Location: Sustainable Perennial Crops Laboratory
Project Number: 8042-21000-267-14-S
Project Type: Specific Cooperative Agreement
Start Date: Jul 1, 2010
End Date: Jul 1, 2015
This project will address the critical need in understanding functional diversity in cacao germplasm and identify likely sources of new genes for breeders. So far the efforts to manage the genetic diversity of cacao have mainly concentrated on germplasm maintenance. Characterization and evaluation of these resources have received much less attention. Most of the phenotyping in this collection has been limited in morphological descriptors. Only a small fraction of the germplasm held in the international collections has been evaluated for major agronomic traits. New sources of variations in resistance to diseases and pests, for environmental adaptation, and for processing quality are urgently needed for cacao breeding. Through this project we would like to phenotype the major agronomic traits such as yield components and diseases resistance. With improved phenotyping results and high through-put genotyping using functional gene markers, we can better understand and document the functional diversity in this collection. Results of this project will contribute to more efficient management of cacao germplasm and better use of cacao germplasm for varietal development through the identification of new sources and higher levels of resistance to cacao diseases, as well as other important agricultural traits.
Based on the previous phase of collaborations fingerprinting data, mislabeling and duplicates have been identified, genetic relationship among individual accessions, families and populations have been analyzed. A subset of 200-300 cacao germplasm accessions held in CRU/ UWI (approximately 10% of the entire collection) comprising different cacao populations will be selected. The criteria to assemble this working collection will be based on (i) confirmed genetic identity through previous DNA fingerprinting, ii) geographic origin and genotypic diversity, and (iii) historical importance of the accessions. Phenotyping will be carried out in the International Cocoa Genebank, Trinidad. Within plot (intra-clone) mislabeling will be identified using 40-50 SNPs. Phenotypic traits related to yield components and diseases resistance, including number of pods, pod index, bean number, bean size, bean weight, butter fat content, field resistance to witches’ broom disease (Moniliophthora perniciosa) and black pod (Phytophthora) will be evaluated. Previously recorded data on morphological variations, (Bekele et al., 2006); will be combined with new observations. In case duplicated trees in different environment (e.g. in the UWI campus) and Marper Farm) are available, phenotypic data will be recorded from different environments. Based on the multi-year field observations, resistance to witches’ broom disease will be further evaluated using the agar-drop inoculation method (Surujdeo-Maharaj et al., 2003). Genotyping will be done with the Illumina Golden Gate’s Bead Array platform and a total of 3000-4000 SNP markers will be genotyped. Part of these SNP have been developed through an agreement (SCA: 1275-21000-264-04S; CIRAD). These SNP markers were developed based on forty-five cDNA libraries constructed from a wide range of cacao organs and tissues, including flowers, cherels, pod cortex, shoot, root, germinated seeds and embryos from diverse genotypes (Lanaud et al., 2009). The first panel of 1536 SNPs has been tested. In addition, two to three hundreds candidate gene SNPs for disease resistance will be developed based on the expression studies of ARS Project (Project number: 1275-21220-225-00D). More SNPs will be available from the ARS cacao genome project in Miami, FL. Population structure patterns will be investigated using the Bayesian clustering method implemented in the STRUCTURE program (Pritchard et al. 2000). Both single-marker models and hyplotype-based tests will be applied for all SNP–trait combinations. Normality of traits and variance homogeneity will be tested using the UNIVARIATE and GLM Procedure of SAS, respectively (SAS Institute Inc., 1998). Marker data will be compared with trait values by three methods: a one-way ANOVA using SAS, a non-parametric Kruskal–Wallis test and a likelihood of the odds (LOD) score test both performed by MapQTL 4 (Van Ooijen and Maliepaard, 1996). Patterns of linkage disequilibrium (LD) among SNPs that are significantly associated with the same trait will be assessed using the maximum-likelihood approach implanted in GENETICS (WARNES and LEISCH 2006; http://cran.r-project.org).