Variation for heterodimerization and nuclear localization among known and novel oil palm SHELL alleles.

Oil palm breeding involves crossing dura and pisifera palms to produce tenera progeny with greatly improved oil yield. Oil yield is controlled by variant alleles of a type II MADS-box gene, SHELL, that impact the presence and thickness of the endocarp, or shell, surrounding the fruit kernel. We identified six novel SHELL alleles in noncommercial African germplasm populations from the Malaysian Palm Oil Board. These populations provide extensive diversity to harness genetic, mechanistic and phenotypic variation associated with oil yield in a globally critical crop. We investigated phenotypes in heteroallelic combinations, as well as SHELL heterodimerization and subcellular localization by yeast two-hybrid, bimolecular fluorescence complementation and gene expression analyses. Four novel SHELL alleles were associated with fruit form phenotype. Candidate heterodimerization partners were identified, and interactions with EgSEP3 and subcellular localization were SHELL allele-specific. Our findings reveal allele-specific mechanisms by which variant SHELL alleles impact yield, as well as speculative insights into the potential role of SHELL in single-gene oil yield heterosis. Future field trials for combinability and introgression may further optimize yield and improve sustainability.

Fine-mapping and cross-validation of QTLs linked to fatty acid composition in multiple independent interspecific crosses of oil palm.

BACKGROUND: The commercial oil palm (Elaeis guineensis Jacq.) produces a mesocarp oil (commonly called 'palm oil') with approximately equal proportions of saturated and unsaturated fatty acids (FAs). An increase in unsaturated FAs content or iodine value (IV) as a measure of the degree of unsaturation would help to open up new markets for the oil. One way to manipulate the fatty acid composition (FAC) in palm oil is through introgression of favourable alleles from the American oil palm, E. oleifera, which has a more unsaturated oil. RESULTS: In this study, a segregating E. oleifera x E. guineensis (OxG) hybrid population for FAC is used to identify quantitative trait loci (QTLs) linked to IV and various FAs. QTL analysis revealed 10 major and two putative QTLs for IV and six FAs, C14:0, C16:0, C16:1, C18:0, C18:1 and C18:2 distributed across six linkage groups (LGs), OT1, T2, T3, OT4, OT6 and T9. The major QTLs for IV and C16:0 on LGOT1 explained 60.0 - 69.0 % of the phenotypic trait variation and were validated in two independent BC2 populations. The genomic interval contains several key structural genes in the FA and oil biosynthesis pathways such as PATE/FATB, HIBCH, BASS2, LACS4 and DGAT1 and also a relevant transcription factor (TF), WRI1. The literature suggests that some of these genes can exhibit pleiotropic effects in the regulatory networks of these traits. Using the whole genome sequence data, markers tightly linked to the candidate genes were also developed. Clustering trait values according to the allelic forms of these candidate markers revealed significant differences in the IV and FAs of the palms in the mapping and validation crosses. CONCLUSIONS: The candidate gene approach described and exploited here is useful to identify the potential causal genes linked to FAC and can be adopted for marker-assisted selection (MAS) in oil palm.

Genetic variability of oil palm parental genotypes and performance of its' progenies as revealed by molecular markers and quantitative traits.

Studies were conducted to assess the genetic relationships between the parental palms (dura and pisifera) and performance of their progenies based on nine microsatellite markers and 29 quantitative traits. Correlation analyses between genetic distances and hybrids performance were estimated. The coefficients of correlation values of genetic distances with hybrid performance were non-significant, except for mean nut weight and leaf number. However, the correlation coefficient of genetic distances with these characters was low to be used as predicted value. These results indicated that genetic distances based on the microsatellite markers may not be useful for predicting hybrid performance. The genetic distance analysis using UPGMA clustering system generated 5 genetic clusters with coefficient of 1.26 based on quantitative traits of progenies. The genotypes, DP16, DP14, DP4, DP13, DP12, DP15, DP8, DP1 and DP2 belonging to distant clusters and greater genetic distances could be selected for further breeding programs.