Cells of the body are decorated with a variety of carbohydrates (sugars) that serve many diverse functions. These sugars not only act as a protective barrier on the outside of the cell, but are also involved in cell adhesion, migration, communication and signaling events in many organisms. Our group studies one type of sugar addition to proteins, known as mucin-type O-linked glycosylation, which is initiated by the polypeptide GalNAc transferase (ppGalNAcT or PGANT) enzyme family. This sugar addition is seen in most eukaryotic organisms including mammals, fish, insects, worms and some types of fungi. The conservation of this protein modification across species suggests that it plays crucial roles during many aspects of development. It is known that there are as many as 20 family members encoding functional ppGalNAcTs in mammals. Given the size of the family and the complexity it generates, we sought an alternative, simpler model system to investigate the biological role of glycosylation. Analysis of the genome databases from other organisms indicated that the fruit fly (Drosophila melanogaster) had only 12 potential members and may therefore be a more tractable experimental system. Additionally, the fruit fly offers more sophisticated genetic techniques, shorter generation times and a wealth of well-characterized stocks on which to build future studies. ? ? We began these studies by cloning and characterizing the genes responsible for O-linked glycosylation in Drosophila. We demonstrated that there are at least 9 functional transferase genes in Drosophila (potentially 12 members total) and that at least one (pgant35A) is required for viability. These studies provided the first evidence that a member of this multigene family is required for development and viability in any eukaryote. Additionally, we defined the spatial and temporal patterns of expression of all the pgant family members throughout Drosophila development and elucidated the developmental profile of specific O-glycans. Of particular interest is the abundance of O-glycans along the presumptive apical and luminal regions of developing tubular structures in the fly. ? ? Examination of mutations in members of this glycosyltransferase family demonstrated that one gene is required at multiple distinct times during development. Specifically, pgant35A is required during embryogenesis; homozygous mutants devoid of wild type maternal RNA show abnormal tracheal tube formation and migration of secondary branches in developing embryos. This is particularly interesting given that the Drosophila tracheal system serves as a model for branching morphogenesis in many mammalian organ systems, including the salivary gland, lung, kidney and vasculature. We are now defining the subcellular changes that are taking place in pgant35A mutants. Electron microscopy has revealed significant changes in the apical and luminal surfaces of the tracheal system, with loss of certain portions of the secreted cuticle and disruption of the taenidial folds that line the inside of the tracheal tube. We are currently performing these studies in a cell culture system to visualize subcellular changes more readily.? ? In addition to pgant35A, we are also examining the developmental role of other members of this family. We have found that mutations in pgant3 alter epithelial cell adhesion in the Drosophila wing blade. A transposon insertion mutation in pgant3 or RNAi to pgant3 resulted in blistered wings, a phenotype characteristic of genes involved in integrin-mediated cell interactions. Precise excision of the tranposon restored pgant3 expression and wing integrity. Expression of wild type pgant3 in the mutant background also rescued the wing blistering phenotype, whereas expression of another family member did not, revealing a unique requirement for pgant3. pgant3 mutants displayed reduced O-glycosylation along the basal surface of wing imaginal discs, suggesting that reduced glycosylation of basal proteins in pgant3 mutants is responsible for disruption of adhesion in the adult wing blade. We have now identified one of the main O-glycosylated proteins in the wing disc using a combination of affinity purification, biochemistry and bioinformatics. This protein is specifically O-glycosylated in wild type wing discs but not in pgant3 mutants. Interestingly, this protein is known to mediate cell adhesion events during other stages of fly development as well. ? ? We are also investigating the role of each transferase using RNA interference (RNAi) in vivo to knockdown the transcript levels of each isoform. Expression of this dsRNA has recapitulated the phenotypes discussed above for pgant35A and pgant3 lending further support for the role of these genes in various aspects of epithelial tube formation and cell adhesion, respectively, as well as verifying the use of RNAi to specifically knockdown transferase gene expression in vivo. Interrogation of other isoforms by this approach indicates that additional pgant genes are required for viability. We are continuing to systematically analyze the consequences of loss of each pgant family member both in vivo as well as in cell culture.? ? Given that this family is evolutionarily conserved and that mammalian and fly orthologues display similar enzymatic activities and substrates preferences in vitro, we are also investigating the role of O-glycans in mammalian organ system development. We are currently interrogating ppGaNAcT expression patterns during embryonic salivary gland development using microarray data sets generated by Dr. Matthew Hoffman as part of the NIDCR Salivary Gland Initiative. We are assessing the expression levels of each of the 18 mouse ppGalNAcTs during various stages of gland development to catalogue when each gene is expressed. We are also examining whether each gene is expressed in epithelial and/or mesenchymal tissue, to assess what unique roles each enzyme may be playing in the development of this organ. Preliminary analyses indicate expression of certain isoforms at key stages of glandular development. Additionally, there are a number of isoforms expressed specifically in either the mesenchyme or epithelium. From here, the functional role of isoforms will be interrogated using siRNA in submandibular organ culture. Additionally, mice deficient in a number of ppGalNAcT family members already exist. Because these lines are homozygous viable, we are using the organ culture system to examine the development of the salivary glands from these knockout mice. ? ? In summary, we are using information gleaned from Drosophila to better focus on crucial aspects of development affected by O-glycosylation in more complex mammalian systems. Our hope is that the cumulative results of the studies described above will elucidate why O-linked glycosylation is necessary and what role sugars play in cellular communication and interactions occurring during eukaryotic development.
