O-GlcNAcase: promiscuous hexosaminidase or key regulator of O-GlcNAc signaling?
Bottom Line: O-GlcNAc signaling is regulated by an opposing pair of enzymes: O-GlcNAc transferase installs and O-GlcNAcase (OGA) removes the modification from proteins.The dynamics and regulation of this process are only beginning to be understood as the physiological functions of both enzymes are being probed using genetic and pharmacological approaches.We identify several areas of "known unknowns" that would benefit from future research, such as the enigmatic C-terminal domain of OGA.
Affiliation: From the Medical Research Council Protein Phosphorylation and Ubiquitylation Unit and.Show MeSH
Mentions: The O-GlcNAc hydrolase OGA was first purified from rat spleen by Dong and Hart in 1994 (15), a decade after the discovery of the reversible modification of proteins with O-GlcNAc. A 1998 study identified it as an antigen expressed by meningiomas and named the gene MGEA5 (16). The sera of meningioma patients showed immunoreactivity against two apparent alternative splicing isoforms, and initial sequence analyses and biochemical assays suggested hyaluronidase activity (16). However, in 2001, Gao et al. (17) performed a thorough biochemical characterization of the recombinantly produced human enzyme, establishing the neutral pH optimum and selectivity for GlcNAc over GalNAc that distinguishes OGA from the lysosomal hexosaminidases HexA and HexB. The existence of an additional hexosaminidase activity (HexC) in mammals had been known since the 1970s (18–20) and was attributed to OGA. Consistent with their subcellular localization, the lysosomal hexosaminidases have an acidic pH optimum, and they cleave terminal O-linked β-d-N-acetylhexosamine residues irrespective of the C4 configuration (GlcNAc and GalNAc). Their biological role is the degradation of glycans on proteins and lipids, and their dysfunction is associated with lysosomal storage disorders, particularly Tay-Sachs and Sandhoff diseases (21). Despite their distinct physiological functions, the existence of these mechanistically related enzymes pertains to O-GlcNAc research, as many frequently used small molecule inhibitors of OGA (most notably PUGNAc) possess little specificity and simultaneously inhibit the lysosomal hexosaminidases. Sequence similarities reveal that OGA belongs to glycoside hydrolase family 84 (GH84) in the CAZy Database (22). A single gene encodes OGA, giving rise to two main isoforms in vertebrates (23) and a single isoform in lower eukaryotes, e.g. Drosophila (24) or Caenorhabditis (25). The shorter isoform lacks the C-terminal domain but retains some OGA activity (26) and is reported to be localized in the nucleus (27) and in lipid droplets (28). Full-length OGA is a predominantly cytoplasmic and nuclear enzyme (29). In addition to the N-terminal GH84 domain, OGA was observed to possess another domain with potential catalytic activity that belongs to the family of GCN5-related acetyltransferases (Fig. 1) (30). One laboratory has reported that the OGA C terminus possesses histone acetyltransferase (HAT) activity in semipure fractions (27, 31). OGA and HAT activities were speculated to act synergistically, opening up the chromatin structure (acetylation) and activating transcription factors through removal of O-GlcNAc (27). However, later studies (32–34) failed to reproduce these observations. Although OGA is post-translationally modified (as reviewed in Ref. 35), the function of these modifications is largely unknown. Similarly, although O-GlcNAc levels and OGA/OGT transcription are tightly linked, the mechanisms underlying this are unexplored.
Affiliation: From the Medical Research Council Protein Phosphorylation and Ubiquitylation Unit and.