Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance

A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance CMLS, Cell. Mol. Life Sci. 60 (2003) 222 – 228 1420-682X/03/020222-07 CMLS Cellular and Molecular Life Sciences © Birkhäuser Verlag, Basel, 2003 A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance L. Wells, K. Vosseller and G. W. Hart * Department of Biological Chemistry, Johns Hopkins School of Medicine, 517 Woods Basic Science Building, 725 N. Wolfe Street, Baltimore, Maryland 21205 (USA), e-mail: [email protected] Abstract. The ability to regulate energy balance at both which produces the acetylated aminosugar nucleotide the cellular and whole body level is an essential process uridine 5¢-diphospho-N-acetylglucosamine (UDP-Glc- of life. As western society has shifted to a higher caloric NAc) as its end product. Since UDP-GlcNAc is the donor diet and more sedentary lifestyle, the incidence of type 2 substrate for modification of nucleocytoplasmic proteins diabetes (non-insulin-dependent diabetes mellitus) has at serine and threonine residues with N-acetylglu- increased to epidemic proportions. Thus, type 2 diabetes cosamine (O-GlcNAc), the possibility of this posttransla- has been described as a disease of ‘chronic overnutrition’. tional modification serving as the nutrient sensor has There are abundant data to support the relationship be- been proposed. We have recently directly tested this tween nutrient availability and insulin action. However, model in adipocytes by examining the effect of elevated there have been multiple hypotheses and debates as to the levels of O-GlcNAc on insulin-stimulated glucose up- mechanism by which nutrient availability modulates in- take. In this review, we summarize the existing work that sulin signaling and how excess nutrients lead to insulin implicates the HSP and O-GlcNAc modif ication as nutri- resistance. One well-established pathway for nutrient ent sensors and regulators of insulin signaling. sensing is the hexosamine biosynthetic pathway (HSP), Key words. O-GlcNAc; glycosylation; posttranslational modification; hexosamine biosynthetic pathway; glucosamine; hyperglycemia; insulin resistance; type 2 diabetes. Type 2 diabetes and insulin resistance the b- cell and upregulating gluconeogenesis and glucose output in the liver will only be touched upon briefly in Chronic hyperglycemia is the hallmark of all types of di- this review and has been reviewed elsewhere [1 – 4]. In abetes (see [1, 2] for recent reviews). In type 2 diabetes, skeletal muscle and adipose tissue insulin resistance in- insulin resistance is the primary feature, and it is believed hibits insulin-responsive glucose uptake and glycogen that this insulin resistance coupled with ‘glucose toxicity’ synthesis. It has been widely proposed that insulin resis- is responsible for the plethora of complications seen in tance in insulin-responsive tissues is an ‘adaptation’ to patients, including microvascular and macrovasular dis- nutrient excess [4 – 6]. If this hypothesis is correct, there orders [3]. While the genesis of type 2 diabetes is still un- must be a sensor(s) capable of detecting changes in nutri- clear and under intense study, it appears that certain ge- ent levels and initiating a proper response. In order to elu- netic traits predispose individuals for development of the cidate these sensors, many investigators have asked, disease when exposed to certain environmental factors, ‘How does hyperglycemia induce peripheral insulin re- namely chronic nutrient excess and low energy expendi- sistance?’ The answer to this question should elucidate an ture [4]. Insulin resistance occurs in three separate sys- ‘energy’ or glucose sensor in muscle and adipose tissue. tems: pancreatic b-cells, liver and peripheral insulin-re- Various hypotheses have been put forward and have been sponsive tissues (adipocytes and skeletal muscle). The reviewed elsewhere [4, 7 – 11]. One key observation made role of insulin resistance modulating insulin secretion in by Traxinger and Marshall was that induction of insulin resistance in cultured adipocytes requires three key com- ponents: glucose, insulin and glutamine [12]. Since glut- * Corresponding author. CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 223 amine:fructose-6-phosphate-amidotransferase (GFAT), the first and rate-limiting enzyme of the hexosamine biosynthetic pathway (HSP), requires glutamine and the glucose metabolite fructose-6-phosphate, researchers be- gan to investigate the possibility that the HSP was serv- ing as an energy sensor. The hexosamine biosynthetic pathway and periph- eral insulin resistance (see fig. 1) Figure 2. UDP-GlcNAc is well positioned for serving as a glucose Once glucose enters a cell, it is rapidly converted to glu- sensor in that it is a high-energy compound that is required for syn- cose-6-phosphate that can be converted to glucose-1- thesis and responds to glucose, amino acid, fatty acid and nu- phosphate for glycogen synthesis or converted to fruc- cleotide metabolism. tose-6-phosphate. Fructose-6-phosphate is preferentially used for glycolysis, but a small percentage is converted to thesis, and N-acetylglucosmine (O-GlcNAc) modifica- glucosamine-6-phosphate with the concomitant conver- tion of nuclear and cytosolic proteins. sion of glutamine to glutamate by the rate-limiting en- Investigators have studied the HSP in a variety of sys- zyme in the HSP GFAT [13]. Glucosamine-6-phosphate tems, using both genetic and pharmacological methods is then rapidly converted to uridine 5¢-diphospho N- (reviewed in [6, 9]). One of the initial observations was acetylglucosamine (UDP-GlcNAc) [13]. Due to the that glucosamine that enters the HSP downstream of chemical makeup of UDP-GlcNAc, it is well positioned GFAT and is rapidly converted to UDP-GlcNAc could to serve as a glucose sensor in that it is a high-energy take the place of hyperglycemia in inducing insulin resis- compound that requires and/or responds to glucose, tance in cultured adipocytes [14]. Glucosamine has now amino acid, fatty acid and nucleotide metabolism for syn- been used to induce insulin resistance in a variety of cells, thesis (fig. 2). UDP-GlcNAc serves as the donor sugar tissues and whole organisms, including humans [6, 9]. In nucleotide for lipid and secretory protein complex glyco- 1998, Hresko and colleagues attributed the effects of glu- sylation, glycosyl phosphatidylinositol (GPI) anchor syn- Figure 1. Glucose metabolism and insulin signal transduction in adipocytes. Small molecules are depicted in black, proteins in red and bi- ological processes in purple. Increased flux through the HSP (fruc-6-p to UDP-GlcNAc) results in increased O-GlcNAc modification of nucleocytoplasmic proteins and inhibition of insulin-stimulated glucose uptake. 224 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor cosamine to ATP depletion [15]. While several other lab- have shown that insulin-dependent Glut4 translocation is oratories have disproved this simple explanation inhibited in response to increased flux through the HSP [16 – 18], it is important to note that excessive concentra- [31, 32]; however, alternative mechanisms including ac- tions of glucosamine can in fact deplete ATP levels, giv- tivity and total protein levels of Glut 4 have been pro- ing rise to secondary toxic effects. While glucosamine posed [33, 34]. The protein munc-18 also translocates to leading to insulin resistance was a very exciting finding the plasma membrane in response to insulin and is be- that implicated the HSP in insulin regulation, the biolog- lieved to play a role in Glut4 vesicles fusing with the ically more relevant finding that the effects of hyper- plasma membrane [35]. At the other end of the signaling glycemia-induced insulin resistance could be blocked by pathway, binding of insulin activates the intrinsic tyrosine inhibition of activity or suppression of expression of kinase activity of the insulin receptor. Insulin receptor GFAT validated the HSP as a sensor [12, 14]. Further, substrate (IRS) proteins bind activated receptor, are tyro- several groups have now shown in cell culture as well as sine phosphorylated and recruit active phosphoinositide in animal models and type 2 diabetic patients that hyper- 3-kinase (PI3)-kinase to the plasma membrane. This glycemia and hyperinsulinemia lead to elevated levels of leads to activation of PDK-1, which phosphorylates and UDP-GlcNAc [6, 19–21]. Interestingly, increased free activates AKT. AKT has a variety of substrates, including fatty acids have also been shown to upregulate the HSP, GSK3b. While it is unknown exactly how AKT activation presumably by inhibiting glycolysis and increasing fruc- leads to Glut4 vesicle translocation, several lines of evi- tose-6-phosphate levels (see fig. 1) [7, 22, 23]. Both dence have clearly established the importance of AKT, in- hypo- and hypercaloric intake have also been negatively cluding akt2 knockout mice that are insulin resistant [36]. and positively correlated, respectively, with increased The molecular defects leading to insulin resistance are an flux through the HSP [23, 24]. Several lines of evidence area of intense study, and there are conflicting reports as from genetically engineered rodents also support the role to where the defect(s) and even who the ‘players’ are in of HSP in modulating insulin resistance and serving as an the signal cascade. For reasons of clarity and brevity, we energy sensor. In mice, targeted overexpression of GFAT have elected to summarize what we believe is the most to skeletal muscle and adipose tissue leads to peripheral commonly held viewpoint at this time. A number of insulin resistance [25]. It is also interesting to note that groups have shown a defect in insulin-dependent glut4 targeted overexpression of GFAT to b-cells of mice leads translocation, presumably due to a defect in AKT phos- to hyperinsulinemia and insulin resistance, implicating phorylation and activation in response to hyperglycemia the HSP in insulin regulation in b-cells as well [26]. Fur- or glucosamine treatment, and this defect is also observed thermore, ob/ob mice, which lack leptin and are insulin for munc-18 translocation in both cases [32, 37 – 39]. Fur- resistant, have elevated UDP-GlcNAc levels [27]. In con- thermore, proximal insulin signaling events, such as in- junction with this, increased levels of hexosamines lead sulin receptor activation and IRS tyrosine phosphoryla- to an increase in leptin release from adipocytes, and glu- tion, appear normal in hyperglycemia or glucosamine- cose-stimulated release of leptin can be reduced by inhi- treated cells. Thus, it would appear that the nutrient bition of GFAT [28, 29]. Leptin, an adipocyte-derived sensor (HSP) is acting at or upstream of AKT and down- signal, alters nutrient flux such that energy expenditure is stream of the insulin receptor. Since active AKT isoforms favored over energy storage [30]. HSP flux regulating have also been implicated in preventing apoptosis [40, leptin secretion is in agreement with the model of the 41], reduced AKT activation under insulin-resistant con- HSP serving as an energy sensor and a negative feedback ditions may contribute to b-cell death in diabetes. Exces- system to limit uptake of glucose under hyperglycemic sive HSP flux has also been shown to induce apoptosis in and hyperinsulinemic conditions. retinal neurons but not in L6 muscle cells [42]. These data If the HSP hypothesis is correct, the next important ques- are consistent with the observed retinopathy often seen in tion is, How does the energy sensor (increased flux type 2 diabetes [2]. Also, Boehmelt and colleagues through the HSP resulting in elevated UDP-GlcNAc lev- showed that cells from Emeg32-deficient mice that are els) transduce the signal to cause insulin resistance? Un- defective in the synthesis of UDP-GlcNAc via the HSP der normal conditions, insulin induces glucose uptake have dramatically decreased UDP-GlcNAc levels, ex- and glycogen synthesis in skeletal muscle and adipocytes press activated AKT and have an increased capacity to (reviewed in [4]). While the mechanism of this signal withstand apoptotic stimuli [43]. The reason increased transduction cascade has not been completely elucidated, HSP flux may be inducing apoptosis in some cell types several key components and pathways have been identi- (retinal neurons and b-cells) but not others (skeletal mus- fied (fig. 1). Glut4-containing vesicles translocate and cle and adipocytes) is completely unknown and is a ques- fuse with the plasma membrane in response to insulin tion open for investigation. Insulin resistance also impairs stimulation, and it is primarily the Glut4 glucose trans- glycogen synthesis [44]. Under normal conditions, in- porter that is responsible for insulin-dependent glucose sulin stimulation activates AKT, leading to phosphoryla- uptake in adipocytes and skeletal muscle. Several groups tion and deactivation of GSK3b (fig. 1). This results in CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 225 glycogen synthase being active since it is no longer deac- also been shown to elevate O-GlcNAc levels on certain tivated by phosphorylation. Because excess HSP flux proteins [58, 59]. Finally, the diabetes-inducing reagent leads to defective AKT activation upon insulin stimula- streptozotocin raises O-GlcNAc levels on proteins in b- tion, glycogen synthesis is inhibited as well since GSK3b cells [58]. is not efficiently phosphorylated and thus can inhibit As an aside, the attractive hypothesis was put forward that glycogen synthase. streptozotocin was inducing b-cell death by inhibiting O- Excessive flux through the HSP serves as an energy sen- GlcNAcase [60]. However, we and others, using the more sor that appears to be mediating its effect, at least in part, potent O-GlcNAcase inhibitor PUGNAc [61], have shown by inhibiting insulin signal transduction at or upstream of that elevated O-GlcNAc levels alone do not induce apop- AKT. Thus, many investigators have established the HSP tosis in b-cells [62 – 64]. Streptozotocin is a potent alky- pathway as a sensor and implicated specific molecular lating reagent and is thought to induce cell death via DNA defects leading to insulin resistance. But what is the damage [65]. Streptozotocin, however, is also a weak O- mechanism by which increased flux inhibits insulin sig- GlcNAcase inhibitor that is capable of raising O-GlcNAc naling? Since the vast majority of carbohydrate entering levels on proteins in cells [66]. Thus, the possibility exists the HSP is rapidly converted to UDP-GlcNAc, several in- that the combination of O-GlcNAcase inhibition and vestigators have proposed that glycosylation may be the DNA damage is necessary to induce b-cell death. mediator of insulin resistance [4, 31, 45]. O-GlcNAc While there was a strong correlation with elevated O-Glc- modification of nucleocytoplasmic proteins is one possi- NAc levels and insulin resistance, until recently there was ble candidate for the mediator. no direct proof for O-GlcNAc levels modulating insulin action. As a f irst step towards addressing whether O-Glc- NAc modif ication of proteins was directly modulating in- O-GlcNAc’s role in nutrient sensing sulin signaling, we elevated O-GlcNAc levels in 3T3-L1 and insulin resistance adipocytes via treatment of the cells with the O-Glc- NAcase inhibitor PUGNAc [67]. PUGNAc treatment sig- O-GlcNAc modification has several features that distin- nificantly elevated the O-GlcNAc modification on many guish it from classical glycosylation and make it an at- nucleocytoplasmic proteins. More important, elevation of tractive target for the molecular mechanism by which in- O-GlcNAc levels impaired insulin-stimulated glucose crease flux via the HSP could inhibit insulin signaling uptake in the cells. Thus, using a pharmacological ap- [46]. The covalent modification of serine and threonine proach, we have established a direct causal relationship hydroxyls on nuclear and cytosolic proteins by b-linked between elevated O-GlcNAc and insulin resistance in O-GlcNAc was described by Torres and Hart in 1984 3T3-L1 adipocytes. We also found that proximal insulin [47]. Several recent reviews have focused on the proper- signaling was unaffected, while AKT phosphorylation ties of the modif ication and the various proteins modif ied and activation (as measured by GSK3b phosphorylation) [48 – 53]. Briefly, O-GlcNAc has several distinguishing was impaired. Thus, elevation of O-GlcNAc levels was characteristics that make it more analogous to phospho- not only causing insulin resistance but appeared to be in- rylation than to classical complex glycosylation. Namely hibiting insulin signaling at the same point in the pathway (i) O-GlcNAc is attached to nucleocytoplasmic proteins as increased flux through the HSP [37, 67]. We were also or to the cytosolic portions of membrane bound proteins, able to show that IRS-1 and b-catenin were modified by (ii) the modification involves the attachment of a single O-GlcNAc in a PUGNAc-dependent fashion in the 3T3- sugar from a high-energy donor (UDP-GlcNAc) that is L1 adipocytes. These findings allow us to put forth a not elongated, (iii) the modification is dynamic, (iv) the working model in which O-GlcNAc acts as a mediator of enzymes responsible for its attachment (O-GlcNAc trans- insulin resistance as well as a metabolic sensor. ferase, OGT) and removal (O-GlcNAcase) are nucleocy- toplasmic, (v) the modification is inducible and (vi) the modification competes with phosphorylation for the Future directions same sites on certain proteins. Underscoring the impor- tance of this modification, embryonic stem cells of mice One interesting question that remains to be answered is, lacking OGT fail to survive [54]. OGT has been found to What is the role of hyperinsulinemia in peripheral insulin be responsive to a wide range of UDP-GlcNAc concen- resistance? Marshall and colleagues established that glu- trations [55]. Yki-Jarvinen and colleagues found that in- cose, glutamine and insulin were necessary for the induc- ducing insulin resistance in rats by glucosamine and hy- tion of insulin resistance in adipocytes [13]. Glucose and perinsulinemia led to elevated levels of O-GlcNAc on glutamine are necessary for elevation of flux through the skeletal muscle proteins [56]. It was also established that HSP, leading to elevated UDP-GlcNAc levels, and can be IRS-1 became O-GlcNAc modified in response to in- substituted for by glucosamine; however, chronic insulin crease flux through the HSP [57]. Hyperglycemia has treatment is still necessary for inducing insulin resis- 226 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor tance. Interestingly, we found that PUGNAc treatment GlcNAc modification and leptin expression, secretion alone was suff icient to induce insulin resistance, and glu- and action remains to be elucidated. cosamine treatment plus insulin elevated O-GlcNAc lev- O-GlcNAc has begun to emerge as an important posttrans- els higher than either alone [67]. Buse and colleagues lational modification in cellular regulation [46, 50]. Many showed that glucosamine treatment alone, which did not tools to facilitate study of this modification have only re- cause insulin resistance, was able to substantially elevate cently become available, including the cloning of OGT in UDP-GlcNAc levels in adipocytes and that the addition 1997 [68, 69] and O-GlcNAcase in 2001 [70], the devel- of insulin induced a two-fold increase in the levels of opment of a general O-GlcNAc specific antibody in 2001 UDP-GlcNAc [18]. Thus, it would appear that PUGNAc [71] and O-GlcNAc site-specific antibodies in 2002 [72], is acting at a point after the convergence of hyper- and mass spectrometry based techniques for site mapping glycemia and hyperinsulinemia signaling. The fact that [73 – 75]. All of these developments should help to acceler- glucosamine treatment still requires hyperinsulinemia ate our understanding of the O-GlcNAc modification. would suggest that insulin is acting downstream of GFAT. In conclusion, we and others are using a working model Based on the above data and since glucosamine-6-phos- that O-GlcNAc modification of proteins serves as a nu- phate is rapidly converted to UDP-GlcNAc, we would hy- trient sensor to modulate extracellular signal transduction pothesize that insulin is acting directly on either OGT or cascades. In this model, the cell is not blindly responding O-GlcNAcase. This hypothesis is currently under investi- to extracellular signals but instead is taking into account gation, and preliminary experiments show that OGT be- its own energy stores and responding appropriately. Fur- comes tyrosine phosphorylated in response to insulin ther, O-GlcNAc, serving as a metabolic sensor, may be stimulation [L. Wells, K. Vosseller, G. W. Hart, unpub- modulating the expression, activity, localization and/or lished results]. secretion of proteins. There have been several other mol- Testing the role of O-GlcNAc levels regulating insulin ecules and mechanisms proposed as nutrient sensors [4, signaling by genetic means as opposed to pharmacologi- 7– 11], and it is likely that cells use multiple sensors and cal methods would greatly strengthen the argument [67 a, mechanisms to regulate such an important process as en- published during revision of this manuscript]. In this re- ergy homeostasis. Deciphering how these sensors are me- gard, reduced expression and overexpression of OGT and diating their effects is an important goal if we are to un- O-GlcNAcase in adipocytes is under development. Test- derstand the pathophysiology of complex pleiotropic ing the role of O-GlcNAc on insulin action in other tissue metabolic diseases, such as type 2 diabetes. An under- types as well as whole animals is also an important area standing of the relationship between O-GlcNAc and O- of future research. Obviously, identification of proteins phosphate in metabolic sensing, signaling cascades and that become modified by O-GlcNAc in response to hy- transcriptional regulation should lead to rational new tar- perglycemia and hyperinsulinemia is important, as well gets for drug development. as the identif ication of proteins that are hyper-O-GlcNAc Acknowledgements. We would like to thank Natasha Zachara and modified in rodent models of type 2 diabetes. Once can- Karen Wells for critical reading of the manuscript. Because the f ield didate proteins are identified, the elucidation of the mol- of metabolism and diabetes is so well studied, we had to pick and ecular mechanism by which O-GlcNAc modification is choose articles to cite. We apologize to our colleagues whose con- inhibiting insulin signaling will be of great importance in tributions to the field were not cited. This work was supported by national research service award fellowships to L. W. (CA83261) understanding this posttranslational modification and to and K. V. (GM20528) and by the national institutes of health future therapies for diabetes. NIDDK grant to G. W. H. (DK61671). The role of O-GlcNAc in the liver and the b-cell, as well as in microvascular and macrovascular disease, remains to be determined. In this regard, exploring whether O- 1Kruszynska Y. T. and Olefsky J. M. (1996) Cellular and molec- GlcNAc plays a role in apoptosis is especially important. ular mechanisms of non-insulin dependent diabetes mellitus. J. Investig. Med. 44: 413 – 428 The HSP appears to be involved in b-cell death and 2 Brownlee M. (2001) Biochemistry and molecular cell biology retinopathy [42, 58]. Also, we have determined that O- of diabetic complications. Nature 414: 813 – 820 GlcNAcase is cleaved in vitro by caspase-3, an execu- 3 O’Doherty R., Stein D. and Foley J. (1997) Insulin resistance. tioner caspase in programmed cell death [52]. Activation Diabetologia 40 Suppl. 3: B10–15 4Patti M. E. (1999) Nutrient modulation of cellular insulin ac- of AKT is also inhibited by elevated flux through the HSP tion. Ann. N. Y. Acad. Sci. 892: 187 – 203 and elevation of O-GlcNAc levels [37, 67]. Active AKT 5 McClain D. A. and Crook E. D. (1996) Hexosamines and in- has been demonstrated to be antiapoptotic [40, 41], and sulin resistance. Diabetes 45: 1003 – 1009 cells with reduced UDP-GlcNAc levels have elevated lev- 6 McClain D. A. (2002) Hexosamines as mediators of nutrient sensing and regulation in diabetes. J. Diabetes Complications els of active AKT and are refractory to apoptotic stimuli 16: 72–80 [43]. Thus, there is compelling evidence for the investi- 7Schmitz-Peiffer C. (2000) Signalling aspects of insulin resis- gation of a possible role for O-GlcNAc in apoptosis. Also, tance in skeletal muscle: mechanisms induced by lipid over- determining whether there is a connection between O- supply. Cell Signal. 12: 583 – 594 CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 227 8 Hussain M. A. (1998) A case for glucosamine. Eur. J. En- thesis pathway in muscles of insulin-resistant ob/ob mice. Am. docrinol. 139: 472 – 475 J. Physiol. 272: E1080 – 1088 9 Rossetti L. (2000) Perspective: hexosamines and nutrient sens- 28 Wang J., Liu R., Hawkins M., Barzilai N. and Rossetti L. (1998) ing. Endocrinology 141: 1922 – 1925 A nutrient-sensing pathway regulates leptin gene expression in 10 Ruderman N. B. and Dean D. (1998) Malonyl CoA, long chain muscle and fat. Nature 393: 684 – 688 fatty acyl CoA and insulin resistance in skeletal muscle. J. Ba- 29 McClain D. A., Alexander T., Cooksey R. C. and Considine R. sic Clin. Physiol. Pharmacol. 9: 295 – 308 V. (2000) Hexosamines stimulate leptin production in trans- 11 Ukropec J., Sebokova E. and Klimes I. (2001) Nutrient sensing, genic mice. Endocrinology 141: 1999 – 2002 leptin and insulin action. Arch. Physiol. Biochem. 109: 38–51 30 Flier J. S. (1997) Leptin expression and action: new experi- 12 Marshall S., Bacote V. and Traxinger R. R. (1991) Complete in- mental paradigms. Proc. Natl. Acad. Sci. USA 94: 4242 – 4245 hibition of glucose-induced desensitization of the glucose 31 Robinson K. A., Sens D. A. and Buse M. G. (1993) Pre-expo- transport system by inhibitors of mRNA synthesis. Evidence sure to glucosamine induces insulin resistance of glucose trans- for rapid turnover of glutamine:fructose-6-phosphate amido- port and glycogen synthesis in isolated rat skeletal muscles. transferase. J. Biol. Chem. 266: 10155 – 10161 Study of mechanisms in muscle and in rat-1 f ibroblasts overex- 13 Marshall S., Garvey W. T. and Traxinger R. R. (1991) New in- pressing the human insulin receptor. Diabetes 42: 1333 – 1346 sights into the metabolic regulation of insulin action and insulin 32 Ross S. A., Chen X., Hope H. R., Sun S., McMahon E. G., resistance: role of glucose and amino acids. FASEB J. 5: 3031 – Broschat K. et al. (2000) Development and comparison of two 3036 3T3-L1 adipocyte models of insulin resistance: increased glu- 14 Marshall S., Bacote V. and Traxinger R. R. (1991) Discovery of cose flux vs glucosamine treatment. Biochem. Biophys. Res. a metabolic pathway mediating glucose-induced desensitiza- Commun. 273: 1033 – 1041 tion of the glucose transport system. Role of hexosamine 33 Chen H., Ing B. L., Robinson K. A., Feagin A. C., Buse M. G. biosynthesis in the induction of insulin resistance. J. Biol. and Quon M. J. (1997) Effects of overexpression of gluta- Chem. 266: 4706 – 4712 mine : fructose-6-phosphate amidotransferase (GFAT) and glu- 15. Hresko R. C., Heimberg H., Chi M. M. and Mueckler M. cosamine treatment on translocation of GLUT4 in rat adipose (1998) Glucosamine-induced insulin resistance in 3T3-L1 cells. Mol. Cell. Endocrinol. 135: 67–77 adipocytes is caused by depletion of intracellular ATP. J. Biol. 34 Thomson M. J., Williams M. G. and Frost S. C. (1997) Devel- Chem. 273: 20658 – 20668 opment of insulin resistance in 3T3-L1 adipocytes. J. Biol. 16 Kang J., Heart E. and Sung C. K. (2001) Effects of cellular ATP Chem. 272: 7759 – 7764 depletion on glucose transport and insulin signaling in 3T3-L1 35 Nelson B. A., Robinson K. A. and Buse M. G. (2002) Insulin adipocytes. Am. J. Physiol. Endocrinol. Metab. 280: E428 – 435 acutely regulates Munc18-c subcellular trafficking: altered re- 17 Vemuri S. and Marchase R. B. (1999) The inhibition of capac- sponse in insulin-resistant 3T3-L1 adipocytes. J. Biol. Chem. itative calcium entry due to ATP depletion but not due to glu- 277: 3809 – 3812 cosamine is reversed by staurosporine. J. Biol. Chem. 274: 36 Cho H., Mu J., Kim J. K., Thorvaldsen J. L., Chu Q., Crenshaw 20165 – 20170 E. B. 3rd et al. (2001) Insulin resistance and a diabetes melli- 18 Nelson B. A., Robinson K. A. and Buse M. G. (2000) High glu- tus-like syndrome in mice lacking the protein kinase Akt2 cose and glucosamine induce insulin resistance via different (PKB beta). Science 292: 1728 – 1731 mechanisms in 3T3-L1 adipocytes. Diabetes 49: 981 – 991 37 Heart E., Choi W. S. and Sung C. K. (2000) Glucosamine-in- 19 Robinson K. A., Weinstein M. L., Lindenmayer G. E. and Buse duced insulin resistance in 3T3-L1 adipocytes. Am. J. Physiol. M. G. (1995) Effects of diabetes and hyperglycemia on the hex- Endocrinol. Metab. 278: E103 – 112 osamine synthesis pathway in rat muscle and liver. Diabetes 44: 38 Nelson B. A., Robinson K. A. and Buse M. G. (2002) Defective 1438 – 1446 Akt activation is associated with glucose- but not glucosamine- 20 Yki-Jarvinen H. and Makimattila S. (1997) Insulin resistance induced insulin resistance. Am. J. Physiol. Endocrinol. Metab. due to hyperglycaemia: an adaptation protecting insulin-sensi- 282: E497 – 506 tive tissues. Diabetologia 40 Suppl. 2: S141 – 144 39 Singh L. P., Gennerette D., Simmons S. and Crook E. D. (2001) 21 Yki-Jarvinen H., Daniels M. C., Virkamaki A., Makimattila S., Glucose-induced insulin resistance of phosphatidylinositol 3¢- DeFronzo R. A. and McClain D. (1996) Increased glu- OH kinase and AKT/PKB is mediated by the hexosamine tamine:fructose-6-phosphate amidotransferase activity in biosynthesis pathway. J. Diabetes Complications 15: 88–96 skeletal muscle of patients with NIDDM. Diabetes 45: 302 – 40 Lawlor M. A. and Alessi D. R. (2001) PKB/Akt: a key media- 307 tor of cell proliferation, survival and insulin responses? J. Cell. 22 Randle P. J., Garland P. B., Newsholme E. A. and Hales C. N. Sci. 114: 2903 – 2910 (1965) The glucose fatty acid cycle in obesity and maturity on- 41 Kandel E. S. and Hay N. (1999) The regulation and activities of set diabetes mellitus. Ann. N. Y. Acad. Sci. 131: 324 – 333 the multifunctional serine/threonine kinase Akt/PKB. Exp. Cell 23 Hawkins M., Barzilai N., Liu R., Hu M., Chen W. and Rossetti Res. 253: 210 – 229 L. (1997) Role of the glucosamine pathway in fat-induced in- 42 Nakamura M., Barber A. J., Antonetti D. A., LaNoue K. F., sulin resistance. J. Clin. Invest. 99: 2173 – 2182 Robinson K. A., Buse M. G. et al. (2001) Excessive hex- 24 Gazdag A. C., Wetter T. J., Davidson R. T., Robinson K. A., osamines block the neuroprotective effect of insulin and induce Buse M. G., Yee A. J. et al. (2000) Lower calorie intake en- apoptosis in retinal neurons. J. Biol. Chem. 276: 43748 – 43755 hances muscle insulin action and reduces hexosamine levels. 43 Boehmelt G., Wakeham A., Elia A., Sasaki T., Plyte S., Potter J. Am. J. Physiol. Regul. Integr. Comp. Physiol. 278: R504 – 512 et al. (2000) Decreased UDP-GlcNAc levels abrogate prolifer- 25 Hebert L. F. Jr, Daniels M. C., Zhou J., Crook E. D., Turner R. ation control in EMeg32-deficient cells. EMBO J. 19: 5092 – L., Simmons S. T. et al. (1996) Overexpression of glu- 5104 tamine:fructose-6-phosphate amidotransferase in transgenic 44 Ciaraldi T. P., Carter L., Nikoulina S., Mudaliar S., McClain D. mice leads to insulin resistance. J. Clin. Invest. 98: 930 – 936 A. and Henry R. R. (1999) Glucosamine regulation of glucose 26 Tang J., Neidigh J. L., Cooksey R. C. and McClain D. A. (2000) metabolism in cultured human skeletal muscle cells: divergent Transgenic mice with increased hexosamine flux specifically effects on glucose transport/phosphorylation and glycogen syn- targeted to beta-cells exhibit hyperinsulinemia and peripheral thase in non-diabetic and type 2 diabetic subjects. Endocrinol- insulin resistance. Diabetes 49: 1492 – 1499 ogy 140: 3971 – 3980 27 Buse M. G., Robinson K. A., Gettys T. W., McMahon E. G. and 45 Rossetti L., Hawkins M., Chen W., Gindi J. and Barzilai N. Gulve E. A. (1997) Increased activity of the hexosamine syn- (1995) In vivo glucosamine infusion induces insulin resistance 228 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor in normoglycemic but not in hyperglycemic conscious rats. J. 63 Kaneto H., Xu G., Song K. H., Suzuma K., Bonner-Weir S., Clin. Invest. 96: 132 – 140 Sharma A. et al. (2001) Activation of the hexosamine pathway 46 Wells L., Vosseller K. and Hart G. W. (2001) Glycosylation of leads to deterioration of pancreatic beta-cell function through nucleocytoplasmic proteins: signal transduction and O-Glc- the induction of oxidative stress. J. Biol. Chem. 276: 31099 – NAc. Science 291: 2376 – 2378 31104 47 Torres C. R. and Hart G. W. (1984) Topography and polypep- 64 Gao Y., Parker G. J. and Hart G. W. (2000) Streptozotocin-in- tide distribution of terminal N-acetylglucosamine residues on duced beta-cell death is independent of its inhibition of O-Glc- the surfaces of intact lymphocytes. Evidence for O-linked Glc- NAcase in pancreatic Min6 cells. Arch. Biochem. Biophys. NAc. J. Biol. Chem. 259: 3308 – 3317 383: 296 – 302 48 Hart G. W., Haltiwanger R. S., Holt G. D. and Kelly W. G. 65 Elsner M., Guldbakke B., Tiedge M., Munday R. and Lenzen S. (1989) Nucleoplasmic and cytoplasmic glycoproteins. Ciba (2000) Relative importance of transport and alkylation for pan- Found. Symp. 145: 102 – 112 creatic beta-cell toxicity of streptozotocin. Diabetologia 43: 49 Hanover J. A. (2001) Glycan-dependent signaling: O-linked N- 1528 – 1533 acetylglucosamine. FASEB J. 15: 1865 – 1876 66 Roos, M. D., Xie, W., Su, K., Clark, J. A., Yang, X., Chin, E. et 50 Zachara N. E. and Hart G. W. (2002) The emerging signif icance al. (1998) Streptozotocin, an analog of N-acetylglucosamine, of O-GlcNAc in cellular regulation. Chem. Rev. 102: 431 – 438 blocks the removal of O-GlcNAc from intracellular proteins. 51 Vosseller K., Wells L. and Hart G. W. (2001) Nucleocytoplas- Proc. Assoc. Am. Physicians 110: 422 – 432 mic O-glycosylation: O-GlcNAc and functional proteomics. 67 Vosseller K., Wells L., Lane M. D. and Hart G. W. (2002) Ele- Biochimie 83: 575 – 581 vated nucleocytoplasmic glycosylation by O-GlcNAc results in 52 Wells L., Gao Y., Mahoney J. A., Vosseller K., Chen C., Rosen insulin resistance associated with defects in Akt activation in A. et al. (2001) Dynamic O-glycosylation of nuclear and cy- 3T3-L1 adipocytes. Proc. Natl. Acad. Sci. USA 99: 5313 – 5318 tosolic proteins: further characterization of the nucleocyto- 67 a McClain D. A., Lubas W. A., Cooksey R. C., Hazel M., Parker plasmic beta-N-acetylglucosaminidase, O-GlcNAcase. J. Biol. G. J., Love D. C. et al. (2002) Altered glycan-dependent signal- Chem. 277: 5313 – 5318 ing induces insulin resistance and hyperleptinemia. Proc. Natl. 53 Comer F. I. and Hart G. W. (2000) O-Glycosylation of nuclear Acad. Sci. USA epub ahead of print and cytosolic proteins. Dynamic interplay between O-GlcNAc 68 Kreppel L. K., Blomberg M. A. and Hart G. W. (1997) Dynamic and O-phosphate. J. Biol. Chem. 275: 29179 – 29182 glycosylation of nuclear and cytosolic proteins. Cloning and 54 Shafi R., Iyer S. P., Ellies L. G., O’Donnell N., Marek K. W., Chui characterization of a unique O-GlcNAc transferase with multi- D. et al. (2000) The O-GlcNAc transferase gene resides on the X ple tetratricopeptide repeats. J. Biol. Chem. 272: 9308 – 9315 chromosome and is essential for embryonic stem cell viability and 69 Lubas W. A., Frank D. W., Krause M. and Hanover J. A. (1997) mouse ontogeny. Proc. Natl. Acad. Sci. USA 97: 5735 – 5739 O-Linked GlcNAc transferase is a conserved nucleocytoplas- 55 Kreppel L. K. and Hart G. W. (1999) Regulation of a cytosolic mic protein containing tetratricopeptide repeats. J. Biol. Chem. and nuclear O-GlcNAc transferase. Role of the tetratricopep- 272: 9316 – 9324 tide repeats. J. Biol. Chem. 274: 32015 – 32022 70 Gao Y., Wells L., Comer F. I., Parker G. J. and Hart G. W. (2001) 56 Yki-Jarvinen H., Virkamaki A., Daniels M. C., McClain D. and Dynamic O-glycosylation of nuclear and cytosolic proteins: Gottschalk W. K. (1998) Insulin and glucosamine infusions in- cloning and characterization of a neutral, cytosolic beta-N- crease O-linked N-acetyl-glucosamine in skeletal muscle pro- acetylglucosaminidase from human brain. J. Biol. Chem. 276: teins in vivo. Metabolism 47: 449 – 455 9838 – 9845 57 Patti M. E., Virkamaki A., Landaker E. J., Kahn C. R. and Yki- 71 Comer F. I., Vosseller K., Wells L., Accavitti M. A. and Hart G. Jarvinen H. (1999) Activation of the hexosamine pathway by W. (2001) Characterization of a mouse monoclonal antibody glucosamine in vivo induces insulin resistance of early postre- specific for O-linked N-acetylglucosamine. Anal. Biochem. ceptor insulin signaling events in skeletal muscle. Diabetes 48: 293: 169 – 177 1562 – 1571 72 Kamemura K., Hayes B. K., Comer F. I. and Hart G. W. (2002) 58 Hanover J. A., Lai Z., Lee G., Lubas W. A. and Sato S. M. Dynamic interplay between O-glycosylation and O-phosphory- (1999) Elevated O-linked N-acetylglucosamine metabolism in lation of nucleocytoplasmic proteins. Alternative glycosyla- pancreatic beta-cells. Arch. Biochem. Biophys. 362: 38–45 tion/phosphorylation of Thr58, a known mutational hot spot of 59 Han I., Oh E. S. and Kudlow J. E. (2000) Responsiveness of the c-Myc in lymphomas, is regulated by mitogens. J. Biol. Chem. state of O-linked N-acetylglucosamine modif ication of nuclear 277: 19229 – 19235 pore protein p62 to the extracellular glucose concentration. 73 Greis K. D., Hayes B. K., Comer F. I., Kirk M., Barnes S., Biochem. J. 350 Pt 1: 109 – 114 Lowary T. L. et al. (1996) Selective detection and site-analysis 60 Liu K., Paterson A. J., Chin E. and Kudlow J. E. (2000) Glucose of O-GlcNAc-modified glycopeptides by beta-elimination and stimulates protein modification by O-linked GlcNAc in pan- tandem electrospray mass spectrometry. Anal. Biochem. 234: creatic beta cells: linkage of O-linked GlcNAc to beta cell 38–49 death. Proc. Natl. Acad. Sci. USA 97: 2820 – 2825 74 Haynes P. A. and Aebersold R. (2000) Simultaneous detection 61 Haltiwanger R. S., Grove K. and Philipsberg G. A. (1998) Mod- and identification of O-GlcNAc-modified glycoproteins us- ulation of O-linked N-acetylglucosamine levels on nuclear and ing liquid chromatography-tandem mass spectrometry. Anal. cytoplasmic proteins in vivo using the peptide O-GlcNAc-beta- Chem. 72: 5402 – 5410 N-acetylglucosaminidase inhibitor O-(2-acetamido-2-deoxy- 75 Wells L., Vosseller K., Cole R. N., Cronshaw J. M., Matunis M. D-glucopyranosylidene)amino-N-phenylcarbamate. J. Biol. J. and Hart G. W. (2002) Mapping sites of O-GlcNAc modifi- Chem. 273: 3611 – 3617 cation using affinity tags for serine and threonine post-transla- 62 Okuyama R. and Yachi M. (2001) Cytosolic O-GlcNAc accu- tional modifications. Mol. Cell Proteomics 1: 791 – 804 mulation is not involved in beta-cell death in HIT-T15 or Min6. Biochem. Biophys. Res. Commun. 287: 366 – 371 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cellular and Molecular Life Sciences Springer Journals

A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance

Download PDF
7 pages

Loading next page...
 
/lp/springer-journals/a-role-for-n-acetylglucosamine-as-a-nutrient-sensor-and-mediator-of-Fm0ncIWT0W

References (77)

Publisher
Springer Journals
Copyright
Copyright © Birkhäuser Verlag, 2003
Subject
Life Sciences; Cell Biology; Biomedicine, general; Life Sciences, general; Biochemistry, general
ISSN
1420-682X
eISSN
1420-9071
DOI
10.1007/s000180300017
Publisher site
See Article on Publisher Site

Abstract

CMLS, Cell. Mol. Life Sci. 60 (2003) 222 – 228 1420-682X/03/020222-07 CMLS Cellular and Molecular Life Sciences © Birkhäuser Verlag, Basel, 2003 A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance L. Wells, K. Vosseller and G. W. Hart * Department of Biological Chemistry, Johns Hopkins School of Medicine, 517 Woods Basic Science Building, 725 N. Wolfe Street, Baltimore, Maryland 21205 (USA), e-mail: [email protected] Abstract. The ability to regulate energy balance at both which produces the acetylated aminosugar nucleotide the cellular and whole body level is an essential process uridine 5¢-diphospho-N-acetylglucosamine (UDP-Glc- of life. As western society has shifted to a higher caloric NAc) as its end product. Since UDP-GlcNAc is the donor diet and more sedentary lifestyle, the incidence of type 2 substrate for modification of nucleocytoplasmic proteins diabetes (non-insulin-dependent diabetes mellitus) has at serine and threonine residues with N-acetylglu- increased to epidemic proportions. Thus, type 2 diabetes cosamine (O-GlcNAc), the possibility of this posttransla- has been described as a disease of ‘chronic overnutrition’. tional modification serving as the nutrient sensor has There are abundant data to support the relationship be- been proposed. We have recently directly tested this tween nutrient availability and insulin action. However, model in adipocytes by examining the effect of elevated there have been multiple hypotheses and debates as to the levels of O-GlcNAc on insulin-stimulated glucose up- mechanism by which nutrient availability modulates in- take. In this review, we summarize the existing work that sulin signaling and how excess nutrients lead to insulin implicates the HSP and O-GlcNAc modif ication as nutri- resistance. One well-established pathway for nutrient ent sensors and regulators of insulin signaling. sensing is the hexosamine biosynthetic pathway (HSP), Key words. O-GlcNAc; glycosylation; posttranslational modification; hexosamine biosynthetic pathway; glucosamine; hyperglycemia; insulin resistance; type 2 diabetes. Type 2 diabetes and insulin resistance the b- cell and upregulating gluconeogenesis and glucose output in the liver will only be touched upon briefly in Chronic hyperglycemia is the hallmark of all types of di- this review and has been reviewed elsewhere [1 – 4]. In abetes (see [1, 2] for recent reviews). In type 2 diabetes, skeletal muscle and adipose tissue insulin resistance in- insulin resistance is the primary feature, and it is believed hibits insulin-responsive glucose uptake and glycogen that this insulin resistance coupled with ‘glucose toxicity’ synthesis. It has been widely proposed that insulin resis- is responsible for the plethora of complications seen in tance in insulin-responsive tissues is an ‘adaptation’ to patients, including microvascular and macrovasular dis- nutrient excess [4 – 6]. If this hypothesis is correct, there orders [3]. While the genesis of type 2 diabetes is still un- must be a sensor(s) capable of detecting changes in nutri- clear and under intense study, it appears that certain ge- ent levels and initiating a proper response. In order to elu- netic traits predispose individuals for development of the cidate these sensors, many investigators have asked, disease when exposed to certain environmental factors, ‘How does hyperglycemia induce peripheral insulin re- namely chronic nutrient excess and low energy expendi- sistance?’ The answer to this question should elucidate an ture [4]. Insulin resistance occurs in three separate sys- ‘energy’ or glucose sensor in muscle and adipose tissue. tems: pancreatic b-cells, liver and peripheral insulin-re- Various hypotheses have been put forward and have been sponsive tissues (adipocytes and skeletal muscle). The reviewed elsewhere [4, 7 – 11]. One key observation made role of insulin resistance modulating insulin secretion in by Traxinger and Marshall was that induction of insulin resistance in cultured adipocytes requires three key com- ponents: glucose, insulin and glutamine [12]. Since glut- * Corresponding author. CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 223 amine:fructose-6-phosphate-amidotransferase (GFAT), the first and rate-limiting enzyme of the hexosamine biosynthetic pathway (HSP), requires glutamine and the glucose metabolite fructose-6-phosphate, researchers be- gan to investigate the possibility that the HSP was serv- ing as an energy sensor. The hexosamine biosynthetic pathway and periph- eral insulin resistance (see fig. 1) Figure 2. UDP-GlcNAc is well positioned for serving as a glucose Once glucose enters a cell, it is rapidly converted to glu- sensor in that it is a high-energy compound that is required for syn- cose-6-phosphate that can be converted to glucose-1- thesis and responds to glucose, amino acid, fatty acid and nu- phosphate for glycogen synthesis or converted to fruc- cleotide metabolism. tose-6-phosphate. Fructose-6-phosphate is preferentially used for glycolysis, but a small percentage is converted to thesis, and N-acetylglucosmine (O-GlcNAc) modifica- glucosamine-6-phosphate with the concomitant conver- tion of nuclear and cytosolic proteins. sion of glutamine to glutamate by the rate-limiting en- Investigators have studied the HSP in a variety of sys- zyme in the HSP GFAT [13]. Glucosamine-6-phosphate tems, using both genetic and pharmacological methods is then rapidly converted to uridine 5¢-diphospho N- (reviewed in [6, 9]). One of the initial observations was acetylglucosamine (UDP-GlcNAc) [13]. Due to the that glucosamine that enters the HSP downstream of chemical makeup of UDP-GlcNAc, it is well positioned GFAT and is rapidly converted to UDP-GlcNAc could to serve as a glucose sensor in that it is a high-energy take the place of hyperglycemia in inducing insulin resis- compound that requires and/or responds to glucose, tance in cultured adipocytes [14]. Glucosamine has now amino acid, fatty acid and nucleotide metabolism for syn- been used to induce insulin resistance in a variety of cells, thesis (fig. 2). UDP-GlcNAc serves as the donor sugar tissues and whole organisms, including humans [6, 9]. In nucleotide for lipid and secretory protein complex glyco- 1998, Hresko and colleagues attributed the effects of glu- sylation, glycosyl phosphatidylinositol (GPI) anchor syn- Figure 1. Glucose metabolism and insulin signal transduction in adipocytes. Small molecules are depicted in black, proteins in red and bi- ological processes in purple. Increased flux through the HSP (fruc-6-p to UDP-GlcNAc) results in increased O-GlcNAc modification of nucleocytoplasmic proteins and inhibition of insulin-stimulated glucose uptake. 224 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor cosamine to ATP depletion [15]. While several other lab- have shown that insulin-dependent Glut4 translocation is oratories have disproved this simple explanation inhibited in response to increased flux through the HSP [16 – 18], it is important to note that excessive concentra- [31, 32]; however, alternative mechanisms including ac- tions of glucosamine can in fact deplete ATP levels, giv- tivity and total protein levels of Glut 4 have been pro- ing rise to secondary toxic effects. While glucosamine posed [33, 34]. The protein munc-18 also translocates to leading to insulin resistance was a very exciting finding the plasma membrane in response to insulin and is be- that implicated the HSP in insulin regulation, the biolog- lieved to play a role in Glut4 vesicles fusing with the ically more relevant finding that the effects of hyper- plasma membrane [35]. At the other end of the signaling glycemia-induced insulin resistance could be blocked by pathway, binding of insulin activates the intrinsic tyrosine inhibition of activity or suppression of expression of kinase activity of the insulin receptor. Insulin receptor GFAT validated the HSP as a sensor [12, 14]. Further, substrate (IRS) proteins bind activated receptor, are tyro- several groups have now shown in cell culture as well as sine phosphorylated and recruit active phosphoinositide in animal models and type 2 diabetic patients that hyper- 3-kinase (PI3)-kinase to the plasma membrane. This glycemia and hyperinsulinemia lead to elevated levels of leads to activation of PDK-1, which phosphorylates and UDP-GlcNAc [6, 19–21]. Interestingly, increased free activates AKT. AKT has a variety of substrates, including fatty acids have also been shown to upregulate the HSP, GSK3b. While it is unknown exactly how AKT activation presumably by inhibiting glycolysis and increasing fruc- leads to Glut4 vesicle translocation, several lines of evi- tose-6-phosphate levels (see fig. 1) [7, 22, 23]. Both dence have clearly established the importance of AKT, in- hypo- and hypercaloric intake have also been negatively cluding akt2 knockout mice that are insulin resistant [36]. and positively correlated, respectively, with increased The molecular defects leading to insulin resistance are an flux through the HSP [23, 24]. Several lines of evidence area of intense study, and there are conflicting reports as from genetically engineered rodents also support the role to where the defect(s) and even who the ‘players’ are in of HSP in modulating insulin resistance and serving as an the signal cascade. For reasons of clarity and brevity, we energy sensor. In mice, targeted overexpression of GFAT have elected to summarize what we believe is the most to skeletal muscle and adipose tissue leads to peripheral commonly held viewpoint at this time. A number of insulin resistance [25]. It is also interesting to note that groups have shown a defect in insulin-dependent glut4 targeted overexpression of GFAT to b-cells of mice leads translocation, presumably due to a defect in AKT phos- to hyperinsulinemia and insulin resistance, implicating phorylation and activation in response to hyperglycemia the HSP in insulin regulation in b-cells as well [26]. Fur- or glucosamine treatment, and this defect is also observed thermore, ob/ob mice, which lack leptin and are insulin for munc-18 translocation in both cases [32, 37 – 39]. Fur- resistant, have elevated UDP-GlcNAc levels [27]. In con- thermore, proximal insulin signaling events, such as in- junction with this, increased levels of hexosamines lead sulin receptor activation and IRS tyrosine phosphoryla- to an increase in leptin release from adipocytes, and glu- tion, appear normal in hyperglycemia or glucosamine- cose-stimulated release of leptin can be reduced by inhi- treated cells. Thus, it would appear that the nutrient bition of GFAT [28, 29]. Leptin, an adipocyte-derived sensor (HSP) is acting at or upstream of AKT and down- signal, alters nutrient flux such that energy expenditure is stream of the insulin receptor. Since active AKT isoforms favored over energy storage [30]. HSP flux regulating have also been implicated in preventing apoptosis [40, leptin secretion is in agreement with the model of the 41], reduced AKT activation under insulin-resistant con- HSP serving as an energy sensor and a negative feedback ditions may contribute to b-cell death in diabetes. Exces- system to limit uptake of glucose under hyperglycemic sive HSP flux has also been shown to induce apoptosis in and hyperinsulinemic conditions. retinal neurons but not in L6 muscle cells [42]. These data If the HSP hypothesis is correct, the next important ques- are consistent with the observed retinopathy often seen in tion is, How does the energy sensor (increased flux type 2 diabetes [2]. Also, Boehmelt and colleagues through the HSP resulting in elevated UDP-GlcNAc lev- showed that cells from Emeg32-deficient mice that are els) transduce the signal to cause insulin resistance? Un- defective in the synthesis of UDP-GlcNAc via the HSP der normal conditions, insulin induces glucose uptake have dramatically decreased UDP-GlcNAc levels, ex- and glycogen synthesis in skeletal muscle and adipocytes press activated AKT and have an increased capacity to (reviewed in [4]). While the mechanism of this signal withstand apoptotic stimuli [43]. The reason increased transduction cascade has not been completely elucidated, HSP flux may be inducing apoptosis in some cell types several key components and pathways have been identi- (retinal neurons and b-cells) but not others (skeletal mus- fied (fig. 1). Glut4-containing vesicles translocate and cle and adipocytes) is completely unknown and is a ques- fuse with the plasma membrane in response to insulin tion open for investigation. Insulin resistance also impairs stimulation, and it is primarily the Glut4 glucose trans- glycogen synthesis [44]. Under normal conditions, in- porter that is responsible for insulin-dependent glucose sulin stimulation activates AKT, leading to phosphoryla- uptake in adipocytes and skeletal muscle. Several groups tion and deactivation of GSK3b (fig. 1). This results in CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 225 glycogen synthase being active since it is no longer deac- also been shown to elevate O-GlcNAc levels on certain tivated by phosphorylation. Because excess HSP flux proteins [58, 59]. Finally, the diabetes-inducing reagent leads to defective AKT activation upon insulin stimula- streptozotocin raises O-GlcNAc levels on proteins in b- tion, glycogen synthesis is inhibited as well since GSK3b cells [58]. is not efficiently phosphorylated and thus can inhibit As an aside, the attractive hypothesis was put forward that glycogen synthase. streptozotocin was inducing b-cell death by inhibiting O- Excessive flux through the HSP serves as an energy sen- GlcNAcase [60]. However, we and others, using the more sor that appears to be mediating its effect, at least in part, potent O-GlcNAcase inhibitor PUGNAc [61], have shown by inhibiting insulin signal transduction at or upstream of that elevated O-GlcNAc levels alone do not induce apop- AKT. Thus, many investigators have established the HSP tosis in b-cells [62 – 64]. Streptozotocin is a potent alky- pathway as a sensor and implicated specific molecular lating reagent and is thought to induce cell death via DNA defects leading to insulin resistance. But what is the damage [65]. Streptozotocin, however, is also a weak O- mechanism by which increased flux inhibits insulin sig- GlcNAcase inhibitor that is capable of raising O-GlcNAc naling? Since the vast majority of carbohydrate entering levels on proteins in cells [66]. Thus, the possibility exists the HSP is rapidly converted to UDP-GlcNAc, several in- that the combination of O-GlcNAcase inhibition and vestigators have proposed that glycosylation may be the DNA damage is necessary to induce b-cell death. mediator of insulin resistance [4, 31, 45]. O-GlcNAc While there was a strong correlation with elevated O-Glc- modification of nucleocytoplasmic proteins is one possi- NAc levels and insulin resistance, until recently there was ble candidate for the mediator. no direct proof for O-GlcNAc levels modulating insulin action. As a f irst step towards addressing whether O-Glc- NAc modif ication of proteins was directly modulating in- O-GlcNAc’s role in nutrient sensing sulin signaling, we elevated O-GlcNAc levels in 3T3-L1 and insulin resistance adipocytes via treatment of the cells with the O-Glc- NAcase inhibitor PUGNAc [67]. PUGNAc treatment sig- O-GlcNAc modification has several features that distin- nificantly elevated the O-GlcNAc modification on many guish it from classical glycosylation and make it an at- nucleocytoplasmic proteins. More important, elevation of tractive target for the molecular mechanism by which in- O-GlcNAc levels impaired insulin-stimulated glucose crease flux via the HSP could inhibit insulin signaling uptake in the cells. Thus, using a pharmacological ap- [46]. The covalent modification of serine and threonine proach, we have established a direct causal relationship hydroxyls on nuclear and cytosolic proteins by b-linked between elevated O-GlcNAc and insulin resistance in O-GlcNAc was described by Torres and Hart in 1984 3T3-L1 adipocytes. We also found that proximal insulin [47]. Several recent reviews have focused on the proper- signaling was unaffected, while AKT phosphorylation ties of the modif ication and the various proteins modif ied and activation (as measured by GSK3b phosphorylation) [48 – 53]. Briefly, O-GlcNAc has several distinguishing was impaired. Thus, elevation of O-GlcNAc levels was characteristics that make it more analogous to phospho- not only causing insulin resistance but appeared to be in- rylation than to classical complex glycosylation. Namely hibiting insulin signaling at the same point in the pathway (i) O-GlcNAc is attached to nucleocytoplasmic proteins as increased flux through the HSP [37, 67]. We were also or to the cytosolic portions of membrane bound proteins, able to show that IRS-1 and b-catenin were modified by (ii) the modification involves the attachment of a single O-GlcNAc in a PUGNAc-dependent fashion in the 3T3- sugar from a high-energy donor (UDP-GlcNAc) that is L1 adipocytes. These findings allow us to put forth a not elongated, (iii) the modification is dynamic, (iv) the working model in which O-GlcNAc acts as a mediator of enzymes responsible for its attachment (O-GlcNAc trans- insulin resistance as well as a metabolic sensor. ferase, OGT) and removal (O-GlcNAcase) are nucleocy- toplasmic, (v) the modification is inducible and (vi) the modification competes with phosphorylation for the Future directions same sites on certain proteins. Underscoring the impor- tance of this modification, embryonic stem cells of mice One interesting question that remains to be answered is, lacking OGT fail to survive [54]. OGT has been found to What is the role of hyperinsulinemia in peripheral insulin be responsive to a wide range of UDP-GlcNAc concen- resistance? Marshall and colleagues established that glu- trations [55]. Yki-Jarvinen and colleagues found that in- cose, glutamine and insulin were necessary for the induc- ducing insulin resistance in rats by glucosamine and hy- tion of insulin resistance in adipocytes [13]. Glucose and perinsulinemia led to elevated levels of O-GlcNAc on glutamine are necessary for elevation of flux through the skeletal muscle proteins [56]. It was also established that HSP, leading to elevated UDP-GlcNAc levels, and can be IRS-1 became O-GlcNAc modified in response to in- substituted for by glucosamine; however, chronic insulin crease flux through the HSP [57]. Hyperglycemia has treatment is still necessary for inducing insulin resis- 226 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor tance. Interestingly, we found that PUGNAc treatment GlcNAc modification and leptin expression, secretion alone was suff icient to induce insulin resistance, and glu- and action remains to be elucidated. cosamine treatment plus insulin elevated O-GlcNAc lev- O-GlcNAc has begun to emerge as an important posttrans- els higher than either alone [67]. Buse and colleagues lational modification in cellular regulation [46, 50]. Many showed that glucosamine treatment alone, which did not tools to facilitate study of this modification have only re- cause insulin resistance, was able to substantially elevate cently become available, including the cloning of OGT in UDP-GlcNAc levels in adipocytes and that the addition 1997 [68, 69] and O-GlcNAcase in 2001 [70], the devel- of insulin induced a two-fold increase in the levels of opment of a general O-GlcNAc specific antibody in 2001 UDP-GlcNAc [18]. Thus, it would appear that PUGNAc [71] and O-GlcNAc site-specific antibodies in 2002 [72], is acting at a point after the convergence of hyper- and mass spectrometry based techniques for site mapping glycemia and hyperinsulinemia signaling. The fact that [73 – 75]. All of these developments should help to acceler- glucosamine treatment still requires hyperinsulinemia ate our understanding of the O-GlcNAc modification. would suggest that insulin is acting downstream of GFAT. In conclusion, we and others are using a working model Based on the above data and since glucosamine-6-phos- that O-GlcNAc modification of proteins serves as a nu- phate is rapidly converted to UDP-GlcNAc, we would hy- trient sensor to modulate extracellular signal transduction pothesize that insulin is acting directly on either OGT or cascades. In this model, the cell is not blindly responding O-GlcNAcase. This hypothesis is currently under investi- to extracellular signals but instead is taking into account gation, and preliminary experiments show that OGT be- its own energy stores and responding appropriately. Fur- comes tyrosine phosphorylated in response to insulin ther, O-GlcNAc, serving as a metabolic sensor, may be stimulation [L. Wells, K. Vosseller, G. W. Hart, unpub- modulating the expression, activity, localization and/or lished results]. secretion of proteins. There have been several other mol- Testing the role of O-GlcNAc levels regulating insulin ecules and mechanisms proposed as nutrient sensors [4, signaling by genetic means as opposed to pharmacologi- 7– 11], and it is likely that cells use multiple sensors and cal methods would greatly strengthen the argument [67 a, mechanisms to regulate such an important process as en- published during revision of this manuscript]. In this re- ergy homeostasis. Deciphering how these sensors are me- gard, reduced expression and overexpression of OGT and diating their effects is an important goal if we are to un- O-GlcNAcase in adipocytes is under development. Test- derstand the pathophysiology of complex pleiotropic ing the role of O-GlcNAc on insulin action in other tissue metabolic diseases, such as type 2 diabetes. An under- types as well as whole animals is also an important area standing of the relationship between O-GlcNAc and O- of future research. Obviously, identification of proteins phosphate in metabolic sensing, signaling cascades and that become modified by O-GlcNAc in response to hy- transcriptional regulation should lead to rational new tar- perglycemia and hyperinsulinemia is important, as well gets for drug development. as the identif ication of proteins that are hyper-O-GlcNAc Acknowledgements. We would like to thank Natasha Zachara and modified in rodent models of type 2 diabetes. Once can- Karen Wells for critical reading of the manuscript. Because the f ield didate proteins are identified, the elucidation of the mol- of metabolism and diabetes is so well studied, we had to pick and ecular mechanism by which O-GlcNAc modification is choose articles to cite. We apologize to our colleagues whose con- inhibiting insulin signaling will be of great importance in tributions to the field were not cited. This work was supported by national research service award fellowships to L. W. (CA83261) understanding this posttranslational modification and to and K. V. (GM20528) and by the national institutes of health future therapies for diabetes. NIDDK grant to G. W. H. (DK61671). The role of O-GlcNAc in the liver and the b-cell, as well as in microvascular and macrovascular disease, remains to be determined. In this regard, exploring whether O- 1Kruszynska Y. T. and Olefsky J. M. (1996) Cellular and molec- GlcNAc plays a role in apoptosis is especially important. ular mechanisms of non-insulin dependent diabetes mellitus. J. Investig. Med. 44: 413 – 428 The HSP appears to be involved in b-cell death and 2 Brownlee M. (2001) Biochemistry and molecular cell biology retinopathy [42, 58]. Also, we have determined that O- of diabetic complications. Nature 414: 813 – 820 GlcNAcase is cleaved in vitro by caspase-3, an execu- 3 O’Doherty R., Stein D. and Foley J. (1997) Insulin resistance. tioner caspase in programmed cell death [52]. Activation Diabetologia 40 Suppl. 3: B10–15 4Patti M. E. (1999) Nutrient modulation of cellular insulin ac- of AKT is also inhibited by elevated flux through the HSP tion. Ann. N. Y. Acad. Sci. 892: 187 – 203 and elevation of O-GlcNAc levels [37, 67]. Active AKT 5 McClain D. A. and Crook E. D. (1996) Hexosamines and in- has been demonstrated to be antiapoptotic [40, 41], and sulin resistance. Diabetes 45: 1003 – 1009 cells with reduced UDP-GlcNAc levels have elevated lev- 6 McClain D. A. (2002) Hexosamines as mediators of nutrient sensing and regulation in diabetes. J. Diabetes Complications els of active AKT and are refractory to apoptotic stimuli 16: 72–80 [43]. Thus, there is compelling evidence for the investi- 7Schmitz-Peiffer C. (2000) Signalling aspects of insulin resis- gation of a possible role for O-GlcNAc in apoptosis. Also, tance in skeletal muscle: mechanisms induced by lipid over- determining whether there is a connection between O- supply. Cell Signal. 12: 583 – 594 CMLS, Cell. Mol. Life Sci. Vol. 60, 2003 Multi-author Review Article 227 8 Hussain M. A. (1998) A case for glucosamine. Eur. J. En- thesis pathway in muscles of insulin-resistant ob/ob mice. Am. docrinol. 139: 472 – 475 J. Physiol. 272: E1080 – 1088 9 Rossetti L. (2000) Perspective: hexosamines and nutrient sens- 28 Wang J., Liu R., Hawkins M., Barzilai N. and Rossetti L. (1998) ing. Endocrinology 141: 1922 – 1925 A nutrient-sensing pathway regulates leptin gene expression in 10 Ruderman N. B. and Dean D. (1998) Malonyl CoA, long chain muscle and fat. Nature 393: 684 – 688 fatty acyl CoA and insulin resistance in skeletal muscle. J. Ba- 29 McClain D. A., Alexander T., Cooksey R. C. and Considine R. sic Clin. Physiol. Pharmacol. 9: 295 – 308 V. (2000) Hexosamines stimulate leptin production in trans- 11 Ukropec J., Sebokova E. and Klimes I. (2001) Nutrient sensing, genic mice. Endocrinology 141: 1999 – 2002 leptin and insulin action. Arch. Physiol. Biochem. 109: 38–51 30 Flier J. S. (1997) Leptin expression and action: new experi- 12 Marshall S., Bacote V. and Traxinger R. R. (1991) Complete in- mental paradigms. Proc. Natl. Acad. Sci. USA 94: 4242 – 4245 hibition of glucose-induced desensitization of the glucose 31 Robinson K. A., Sens D. A. and Buse M. G. (1993) Pre-expo- transport system by inhibitors of mRNA synthesis. Evidence sure to glucosamine induces insulin resistance of glucose trans- for rapid turnover of glutamine:fructose-6-phosphate amido- port and glycogen synthesis in isolated rat skeletal muscles. transferase. J. Biol. Chem. 266: 10155 – 10161 Study of mechanisms in muscle and in rat-1 f ibroblasts overex- 13 Marshall S., Garvey W. T. and Traxinger R. R. (1991) New in- pressing the human insulin receptor. Diabetes 42: 1333 – 1346 sights into the metabolic regulation of insulin action and insulin 32 Ross S. A., Chen X., Hope H. R., Sun S., McMahon E. G., resistance: role of glucose and amino acids. FASEB J. 5: 3031 – Broschat K. et al. (2000) Development and comparison of two 3036 3T3-L1 adipocyte models of insulin resistance: increased glu- 14 Marshall S., Bacote V. and Traxinger R. R. (1991) Discovery of cose flux vs glucosamine treatment. Biochem. Biophys. Res. a metabolic pathway mediating glucose-induced desensitiza- Commun. 273: 1033 – 1041 tion of the glucose transport system. Role of hexosamine 33 Chen H., Ing B. L., Robinson K. A., Feagin A. C., Buse M. G. biosynthesis in the induction of insulin resistance. J. Biol. and Quon M. J. (1997) Effects of overexpression of gluta- Chem. 266: 4706 – 4712 mine : fructose-6-phosphate amidotransferase (GFAT) and glu- 15. Hresko R. C., Heimberg H., Chi M. M. and Mueckler M. cosamine treatment on translocation of GLUT4 in rat adipose (1998) Glucosamine-induced insulin resistance in 3T3-L1 cells. Mol. Cell. Endocrinol. 135: 67–77 adipocytes is caused by depletion of intracellular ATP. J. Biol. 34 Thomson M. J., Williams M. G. and Frost S. C. (1997) Devel- Chem. 273: 20658 – 20668 opment of insulin resistance in 3T3-L1 adipocytes. J. Biol. 16 Kang J., Heart E. and Sung C. K. (2001) Effects of cellular ATP Chem. 272: 7759 – 7764 depletion on glucose transport and insulin signaling in 3T3-L1 35 Nelson B. A., Robinson K. A. and Buse M. G. (2002) Insulin adipocytes. Am. J. Physiol. Endocrinol. Metab. 280: E428 – 435 acutely regulates Munc18-c subcellular trafficking: altered re- 17 Vemuri S. and Marchase R. B. (1999) The inhibition of capac- sponse in insulin-resistant 3T3-L1 adipocytes. J. Biol. Chem. itative calcium entry due to ATP depletion but not due to glu- 277: 3809 – 3812 cosamine is reversed by staurosporine. J. Biol. Chem. 274: 36 Cho H., Mu J., Kim J. K., Thorvaldsen J. L., Chu Q., Crenshaw 20165 – 20170 E. B. 3rd et al. (2001) Insulin resistance and a diabetes melli- 18 Nelson B. A., Robinson K. A. and Buse M. G. (2000) High glu- tus-like syndrome in mice lacking the protein kinase Akt2 cose and glucosamine induce insulin resistance via different (PKB beta). Science 292: 1728 – 1731 mechanisms in 3T3-L1 adipocytes. Diabetes 49: 981 – 991 37 Heart E., Choi W. S. and Sung C. K. (2000) Glucosamine-in- 19 Robinson K. A., Weinstein M. L., Lindenmayer G. E. and Buse duced insulin resistance in 3T3-L1 adipocytes. Am. J. Physiol. M. G. (1995) Effects of diabetes and hyperglycemia on the hex- Endocrinol. Metab. 278: E103 – 112 osamine synthesis pathway in rat muscle and liver. Diabetes 44: 38 Nelson B. A., Robinson K. A. and Buse M. G. (2002) Defective 1438 – 1446 Akt activation is associated with glucose- but not glucosamine- 20 Yki-Jarvinen H. and Makimattila S. (1997) Insulin resistance induced insulin resistance. Am. J. Physiol. Endocrinol. Metab. due to hyperglycaemia: an adaptation protecting insulin-sensi- 282: E497 – 506 tive tissues. Diabetologia 40 Suppl. 2: S141 – 144 39 Singh L. P., Gennerette D., Simmons S. and Crook E. D. (2001) 21 Yki-Jarvinen H., Daniels M. C., Virkamaki A., Makimattila S., Glucose-induced insulin resistance of phosphatidylinositol 3¢- DeFronzo R. A. and McClain D. (1996) Increased glu- OH kinase and AKT/PKB is mediated by the hexosamine tamine:fructose-6-phosphate amidotransferase activity in biosynthesis pathway. J. Diabetes Complications 15: 88–96 skeletal muscle of patients with NIDDM. Diabetes 45: 302 – 40 Lawlor M. A. and Alessi D. R. (2001) PKB/Akt: a key media- 307 tor of cell proliferation, survival and insulin responses? J. Cell. 22 Randle P. J., Garland P. B., Newsholme E. A. and Hales C. N. Sci. 114: 2903 – 2910 (1965) The glucose fatty acid cycle in obesity and maturity on- 41 Kandel E. S. and Hay N. (1999) The regulation and activities of set diabetes mellitus. Ann. N. Y. Acad. Sci. 131: 324 – 333 the multifunctional serine/threonine kinase Akt/PKB. Exp. Cell 23 Hawkins M., Barzilai N., Liu R., Hu M., Chen W. and Rossetti Res. 253: 210 – 229 L. (1997) Role of the glucosamine pathway in fat-induced in- 42 Nakamura M., Barber A. J., Antonetti D. A., LaNoue K. F., sulin resistance. J. Clin. Invest. 99: 2173 – 2182 Robinson K. A., Buse M. G. et al. (2001) Excessive hex- 24 Gazdag A. C., Wetter T. J., Davidson R. T., Robinson K. A., osamines block the neuroprotective effect of insulin and induce Buse M. G., Yee A. J. et al. (2000) Lower calorie intake en- apoptosis in retinal neurons. J. Biol. Chem. 276: 43748 – 43755 hances muscle insulin action and reduces hexosamine levels. 43 Boehmelt G., Wakeham A., Elia A., Sasaki T., Plyte S., Potter J. Am. J. Physiol. Regul. Integr. Comp. Physiol. 278: R504 – 512 et al. (2000) Decreased UDP-GlcNAc levels abrogate prolifer- 25 Hebert L. F. Jr, Daniels M. C., Zhou J., Crook E. D., Turner R. ation control in EMeg32-deficient cells. EMBO J. 19: 5092 – L., Simmons S. T. et al. (1996) Overexpression of glu- 5104 tamine:fructose-6-phosphate amidotransferase in transgenic 44 Ciaraldi T. P., Carter L., Nikoulina S., Mudaliar S., McClain D. mice leads to insulin resistance. J. Clin. Invest. 98: 930 – 936 A. and Henry R. R. (1999) Glucosamine regulation of glucose 26 Tang J., Neidigh J. L., Cooksey R. C. and McClain D. A. (2000) metabolism in cultured human skeletal muscle cells: divergent Transgenic mice with increased hexosamine flux specifically effects on glucose transport/phosphorylation and glycogen syn- targeted to beta-cells exhibit hyperinsulinemia and peripheral thase in non-diabetic and type 2 diabetic subjects. Endocrinol- insulin resistance. Diabetes 49: 1492 – 1499 ogy 140: 3971 – 3980 27 Buse M. G., Robinson K. A., Gettys T. W., McMahon E. G. and 45 Rossetti L., Hawkins M., Chen W., Gindi J. and Barzilai N. Gulve E. A. (1997) Increased activity of the hexosamine syn- (1995) In vivo glucosamine infusion induces insulin resistance 228 L. Wells, K. Vosseller and G. W. Hart O-GlcNAc as a nutrient sensor in normoglycemic but not in hyperglycemic conscious rats. J. 63 Kaneto H., Xu G., Song K. H., Suzuma K., Bonner-Weir S., Clin. Invest. 96: 132 – 140 Sharma A. et al. (2001) Activation of the hexosamine pathway 46 Wells L., Vosseller K. and Hart G. W. (2001) Glycosylation of leads to deterioration of pancreatic beta-cell function through nucleocytoplasmic proteins: signal transduction and O-Glc- the induction of oxidative stress. J. Biol. Chem. 276: 31099 – NAc. Science 291: 2376 – 2378 31104 47 Torres C. R. and Hart G. W. (1984) Topography and polypep- 64 Gao Y., Parker G. J. and Hart G. W. (2000) Streptozotocin-in- tide distribution of terminal N-acetylglucosamine residues on duced beta-cell death is independent of its inhibition of O-Glc- the surfaces of intact lymphocytes. Evidence for O-linked Glc- NAcase in pancreatic Min6 cells. Arch. Biochem. Biophys. NAc. J. Biol. Chem. 259: 3308 – 3317 383: 296 – 302 48 Hart G. W., Haltiwanger R. S., Holt G. D. and Kelly W. G. 65 Elsner M., Guldbakke B., Tiedge M., Munday R. and Lenzen S. (1989) Nucleoplasmic and cytoplasmic glycoproteins. Ciba (2000) Relative importance of transport and alkylation for pan- Found. Symp. 145: 102 – 112 creatic beta-cell toxicity of streptozotocin. Diabetologia 43: 49 Hanover J. A. (2001) Glycan-dependent signaling: O-linked N- 1528 – 1533 acetylglucosamine. FASEB J. 15: 1865 – 1876 66 Roos, M. D., Xie, W., Su, K., Clark, J. A., Yang, X., Chin, E. et 50 Zachara N. E. and Hart G. W. (2002) The emerging signif icance al. (1998) Streptozotocin, an analog of N-acetylglucosamine, of O-GlcNAc in cellular regulation. Chem. Rev. 102: 431 – 438 blocks the removal of O-GlcNAc from intracellular proteins. 51 Vosseller K., Wells L. and Hart G. W. (2001) Nucleocytoplas- Proc. Assoc. Am. Physicians 110: 422 – 432 mic O-glycosylation: O-GlcNAc and functional proteomics. 67 Vosseller K., Wells L., Lane M. D. and Hart G. W. (2002) Ele- Biochimie 83: 575 – 581 vated nucleocytoplasmic glycosylation by O-GlcNAc results in 52 Wells L., Gao Y., Mahoney J. A., Vosseller K., Chen C., Rosen insulin resistance associated with defects in Akt activation in A. et al. (2001) Dynamic O-glycosylation of nuclear and cy- 3T3-L1 adipocytes. Proc. Natl. Acad. Sci. USA 99: 5313 – 5318 tosolic proteins: further characterization of the nucleocyto- 67 a McClain D. A., Lubas W. A., Cooksey R. C., Hazel M., Parker plasmic beta-N-acetylglucosaminidase, O-GlcNAcase. J. Biol. G. J., Love D. C. et al. (2002) Altered glycan-dependent signal- Chem. 277: 5313 – 5318 ing induces insulin resistance and hyperleptinemia. Proc. Natl. 53 Comer F. I. and Hart G. W. (2000) O-Glycosylation of nuclear Acad. Sci. USA epub ahead of print and cytosolic proteins. Dynamic interplay between O-GlcNAc 68 Kreppel L. K., Blomberg M. A. and Hart G. W. (1997) Dynamic and O-phosphate. J. Biol. Chem. 275: 29179 – 29182 glycosylation of nuclear and cytosolic proteins. Cloning and 54 Shafi R., Iyer S. P., Ellies L. G., O’Donnell N., Marek K. W., Chui characterization of a unique O-GlcNAc transferase with multi- D. et al. (2000) The O-GlcNAc transferase gene resides on the X ple tetratricopeptide repeats. J. Biol. Chem. 272: 9308 – 9315 chromosome and is essential for embryonic stem cell viability and 69 Lubas W. A., Frank D. W., Krause M. and Hanover J. A. (1997) mouse ontogeny. Proc. Natl. Acad. Sci. USA 97: 5735 – 5739 O-Linked GlcNAc transferase is a conserved nucleocytoplas- 55 Kreppel L. K. and Hart G. W. (1999) Regulation of a cytosolic mic protein containing tetratricopeptide repeats. J. Biol. Chem. and nuclear O-GlcNAc transferase. Role of the tetratricopep- 272: 9316 – 9324 tide repeats. J. Biol. Chem. 274: 32015 – 32022 70 Gao Y., Wells L., Comer F. I., Parker G. J. and Hart G. W. (2001) 56 Yki-Jarvinen H., Virkamaki A., Daniels M. C., McClain D. and Dynamic O-glycosylation of nuclear and cytosolic proteins: Gottschalk W. K. (1998) Insulin and glucosamine infusions in- cloning and characterization of a neutral, cytosolic beta-N- crease O-linked N-acetyl-glucosamine in skeletal muscle pro- acetylglucosaminidase from human brain. J. Biol. Chem. 276: teins in vivo. Metabolism 47: 449 – 455 9838 – 9845 57 Patti M. E., Virkamaki A., Landaker E. J., Kahn C. R. and Yki- 71 Comer F. I., Vosseller K., Wells L., Accavitti M. A. and Hart G. Jarvinen H. (1999) Activation of the hexosamine pathway by W. (2001) Characterization of a mouse monoclonal antibody glucosamine in vivo induces insulin resistance of early postre- specific for O-linked N-acetylglucosamine. Anal. Biochem. ceptor insulin signaling events in skeletal muscle. Diabetes 48: 293: 169 – 177 1562 – 1571 72 Kamemura K., Hayes B. K., Comer F. I. and Hart G. W. (2002) 58 Hanover J. A., Lai Z., Lee G., Lubas W. A. and Sato S. M. Dynamic interplay between O-glycosylation and O-phosphory- (1999) Elevated O-linked N-acetylglucosamine metabolism in lation of nucleocytoplasmic proteins. Alternative glycosyla- pancreatic beta-cells. Arch. Biochem. Biophys. 362: 38–45 tion/phosphorylation of Thr58, a known mutational hot spot of 59 Han I., Oh E. S. and Kudlow J. E. (2000) Responsiveness of the c-Myc in lymphomas, is regulated by mitogens. J. Biol. Chem. state of O-linked N-acetylglucosamine modif ication of nuclear 277: 19229 – 19235 pore protein p62 to the extracellular glucose concentration. 73 Greis K. D., Hayes B. K., Comer F. I., Kirk M., Barnes S., Biochem. J. 350 Pt 1: 109 – 114 Lowary T. L. et al. (1996) Selective detection and site-analysis 60 Liu K., Paterson A. J., Chin E. and Kudlow J. E. (2000) Glucose of O-GlcNAc-modified glycopeptides by beta-elimination and stimulates protein modification by O-linked GlcNAc in pan- tandem electrospray mass spectrometry. Anal. Biochem. 234: creatic beta cells: linkage of O-linked GlcNAc to beta cell 38–49 death. Proc. Natl. Acad. Sci. USA 97: 2820 – 2825 74 Haynes P. A. and Aebersold R. (2000) Simultaneous detection 61 Haltiwanger R. S., Grove K. and Philipsberg G. A. (1998) Mod- and identification of O-GlcNAc-modified glycoproteins us- ulation of O-linked N-acetylglucosamine levels on nuclear and ing liquid chromatography-tandem mass spectrometry. Anal. cytoplasmic proteins in vivo using the peptide O-GlcNAc-beta- Chem. 72: 5402 – 5410 N-acetylglucosaminidase inhibitor O-(2-acetamido-2-deoxy- 75 Wells L., Vosseller K., Cole R. N., Cronshaw J. M., Matunis M. D-glucopyranosylidene)amino-N-phenylcarbamate. J. Biol. J. and Hart G. W. (2002) Mapping sites of O-GlcNAc modifi- Chem. 273: 3611 – 3617 cation using affinity tags for serine and threonine post-transla- 62 Okuyama R. and Yachi M. (2001) Cytosolic O-GlcNAc accu- tional modifications. Mol. Cell Proteomics 1: 791 – 804 mulation is not involved in beta-cell death in HIT-T15 or Min6. Biochem. Biophys. Res. Commun. 287: 366 – 371

Journal

Cellular and Molecular Life Sciences – Springer Journals

Published: Feb 1, 2003

Keywords: Key words. O-GlcNAc; glycosylation; posttranslational modification; hexosamine biosynthetic pathway; glucosamine; hyperglycemia; insulin resistance; type 2 diabetes.

There are no references for this article.