Glucose Deprivation Stimulates O-GlcNAc Modification of Proteins through Up-regulation of O-Linked N-Acetylglucosaminyltransferase*

O-Linked N-acetylglucosamine (O-GlcNAc) is a post-translational modification of proteins that functions as a nutrient sensing mechanism. Here we report on regulation of O-GlcNAcylation over a broad range of glucose concentrations. We have discovered a significant induction of O-GlcNAc modification of a limited number of proteins under conditions of glucose deprivation. Beginning 12 h after treatment, glucose-deprived human hepatocellular carcinoma (HepG2) cells demonstrate a 7.8-fold increase in total O-GlcNAc modification compared with cells cultured in normal glucose (5 mm; p = 0.008). Some of the targets of glucose deprivation-induced O-GlcNAcylation are distinct from those modified in response to high glucose (20 mm) or glucosamine (10 mm) treatment, suggesting differential targeting with glucose deprivation and glucose excess. O-GlcNAcylation of glycogen synthase is significantly increased with glucose deprivation, and this O-GlcNAc increase contributes to a 60% decrease (p = 0.004) in glycogen synthase activity. Increased O-GlcNAc modification is not mediated by increased UDP-GlcNAc, the rate-limiting substrate for O-GlcNAcylation. Rather, the mRNA for nucleocytoplasmic O-linked N-acetylglucosaminyltransferase (OGT) increases 3.4-fold within 6 h of glucose deprivation (p = 0.006). Within 12 h, OGT protein increases 1.7-fold (p = 0.01) compared with normal glucose-treated cells. In addition, 12-h glucose deprivation leads to a 49% decrease in O-GlcNAcase protein levels (p = 0.03). We conclude that increased O-GlcNAc modification stimulated by glucose deprivation results from increased OGT and decreased O-GlcNAcase levels and that these changes affect cell metabolism, thus inactivating glycogen synthase.

Growth, Treatment, and Extraction of HepG2 Cells-HepG2 cells were grown in 10 ml of Dulbecco's modified Eagle's medium containing 20 mM glucose, 10% fetal calf serum, 100 units/ml penicillin G sodium, and 100 g/ml streptomycin sulfate in 10-cm plates (Corning Glass) at 37°C in 5% CO 2 . The medium was replaced 1 day prior to experimental treatment initiation; media glucose concentrations at treatment initiation averaged 10 mM. Experimental treatments were initiated once cells reached 70% confluence. We found 70% confluence to be optimal for promoting the glucose deprivation effect; underand over-confluent cells demonstrated a diminished glucose deprivation effect. Experimental treatment of each plate comprised 10 ml of glucose-free Dulbecco's modified Eagle's medium, 1% fetal calf serum, 1 mM sodium pyruvate, 4 mM L-glutamine, and 0 -40 mM glucose. Glucosamine treatments included glucose-free Dulbecco's modified Eagle's medium, 1% fetal calf serum, 1 mM sodium pyruvate, 4 mM L-glutamine, 2.5 mM glucose, and 10 mM D-glucosamine. Because media glucose concentrations depleted significantly over time in pilot experiments, media glucose concentrations were assayed every 3 h (using the Beckman Glucose Analyzer II), and glucose was replenished to achieve consistent glucose concentrations throughout treatment. Experimental treatment lasted 0 -24 h. No cell death was observed for any of the treatment durations. For protein extracts, plates were placed on ice and washed twice with ice-cold Krebs-Ringer bicarbonate HEPES buffer (25 mM HEPES, pH 7.4, 150 mM sodium chloride, 4.4 mM potassium chloride, 1.2 mM sodium phosphate, pH 7.4, 1 mM magnesium chloride, and 1.9 mM calcium chloride), and then the cells were harvested in 0.75 ml of extraction buffer (50 mM HEPES, pH 7.4, 100 mM sodium chloride, 5% (v/v) glycerol, 50 M PUGNAc, and protease inhibitors). The resulting cell suspension was sonicated with a Sonic Dismembrator F60 for 15 s at setting 6 (Thermo) and centrifuged at 20,000 ϫ g for 2 min at 4°C. Supernatant aliquots were immediately frozen in liquid nitrogen. Cells whose lysates were subsequently digested with hexosaminidase were harvested as above, but in a mod-ified extraction buffer lacking PUGNAc and EDTA to prevent hexosaminidase inhibition. For cells used for RNA determination, the medium was aspirated/discarded, and 1 ml of TRI Reagent was immediately applied to the cells. Cells were scraped, disrupted by repeated pipetting, and immediately frozen in liquid nitrogen.
Western Blotting-Protein concentrations of HepG2 lysates were determined using Bio-Rad protein reagent. Lysates were prepared for gel electrophoresis by dilution with extraction buffer and 5ϫ Laemmli buffer. 10 g of protein was added to each lane. SDS-PAGE was conducted using the Bio-Rad Mini-PROTEAN 3 electrophoresis cell, and resolved proteins were transferred to an Immobilon-P SQ transfer membrane (Millipore Corp., Bedford, MA). Resulting blots were blocked with TBST (20 mM Tris, pH 7.4, 150 mM sodium chloride, and 0.5% Tween 20) containing 4% (w/v) nonfat dried milk for 1 h at room temperature or overnight at 4°C. 4% (w/v) bovine serum albumin was used in lieu of dried milk for detection with the anti-O-GlcNAc antibody. Blots were incubated with primary antibodies for 1 h at room temperature or overnight at 4°C, washed three times in TBST, and then incubated with the appropriate horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. The blots were washed five times in TBST and imaged by treatment with Super Signal West Dura reagents (Pierce) and by exposure to Classic Blue BX autoradiography film (Molecular Technologies, St. Louis, MO). Densitometry measurements were obtained using an Epson Perfection 3200 photo scanner and NIH Image version 1.62 software (rsb.info.nih.gov/ij). In all experiments, glyceraldehyde-3-phosphate dehydrogenase protein levels were used to normalize changes in protein/modification. Glyceraldehyde-3phosphate dehydrogenase protein levels were not affected by the various cell treatments of these studies (see Fig. 1, A and C; and Fig. 2C).
Immunoprecipitation of Glycogen Synthase-We followed the glycogen synthase antibody manufacturer's protocol for immunoprecipitation. Briefly, 300 l of cell lysate (600 g of protein) combined with glycogen synthase antibody (1:25) was rotated overnight at 4°C. Incubation with rabbit IgG served as the negative control. 20 l of Protein A/G PLUS-agarose beads (50% bead slurry) was added, and mixtures were rotated for 3 h at 4°C. Beads were pelleted by centrifugation (20,000 ϫ g for 30 s at 4°C), and the supernatant was discarded. Some supernatant was retained and run as a positive control. Beads were washed five times with 500 l of 1ϫ lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , and 1 g/ml leupeptin). Pellet was resuspended in 20 l of 3ϫ SDS sample buffer. Samples were boiled at 95°C for 5 min and loaded onto SDS-polyacrylamide gel (4 -15%). Western-blotted membranes were probed with ␣-O-GlcNAc (CTD110.6) and ␣-GS.
Quantitation of mRNA by Reverse Transcription-PCR-RNA was prepared from Ϫ70°C frozen TRI Reagent/cell suspensions according to the manufacturer's protocol (Molecular Research Center, Inc.) and dissolved in water. RNA concentrations were measured spectrophotometrically. First-strand cDNA synthesis was carried out using 3 g of RNA, oligo(dT) primers (Invitrogen), and SuperScript III reverse transcriptase (Invitrogen) according to the manufacturer's protocol. Realtime PCR was performed with a rapid thermal cycler (Light-Cycler, Roche Diagnostics). Reactions (10 l) were performed using ϳ16 ng of cDNA as a template with 0.5 M each primer, 200 M each deoxynucleotide triphosphate, 50 mM Tris, pH 8.3, 500 g/ml nonacetylated bovine serum albumin (Sigma), 3.0 mM MgCl 2 , 0.04 units/l Platinum Taq DNA polymerase (Invitrogen), and a 1:30,000 dilution of SYBR Green I fluorescent dye (Molecular Probes, Eugene, OR). Primers based on human sequences were chosen using the Primer3 program: nucleocytoplasmic OGT, 5Ј-CTTTAGCACTCTGGCAAT-TAAACAG-3Ј and 5Ј-TCAAATAACATGCCTTGGCTTC-3Ј; mitochondrial OGT, 5Ј-TTTACCTCCTTTCCCTCCC-ATC-3Ј and 5Ј-CTGTCAAAAATGCGTGCCTCT-3Ј; all OGT isoforms, 5Ј-CTGCCCCAGAACCGTATCA-3Ј and 5Ј-TTCCAGACTTTGCCACGAACT-3Ј; O-GlcNAcase, 5Ј-AGC-CTTGAGTGGTGAGCCTA-3Ј and 5Ј-TCTGGGGATTTTG-ATTCAGC-3Ј; and NONO (non-POU domain-containing octamer-binding protein), 5Ј-CAAGTGGACCGCAACATCA-3Ј and 5Ј-CGCCGCATCTCTTCTTCAC-3Ј. We assayed the expression of six different potential normalizer genes and found that NONO expression was consistent across all cell treatments. Amplification occurred over 26 -45 four-step cycles, with a rate of temperature change between steps of 20°C/s. The steps were 95°C with a 0-s hold, 60°C with a 0-s hold, 72°C with an 11-s hold, and 80°C with a 1-s hold. Fluorescence was detected during the fourth step at a temperature determined previously to be below the melting temperature of the PCR products. After amplification, a melting curve was generated by slowly heating the double-stranded DNA product. Analysis of the postamplification melting curves confirmed the absence of nonspecific DNA products. For each amplification's fluorescence versus cycle line, the LightCycler software determined the second derivative maximum (the threshold cycle at which fluorescence clearly increased above background). Standard curves of log cDNA versus second derivative maximum (fractional cycle number) were constructed for each quantitated transcript and for the NONO normalization transcript, from cDNA mixtures comprising equal amounts of all cell treatment condition cDNAs. Standard curve points of 0, 6, 10, 16, 26, and 32 ng of combined cell cDNA were always included with the same PCR run with the entire set of individual cDNA amplifications of the same transcript. Results for each individual cDNA were normalized by dividing the relative amount of each transcript by the relative amount of NONO transcript from the same experiment. Within each experiment the same mixture was used, containing everything but the specific primers.
UDP-N-Acetylhexosamine Assay-Levels of UDP-N-acetylhexosamines (consisting of UDP-GlcNAc and UDP-GalNAc), products of the hexosamine biosynthesis pathway, were measured in cell extracts as described previously (11). Cell extracts were homogenized at 4°C in 4 volumes of perchloric acid (300 mM). The precipitates were centrifuged (10,000 ϫ g for 15 min at 4°C), and the lipid was extracted from the supernatants with 2 volumes of tri-n-octylamine:1,1,2-trichlorofluoroethane (1:4). The aqueous phase was stored at Ϫ80°C until analysis by HPLC. The extracts were filtered (0.45 m); HPLC was performed on a Partisil 10.5Ax column (25 cm x 4.6 mm, Waters Corp., Taunton, MA); and the extracts were eluted with a concave gradient from 5 mM potassium phosphate (pH 7.2) to 750 mM potassium phosphate (pH 7.2) over 48 min at a flow rate of 1 ml/min. UDP-N-acetylhexosamine levels were quantified by UV absorption at 254 nm and compared with external standards.
Glycogen Synthase Assay-The assay for glycogen synthase was performed as described previously (8). HepG2 lysate (7.5 g of protein) was incubated in a 100-l final volume with 100 mM HEPES, pH 7.4, 5 mM EDTA, pH 7.4, 0.8 mg of glycogen (type III from rabbit liver), 2 mM UDP-glucose, 10 l of glycerol, 0 or 10 mM glucose 6-phosphate, and 0.4 Ci of UDP-[6-3 H]glucose for 45 min at 37°C. The incubation was terminated by application to Whatman qualitative filter paper (No. 3MM, Maidstone, UK) and immersion in 60% (v/v) ethanol. After five washes in 400 ml of 60% ethanol, the paper squares were washed once in acetone, dried, and assayed for tritium. All assays were done in duplicate. The incorporation of tritium was found to be optimal at 37°C and linear for 120 min. Total GS activity was defined as the activity at maximal glucose 6-phosphate (10 mM).
Digestion of HepG2 Extracts with Hexosaminidase-Hexosaminidase digestions were conducted by treating cell lysates (50 g of protein) with 1 unit of N-acetylglucosaminidase from jack beans (Sigma) and 25 l of hexosaminidase buffer (20 mM sodium citrate, pH 4.5, 10% glycerol, 100 mM sodium chloride, and protease inhibitors) and incubating the preparation for 1 h at 30°C. Lysates not digested with hexosaminidase were protected from deglycosylation by the addition of PUGNAc (50 M); PUGNAc inhibits endogenous O-GlcNAcase. Immediately following digestion, glycogen synthase activation was measured as described above.
Statistics-Descriptive statistics are represented as mean Ϯ S.E. Each mean represents data from at least three independent experiments. Student's t test (two-tail) was used to compare differences between groups. compared with cells cultured in normal and high glucose, respectively (Fig. 1). Appreciable increases of O-GlcNAc with glucose deprivation first appear at 12 h and continue to increase through 18 and 24 h (Fig. 1C). Glucosamine treatment has been shown to stimulate O-GlcNAc modification even more potently than high glucose in a variety of systems (12). However, we observed 3-fold greater O-glycosylation with glucose deprivation than with glucosamine treatment (p ϭ 0.02; Fig. 1, A and B).

Glucose Deprivation of HepG2 Cells Stimulates O-GlcNAc
Although some protein targets demonstrate increased O-GlcNAc modification with both glucose deprivation and high glucose/glucosamine conditions, others are highly glycosylated exclusively with glucose deprivation. For two protein bands (Fig. 1A, bands b and c, ϳ250 and ϳ150 kDa, respectively), O-glycosylation is maximal with glucose deprivation and also increases with high glucose/GlcN treatment, with the lowest glycosylation levels at 1 mM glucose. Other proteins (Fig.  1A, band a, Ͼ250 kDa) fail to show glycosylation with high glucose/glucosamine treatment, but demonstrate robust glycosylation exclusively with glucose deprivation. Yet other targets (Fig. 1A, band d, ϳ75 kDa) demonstrate bimodal modification in which glycosylation is lowest at normal (5 mM) glucose and increases with both lower and higher glucose concentrations, suggesting that some targets are regulated by O-GlcNAc modification throughout the entire range of physiologic glucose concentrations.
Increased O-GlcNAcylation of GS in Glucose-deprived Cells Contributes to a Decrease in Activity-The O-GlcNAc modification induced by glucose deprivation is functionally significant. We have shown previously in adipocytes that O-Gl-cNAc modification of GS in conditions of high glucose decreases GS activity (8). This is also the case with glucose deprivation of HepG2 cells. HepG2 cells cultured in 0 mM glucose for 12 h exhibit a 60% decrease in maximal glycogen synthase activity compared with 5 mM glucose treatment (p ϭ 0.004; Fig.  2A). We observed no differences in GS activity among cells treated with 5 and 20 mM glucose or glucosamine (Fig. 2A). The decrease in activity observed with glucose deprivation corre- lates with a 675% increase in O-GlcNAc-modified GS levels (p ϭ 0.05; Fig. 2B) and a 330% increase in the amount of GS precipitated by sWGA (p ϭ 0.002; Fig. 2C), a lectin that specifically binds terminal GlcNAc. We observed no difference in glycogen synthase protein amounts across glucose treatments or phosphoglycogen synthase levels among treatments (Fig. 2, B  and D). We also detected no change in the K m of GS from glucose-deprived cells for its substrate UDP-glucose or its allosteric activator glucose-6-phosphate (data not shown).
To demonstrate directly that O-GlcNAc modification contributes to the observed decrease in GS activity, we treated cell lysates with hexosaminidase to reduce the levels of O-GlcNAc modification. This treatment resulted in a 40% rescue of total GS activity after glucose deprivation (p ϭ 0.003), whereas digestion of lysates from cells treated with normal glucose demonstrated no change in total activity (Fig. 2E). We observed a significant decrease in O-GlcNAc modification of proteins to background levels after hexosaminidase digestion, confirming deglycosylation (data not shown).

Increased O-GlcNAc with Glucose Deprivation Does Not Result from Increased HBP Flux-Previous reports have mainly attributed increases of cellular O-GlcNAc to increased HBP
flux and synthesis of the end product of the pathway, UDP-GlcNAc (1,13). This is not the case in glucose-deprived HepG2 cells. Cells cultured for 12 h in 0 mM glucose, which demonstrate a 7.8-fold increase in O-GlcNAc (Fig. 1, A and B), exhibit a 40% decrease in UDP-GlcNAc levels compared with normal glucose-treated cells (p ϭ 0.01). UDP-GlcNAc levels increased 236% (p ϭ 0.008) with high glucose and 616% (p ϭ 0.01) with glucosamine treatment compared with 0 mM glucose treatment (Fig. 3).
Nucleocytoplasmic OGT mRNA and OGT Protein Levels Are Increased in Glucose-deprived Cells-To explore the mechanisms for increased O-GlcNAc modification in the absence of increased HBP flux, we examined expression of OGT in conditions of glucose deprivation. Multiple OGT variants have been described (14 -16), the best characterized in human liver being the nucleocytoplasmic (nc) and mitochondrial (m) isoforms (16,17). ncOGT and mOGT isoforms are transcribed from a single OGT gene and are generated through alternative splicing and possibly by transcription initiation at a second internal promoter (17).
Increased OGT mRNA was found to correlate with and precede the increase in protein O-GlcNAc modification observed with glucose deprivation. By 6 h, a 3.4-fold increase in the ncOGT transcript was observed (p ϭ 0.006; Fig. 4A). By 12 h, with rabbit IgG confirmed specific GS precipitation (not shown). C, glucosedeprived cells exhibit a 3.3-fold increase in sWGA-bound/O-GlcNAc-modified glycogen synthase (p ϭ 0.002; n ϭ five independent determinations). Inset, sWGA-precipitated proteins immunoblotted with ␣-GS are shown. D, immunoblots of total GS and phospho-GS (pGS) levels demonstrate no change among treatments, suggesting that changes in glycogen synthase activity are not due to changes in GS protein or phospho-GS levels (representative blots of at least three independent determinations per treatment). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was probed as a loading control (n ϭ at least three independent determinations per treatment). E, glucose-deprived lysates digested with hexosaminidase demonstrate a 40% rescue of glycogen synthase total activity (0Ϫ versus 0ϩ; p ϭ 0.003; n ϭ six independent determinations). **, p Յ 0.001. A.I.U., arbitrary intensity units.
ncOGT mRNA levels had returned to levels seen in untreated cells. No change was observed in mOGT mRNA preceding O-GlcNAc induction (Fig. 4B). We tested primers directed to the conserved 3Ј-end of the OGT gene that detect all OGT transcripts. At 6 h, we observed an increase, although blunted, for total OGT; 0 mM glucose treatment promoted a 1.7-fold total OGT mRNA induction compared with 5 mM glucose treatment (p ϭ 0.001; Fig. 4C).
OGT protein levels were significantly increased following OGT transcriptional induction in glucose deprivation. At 12 h, we observed a 1.7-fold induction of OGT protein in glucosedeprived cells compared with normal glucose-treated cells (p ϭ 0.01; Fig. 5). OGT protein levels in high glucose-and glucosamine-treated cells were unchanged relative to normal glucosetreated cells.
O-GlcNAcase Protein Levels Are Significantly Decreased in Glucose-deprived Cells-Cellular O-GlcNAc levels are also determined by the activity of the enzyme O-GlcNAcase. We therefore determined whether changes in O-GlcNAcase might also contribute to the observed increase in protein O-GlcNAc modification with glucose deprivation. At 12 h, O-GlcNAcase protein levels were decreased 49% in glucose-deprived cells compared with cells cultured in 5 mM glucose (p ϭ 0.03). We observed no difference in O-GlcNAcase protein levels in 20 mM glucose-and glucosamine-treated cells compared with 5 mM glucose-treated cells (Fig. 6). We observed no difference in O-GlcNAcase transcript levels among all treatments (data not shown).

DISCUSSION
To date, much work on O-GlcNAc modification has focused on its contribution to metabolic regulation under hyperglycemic conditions. In this study we have shown that HepG2 cells under conditions of glucose deprivation demonstrate an even greater increase in O-GlcNAcylation of a number of proteins. This increase is not mediated by an increase in the rate-limiting substrate UDP-GlcNAc, as is the case in high glucose, but rather by induction of OGT and down-regulation of O-GlcNAcase.
The modification of proteins by O-GlcNAc in glucose deprivation may serve a role in metabolic regulation, part of a general response to conserve energy. For example, we show that GS activity is diminished with glucose deprivation, and this decrease is due in part to O-GlcNAcylation of GS. GS, the ratelimiting enzyme for glycogen synthesis, is highly expressed in liver and is integral to glucose homeostasis (18,19). GS activity is regulated through hormone-mediated phosphorylation and dephosphorylation (20) (21), but also by modification with Glc-NAc. In 3T3-L1 adipocytes and diabetic mice, O-GlcNAc modification of GS inhibits the enzyme in a manner analogous to phosphate, and this inhibition contributes to insulin resistance and the diabetic phenotype (8,9). O-GlcNAcylation of GS in  HepG2 cells were cultured for 0 -24 h in 0 or 5 mM glucose. Transcript levels of ncOGT, mOGT, and all OGT were measured by reverse transcription-PCR and normalized to NONO transcript levels. A, by 6 h, a 3.4-fold induction of ncOGT was observed in glucose-deprived extracts compared with normal glucose extracts (p ϭ 0.006). By 12 h, ncOGT levels had returned to those seen under normal glucose conditions. B, no change was observed for mOGT mRNA preceding O-GlcNAc induction. C, at 6 h, OGT mRNA levels for all isoforms combined were 1.7-fold greater with glucose deprivation than with 5 mM glucose treatment (p ϭ 0.001). Average mRNA levels are based on at least four independent determinations per treatment. *, p Յ 0.05; **, p Յ 0.001. glucose deprivation contributes, however, to the prevention of futile cycling of glycogenesis and glycogenolysis.
The mechanism for the decreased activity of O-Glc-NAcylated GS is not known. Post-translational modification of GS with phosphate and O-GlcNAc has been shown previously to alter the K m of the enzyme for UDP-glucose and glucose 6-phosphate. The O-GlcNAc-mediated decrease in GS total activity reported herein, namely a change in total activity (V max ) without a change in protein level or the K m for substrate or activators, is novel. Why this differs from the regulation of GS in high glucose conditions is not known. Our current data demonstrate differential protein targeting of O-GlcNAc in conditions of high glucose and glucose deprivation, and the same may be true of specific sites of O-GlcNAcylation within proteins. Alternatively, the differences may relate to the degree of glycosylation or to the tissue/cell type in which the glycosylation occurs.
It is well established that low energy states can also mediate down-regulation of anabolic reactions such as protein, fat, and complex carbohydrate synthesis through the AMP-activated kinase cascade (22). The proximal signal by which glucose deprivation signals the induction of OGT is not known, however. Preliminary experiments in which pyruvate failed to rescue the glucose deprivation O-GlcNAc response suggest that the signal for O-GlcNAc induction is not a general decrease in energy availability, but rather a specific response to decreased glucose availability. These results suggest that protein O-GlcNAcylation may therefore be an additional and distinct mechanism contributing to a general energy conservation response.
In addition to playing a role in metabolic adaptation to glucose deprivation, the observed increase in O-GlcNAcylation with glucose deprivation may also represent a stress response aimed at preserving protein structure and function. Recent publications implicate O-GlcNAc modification as a survival response to a variety of cellular stressors (23,24). Hydrogen peroxide, ultraviolet light, ethanol, NaCl, and thermal stress all lead to increased global levels of O-GlcNAcylation that have been shown to improve survival under stress (25)(26)(27). Increased O-GlcNAc modification under these conditions has been attributed in part to stress-induced increased glucose import and a resultant increase in HBP flux (23). It thus represents a response with similar effects on O-GlcNAc, but induction through a different mechanism, particularly as these studies were conducted over a different time frame (23). HSP70, a chaperone protein, is known to respond to cellular stress and has been described as an O-GlcNAc-specific lectin (24,28,29). The lectin binding activity changes with nutrient deprivation and other stressors (27)(28)(29). It is itself a target of O-Glc-NAcylation, although O-GlcNAc does not change the lectin activity of HSP70 (28,30). Manipulation of O-GlcNAc levels through the use of PUGNAc or diazooxonorleucine or through genetic ablation of OGT has been shown to correlate with HSP70 protein levels (23). Consistent with the protective effect of O-GlcNAc on proteins, increased O-GlcNAcylation of the 26 S proteasome has been found to inhibit proteasome-mediated protein degradation (31). Although no consistent consensus amino acid sequence has been identified for O-GlcNAc modification, many O-GlcNAc sites are "PEST" (Pro-Glu-Ser-Thr) sequences, which are known to render proteins more susceptible to proteasomal degradation (32). Modification of PEST sequences with O-GlcNAc is thought to slow protein degradation (33).TheseandotherfindingssuggestthatO-Glc-NAcylation plays a role in protecting its targets against degradation, protection that would benefit stressed, injured, or glucose-deprived cells.
We observe hyper-O-GlcNAcylation of some protein bands with both glucose deprivation and high glucose/GlcN treatment. However, hyper-O-GlcNAcylation of other bands occurs only with glucose deprivation. These findings suggest differential O-GlcNAc targeting. Little is known about OGT targeting, although OGT interaction with other proteins may contribute FIGURE 5. OGT protein is significantly increased in glucose-deprived cells. HepG2 cells were cultured for 12 h in 0, 5, or 20 mM glucose or 10 mM glucosamine. At 12 h, 0 mM glucose promotes an OGT protein increase of 1.7-fold compared with normal glucose conditions (p ϭ 0.01). OGT protein levels in high glucose-and glucosamine-treated cells are unchanged relative to normal glucose-treated cells. OGT bands from the same gel were rearranged for presentation purposes. Average OGT protein densities are based on at least three independent determinations per treatment. *, p Յ 0.05. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. to the regulation of O-GlcNAcylation. For example, OGT is known to occur in complex with protein phosphatase-1, and this interaction has been hypothesized to target OGT to serine and threonine residues that would also be modified by protein phosphatase-1 (34). The binding of OGT to binding partners is mediated largely by its tetratricopeptide repeat interactions with target proteins (35). Although the ncOGT and mOGT isoforms share a conserved C-terminal catalytic region, they differ in the number of N-terminal tetratricopeptide repeats as well as cellular localization targeting sequences (14,15,36,37). We have demonstrated ncOGT induction with glucose deprivation without concomitant mOGT induction, suggesting that differential regulation of OGT transcription, splicing, and transcript turnover may also function to regulate target specificity under varying conditions.
The current data demonstrate that metabolic regulation by O-GlcNAc is not simply a phenomenon seen in hyperglycemic states, but operates in glucose concentrations from below to above normal. This modulation of O-GlcNAc levels is achieved through a number of distinct mechanisms, including substrate flux through the HBP, transcriptional regulation of OGT, regulation of OGT enzymatic activity, post-translational regulation of O-GlcNAcase, and differential substrate targeting. The data on the distinct effects of O-GlcNAc on GS at low and high glucose also underscore the complexities of how protein function is modulated by O-GlcNAc. Combined with previous data demonstrating the effects of O-GlcNAc on subcellular targeting, survival, and intrinsic activity of a great variety of proteins, the current results are consistent with a central role for O-Glc-NAc in the adaptation of cells and tissues both to stress and to changing nutrient availability.