QTL Mapping of Growth-Related Traits in a Full-Sib Family of Rubber Tree (Hevea brasiliensis) Evaluated in a Sub-Tropical Climate

Livia Moura Souza,Rodrigo Gazaffi,Anete Pereira Souza,Carla Cristina Silva,Vincent Le Guen,Saulo Emilio Almeida Cardoso,Camila Campos Mantello,Dominique Garcia,Antonio Augusto Franco Garcia,Rongling Wu

doi:10.1371/journal.pone.0061238

Abstract

The rubber tree (Hevea spp.), cultivated in equatorial and tropical countries, is the primary plant used in natural rubber production. Due to genetic and physiological constraints, inbred lines of this species are not available. Therefore, alternative approaches are required for the characterization of this species, such as the genetic mapping of full-sib crosses derived from outbred parents. In the present study, an integrated genetic map was obtained for a full-sib cross family with simple sequence repeats (SSRs) and expressed sequence tag (EST-SSR) markers, which can display different segregation patterns. To study the genetic architecture of the traits related to growth in two different conditions (winter and summer), quantitative trait loci (QTL) mapping was also performed using the integrated map. Traits evaluated were height and girth growth, and the statistical model was based in an extension of composite interval mapping. The obtained molecular genetic map has 284 markers distributed among 23 linkage groups with a total length of 2688.8 cM. A total of 18 QTLs for growth traits during the summer and winter seasons were detected. A comparison between the different seasons was also conducted. For height, QTLs detected during the summer season were different from the ones detected during winter season. This type of difference was also observed for girth. Integrated maps are important for genetics studies in outbred species because they represent more accurately the polymorphisms observed in the genitors. QTL mapping revealed several interesting findings, such as a dominance effect and unique segregation patterns that each QTL could exhibit, which were independent of the flanking markers. The QTLs identified in this study, especially those related to phenotypic variation associated with winter could help studies of marker-assisted selection that are particularly important when the objective of a breeding program is to obtain phenotypes that are adapted to sub-optimal regions.

Highlights

Showed a polymorphism between the parents of the mapping population. This level of polymorphism observed in the studied population was expected because a similar level of polymorphism was previously reported in a study by Lespinasse et al [9]
simple sequence repeats (SSRs) derived from ESTs have been shown to be less polymorphic (39%) than SSRs derived from genomic sequences (54.5%), they are present in gene-rich regions of the genome and are frequently detected at a relatively high abundance [39]
A chi-squared test performed on the genotyped polymorphic loci revealed that 138 (48.6%) loci exhibited a segregation ratio of 1:1, 31 (10.9%) loci exhibited a ratio of 1:2:1, and 115 (40.5%) loci exhibited a ratio of 1:1:1:1

Summary

Introduction

This mapped region is fully characterized, i.e., the significant effects are identified, along with the linkage phase between markers and QTLs, and the QTL segregation pattern was inferred. The estimate map contained 284 markers assigned to 23 linkage groups and spanned a total of 2688.8 cM; the previous map developed for the rubber tree was considerably smaller [9], more informative genetic linkage was determined because it was possible to use markers with segregation ratios of 1:1:1:1, 1:2:1 and 1:1 in order to estimate an integrated map, better results were achieved.

Results

Conclusion