Impact of GC content on gene expression pattern in chicken.

BACKGROUND: GC content varies greatly between different genomic regions in many eukaryotes. In order to determine whether this organization named isochore organization influences gene expression patterns, the relationship between GC content and gene expression has been investigated in man and mouse. However, to date, this question is still a matter for debate. Among the avian species, chicken (Gallus gallus) is the best studied representative with a complete genome sequence. The distinctive features and organization of its sequence make it a good model to explore important issues in genome structure and evolution. METHODS: Only nuclear genes with complete information on protein-coding sequence with no evidence of multiple-splicing forms were included in this study. Chicken protein coding sequences, complete mRNA sequences (or full length cDNA sequences), and 5' untranslated region sequences (5' UTR) were downloaded from Ensembl and chicken expression data originated from a previous work. Three indices i.e. expression level, expression breadth and maximum expression level were used to measure the expression pattern of a given gene. CpG islands were identified using hgTables of the UCSC Genome Browser. Correlation analysis between variables was performed by SAS Proprietary Software Release 8.1. RESULTS: In chicken, the GC content of 5' UTR is significantly and positively correlated with expression level, expression breadth, and maximum expression level, whereas that of coding sequences and introns and at the third coding position are negatively correlated with expression level and expression breadth, and not correlated with maximum expression level. These significant trends are independent of recombination rate, chromosome size and gene density. Furthermore, multiple linear regression analysis indicated that GC content in genes could explain approximately 10% of the variation in gene expression. CONCLUSIONS: GC content is significantly associated with gene expression pattern and could be one of the important regulation factors in the chicken genome.

Indel segregating within introns in the chicken genome are positively correlated with the recombination rates.

Insertions and deletions (Indel) are important sources of genetic diversity and phenotypic divergence. Many factors such as mutation, recombination, selection and genetic drift can jointly affect the indel distribution across the genome. Studies of the relationship between recombination and indel density can, to a certain extent, reflect the selective constrain on indel. Based on the improved genetic map, genome sequence assembly and the partial (0.25X) shotgun sequencing of three breeds of domestic chicken, we calculated the recombination rates and the indel density segregating within introns and intergenic for 4 Mb windows (n = 210). Regression analyses demonstrated that recombination rates are significantly correlated with intron indel density, but not with the intergenic indel density. After adjusted regional effect, the significant trend was remained. This implies that selection is an important factor to influence the indel distribution within introns in chicken genome. By contrast, the intergenic indel seem to be neutral. Since the intron indel density on Z chromosome is less than half of that on autosomes, we preliminarily deduced that genetic hitchhiking might be more important than background selection in producing the observed correlation. As these two processes are not mutually exclusive, it is most likely that both contribute somewhat to the observed pattern. In result similar to previous study, we also found SNP density is highly correlated with indel density. Based on this characteristic, a hypothesis suggested that there are common effects of mutation and/or selection on the occurrence of indel and point mutations. This hypothesis can not explain our observations.