Proteomic profiling of developing cotton fibers from wild and domesticated Gossypium barbadense

Pima cotton (Gossypium barbadense) is widely cultivated because of its long, strong seed trichomes ('fibers') used for premium textiles. These agronomically advanced fibers were derived following domestication and thousands of years of human-mediated crop improvement. To gain an insight into fiber development and evolution, we conducted comparative proteomic and transcriptomic profiling of developing fiber from an elite cultivar and a wild accession. Analyses using isobaric tag for relative and absolute quantification (iTRAQ) LC-MS/MS technology identified 1317 proteins in fiber. Of these, 205 were differentially expressed across developmental stages, and 190 showed differential expression between wild and cultivated forms, 14.4% of the proteome sampled. Human selection may have shifted the timing of developmental modules, such that some occur earlier in domesticated than in wild cotton. A novel approach was used to detect possible biased expression of homoeologous copies of proteins. Results indicate a significant partitioning of duplicate gene expression at the protein level, but an approximately equal degree of bias for each of the two constituent genomes of allopolyploid cotton. Our results demonstrate the power of complementary transcriptomic and proteomic approaches for the study of the domestication process. They also provide a rich database for mining for functional analyses of cotton improvement or evolution.