot esculenta Crantz): discovery, 
inheritance and variability. Trends in Genetics (in press). 
3.3.1.2 Other Publications 
• Book chapters and Conference Proceedings 
Thro, A.M., M. Fregene, N. Taylor, C.J.J.M. Raembarker, J. Pounti-Kaerlas, C. Schopke, 
R. Visser, l. Potrykus, C. Fauquet, W. Roca, C. Hershey. 1998. Genetic biotechnologies 
111 
and cassava-based development of marginal rural areas: In: T. Hohn and L. Leisinger 
9eds). Gene and biotechnology of food crops in developing countries. Springer; Wien, 
New York (E.S. Dennis et al. (eds). Plant gene research: Basic Knowledge and 
applications. In press. 
Roca, W.; Correa-Victoria, F. Martínez, C. Tohme, J. Lentini, Z. Levy, M. 1998. 
Development of durable resistance to rice blast: Productívíty and environmental 
considerations. In Komen, J. Falconi, C. and Hernandez, H. (eds) Transforrning 
Priorities into Viable Programs, Lima Peru. IBS/!ISNAR, The Hague, The Netherlands. 
Thro, A.M. and M.O. Akoroda (eds). 1997. Proceedings Third International Scientific 
Meeting, Cassava Biotechnology Network, Kampala, Uganda, 26-30 Aug. 1996. African 
J. Root & Tuber Crops, Vol II. Nos. 1&2. 
Thro, A.M. 1998. The Cassava Biotechnology Network: Considerations for effective use 
of biotechnology tools for cassava. pp. 515-525 In: CIA T (R Howeler). Cassava 
breeding, agronomy, and participaroty research in Asia. Proc. Fifth Asían Cassava 
Research Workshop, Danzhou, Haínan, China, 3-8 Nov. 1996. 
Restrepo S., Tohme,. J. and Verdier, V. 1998. Assessment ofthe AFLP technique for the 
genetic study of Xanthomonas axonopodis pv. Manihotis. In Abstr. Of the proceedings 
of the 7th International Congress of Plant Pathology. Edinburg, Scotland, 9-16 August 
1998. 
Restrepo, S. Velez C. Verdier, V. 1998. Measurement of the genetic diversity in 
Xanthomonas axonopodis pv manihotis within different field in Colombia. In Abstract of 
the proceedings of the 7th International Congress of Plant Pathology, Edinburgh, 
Scotland, 9-16 August 1998. 
Lentini, Z. 1998. Biotechnology mediated genetic modification of rice for insect 
resistance: A component for integrated pest management. Latin American Rice Forurn 
4(1):14~16. In Conference Proceedings. 
Lentini, Z., L.Calvert, I. Lozano, andE. Tabares. 1998. Resistance to hoja blanca virus 
encoded by the nucleocapsid transgene in tropical rice. IX Intemational Congress on 
Plant Tissue and Cell Culture. Plant Biotechnology and In Vitre Biology in the 21st 
Century. Jerusalem, Israel. 
Mora A. Z. Lentini, and C. Martinez. 1998. Rice breeding with anther culture. II Latín 
American Meeting on Plant Biotechnology. Havanna, Cuba. 
Lentini, Z., L. Calvert, I. Lozano, and E. Tabares. 1998. Trangenic rice with RHBV 
resistance. 11. Latin American Meeting on Plant Biotechnology. Havanna, Cuba. 
Mora, A., Z. Lentini, and C. Martinez. 1998. Anther culture a tool for rice breeding. 
First Intemational Meeting of Rice. Havanna, Cuba. 
112 
Lentini, Z. L. Calvert, E. Tabares, l. Lozano, M. Cuervo, W. Roca and BC Ramires, 
1997. Resistan ce to rice hoja blanca in transgenic rice using the nucleoprotein gene. 
Fifth lnternationa1 Crongress ofPlant Molecular Biology. Singapore. 
Mejia-Jimenez, H.J. Jacobsen, 1998. Development of an in vitre regeneration system in 
common bean suitable for genetic transformation. Eucarpia Int. symposium, Partevedre, 
Spain. 
Escobar, R.H., Palacio, J .D., Rangel, M.P. and Roca, W.M. 1998. Crioconservación de 
apices de yuca mediante encapsulacion-dehidratacion. In: 111 Latín American meeting on 
plant biotechnology, REDBIO, La Habana, Cuba. 
• Degree Theses 
Galindo, L. 1997. Transformación genética de la gramínea forragera Brachiaria spp 
mediante la técnica de bombardeo de partícula. CIAT. 
Rengifo J. 1997. Caracterización de germoplasma de Phaseolus vulgaris L. cultivado 
Andino por medio de AFLPs 
Reina, G. 1997 Effecto de la congelación rápida de ápices de yuca (Manihot esculenta 
Cranzt) euphorbiaceae, a la crioconservación de nitrógeno líquido. 
Giraldo M. 1997. Effecto del genotipo de yuca yuca (Manihot esculenta Cranzt) 
euphorbiaceae, a la crioconservación de nitrógeno líquido. 
Mancilla L.I. 1997. Efecto del aluminio y bajos niveles de nutrientes sobre la expresión 
de genes y actividad de las enzimas H+ ATPasa y Fosfatasa acida y alcalina de tres 
especies de Brachiaria. 
Suarez M.C. 1998. Desarrollo de la técnica cDNA/AFLPs para mapeo en yuca Manihot 
esculenta Crantz. 
López A. F. 1997. Estudios de la bacteria Xanthomonas campestris pv manihotis: 
expresion de un gen de patogenicidad PthB. 
Dávalos L. 1997. Mapeo de genes de resistencia al virus Hoja Blanca en arroz en la 
población Fanny x Oryzica Llanos 5. 
Segovia, V. 1997. Optimización de la transformación genética de tejido de Yuca 
Manihot esculenta C. utilizando Agrobacterium tumefaciens. 
Palacio, J.D. 1998. Crioconservación de ápices de yuca utilizando la técnica de 
encapsulación deshidratación. CIAT. 
113 
Hernández, A. 1998. Propiedades Químicas de paredes celulares de raíces de Brachiaria 
spp en relación con el bajo suministro de nutrientes y toxicidad de aluminio. 
Almeida, A. 1998. Marcadores microsatelites asociados a QTL;, de Rendimiento en 
Arroz, en un retrocruce avanzado entre una especie silvestre (Oryza barthii) y una 
variedad mejorada (Lemont). 
Santaella, M. 1998. Hacia el aislamiento de genes de resistencia en Arroz a Pyricularia 
grisea. 
Lasso, N.L. 1998. Efecto del nitrogeno, fósforo y alumnio en la estructura del sistema 
radical de tres especies del género Brachiaria. 
Villota, F. 1998. Evaluación de métodos para la determinación de carotenos y 
tocoferoles. 
3.3.1.3 Concept Notes and Project Proposals prepared 
• Gene flow analysis at the ecological frontier: Setting risk assessment for genetically 
modified plants. Donor: BMZ. Participating institutions: CIAT; Federal Biology 
Institute, Germany; ICA and Institute Von Humboldt, (Colombia); UNALM (Peru); 
DANAC (Venezuela); UCR (Costa Rica). Responsible: Z. Lentini, W. Roca, D. 
Debouck. 
• Expanding the range of uses of cassava starch: A source of employment and income 
Fedeyuca, Colombia; and Agropecuaria Mandioca, Venezuela. Responsible: z. 
Lentini, C.Iglesias, W. Roca. 
• Plant genetic engineering to incorporate tolerance to iron toxicity and increase iron 
nutrional institutions: CIAT; and IDEA (lnstitute of Advanced Studies), Venezuela. 
Z.Lentini (CIA T), R. Rangel (IDEA). 
• Genetic transformation of venezuelan rice varieties for resistance to rice hoja blanca 
virus. Donor: Polar. Participating institutions: CIAT; Rutgers University; Danac and 
IDEA, Venezuela. Responsible: Z. Lentini (CIAT), Nilgun Tumer (Rutgers), Maria 
Angelica Santana (IDEA). 
• Increasing cassava starch yields: Stabilíty in synthesis and storage in the roots. 
Donor: Palmaven, Petroleum of Venezuela. Participating institutions: CIAT, IDEA 
(lnstitute of Advanced Studies), Universidad Central de Venezuela (UCV), and 
Agropecuaria Mandioca, Venezuela, Responsible: Z. Lentini and C. Iglesias (CIA n, 
Maria Angelica Santana (IDEA), Juscelino Tovar (UCV). 
• Funding of CBN Second Phase: Latín American Network. Donor: DGIS. 
Responsible: A.M. Thro, W. Roca and partners from LAC countries. 
• Producción de semilla de calidad para mejorar la economía de pequeños agricultores 
de la Costa Norte de Colombia. Donor: DGIS-Colombia. Collaboración con 
CORPOICA. Responsible: W. Roca, A.M. Thro. 
114 
• Genetic Improvement of Common Bean Exotic Germplasm and Biotecnology-
Second Phase. Donor: AGCD. Collaborator: Univ. of Ghent, Belgium, Responsible: 
W. Roca, C. Cardona. 
• Development of a pilot project for cassava under cryopreservation. Donor: to be 
identified. Collaborators: CORPOICA, EMBRAPA, CUBA, liT A, CIP. 
Responsible: W. Roca 
• Fine mapping and cloning and engineering of apomixis gene from Brachiaria. Donor: 
prívate sector. Collaborators: Clemson Univ., USA; CAMBIA, Australia. 
Responsible: J. Tohrne, W. Roca, J. Miles. 
• Genetic improvement/clonal multiplication of tropical fruit crops. Donor: Fontagro. 
Collaborators: CORPOICA, FONAIAP, INIAP. Responsible: I. Sanchez 
(CORPOICA), W. Roca 
• Development of an Artesanal Tissue Culture Propagation Lab. Donor: System Wide 
Program a Participatory Research and Gender Analysis. Collaborators: FIDAC (an 
DGO), small cassava growers, Cauca, Responsible: W. Roca, A.M. Thro. 
DONORS CONTRIBUTING TO PROJECT SB-02 COMPLEMENT ARY 
FUNDING IN 1997-98 
The RF. 
• Rice Biotechnology. (I) Molecular analysis of Pyricularia genetic structure and 
population dynamics; (ii) Mapping of rice blast resistance genes; (iii) transgenic 
resistance to RHBV. 
• Cassava Biotechnology: (I) Saturation of cassava molecular genetic map; (ii) 
mapping of ACMF resistance genes. 
USAID 
• Mapping of QTLs from wild Oryza spp responsible for yield in rice<*> 
BMZ/GTZ Germany 
• Common bean genetic transformation (*) AGCD, Belgium 
• Transfer of econornic traits from P. coccineus to common bean 
DFID, U.K. 
• Genetic analysis of cassava roots post-harvest deterioration 
• Genetic transformation of cassava(*) 
DGIS, The Netherlands 
• Cassava Biotechnology Network (**) 
115 
SDC, Switzerland 
• Microsatellite markers for cassava 
COLCIENCIAS, Colombia 
• Cloning CBB resistance genes 
Institute A.V. Humboldt, Colombia 
• Training Course 
Fundación Polar 
• Publication of Rice anther Culture Manual. 
(*) To end in 1999. 
(**) Ended in mid 1998. 
INTERNALL Y COMMISIONED EXTERNAL REVIEW (ICER) OF PROJECT 
SB-02 
Date of Review: 
Review Panel: 
Nov. 17-23, 1997 
B. Schaal, Washington Univ. St. Louis, MO, USA. (Chair) 
M.J. Sampaio, EMBRAP A, Brazil 
K.I. Hayashi, Japan 
W. Beversdorf, Novartis Co (CIA T BOT) 
The review's outcome was very possitive. Project SB-02 staff, teaming with Project SB-
0 1 staff, ha ve prepared and submitted to the attention of the Review Panel a Strategy for 
future directions in agrobiodiversity and biotechnology research at CIAT. The document 
contain a proposal to focus CIAT biotech research at genomic level, with a Genome 
Research Lab comprising three outputs: genome characterization, genome modification 
and clonal propagation. An additional facility is proposed for regional partnerships in 
geno mi e research, particularly in dealing with crops outside CIA T' s mandate and with 
the private sector. 
116 
PROJECT SB-02 ST AFF (1998) 
N ame 
J . Tohme 
S. Beebe 
S. Singh 
A. Bellotti 
Z. Lentini 
M. Fregene 
A. Mejia 
A.M. Thro 
C. Martínez 
I. Sanchez 
V. Verdier 
D. Debouck 
W. Roca 
Discipline 
Genetícs 
Breeding 
Breeding 
Entomology 
Cell Biology/genetics 
Molec. Genetics 
Biology 
Breeding 
Breedings 
Genetics 
Molec. Pathology 
Botan y 
Physiology 
Genome Modification 
L.F. Galindo 
F. Giraldo 
P . Chavarriaga 
L.I. Mancilla 
V. Segovia 
J.J. Ladino 
R. Escobar 
M.P. Rangel 
E. Tabares 
L. Duque 
C. Flores 
H. Rarnírez 
Pablo Herrera 
M. Valenciano 
Genome Diversity 
G. Gallego 
A. almeida 
C. López 
A.C. Roa 
M.C. Suarez 
E.Gaitan 
E. Barrera 
J.P. Gutierrez 
Area 
(Molec. Markers) 
(Gene mapping) 
( wide crossing) 
(IPM) 
Tissue culture/genetic transformation 
(Molecular markers) 
(Tissue culture. transformation) 
(Nerworking) 
(Germplasm-enhancement) 
(Diversity-CORPOICA) 
Microbial diversity 
(Genetic Resource) 
(Tissue culture/transformation) 
Young Research- COLCIENCIAS 
Assistant 
Associated Scientist 
Assistant 
Y oung Research- COLCIENCIAS 
Student- Univ. Nacional 
Research Assistant 
Student- Univ. Javerlana 
Research Assistant 
Student Univ. Nacional 
Student Doctorado Univ .. Nacional 
Student Doctorado Univ .. Nacional 
Téchnician 
Téchnician 
Coordinador Investigación 
Research Assistant 
Research Assistant 
Research Assistant 
Research Assistant 
Research Assistant 
Research Assistant 
Research Assistant 
117 
Dedication 
o¡o 
1.0 
0.7 
0.3 
0.2 
0.8 
1.0 
l. O 
1.0 
0.75 
1.0 
l. O 
0.2 
0.8 
J. Lopez 
A. Bohorquez 
J. Vargas 
M. Santaella 
D.F. Cortés 
S.M. Garzón 
N. Reyes 
Plant-Stress interactions 
P. Wenzl 
A.L. Chaves 
G. Patiño 
A. Hemández 
J.M. Bedoya 
F. Villota 
Administrative 
O.L. Cruz 
C.S. Zuñiga 
A.V. Humboldt 
J.D. Palacio 
P. Sanchez 
J.F.Femández 
SIN Cm 
F. Rodríguez 
P. Toquica 
Corporación BIOTEC 
N. Royero 
Research Assistant 
Research Assistant 
Research Assistant 
Student 
Student 
Student 
Technician 
Student, PhD. 
Associated Research 
Student 
Student 
Student 
Student 
Bilingual Secretary 
Bilingual Secretary 
Visiting Researcher 
Visiting Researcher 
Visiting Researcher 
Visiting Researcher 
Visiting Researcher 
Visiting Researcher 
118