UC Berkeley

In organisms that determine sex using specialized sex chromosomes, chromosome-wide dosage compensation is required to equalize gene expression between the sexes. In each well-studied dosage compensation system, factors are targeted to the sex chromosomes of only one sex to modulate sex-chromosome expression. In the nematode Caenorhabditis elegans, the dosage compensation complex (DCC) binds both X chromosomes of XX hermaphrodites and reduces X-linked transcript levels by approximately half to equal those of the XO male. My research addresses the mechanisms by which the DCC recognizes X chromosomes and controls transcription once bound. Previous studies determined that the DCC is recruited to the X chromosomes through rex sites that recruit the DCC autonomously. rex sites contain high-scoring matches to several DNA sequence elements that are required for full binding of the DCC, including the motif enriched on X (MEX). In addition to binding to rex sites, the DCC binds to dox sites that reside in the promoters of active genes and are bound only when attached to the X chromosome. In this dissertation, I report advances in understanding how the DCC is targeted to the X chromosome and how it regulates transcription of X-linked genes once it is bound.

In chapter 2, I describe experiments that redefine our understanding of DCC composition, binding and function, performed in collaboration with another graduate student. The protein DPY-30 is shown to be a component of both the repressive DCC and the activating MLL/COMPASS complex. We define fundamental differences in the molecular requirements for DCC binding to rex and dox sites, and describe the relationship between DCC binding at these two classes of sites. Finally, we show that C. elegans dosage compensation controls X-linked transcript levels by regulating transcription.

In chapter 3, I describe work performed in collaboration with John Lis's laboratory at Cornell University using Global Run-On Sequencing (GRO-seq) to determine the step of transcription controlled by the DCC. The mechanism by which the DCC controls transcription has been elusive due, in part, to improperly annotated transcription start sites (TSSs). Co-transcriptional trans-splicing removes the true TSSs for the majority of genes. To enhance our dosage compensation studies, I first annotated TSSs genome-wide using a novel technique to map nascent 5'-capped RNA. This technique, when combined with the requirement of uninterrupted GRO-seq signal between previously annotated TSS and the 5'-cap, permitted unambiguous TSS annotation and showed that TSSs can be up to 14 kb upstream of previously annotated TSSs. Correct gene annotation permits the identification of nematode-specific features of transcription and improves the tractability of C. elegans as a system for studying transcription regulation. After re-annotating TSSs, I compared the location of transcriptionally engaged RNA Polymerase II in wild type and DCC mutant animals. I discovered that dosage compensation mutants show a uniform increase in transcription across the entire gene, indicating that the DCC reduces recruitment of RNA Polymerase II to X-linked genes of hermaphrodites to equalize expression with that of males.

In chapter 4, I report ChIP-seq experiments that map DCC binding and improve the classification of DCC binding sites. Several DNA sequence motifs had been previously shown to be important for DCC binding to rex sites, however because many rex sites lacked known motifs we hypothesized that not all sequence elements involved in DCC recruitment had been identified. These ChIP-seq experiments permitted the identification of additional DNA sequence motifs that are required for full DCC binding to rex sites. Not only are these motifs enriched on the X chromosome, but they also cluster closely together with previously identified motifs in many rex sites. I show, in vivo, that this clustering is required for full DCC binding, suggesting that, within some rex sites, the DCC binds cooperatively to the different sequence motifs. Therefore, the X-specificity of DCC assembly results from strong binding to motifs enriched on the X chromosome and cooperative binding to clustered motifs.