O -linked N -acetylglucosamine transferase (OGT) reversibly modifies serine and threonine residues of many intracellular proteins with a single β- O -linked N -acetylglucosamine residue ( O -GlcNAc), and has been implicated in insulin signaling, neurodegenerative disease, cellular stress response, and other important processes in mammals. OGT also glycosylates RNA polymerase II and various transcription factors, which suggests that it might be directly involved in transcriptional regulation. We report here that the Drosophila OGT is encoded by the Polycomb group (PcG) gene, super sex combs ( sxc ). Furthermore, major sites of O -GlcNAc modification on polytene chromosomes correspond to PcG protein binding sites. Our results thus suggest a direct role for O -linked glycosylation by OGT in PcG-mediated epigenetic gene silencing, which is important in developmental regulation, stem cell maintenance, genomic imprinting, and cancer. In addition, we observe rescue of sxc lethality by a human Ogt cDNA transgene; thus Drosophila may provide an ideal model to study important functional roles of OGT in mammals.

A large portion of the Drosophila melanogaster genome is contained within heterochromatic regions of chromosomes, predominantly at centromeres and telomeres. The remaining euchromatic portions of the genome have been extensively characterized with respect to gene organization, function and regulation. However, it has been difficult to derive similar data for sequences within centromeric (centric) heterochromatin because these regions have not been as amenable to analysis by standard genetic and molecular tools. Here we present an updated genetic and molecular analysis of chromosome 3L centric heterochromatin (3L Het). We have generated and characterized a number of new, overlapping deficiencies (Dfs) which remove regions of 3L Het. These Dfs were critically important reagents in our subsequent genetic analysis for the isolation and characterization of lethal point mutations in the region. The assignment of these mutations to genetically-defined essential loci was followed by matching them to gene models derived from genome sequence data: this was done by using molecular mapping plus sequence analysis of mutant alleles, thereby aligning genetic and physical maps of the region. We also identified putative essential gene sequences in 3L Het by using RNA interference to target candidate gene sequences. We report that at least 25, or just under 2/3 of loci in 3L Het, are essential for viability and/or fertility. This work contributes to the functional annotation of centric heterochromatin in Drosophila, and the genetic and molecular tools generated should help to provide important insights into the organization and functions of gene sequences in 3L Het.