The Conflict within and the Escalating War between the Sex Chromosomes

Selfish genetic elements that distort Mendelian segregation to favor their own transmission are common in eukaryotic genomes [1,2]. Segregation distortion can reduce whole organism fitness, resulting in strong counter selection for genes that suppress distorters. Such intragenomic conflicts have the potential to drive recurrent bouts of antagonistic co-evolution [3]. Theory predicts that genetic conflicts should be particularly intense between the sex chromosomes [4,5]. The expectation that sex-linked conflict should be rampant has led to a renewed emphasis on the importance of antagonistic coevolution for driving genome evolution [6]. However, while numerous examples of genes involved in intragenomic conflict now exist [1], evidence for antagonistic coevolution between the mammalian X and Y chromosomes has remained elusive. In this issue of PLOS Genetics, Cocquet et al. have demonstrated a genetic basis for X–Y conflict acting during a crucial stage of mouse spermatogenesis [7]. The sex chromosomes are silenced via chromatin remodeling during the initiation of meiosis (meiotic sex chromosome inactivation or MSCI) [8]. Gene silencing persists through the remainder of spermatogenesis (postmeiotic sex chromatin or PMSC), save for a subset of genes that escape inactivation [9]. Considerable progress has been made recently on the epigenetic regulation of MSCI and PMSC, including the identification of a multicopy Y-linked gene, Sly, involved in the maintenance of PMSC [10]. Male mice with Sly deficiency show up-regulation of several sex-linked genes during PMSC, are sub-fertile, and produce female-biased litters. Thus, Sly is a strong candidate for being involved in X– Y conflict due to its repressive interaction with other genes and its potential to bias sex chromosome transmission. Intriguingly, there are two X-linked genes (Slx and Slxl1; hereafter Slx/Slxl1) that are closely related to Sly. Both are regulated by Sly, occur in large multicopy clusters on the X, and are crucial for spermatogenesis [11]. To test for genetic conflict between these genes, Cocquet et al. generated transgenic mice expressing short hairpin RNA (shRNA) that knockdown Sly or Slx/ Slxl1 transcript levels without completely knocking out gene function [7]. Both Slyand Slx/Slxl1-deficient mice showed impaired spermatogenesis, but Slx/Slxl1 deficiency led to a slight reduction in sexlinked gene expression in postmeiotic cells and male-biased litters (Figure 1A). Strikingly, mice deficient for both Sly and Slx/ Slxl1I showed a complete rescue of XY expression, male fertility, and sex ratio phenotypes. That is, the genes have antagonistic roles during spermatogenesis: Sly represses XY expression during PMSC and promotes the transmission of the Y, while Slx/Slxl1 activates XY expression and promotes the transmission of the X. The surprising conclusion is that antagonism depends on the relative expression of these genes and not their total abundance. Several questions remain regarding the mechanistic and genetic bases of distortion. For example, segregation distortion in the Sly-Slx/Slxl1 system appears to be caused by the differential fertilization ability of Xand Y-bearing sperm. Distorter genes often skew transmission through epistatic interactions with one or more responder genes [12]. In this context both Sly and Slx/Slxl1 appear to be distorters acting on one or more responder genes to impair the function of the Xor Y-bearing sperm, respectively [7]. Which raises the question, what are the responders? Even more interesting are the evolutionary consequences of recurrent sexlinked conflict. If Sly and Slx/Slxl1I were locked in an antagonistic conflict, then we would predict that each would be rapidly evolving on some level. The relevant metric here seems to be gene copy number. Sly and Slx/Slxl1I are recent additions to the mouse genome, appearing within the past 3 million years (Figure 1B). Since that time they have rapidly expanded in some, but not all, lineages [13]. Why? Is genetic conflict more intense in some species? Or is the antagonistic interaction a consequence of novel functions that have evolved more recently? The Mus musculus X is enriched for dozens of other multicopy gene families expressed primarily in postmeiotic cells, which is thought to be a mechanism for escaping PMSC [14]. This interpretation now appears to be correct, with the added caveat that the entire process may be a side effect of genetic conflict between Sly and Slx/Slxl1I. Most of these X-linked amplicons are repressed by Sly during PMSC. Thus, the rapid expansion of Sly— driven by conflict with Slx/Slxl1I—may in turn drive compensatory expansion of other sex-linked genes in order to maintain proper expression levels [13]. One important consequence of recurrent sex-linked conflict is its potential to drive speciation [6]. Several of the mice presented in Figure 1B can hybridize, often resulting in hybrid male sterility (HMS). In particular, some reciprocal crosses between M. m. musculus and M. m. domesticus yield asymmetric HMS; males are only sterile when a M. m. musculus female is crossed with an M. m. domesticus male. Moreover, sterile males show widespread over-expression of the X, presumably due to a failure of MSCI and/or PMSC [15]. Cocquet et al. [7] propose that interactions between Sly and Slx/