Streptozotocin-Induced b-Cell Death Is Independent of ItsI

a

ivn

Archives of Biochemistry and BiophysicsVol. 383, No. 2, November 15, pp. 296–302, 2000doi:10.1006/abbi.2000.2094, available online at http://www.idealibrary.com on

nhibition of O-GlcNAcase in Pancreatic Min6 Cells

Yuan Gao, Glendon J. Parker, and Gerald W. Hart1

Department of Biological Chemistry, The Johns Hopkins University School of Medicine,725 North Wolfe Street, Baltimore, Maryland 21205-2185

Received August 8, 2000

pancreatic b-cells. Chronic autoimmune reaction, viral

Streptozotocin (STZ) injection into experimental an- infections, and environmental toxins have all been im-

imals selectively causes massive b-cell death. Themechanism of this specific toxicity is not fully under-stood. Recently, it has been discovered that O-linkedN-acetylglucosamine (O-GlcNAc) is enriched in theb-cells. It has been proposed that STZ toxicity may bedue to its inhibition of neutral O-GlcNAcase activity,the enzyme that removes O-GlcNAc from cytosolic pro-teins (K. Liu et al., 2000, Proc. Natl. Acad. Sci. USA 97,2820–2825). To further ascertain the role of O-GlcNA-case in b-cell death, we have used PUGNAc, a potent

nd specific O-GlcNAcase inhibitor, together with STZin pancreatic Min6 cells. Both STZ and PUGNAc in-creased O-GlcNAc to similar levels on intracellularproteins. STZ, but not PUGNAc, decreased cellularprotein synthesis by 66.0% within 8 h, killed 80.9% ofthe cells within 18 h, and decreased insulin secretion.STZ, but not PUGNAc, also caused genomic DNA frag-mentation, suggesting that some of the cells were un-dergoing apoptosis. Prolonged treatment with PUG-NAc (72 h) maintained high intracellular O-GlcNAclevels, but did not result in any apparent cell damage.Furthermore, the toxicity of STZ can be largely re-versed by 3-aminobenzamide, a poly(ADP-ribose) poly-merase inhibitor. These data strongly indicate thatSTZ-induced b-cell death is not caused by elevatedntracellular O-GlcNAc levels, but instead likely in-olves poly(ADP-ribose) polymerase in the mecha-ism. © 2000 Academic Press

Key Words: streptozotocin; O-GlcNAc; O-GlcNAcase;PUGNAc; glucosamine; b-cell death; diabetes.

The hallmark event in the development of type Idiabetes mellitus is the destruction of insulin-secreting

1 To whom correspondence should be addressed. Fax: (410) 614-8804. E-mail: gwhart@jhmi.edu.

296

plicated in this self-destruction (1). In the laboratory,STZ2, a b-cell-specific toxin, has been widely used togenerate diabetic animals as a model system to studytype I diabetes mellitus (2). A single high dose, ormultiple low doses of STZ injection produces massiveb-cell death (3), resulting in a steady increase of bloodglucose levels.

Despite the extensive use of STZ, the mechanism bywhich it specifically causes b-cell death is not com-pletely understood. However, STZ is known to decom-pose intracellularly upon entry and causes damage toDNA either by alkylation or by the generation of nitricoxide (4, 5). This DNA breakage activates the nuclearPARP (EC 2.4.2.30) (6). PARP synthesizes largeamounts of the (ADP-ribose) polymer, using NAD1 as asubstrate, and therefore can deplete intracellularNAD1 and ATP, and leads to cell death by apoptosis ornecrosis (7). PARP inhibitors such as ABA or nicotin-amide prevent the depletion of NAD1 and the STZ-induced cell death (8). Transgenic mice in which thePARP gene is disrupted become resistant to STZ andtherefore fail to develop diabetes (9–11).

Another line of evidence suggests a role for a noveltype of protein posttranslational modification in STZ-induced b-cell death. This modification is the additionof a single O-linked N-acetylglucosamine moiety (O-

2 Abbreviations used: ABA, 3-aminobenzamide; BSA, bovine se-rum albumin; DMSO, dimethylsulfoxide; DON, diazo-ozo-norleucine;GFAT, glutamine:fructose-6-phosphate amidotransferase; FBS, fetalbovine serum; Glut, glucose transporter; IRS, insulin receptor sub-strate; O-GlcNAc, O-linked N-acetylglucosamine; O-GlcNAcase,N-acetyl-b-D-glucosaminidase; OGT, O-GlcNAc transferase; PAGE,polyacrylamide gel electrophoresis; PARP, poly(ADP-ribose) poly-merase; PBS, phosphate-buffered saline; PUGNAc, O-(2-acetamido-2-deoxy-D-glucopyranosylidene)-amino-N-phenylcarbamate; PVDF,polyvinyldene difluoride; SDS, sodium dodecyl sulfate; STZ, strepto-zotozin.

0003-9861/00 $35.00Copyright © 2000 by Academic Press

All rights of reproduction in any form reserved.

GlcNAc) to serine or threonine residues (12, 13). O-GlcNAc transferase (OGT) catalyzes the saccharide’s

mt

oIC(c

m

t

aAf

n

titbRmm

wh0

297STREPTOZOTOCIN-INDUCED b-CELL DEATH AND O-GlcNAc

attachment, and O-GlcNAcase catalyzes its detach-ent in a manner analogous to protein phosphoryla-

ion/dephosphorylation. O-GlcNAc has been found on anumber of proteins, such as transcription factors, on-cogenes, RNA polymerase, and nuclear pore proteins(13, 14). Evidence has been mounting that O-GlcNAcplays a regulatory role, such as mediating in protein–protein interaction, assisting in protein translocationinto the nucleus and competing with phosphorylationsites. The importance of this modification is furthermanifested by the lethality of OGT gene knockout inmouse embryonic stem cells (15). Recently, it has beendiscovered that the endocrine cells of the pancreas areenriched for OGT transcripts and OGT protein relativeto other tissues examined (16–18). Furthermore, STZis a close analog of GlcNAc (Fig. 1a) and an inhibitor ofO-GlcNAcase (17, 19). It has also been reported thatSTZ-induced b-cell death may be the consequence ofaccumulation of O-GlcNAc on proteins through theinhibition of O-GlcNAcase activity (20). In the presentstudy, we further examined this appealing hypothesisby using a potent and specific O-GlcNAcase inhibitor,PUGNAc. We found that there is no correlation be-tween elevated intracellular O-GlcNAc levels andb-cell toxicity. It is concluded that STZ-induced celldeath is largely caused by a mechanism other than theinhibition of O-GlcNAcase.

MATERIALS AND METHODS

Materials. STZ and ABA were purchased from Sigma (St. Louis,MO). PUGNAc was purchased from Carbogen (Switzerland). Stocksolutions of STZ (0.5 M), ABA (2 M in DMSO), and PUGNAc (20 mM)were made immediately prior to use. All chromatographic materialsfor O-GlcNAcase purification and in vivo labeling grade [35S]methi-nine were from Amersham Pharmacia Biotech (Piscataway, NJ).nsulin radioimmunoassay kit was from Linco Research Inc. (St.harles, MI). Tissue culture reagents were from Life Technologies

Gaithersburg, MD). All other chemicals were of the highest purityommercially available.Tissue culture. Min6 cells, a mouse b-cell line (21), were grown inonolayer in DMEM (4.5 g/L glucose), 15% heat-inactivated FBS, 60

mM 2-mercaptoethanol and antibiotics (penicillin, 75 mg/mL; strep-omycin, 50 mg/mL) in 5% CO2/95% air at 37°C. The cells were split

weekly at a ratio of 1:5. All cells used in this work were at passages18–23.

Purification and assay of neutral O-GlcNAcase. Cytosolic neutralO-GlcNAcase was purified from bovine brain according to a modifiedprocedure (22). Briefly, O-GlcNAcase in bovine brain homogenatewas purified by precipitation with 30–50% ammonium sulfate, fol-lowed by sequential Sepharose Q ion exchange, Con A Sepharose 4B,Phenyl-Sepharose, and Mono-Q chromatography. O-GlcNAcase ac-tivity was assayed in 50 mM sodium cacodylate (pH 6.5), 2 mMp-nitrophenyl-N-acetyl-b-D-glucosaminide (pNP-GlcNAc), 50 mM N-cetylgalactosamine and 3% BSA for 1 h at 37°C as described (22).ppropriate concentrations of effectors were included in the assays

or inhibition studies where indicated.Isolation of nuclear extract and Western blot. Cytoplasmic and

uclear proteins were prepared as described (23). Protein concentra-

ion was determined by Bio-Rad assay using BSA as standards. Formmunoblot, the proteins were separated on 10% SDS–PAGE andhen electroblotted onto PVDF membrane. The membrane waslocked in 5% (w/v) nonfat milk, incubated with monoclonal antibodyL-2, and subsequently with appropriate secondary antibody. Im-unoreactive proteins were visualized using an enhanced chemilu-inescent (ECL) detection system.DNA fragmentation analysis. Following treatment of Min6 cellsith STZ or PUGNAc for 8 h, cells were washed twice with cold PBS,arvested, and lysed with 20 mM Tris (pH 7.8), 2 mM EDTA and.5% (v/v) Triton X-100 on ice for 15 min with intermittent vortexing

FIG. 1. The chemical structures (a) of GlcNAc, STZ, and PUGNAcand their inhibition of neutral O-GlcNAcase activity (b). The bondlengths in the schematic structures are not drawn to scale. O-Glc-NAcase was partially purified from bovine brain and the activity wasassayed with 2 mM pNP-GlcNAc as substrate at pH 6.5 as describedunder Materials and Methods.

(24). The lysate was centrifuged at 15,000g for 10 min, low-molecu-lar-weight DNA in the supernatant was extracted with phenol/chlo-roform and precipitated with ethanol. Contaminating RNA was elim-

ePuwa

298 GAO, PARKER, AND HART

inated by RNase digestion. Fragmented DNA was separated on 1.5%agarose gel, stained with ethidium bromide and photographed underUV illumination.

Protein synthesis analysis and cell survival. Following treatmentof subconfluent Min6 cells with STZ or PUGNAc for 8 h, proteinsynthesis was performed for 20 min in six-well plates at 37°C using[35S]methionine at 25 mCi/mL as described (25). The incorporationfficiency was 5–10%. After labeling, the cells were washed in coldBS, harvested, and lysed by three cycles of freezing/thawing. Sol-ble proteins were precipitated with 10% (v/v) TCA. The pellet wasashed once in 10% TCA and resuspended in 0.1 N NaOH. Anliquote was counted for 35S using a liquid scintillation counter. For

cell survival assay, subconfluent cells were treated with indicatedeffectors for 18 h, dead cells were washed off with PBS, and live cellswere trypsinized and counted using trypan blue assay. The averagevalues in the absence of any effector were arbitrarily set as 100%. Alltreatments had four replicates.

Insulin secretion assay. Fifty to 60% confluent Min6 cells in 12-well plates were pretreated with STZ or PUGNAc for 8 or 18 h. Thecells were washed twice with DMEM containing 0.5 mM glucose and15% FBS and incubated in this medium for 1 h and then aspirated.DMEM (0.6 mL) containing 25 mM glucose and 5% FBS was addedto each well and the cells were maintained in this medium for 60 minat 37°C before the medium was harvested for insulin assay using arat insulin radioimmunoassay kit.

RESULTS

STZ and PUGNAc, both GlcNAc Analogs, InhibitO-GlcNAcase Activity

The dynamics of the posttranslational O-GlcNAcmodification is catalyzed by OGT and O-GlcNAcase,which attach and remove the O-GlcNAc moiety fromproteins, respectively (12, 13). Previous Northern blot,in situ hybridization, and fluorescence studies haveshown that OGT is highly expressed in pancreaticb-cells (16–18). This high OGT activity leads to en-riched O-GlcNAc levels in these cells, as shown by RL-2staining (19). This evidence suggests a role for O-Glc-NAc in b-cell functions.

Although O-GlcNAcase has yet to be cloned, the en-zyme has been extensively purified and characterized(22). O-GlcNAcase is selectively inhibited by GlcNAcand its analogs with varying potencies. Recently, it hasbeen realized that the b-cell toxin STZ is more pre-cisely a GlcNAc analog with a nitroso group, ratherthan a glucose analog (Fig. 1a). It is, as expected, amoderate inhibitor of O-GlcNAcase from differentsources such as spleen and brain (17, 19). With thepartially purified O-GlcNAcase from bovine brain, wewere able to confirm the relative effectiveness of Glc-NAc, STZ, and PUGNAc as inhibitors. GlcNAc and STZhad Ki values of 16.3 and 2.5 mM, respectively (Fig.1b). PUGNAc, a GlcNAc analog whose efficacy in vitroand in vivo has been documented (22, 26), is a potentinhibitor with a Ki value of 0.22 mM (Fig. 1b). In sub-sequent studies, PUGNAc, together with STZ, were

employed to upregulate intracellular O-GlcNAc levelsand to compare their relative toxicity on b-cells.

Both STZ and PUGNAc Increased IntracellularO-GlcNAc Levels in Min6 Cells

It has been shown previously that cell-permeableSTZ and PUGNAc increase protein O-GlcNAcylation invivo through the inhibition of O-GlcNAcase. PUGNAchas been used successfully in a number of cell lines (26)and infusion of STZ increases O-GlcNAc levels in ani-mals (19). In the present study, subconfluent Min6cells were treated with 5 mM STZ or 100 mM PUGNAcfor 8 h. O-GlcNAc levels were determined on nuclearproteins using RL-2, an antibody that specifically rec-ognizes the O-GlcNAc moiety attached to a subset ofproteins (27). As shown in Fig. 2a, STZ and PUGNActreatments increased O-GlcNAc levels significantlyand to similar levels.

In addition to elevating O-GlcNAc levels, STZ alsoinduced significant protein changes in the nucleus (Fig.2b) but caused little change in the cytoplasm (data notshown). Specifically, STZ decreased the intensities of

FIG. 2. Both STZ and PUGNAc increased GlcNAc levels on intra-cellular proteins (a) and STZ, but not PUGNAc, induced proteinchanges (b). Cells were treated with 0, 2, or 5 mM STZ or 100 mMPUGNAc for 8 h in DMEM containing 2 mM glucose and 15% FBS.Nuclear extract was prepared, separated on 10% SDS–PAGE, andstained by Commassie brilliant blue R-250 (b), or the proteins wereblotted to PVDF membrane and terminal GlcNAc was detected usingmonoclonal RL-2 antibody (a). 20 and 10 mg nuclear protein wasloaded per lane in a and b, respectively. Arrows indicate changes ofspecific proteins.

two proteins with apparent sizes of 60 and 78 kDa,respectively, while it increased the intensity of a 48-

299STREPTOZOTOCIN-INDUCED b-CELL DEATH AND O-GlcNAc

kDa protein. PUGNAc did not cause any apparentchanges in the protein patterns.

STZ, But Not PUGNAc, Was Toxic to Min6 Cells

In experimental animals and insulinoma cells, STZoften causes b-cell damage and death by apoptosis ornecrosis, depending on the dosage (3, 7). So we com-pared the effect of STZ and PUGNAc on cell viability byassaying protein synthesis, cell survival and their in-sulin secretion. We first tested the effectiveness ofthese inhibitors on cells grown in DMEM media con-taining high (25 mM) or low (2 mM) glucose. STZ (5mM) in the presence of 25 mM glucose had limitedeffect on protein synthesis of the cells (12% decrease).This was in contrast to low glucose in which STZ de-creased protein synthesis by 66% (Figs. 3a and 3b).This disparity can be explained by the fact that bothglucose and STZ compete for transport by Glut2 inb-cells (28), so high glucose concentration decreasesthe uptake of STZ. When the dosage of STZ was exam-ined at 2 mM glucose, STZ steadily decreased proteinsynthesis with a half maximal concentration of 3.0 mM(Fig. 3c). In contrast to STZ, PUGNAc at both glucoseconcentrations did not have any effect on protein syn-thesis (Figs. 3a and 3b).

Although STZ caused a dramatic decrease in proteinsynthesis, it killed only about 10% of the cells withinan 8-h incubation period as measured by standardtrypan blue assay (data not shown). However, extend-ing the incubation time to 18 h resulted in massive celldeath, as evident under a light microscope. The extentof cell survival was 69.6 and 19.9% with 2 and 5 mMSTZ, respectively. Again, PUGNAc had no effect on celldeath (Fig. 3d).

Despite the drastic decreases in protein synthesis,insulin secretion by the cells was affected to a muchlesser extent following 8 h STZ treatment (Fig. 3e). Thedata seem to suggest that STZ has a differential effecton general protein synthesis and insulin secretion. Infact, previous studies have shown that the insulin se-cretory apparatus is intact in STZ-treated islets undersome circumstances (29). Following an 18-h treatment,insulin secretion was decreased by 26.1 and 66.2% with2 and 5 mM STZ, respectively. This decrease mayprimarily represent a change in number of cells sur-viving the treatment. PUGNAc had little effect on in-sulin secretion (Figs. 3e and 3f).

STZ, But Not PUGNAc, Caused DNA Fragmentationin Min6 Cells

Next, we compared the effect of STZ and PUGNAc onDNA fragmentation. STZ often induces DNA damage

in the b-cells, resulting in a ladder pattern of DNA onagarose gel typical for apoptosis (7, 30). This effect ofSTZ was confirmed in Min6 cells, with higher STZconcentration causing more severe fragmentation (Fig.4). The background fragmentation in the control cellswas likely to be due to the DNA isolation procedureadopted, as has been observed previously (24). In cellstreated with PUGNAc, the fragmentation was only atthe level of the background.

Prolonged PUGNAc Treatment Maintained HighO-GlcNAc Levels without Any Cell Toxicity

Although PUGNAc is nontoxic in a number of celllines examined (26) and appears to be so in Min6 cellsin the short term, its long-term effect on b-cells has notbeen determined. As shown in Fig. 5a, PUGNAc main-tained high O-GlcNAc levels throughout the time

FIG. 3. STZ, but not PUGNAc treatment, decreased protein syn-thesis (a, b, c), caused massive cell death (d), and reduced insulinsecretion (e, f) in Min6 cells. Subconfluent Min6 cells were treatedwith indicated effectors in DMEM containing 25 mM (a) or 2 mMglucose (b, c, d, e, f) for 8 h (a, b, c, e) or 18 h (d, f). [35S]methionineincorporation into protein synthesis, cell survival rate, and insulinsecretion were determined as described under Materials andMethods.

spmfip

transferase (GFAT) being the rate-limiting step (32). Inperipheral tissues such as muscles and adipocytes, thehrcmpiG(

MPw

300 GAO, PARKER, AND HART

course. However, protein synthesis of the cells showedno significant changes (Fig. 5b). Light microscopic ex-amination of cells treated with or without PUGNAcshowed no difference either (data not shown).

STZ-Induced Cell Death Can Be Reversed by ABA

It is known that STZ upregulates the activity ofPARP, which may be the cause of b-cell death (30).PARP activity can be inhibited by compounds such asABA and nicotinamide (8, 30, 31). We tested if ABAcould reverse the toxic effect of STZ in Min6 cells. Overshort incubation periods (8 h), 5 mM STZ reducedcellular protein synthesis to 38.7% of the control rate(Fig. 6a). However, this toxicity was completely pre-vented by 10 mM ABA. Over longer incubation periods(18 h), 10 mM ABA increased the cell survival ratefrom 27.8 to 74.8% (Fig. 6b). This protective effect ofABA was not a result of altered cellular O-GlcNAclevels, as ABA had no effect on O-GlcNAcase activity(Fig. 6c), and ABA did not change intracellular O-GlcNAc levels as determined by RL-2 Western blot ofthe nuclear proteins (Fig. 6d).

DISCUSSION

The high expression of OGT in the b-cells (17, 18),as well as the dynamic nature of O-GlcNAc in re-ponse to effectors (19), suggest a role for O-linkedrotein GlcNAcylation in glucose sensing in order toaintain glucose homeostasis. The donor substrate

or this protein modification is UDP-GlcNAc, whichs synthesized by the hexsosamine biosyntheticathway with glutamine:fructose-6-phosphate amido-

FIG. 4. STZ, but not PUGNAc, induced DNA fragmentation inin6 cells. Min6 cells were treated with 0, 2, or 5 mM STZ or 100 mMUGNAc for 8 h in DMEM containing 2 mM glucose. Low-molecular-eight genomic DNA was isolated and analyzed on 1.5% agarose gel.

exsosamine biosynthetic pathway may regulate theate of glucose transport. Treatment of primary adipo-ytes with glucosamine or overexpression GFAT inice, both of which increase the flux of the hexosamine

athway, results in insulin resistance (33–35). In somenstances the insulin resistance can be reversed by theFAT inhibitor azaserine or diazo-oxo-norleucine

DON) (33). Despite this abundant information, O-Gl-cNAc has not yet been directly linked to the develop-ment of insulin resistance. Nevertheless, some keycomponents in the insulin signaling pathway, such asIRS-1 and IRS-2, have been shown to be modified byO-GlcNAc (34).

Recently it has been shown that STZ blocks theremoval of O-GlcNAc from proteins, resulting in anaccumulation of intracellular O-GlcNAc in cells (17,19). It has been hypothesized that STZ-induced b-celltoxicity is the consequence of an elevation in O-GlcNAc(20). We have further examined this hypothesis using apotent O-GlcNAcase inhibitor, PUGNAc. Unlike STZ,

FIG. 5. Prolonged treatment of Min6 cells with PUGNAc main-tained high GlcNAc levels but did not produce any apparent celldamage. Min6 cells were treated with 100 mM PUGNAc for 24, 48, or72 h in DMEM containing 2 mM glucose. The medium was replen-ished daily. O-GlcNAc on nuclear proteins was detected by Westernblot with antibody RL-2 (a) and cell damage was assessed by [35S]me-thionine incorporation into protein synthesis (b).

asSO

transgenic mice is an important consideration. Theexpression and/or translocation of glucose transporters

A

t(po

301STREPTOZOTOCIN-INDUCED b-CELL DEATH AND O-GlcNAc

which is also an alkalyting agent and releases nitricoxide, PUGNAc is not highly chemically reactive incells (26). We have shown that both STZ and PUGNAcupregulate intracellular O-GlcNAc levels to a similarextent in Min6 b-cells. However, the toxic effects, suchs breaking DNA strands, decreasing protein synthe-is, and killing the cells, are exclusively confined toTZ. The data strongly argue against a role of elevated-GlcNAc in STZ-induced b-cell death. The absence of

a role of GlcNAc in b-cell dysfunction induced by STZ isfurther supported by mice overexpressing GFAT.These mice exhibit an increase in cellular UDP-GlcNAclevels and O-GlcNAc-modified proteins, but the b-cellsdo not undergo pathogenesis. In contrast, they arecapable of secreting twice as much insulin than in thewild-type mice (35).

The work of Liu et al. (20) shows that mice overex-pressing GFAT antisense are resistant to STZ toxicityand the toxic effect of STZ can be potentiated by infu-sion of high concentrations of glucose or glucosamine.This study leaves two open questions that need to befurther studied. First, since STZ is taken up by Glut2in the b-cells (28), the status of Glut2 in the GFAT

FIG. 6. STZ-induced cell damage can be effectively protected byBA. Min6 cells were treated for 8 h with ABA and STZ before

protein synthesis were assayed by [35S]methionine labeling (a) orreated for 18 h to determine cell survival rate by trypan blue assayb). ABA had no effect on the activity of O-GlcNAcase partiallyurified from bovine brain (c). RL-2 immunoblot of nuclear extractsf Min6 cells treated with STZ and ABA for 8 h (d).

are affected by a range of factors. For example, admin-istration of STZ to mice decreases Glut2 gene expres-sion in the b-cells (36, 37), and the translocation ofGlut4 to the plasma membrane is impaired in trans-genic mice overexpressing GFAT or adipose cellstreated with glucosamine (38, 39). In view of this evi-dence, the expression of Glut2 and its translocation tothe plasma membrane should be examined in theb-cells of the mice overexpressing GFAT antisense. Itis possible that overexpression of GFAT antisense al-ters Glut2 expression and/or its translocation so thetransport of STZ is impaired, which in turn, leads toSTZ resistance.

Second, the suggestion that the effect of STZ is po-tentiated by high glucose or glucosamine infusion maynot be justified (20). High-glucose concentration alonehas been shown to induce b-cell apoptosis in rodents orin cultured islets (40, 41). Glucosamine infusion alsoexerts general toxicity, perhaps through the inhibitionof glucokinase (42). In fact, Liu et al. (20) began toobserve islet pathology 8 h after glucose infusion, andglucosamine infusion produced clear islet pathology bythis time. Unfortunately, the effects of high glucoseand glucosamine concentration on islets 26 h aftertheir infusion, the time when the islets were examinedafter STZ infusion, were not reported. Without firstdocumenting the independent effects of high glucoseand glucosamine infusions, the apparent increase inb-cell apoptosis by concurrent treatment with STZ andglucose or glucosamine cannot be ascribed to their po-tentiating effects.

Recent studies provide good clues to the understand-ing of the molecular basis of STZ induced b-cell death.STZ may have multiple effects. First of all, it severelyinhibits Glut2 gene expression and glucokinase activ-ity, resulting in reduced glucose uptake and metabo-lism (37). Furthermore, STZ releases nitric oxide (NO)intracellularly, causes massive DNA damage, and ac-tivates PARP (4, 43). This effect of NO has been repro-duced by structurally unrelated but NO-releasing com-pound such as interleukin-1b (44, 45). The activation ofPARP consumes large amount of NAD1 and ATP. Thisprocess, in conjunction with reduced glucokinase activ-ity and glucose influx, may quickly lead to cell starva-tion and death. Independent studies have shown thatPARP-deficient mice are resistant to STZ (9–11). Theeffect of STZ on cultured b-cells can be effectively re-versed in the short term by the PARP inhibitors suchas ABA (8, Fig. 6), and the sensitivity of mouse strainsto STZ has been correlated to PARP activity (46). WhileO-GlcNAc may play a role in insulin resistance associ-ated with type II diabetes (32), our results suggest thatthe increase in O-GlcNAc in b-cells is not directly in-volved in the mechanism of STZ toxicity.

ACKNOWLEDGMENTS

The authors thank Dr. J. Miyazaki for providing Min6 cells and

1

1

11

1

1

1

1

22. Dong, D. L., and Hart, G. W. (1994) J. Biol. Chem. 269, 19321–19330.

2

2

2

2

2

3

3

3

3

3

3

33

3

4

4

44

302 GAO, PARKER, AND HART

Dr. N. Zachara for reading the manuscript. Supported by the Juve-nile Diabetes Foundation and Fifty-Fifty Foods, and by NIH grantRO1HD13563 to G.W.H. G.W.H. discloses that he is a member of theScientific Advisory Board of Oxford Glycosciences.

REFERENCES

1. Simone, E., and Eisenbarth, G. S. (1996) Horm. Metab. Res. 28,332–336.

2. Kolb, H. (1987) Diabetes Metab. Rev. 3, 751–778.3. O’Brien, B. A., Harmon, B. V., Cameron, D. P., and Allan, D. J.

(1996) J. Pathol. 178, 176–181.4. Turk, J., Corbett, J. A., Ramanadham, S., Bohrer, A., and Mc-

Daniel, M. L. (1993) Biochem. Biophys. Res. Commun. 197,1458–1464.

5. Kaneto, H., Fuji, J., Seo, H. G., Suzuki, K., Matsuoka, T., Naka-mura, M., Tatsumi, H., Yamasaki, Y., Kamada, T., and Tanigu-chi, N. (1995) Diabetes 44, 733–738.

6. Radons, J., Heller, B., Burkle, A., Hartmann, B., Rodriguez,M. L., Kroncke, K. D., Burkart, V., and Kolb, H. (1994) Biochem.Biophys. Res. Commun. 199, 1270–1277.

7. Saini, K. S., Thompson, C., Winterford, C. M., Walker, N. I., andCameron, D. P. (1996) Biochem. Mol. Biol. Int. 39, 1229–1236.

8. Uchigata, Y., Yamamoto, H., Kawamura, A., and Okamoto, H.(1982) J. Biol. Chem. 257, 6084–6088.

9. Burkart, V., Wang, Z. Q., Radons, J., Heller, B., Herceg, Z., Stingl,L., Wagner, E. F., and Kolb, H. (1999) Nat. Med. 5, 314–319.

0. Masutani, M., Suzuki. H., Kamada, N., Watanabe, M., Ueda, O.,Nozaki, T., Jishage, K., Watanabe, T., Sugimoto, T., Nakagama,H., Ochiya, T., and Sugimura, T. (1999) Proc. Natl. Acad. Sci.USA 96, 2301–2304.

1. Pieper, A. A., Brat, D. J., Krug, D. K., Watkins, C. C., Gupta, A.,Blackshaw, S., Verma, A., Wang, Z. Q., and Snyder, S. H. (1999)Proc. Natl. Acad. Sci. USA 96, 3059–3064.

2. Hart, G. W. (1997) Annu. Rev. Biochem. 66, 315–335.3. Comer, F. I., and Hart, G. W. (2000) J. Biol. Chem. 275, 29179–

29182.4. Comer, F. I., and Hart, G. W. (1999) Biochim. Biophys. Acta

1473, 161–171.5. Shafi, R., Iyer, S. P., Ellies, L. G., O’Donnell, N., Marek, K. W.,

Chui, D., Hart, G. W., and Marth, J. D. (2000) Proc. Natl. Acad.Sci. USA 97, 5735–5739.

6. Lubas, W. A., Frank, D. W., Krause, M., and Hanover, J. A.(1997) J. Biol. Chem. 272, 9316–9324.

7. Hanover, J. A., Lai, Z., Lee. G., Lubas, W. A., and Sato, S. M.(1999) Arch. Biochem. Biophys. 362, 38–45.

18. Akimoto, Y., Kreppel, L. K., Hirano, H., and Hart, G. W. (1999)Diabetes 48, 2407–2413.

19. Roos, M. D., Xie, W., Su, K., Clark, J. A., Yang, X., Chin, E.,Paterson, A. J., and Kudlow, J. E. (1998) Proc. Assoc. Am. Phy-sicians 110, 422–432.

20. Liu, K., Paterson, A. J., Chin, E., and Kudlow, J. E. (2000) Proc.Natl. Acad. Sci. USA 97, 2820–2825.

21. Miyazaki, J., Araki, K., Yamato, E., Ikegami, H., Asano, T.,Shibasaki, Y., Oka, Y., and Yamamura, K. (1990) Endocrinology127, 126–132.

23. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W.(1989) Nucl. Acids Res. 17, 6419.

24. de Murcia, J. M., Niedergang, C., Trucco, C., Ricoul, M.,Dutrillaux, B., Mark, M., Oliver, F. J., Masson, M., Die-rich, A., LeMeur, M., Walztinger, C., Chambon, P., and deMurcia, G. (1997) Proc. Natl. Acad. Sci. USA 94, 7303–7307.

5. Bonifacino, J. S. (1998) in Current Protocol in Molecular Bio-lolgy (Ausubel, F. M., et al., Eds.), Chap. 10.18. Wiley, NewYork.

6. Haltiwanger, R. S., Grove, K., and Philipsberg, G. A. (1998)J. Biol. Chem. 273, 3611–3617.

7. Holt, G. D., Snow, C. M., Senior, A., Haltiwanger, R. S., andHart, G. W. (1987) J. Cell Biol. 104, 1157–1164.

8. Schnedl, W. J., Ferber, S., Johnson, J., and Newgard, C. B.(1994) Diabetes 43, 1326–1333.

9. Eizirik, D. L., Sandler, S., Welsh, N., and Hellerstrom, C. (1988)Endocrinology 122, 1242–1249.

0. Yamamoto, H., Uchigata, Y., and Okamoto, H. (1981) Nature294, 284–286.

1. Mizumoto, K., and Farber, J. L. (1995) Arch. Biochem. Biophys.319, 512–518.

2. McClain, D. A., and Crook, E. D. (1996) Diabetes 45, 1003–1009.

3. Marshall, S., Bacote, V., and Traxinger, R. R. (1991) J. Biol.Chem. 266, 4706–4712.

4. Patti, M. E., Virkamaki, A., Landaker, E. J., Kahn, C. R., andYki-Jarvinen, H. Diabetes 48, 1562–1571.

5. Hebert, L. F. Jr, Daniels, M. C., Zhou, J., Crook, E. D., Turner,R. L., Simmons, S. T., Neidigh, J. L., Zhu, J. S., Baron, A. D., andMcClain, D. A. (1996) J. Clin. Invest. 98, 930–936.

6. Wang, Z., and Gleichmann, H. (1998) Diabetes 47, 50–56.7. Park, C., Kim, J. R., Shim, J. K., Kang, B. S., Park, Y. G., Nam,

K. S., Lee, Y. C., and Kim, C. H. (1999) Arch. Biochem. Biophys.362, 217–224.

8. Chen, H., Ing, B. L., Robinson, K. A., Feagin, A. C., Buse, M. G.,and Quon, M. J. (1997) Mol. Cell Endocrinol. 135, 67–77.

39. Cooksey, R. C., Hebert, L. F. Jr, Zhu, J. H., Wofford, P.,Garvey, W. T., and McClain, D. A. (1999) Endocrinology 140,1151–1157.

0. Donath, M. Y., Gross, D. J., Cerasi, E., and Kaiser, N. (1999)Diabetes 48, 738–744.

1. Efanova, I. B., Zaitsev, S. V., Zhivotovsky, B., Kohler, M., Efen-dic, S., Orrenius, S., and Berggren, P. O. (1998) J. Biol. Chem.273, 33501–33507.

2. Balkan, B., and Dunning, B. E. (1994) Diabetes 43, 1173–1179.3. Kroncke, K. D., Fehsel, K., Sommer, A., Rodriguez, M. L., and

Kolb-Bachofen, V. (1995) Biol. Chem. 376, 179–185.44. Suarez-Pinzon, W. L., Strynadka, K., Schulz, R., and Rabino-

vitch, A. (1994) Endocrinology 134, 1006–1010.45. Delaney, C. A., Green, M. H., Lowe, J. E., and Green, I. C. (1993)

FEBS Lett. 333, 291–295.46. Cardinal, J. W., Allan, D. J., and Cameron, D. P. (1999) J. Mol.

Endocrinol. 22, 65–70.