Skip to main content
ARS Home » Southeast Area » Miami, Florida » Subtropical Horticulture Research » Research » Publications at this Location » Publication #199967
Title: Mapping QTLs in breeding populations of cacao (Theobroma cacao L.)

Author
item Cervantesmartinez, Cuauhtemoc
item PHILLIPS-MORA, WILBERT - CATIE
item Brown, James
item Schnell Ii, Raymond
item Douglas, Stephanie

Submitted to: Proceedings of the XVth International Cacao Conference
Publication Type: Proceedings
Publication Acceptance Date: 6/15/2006
Publication Date: 10/10/2006
Citation: Cervantesmartinez, C., Phillips-Mora, W., Brown, J.S., Schnell Ii, R.J., Douglas, S.L. 2006. Mapping QTLs in breeding populations of cacao (Theobroma cacao L.). Proceedings of the XVth International Cacao Conference.

Interpretive Summary: In the present research we develop the analysis of QTLs for yield and frosty pod resistance (Moniliophthora roreri Cif. and Par.) of the cacao (Theobroma cacao L.) clone ‘UF 712’. The study is performed in a set of crosses of ‘UF 712’ with other nine elite clones. These crosses were made at CATIE, Turrialba, Costa Rica, as a part of the cacao breeding program to increase the genetic base of resistance in Central America. Our approach increases the precision of QTL estimation by pooling populations with a common parent (‘UF 712’). More reliable QTLs are expected to be obtained with his method due to a larger sample size of the progeny and an increase in the number of genetic backgrounds where the actual QTLs are tested. However, low polymorphism of the parental clones and nonadditive gene action of the trait of interest may diminish the efficiency of our approach. In this report we show the preliminary results. A composite molecular map was built with 58 SSR markers previously screened and selected by their polymorphic state in ‘UF 712’. Thirty-nine markers were located in the same linkage group and sequence than the reference map developed at CIRAD, FR. The correlation of map distances of adjacent markers between the two maps was .69 (P < .001). The average distance between markers was approximately 10 cM. Theoretical results indicate that markers spaced for more than 18 cM are unlikely to be mapped due to the small sample size per each cross. A larger number of markers would shorten the distance among them making it possible to incorporate them into the composite linkage map. However, this is a challenging task given the low percentage of polymorphism of the clone ‘UF 712’. Two markers were associated with the resistance to frosty pod using single factor analysis. The SSRs were located at linkage groups 2 and 8. Two favorable alleles, one from each of the two markers, were responsible of an increase of approximately 12% in the resistance of frosty pod. Four markers were associated with the total number of pods in linkage groups 1, 2, 3, and 4. A set composed of one favorable allele of each of theses markers was responsible of an increase of 12 pods per tree. A larger number of markers would shorten the distance among them making it possible to incorporate them into the composite linkage map. However, this is a challenging task given the low percentage of polymorphism of the clone ‘UF 712’.

Technical Abstract: Genomic regions of cacao (Theobroma cacao L.) responsible of phenotypic expression have been identified by the use of molecular markers with quantitative trait locus (QTL) analysis, and with molecular marker-trait ANOVA, which is used to implement marker assisted selection (MAS) programs. Developing large mapping populations in cacao with sufficiently large sample sizes (150 to 200 trees) is a difficult task given the size and other biological characteristics of the species. Alternatively, the integration of existing breeding populations into QTL analysis has been proposed as a method to increase the sample size and number of genetic backgrounds which are tested for potential QTLs. Cervantes-Martinez and Brown (2004) proposed a method using F1 populations of cacao for mapping QTL in existing breeding populations. This approach increases the power of QTL detection in one recurrent parent by utilizing crosses made to several other parents in a partial full-sib mating design. Crosses among clones selected for disease resistance and other beneficial horticultural traits were made in the partial full-sib mating design, and evaluated in a field experiment over a five-year period, beginning two years after planting, at CATIE, Costa Rica, as part of the CATIE cacao breeding program. Crosses with two recurrent clones, ‘UF 712’ and ‘UF 273’, were identified as potential candidates for QTL analysis as breeding populations, given their favorable GCA estimates for frosty pod resistance and total pod production, respectively (Cervantes-Martinez et al. 2006). The crosses with ‘UF 712’ were chosen as final candidates, as ‘UF712’ showed more resistance to frosty pod. The objectives of this research were: a) to construct a composite linkage map based on molecular markers from the crosses of ‘UF 712’ with other nine elite clones, and b) to map QTLs for resistance to frosty pod (Moniliophthora roreri Cif. and Par.), yield, vigor, and maturity traits and estimate allelic effects using the haplotypic method, pooling the breeding populations. We present the preliminary results in this paper of single factor analysis for these populations, and discuss the theoretical and practical implications for the implementation of this strategy.