Evolution of transcriptional inactivation on sex chromosomes in birds and mammals

Abstract

Differentiated sex chromosomes (XY or ZW) result in a 2:1 gene dosage imbalance between the X or Z and the autosomes, in the heterogametic sex, because genes have been lost from the usually small and heterochromatic sex-specific element (Y or W). This gene loss also results in a 2:1 dosage difference between the sexes for unpartnered genes on the X or Z. Such dosage imbalances may be compensated for by various stratagies in different taxa. In this thesis I examine dosage compensation of the X chromosomes in a monotreme mammal, and the Z chromosome of a representative bird. In therian mammals (eutherians and marsupials), one X chromosome is inactivated in the somatic cells of females, leaving both sexes with a single upregulated X. The process of X chromosome inactivation (XCI) in eutherian mammals is tightly controlled. In contrast, genes on the orthologous (although independently evolved) platypus X and chicken Z chromosomes are expressed globally at a higher level in the homogametic sex, and there is no apparent chromosome-wide inactivation system on the platypus X or the chicken Z chromosomes. I used RNA-FISH to demonstrate, at a cellular level, inactivation of genes on the chicken Z, and on the multiple X chromosomes of platypus. Thus, transcriptional silencing of one X/Z allele in the homogametic sex is at least partly responsible for to the partial dosage compensation systems in platypus and chicken. To understand how inactivation of Z/X-linked genes is achieved in chicken and platypus, I investigated epigenetic factors that are involved in X chromosome inactivation in eutherian and marsupial mammals, which included DNA methylation and four histone modifications. I examined genome-wide sex-specific DNA methylation in chicken, and discovered that males have increased DNA methylation 2kb 5' of transcription start sites of genes in the valley region (which is thought to bear an increased frequency of dosage compensated genes), indicative of a role for DNA methylation in silencing Z-linked genes. I also explored histone modifications associated with the eutherian Xi and constitutive heterochromatin on the platypus X chromosomes. I found H3K27me3 and H3K9me3 to be enriched on X5 and X3 respectively in females, suggesting a role in transcriptional silencing. Both modifications are also enriched on the marsupial Xi, but H3K9me3 is not enriched on the eutherian Xi, and is instead found on constitutive heterochromatin. My results suggest that H3K27me3 is involved in transcriptional silencing on independently evolved Xs in mammals. Additionally, I observed an enrichment of RNA polymerase II on the X chromosomes in males, providing evidence for transcriptional upregulation. The results from my studies suggest that probabilistic silencing of the X or Z chromosome in the homogametic sex was exapted independently in therian mammals, monotreme mammals and birds, from a common epigenetic toolbox, as an early step in the evolution of sex chromosome inactivation. This convergent evolution was perhaps constrained by limited solutions for the common problem of sex chromosome gene dosage imbalance that resulted from a degrading Y, or W.