The nearly neutral theory of molecular evolution posits variation among species in the effectiveness of selection. In an idealized model, the census population size determines both this minimum magnitude of the selection coefficient required for deleterious variants to be reliably purged, and the amount of neutral diversity. Empirically, an “effective population size” is often estimated from the amount of putatively neutral genetic diversity, and is assumed to also capture a species’ effectiveness of selection. The degree to which selection maintains preferred codons has the potential to more directly quantify the effectiveness of selection. However, past metrics that compare codon bias across species are confounded by among-species variation in %GC content and/or amino acid composition. Here we propose a new Codon Adaptation Index of Species (CAIS) that corrects for both confounders. Unlike previous metrics of codon bias, CAIS yields the expected relationship with adult vertebrate body mass. We demonstrate the use of CAIS correlations to show that the protein domains of more highly adapted vertebrate species evolve higher intrinsic structural disorder.

The mammalian nuclear hormone receptors LRH1 (NR5A2) and SF1 (NR5A1) are close paralogs that can bind the same DNA motif and play crucial roles in gonadal development and function. Lrh1 is essential for follicle development in the ovary and has been proposed to regulate steroidogenesis in the testis. Lrh1 expression in the testis is highly elevated by loss of the sex regulator Dmrt1, which triggers male-to-female transdifferentiation of Sertoli cells. While Sf1 has a well-defined and crucial role in testis development, no function for Lrh1 in the male gonad has been reported. Here we use conditional genetics to examine Lrh1 requirements both in gonadal cell fate reprogramming and in normal development of the three major cell lineages of the mouse testis. We find that loss of Lrh1 suppresses sexual transdifferentiation, confirming that Lrh1 can act as a key driver in reprogramming sexual cell fate. In otherwise wild-type testes, we find that Lrh1 is dispensable in Leydig cells but is required in Sertoli cells for their proliferation, for seminiferous tubule morphogenesis, for maintenance of the blood-testis barrier, for feedback regulation of androgen production, and for support of spermatogenesis. Expression profiling identified misexpressed genes likely underlying most aspects of the Sertoli cell phenotype. In the germ line we found that Lrh1 is required for maintenance of functional spermatogonia, and hence mutants progressively lose spermatogenesis. Reduced expression of the RNA binding factor Nxf2 likely contributes to the SSC defect. Unexpectedly, however, over time the Lrh1 mutant germ line recovered abundant spermatogenesis and fertility. This finding indicates that severe germ line depletion triggers a response allowing mutant spermatogonia to recover the ability to undergo complete spermatogenesis. Our results demonstrate that Lrh1, like Sf1, is an essential regulator of testis development and function but has a very distinct repertoire of functions.

Sex determination in mammals hinges on a cell fate decision in the fetal bipotential gonad between formation of male Sertoli cells or female granulosa cells. While this decision normally is permanent, loss of key cell fate regulators such as the transcription factors Dmrt1 and Foxl2 can cause postnatal transdifferentiation from Sertoli to granulosa-like (Dmrt1) or vice versa (Foxl2). Here we examine the mechanism of male-to-female transdifferentiation in mice carrying either a null mutation of Dmrt1 or a point mutation, R111G, that alters the DNA binding motif and causes human XY gonadal dysgenesis and sex reversal. We first define genes misexpressed during transdifferentiation and then show that female transcriptional regulators driving transdifferentiation in the mutant XY gonad (ESR2, LRH1, FOXL2) bind chromatin sites related to those normally bound in the XX ovary. We next define gene expression changes and abnormal chromatin compartments at the onset of transdifferentiation that may help destabilize cell fate and initiate the transdifferentiation process. We model the R111G mutation in mice and show that it causes dominant gonadal dysgenesis, analogous to its human phenotype but less severe. We show that R111G partially feminizes the testicular transcriptome and causes dominant disruption of DMRT1 binding specificity in vivo. These data help illuminate how transdifferentiation occurs when sexual cell fate maintenance is disrupted and identify chromatin sites and transcripts that may play key roles in the transdifferentiation process.