Novel patterns of expression and recruitment of new genes on the t-haplotype, a mouse selfish chromosome

Abstract

The t-haplotype of mice is a classical model for autosomal transmission distortion. A largely non-recombining variant of the proximal region of chromosome 17, it is transmitted to more than 90% of the progeny of heterozygous males through the disabling of sperm carrying a standard chromosome. While extensive genetic and functional work has shed light on individual genes involved in drive, much less is known about the evolution and function of the rest of its hundreds of genes. Here, we characterize the sequence and expression of dozens of t-specific transcripts and of their chromosome 17 homologues. Many genes showed reduced expression of the t-allele, but an equal number of genes showed increased expression of their t-copy, consistent with increased activity or a newly evolved function. Genes on the t-haplotype had a significantly higher non-synonymous substitution rate than their homologues on the standard chromosome, with several genes harbouring dN/dS ratios above 1. Finally, the t-haplotype has acquired at least two genes from other chromosomes, which show high and tissue-specific expression. These results provide a first overview of the gene content of this selfish element, and support a more dynamic evolutionary scenario than expected of a large genomic region with almost no recombination.

Keywords: gene gain; neofunctionalization; transmission distortion.

Figure 1.

Chromosomal locations of assembled t-specific transcripts. (a) Numbers of genes for which t-specific transcripts were assembled using the three assembly strategies. The proportion of genes mapping to the t-complex (3–42 Mb on chromosome 17) is shown in green, while those mapping elsewhere in the genome are in white. The dashed line indicates the number of genes present in all assemblies. (b) Proportion and location of genes assembled along the t-complex. Light green bars indicate the proportions of genes in 1 Mb windows, for which a t-haplotype-specific sequence was assembled. The grey line shows the average excess heterozygosity of M. m. domesticus +/t mice compared to +/+ mice, adapted from [24]. The locations of t-specific genes are shown as green empty diamonds, so mapping genes can be better visualized. The locations of the four inversions along the t complex, based on the coordinates of genes confirmed to be in each, are shown on top of the figure.

Figure 2.

Relative expression of t-specific alleles in the t complex. Colour coding shows the log₂ ratio of the average expression of the t allele in +/t mice to that of the + allele in +/ + mice. Non-significant differences in expression are coloured grey, while tissues with no expression (on average < 1 TPM for both alleles) are white. Genes whose t allele is underexpressed (purple) in at least one tissue and do not show overexpression (green) in any tissue are shown on the left. Genes with conserved t allele expression in all tissues with expression are shown in the middle. Genes whose t allele is overexpressed in at least one tissue are shown on the right. The numbers of gene copy gain detected in 4 +/t versus 4 +/ + M .m. domesticus DNA samples are indicated on top of the figure, with asterisks denoting fixed copy gain among the four +/t mice.

Figure 3.

Ratios of non-synonymous to synonymous substitution rates (dN/dS) of t (green) and + (black) alleles of 16 t complex genes. Only genes with a CDS alignment of at least 100 base pairs and dS > 0 on both the + and t lineages were included. Black stars on top of the bars mean that dN/dS is significantly different between the t and + alleles, and white stars indicate dN/dS values significantly higher than 1 (using likelihood ratio tests; see Methods). The estimated mean copy number gained by 4 +/t M. m. domesticus mice is indicated on top of the figure, with asterisks denoting fixed copy gain among the four +/t mice. Underlined genes have a premature STOP codon in their t alleles. The numbers of substitutions in the t and + alleles are shown in parentheses on the bottom (for numbers of synonymous and non-synonymous substitutions see electronic supplementary material, data S7). The boxes on top of the figure indicate the t allele’s expression pattern (figure 2). (Online version in colour.)

Figure 4.

Presence, expression and sequence evolution of gained genes on the t-haplotype. (a) PCR bands showing the presence of the t-specific copies of Ppp1cb and Rnpepl1 in 3 +/t mice and their absence in 3 +/+ mice (for all 20 mice tested, see electronic supplementary material, table S2). (b) Expression in the three tissues, where the gained genes are differentially expressed [24]. Dots show Transcripts Per Million (TPM) measured in individual mice, while the horizontal bars show the average of the four mice. Expression is shown in green for the t-specific copy and in black for the paralogues on the other chromosomes. (c) Phylogenetic tree estimated by PAML based on the sequence alignment of Ppp1cb. The ratio of non-synonymous and synonymous substitution rates, dN/dS, was estimated for each branch separately, as this model was superior to one with shared dN/dS (p < 0.0001, likelihood ratio test). dN/dS values are shown above each branch. (Online version in colour.)