Gene expression analysis in Festulolium hybrids using RNA-seq

The advent of Next Generation Sequencing permitted sequencing whole genomes and transcriptomes with a depth and resolution never achieved before. One of the exciting opportunities has been the analysis of gene expression in wide hybrids. The hybridization brings homo(eo)logous genes of parental genomes under one roof and these pairs of genes may evolve according to different scenarios during the first generations after hybrid creation. First, one gene copy may become non-functional by genetic and/or epigenetic changes (nonfunctionalization). The other option is that one copy may acquire a novel, usually beneficial function and will be preserved by natural selection with the other copy retaining the original function (neo-functionalization). The last scenario for duplicated genes is that both copies become partially compromised by mutation accumulation to the point where their total capacity has been reduced to the level of the single-copy ancestral gene (sub-functionalization).
The aim of our project is to study processes described above in initial generations (F1-F3) of hybrids of ryegrasses (Lolium) and fescues (Festuca) – Festuloliums. We set out to develop a SNP platform based on RNA-seq data to reconstruct genomic composition of Festulolium hybrids and to analyze the expression of parental genes in hybrid progenies. We first performed Illumina RNA-seq on six F. pratensis and six L. multiflorum parental genotypes and mapped sequence reads from each genotype separately onto a common transcriptome reference sequence of 19.345 L. perenne genes. This led to identification of 600.000 to 900.000 putative SNPs per genotype. In each pair-wise F. pratensis x L. multiflorum combination, we selected SNPs that were homozygous within each genotype, and polymorphic between species. This
yielded 30.000 to 90.000 species-specific SNPs, depending on the sequence depth and particular pair-wise combination of F. pratensis × L. multiflorum genotypes. We found that the SNPs were localized in 6.000 to 9.000 genes distributed more or less evenly across the genome. The next step is to analyze RNAseq data from reciprocal F1, F2 and F3 hybrids, to quantify gene expression from the respective parental genomes and to assess the putative loss and silencing of parental alleles in hybrid progenies. This work was supported by grant awards P501/11/0504 and CZ.1.05/2.1.00/01.0007.