Dissecting the genetic architecture and candidate genes of latex yield traits in Hevea brasiliensis using integrative ultra-dense QTL mapping and transcriptome analysis

Hevea brasiliensis is the sole commercial source of natural rubber at industrial scale, yet the polygenic architecture underlying latex productivity has constrained the pace of genetic improvement. In this study, we constructed an ultra-dense SNP-based linkage map from whole-genome resequencing data generated for a biparental F 1 population (IAN-873 ×CATAS-106) and employed to map quantitative trait loci (QTLs) for latex yield related traits across two environments. A total of 20,730,159 high-confidence SNPs were condensed into 5069 recombination bin markers spanning 18 linkage groups and genetic length of 10,779.69 cM, with a mean inter-bin interval of 2.35 cM. QTL analysis identified 146 loci associated with the three latex yield related traits, of which 22 were classified as stable and/or colocalized across environments, with favorable additive alleles predominantly contributed by the elite parent (IAN-873). Candidate gene mining within six prioritized stable QTL clusters yielded nine functionally annotated genes with predicted roles in isoprenoid precursor supply and transcriptional regulation of laticifer metabolism. Parallel transcriptome profiling of latex from phenotypically contrasting individuals identified 1573 differentially expressed genes, from which a manually curated subset of 75 genes with established functional annotation to rubber biosynthesis, encompassing mevalonate pathway enzymes, rubber particle proteins, and multiple transcription factor families, was selected for candidate gene prioritization. Notably, four out of nine genes ( HbHMGR1 , HbFPS1 , HbACAT2 , and HbERF-RAP2–12 ) were independently supported by both positional QTL evidence and differential expression data, establishing them as high-priority candidates for functional validation. These findings advance the molecular understanding of latex biosynthesis and deliver valuable genomic resources, including a high-resolution linkage map, stable QTL intervals, and integrated candidate genes, for implementation in marker-assisted and genomic selection programs targeting rubber yield improvement in rubber tree.