Duplicate gene evolution, homoeologous recombination, and transcriptome characterization in allopolyploid cotton

A Long-Read Transcriptome Assembly of Cotton (Gossypium hirsutum L.) and Intraspecific Single Nucleotide Polymorphism Discovery

Upland cotton (Gossypium hirsutum L.) has a narrow germplasm base, which constrains marker development and hampers intraspecific breeding. A pressing need exists for high-throughput single nucleotide polymorphism (SNP) markers that can be readily applied to germplasm in breeding and breeding-related research programs. Despite progress made in developing new sequencing technologies during the past decade, the cost of sequencing remains substantial when one is dealing with numerous samples and large genomes. Several strategies have been proposed to lower the cost of sequencing for multiple genotypes of large-genome species like cotton, such as transcriptome sequencing and reduced-representation DNA sequencing. This paper reports the development of a transcriptome assembly of the inbred line Texas Marker-1 (TM-1), a genetic standard for cotton, its usefulness as a reference for RNA sequencing (RNA-seq)-based SNP identification, and the availability of transcriptome sequences of four other cotton cultivars. An assembly of TM-1 was made using Roche 454 transcriptome reads combined with an assembly of all available public expressed sequence tag (EST) sequences of TM-1. The TM-1 assembly consists of 72,450 contigs with a total of 70 million bp. Functional predictions of the transcripts were estimated by alignment to selected protein databases. Transcriptome sequences of the five lines, including TM-1, were obtained using an Illumina Genome Analyzer-II, and the short reads were mapped to the TM-1 assembly to discover SNPs among the five lines. We identified >14,000 unfiltered allelic SNPs, of which ∼3,700 SNPs were retained for assay development after applying several rigorous filters. This paper reports availability of the reference transcriptome assembly and shows its utility in developing intraspecific SNP markers in upland cotton.

A Transcriptome Profile for Developing Seed of Polyploid Cotton

Cotton ranks among the world's important oilseed crops, yet relative to other oilseeds there are few studies of oil-related biosynthetic and regulatory pathways. We present global transcriptome analyses of cotton seed development using RNA-seq and four developmental time-points. Because Upland cotton (Gossypium hirsutum L.) is an allopolyploid containing two genomes (A/D), we partitioned expression into the individual contributions of each homeologous gene copy. Data were explored with respect to genic and subgenomic patterns of expression, globally and with respect to seed pathways and networks. The most dynamic period of transcriptome change is from 20-30 d postanthesis (DPA), with about 20% of genes showing homeolog expression bias. Co-expression analysis shows largely congruent homeolog networks, but also homeolog-specific divergence. Functional enrichment tests show that flavonoid biosynthesis and lipid related genes were significantly represented early and later in seed development, respectively. An involvement of new features in oil biosynthesis was found, like the contribution of DGAT3 (diacylglycerol acyltransferase) to the total triglyceride expression pool. Also, catechin-based and epicatechin-based proanthocyanidin expression are reciprocally biased with respect to homeolog usage. This study provides the first temporal analysis of duplicated gene expression in cotton seed and a resource for understanding new aspects of oil and flavonoid biosynthetic processes.