On Friday, February 15th, The BCMB Friday Seminar series hosted another faculty member of BCMB. This month, Gerald Hart of the Department of Biological Chemistry gave a talk entitled, “Linking nutrients to signaling and transcription: Roles of O-GlcNAcylation in diabetes, neurodegeneration, and cancer.”
Hart was introduced by a student in the Biological Chemistry department, Teresa Romeo-Luperchio. Luperchio gave a great run-down of Hart’s history as a scientist, including the fact that he used to have a lab in his bedroom when he was a child, complete with a Bunsen burner! Hart also founded the journal “Glycobiology.”
Hart began his talk by thanking BCMB program director Carolyn Machamer for organizing the seminar series, and for the opportunity to find out what his colleagues were doing.
Hart stated that the discovery of O-linked N-acetylglucosamine, better known as O-GlcNAc, occurred almost thirty years ago. Scientists thought that glycans were added in the lumen of the ER and Golgi, and were mostly on protein domains destined for outside of the cell. However, with the discovery of O-GlcNAc, they realized that glycosylation was occurring in the cytoplasm and nucleus as well. There have been approximately 1700 publications about O-GlcNAc and related processes since the 1980s.
O-GlcNAc is a small monomeric sugar that is not elongated. Between 2-5% of glucose in the cell is used to make its precursor molecule, UDP-GlcNAc. The modification occurs in metazoans, some bacteria, fungi, plates, and viruses. UDP-GlcNAc sits at a major node of metabolism: amino acid, fatty acid, and nucleotide synthesis all utilize this precursor molecule.
In mammals, O-GlcNAc is most highly abundant in the beta cells of the pancreas. It can also be found in the brain, liver and other tissues. O-GlcNAcylation exhibits cross-talk with phosphorylation (it modifies the same serine and threonine amino acid residues). It has been shown to play a role in circadian clock regulation, diabetes, transcription, and proteasome regulation. The enzyme that adds the O-GlcNAc modification, O-GlcNAc transferase (OGT) has been shown to be an essential gene.
Hart’s lecture consisted of four main parts:
1. O-GlcNAc and transcription
2. Cycling enzymes
3. Crosstalk with phosphorylation
4. Hyper O-GlcNAcylation of mitochondrial proteins, which is believed to increase ROS production and contribute to diabetes
O-GlcNAc and Transcription
O-GlcNAcylation is important in transcription. It is distributed along the chromosomes, and is most abundant at sites of active gene transcription. The C-terminal domain of RNA Polymerase II is heavily glycosylated by O-GlcNAc. The presence of the sugar modification prevents phosphorylation from occurring on the same residues. Pre-initiation complexes require the O-GlcNAc modifications, which are then removed to allow for phosphorylation and for the process of elongation. Hart discussed work their lab did in collaboration with Brian Lewis (NIH) that showed significant overlap between promoter-localized RNA Pol II and O-GlcNAcylation of the C-terminal domain of the polymerase. These modifications clustered around the transcriptional start site, and the enzymes that add and remove O-GlcNAc localized to the site as well.
Their study also showed that knockdown of OGT by shRNA could reduce levels of this essential enzyme by 50-60%, and that the outcome of this reduction was less overall transcription. Using the adenovirus E3 promoter in vitro, they demonstrated that this decrease was due to the inability of O-GlcNAc and phosphorylation modifications to cycle: the pre-initiation complex required the sugar modification to form, but the elongating ribosome complexes only required phosphorylation. The results of this study and several others were summarized in the Nature Reviews – Cancer journal in 2011.
Hart mentioned several other transcriptional networks that involved O-GlcNAc, such as the signaling pathway containing CREB and p53. He also said that histone proteins are O-GlcNAcylated, and that the modification is part of the histone code, occurring where DNA binds to histones.
Hart discussed some of the structural information that is known about the OGT enzyme and O-GlcNAcase enzyme (responsible for removing the sugar modification). Unlike phosphorylation, only one enzyme has been found to add O-GlcNAc, likewise, only one found to remove it. These enzymes have unique domains and features which assist in their function.
OGT is mostly present in the nucleus, and is a highly conserved enzyme. It is regulated by UDP-GlcNAc levels (the O-GlcNAc precursor), and has been shown to complex with phosphatase, presumably to add O-GlcNAc upon removal of a phosphorylation modification.
O-GlcNAcase looks structurally similar to histone acetyltransferase enzymes, but it doesn’t have this function. It contains a caspase 3 cleavage site, and is present mainly in the cytoplasm.
Hart discussed some recently developed mass spectrometry methods that could look at reciprocal phosphorylation and O-GlcNAcylation. By using iTRAC MS, they were able to quantitatively compare modifications of proteins and calculate a relative occupancy ratio for these modifications. Using NIH 3T3 cells, they showed that almost all sites wich had cycling phosphorylation and O-GlcNAcylation could be affected by elevated O-GlcNAc levels. They also found, surprisingly, that between one-third to one-half of all human kinases are modified by O-GlcNAc.
Collaborating with Heng Zhu, they probed a library of human kinases to identify the exact O-GlcNAc modifications present on these proteins. As an example of the type of data gained from this study, they looked at AMPK IV. This particular kinase was modified by O-GlcNAc on serine 189, while the major phosphorylation site was on threonine 200. There was reciprocal regulation of the kinase activity due to these modifications, leading Hart and his group to propose that they act as a “safety switch” to prevent activation of this kinase. They also showed that casein kinase II had different target specificity when it was O-GlcNAcylated than when it was phosphorylated.
Hart concluded his talk by presenting new work that he believes can explain the mechanism behind Type-II diabetes. He noted that conditions of high glucose caused mitochondria to produce reactive oxygen species (ROS), and that the mechanism behind this was due to O-GlcNAcylation of the mitochondrial proteome in the presence of high glucose. His group showed that O-GlcNAc and OGT are both present and elevated in diabetic mice, and that OGT actually has a mitochondrial-specific splice variant.
Using copper-free “click chemistry” and a reductive cleavable affinity tag coupled with mass spectrometry, they showed that having too much glucose specifically elevated O-GlcNAcylation in mitochondria, leading to a dramatic increase in ROS production. The levels of mitochodrial-specific OGT went up dramatically in mitochondria from diabetic cells. Taken together this evidence suggests a molecular link between hyperglycemia and defective mitochondria.
Hart gave some final remarks to the young scientists in the audience that O-GlcNAc research was wide-open, and had many interesting aspects to be explored. Besides the link to diabetes, O-GlcNAcylation is elevated in nearly all cancers, and likely plays a role in the molecular process leading to cancer.
Hart gave a “nod to the fantastic BCMB students” who helped accomplish the body of work he presented in his talk. He joked that he knows “you think we all work you hard, but science is fun so it’s not really work!”