Heat Shock Inhibits Cytokine-Induced Nitric Oxide ... - CiteSeerX

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society

Vol. 139, No. 12 Printed in U.S.A.

Heat Shock Inhibits Cytokine-Induced Nitric Oxide Synthase Expression by Rat and Human Islets* ANNA L. SCARIM, MONIQUE R. HEITMEIER,

AND

JOHN A. CORBETT†

Edward A. Doisy Department of Biochemistry and Molecular Biology, Saint Louis University School of Medicine, St. Louis, Missouri 63104 ABSTRACT In this study the effects of heat shock on interleukin-1b (IL-1)induced inhibition of islet metabolic function were examined. Treatment of rat islets for 18 h with IL-1 results in a potent inhibition of glucose-stimulated insulin secretion. The inhibitory effects of IL-1 on insulin secretion are completely prevented if islets are pretreated for 60 min at 42 C before cytokine stimulation. Heat shock also prevents IL-1-induced inhibition of insulinoma RINm5F cell mitochondrial aconitase activity. The protective effects of heat shock on islet metabolic function are associated with the inhibition of IL-1-stimulated inducible nitric oxide synthase (iNOS or NOS II) expression. Islets heat shocked for 60 min at 42 C fail to express iNOS (messenger RNA or protein) or produce nitrite in response to IL-1. IL-1-induced iNOS expression by rat islets requires activation of the transcriptional regulator nuclear factor kB (NF-kB). Heat shock prevents IL-1-

D

URING the development of insulin-dependent diabetes mellitus insulin-secreting b-cells found in pancreatic islets of Langerhans are selectively destroyed. Much evidence supports an effector role for cytokines in mediating b-cell destruction (1, 2). Incubation of rat islets with interleukin-1b (IL-1b) results in the expression of the inducible form of nitric oxide synthase (iNOS), increased production of nitric oxide (NO), and potent inhibition of glucose-stimulated insulin secretion (3–7). Inhibitors of iNOS, aminoguanidine (8), and NG-monomethyl-L-arginine (NMMA) (9) completely prevent IL-1-induced inhibition of glucose-stimulated insulin secretion by rat islets. Similar to rat islets, incubation of human islets with a combination of IL-1 1 interferon g (IFN-g) 1 tumor necrosis factor (TNF) results in the expression of iNOS, increased production of NO (9), and an inhibition of insulin secretion (10). NMMA prevents cytokine-induced nitrite production and attenuates the inhibitory effects on insulin secretion, suggesting that NO mediates, in part, the inhibitory actions of cytokines on human islet function (10). NO participates in cytokine-induced inhibition of insulin secretion by targeting and inhibiting the enzymatic activities of mitochondrial iron-sulfur-containing enzymes, specifically aconitase and the electron transport Received May 12, 1998. Address all correspondence and requests for reprints to: John A. Corbett, Saint Louis University School of Medicine, Department of Biochemistry and Molecular Biology, 1402 South Grand Boulevard, St. Louis, Missouri 63104. E-mail: [email protected]. * This work was supported by research grants from the NIH (DK52194) and The Tobacco Research Council † Supported by a Career Development Award from the Juvenile Diabetes Foundation International.

induced NF-kB nuclear localization by inhibiting inhibitory protein kB (IkB) degradation in rat islets. Similar to rat islets, heat shock (stimulated by 90 min incubation at 42 C) prevents IL-1 1 interferon g-induced iNOS expression and NF-kB nuclear localization in human islets. IL-1 also stimulates heat-shock protein 70 (hsp 70) expression by rat islets, and hsp 70 expression is dependent on islet production of nitric oxide. Last, evidence is presented that implicates nitric oxide as a stimulus for the expression of proteins that participate in islet recovery from nitric oxide-mediated damage. These studies indicate that heat shock prevents cytokine-induced islet damage by inhibiting iNOS expression, and suggest that nitric oxide is one effector molecule that stimulates the expression of factors involved in b-cell recovery from nitric oxide-mediated damage. (Endocrinology 139: 5050 –5057, 1998)

chain at complexes I and II (6, 11). Treatment of rat or human islets with IL-1b or IL-1b 1 IFN-g, respectively, results in a potent inhibition of aconitase that is completely prevented by NMMA (12). The inhibitory effects of IL-1 on islet mitochondrial aconitase activity and insulin secretion are reversible. Rat islets incubated with IL-1b for 15 h require a 4-day incubation in cytokine-free media to recover normal glucose-stimulated insulin secretion (13, 14). The recovery of islet secretory function can be reduced to 8 h by the inhibition of iNOS. The addition of NMMA to islets pretreated for 18 h with IL-1b, followed by an 8 h incubation in the presence of both IL-1b and NMMA, results in complete recovery of insulin secretion (15). The recovery of insulin secretion correlates with a simultaneous recovery of mitochondrial aconitase activity. In addition, the recovery of islet metabolic function is prevented by the transcriptional inhibitor, actinomycin D (12, 15). These findings suggest that NO, IL-1, or both, stimulate the expression of yet to be identified proteins that participate in islet recovery from NO-mediated damage. Islets exposed to heat shock appear to be resistant to cellular death induced by NO donor compounds (16). Heatshock protein (hsp) 70 is one member of the heat-shock family of proteins whose expression is highly induced under conditions of cellular stress (17). Kolb and co-workers (18) showed that overexpression of hsp 70 prevents RINm5F cell lysis in response to sodium nitroprusside. Cytokines have also been shown to stimulate hsp 70 expression by rat islets, however the mechanism of cytokine-induced hsp 70 expression is unknown (19, 20). In addition, the effects of heat shock on IL-1-induced iNOS expression and NO production by islets have not been examined. In this study we show that

5050

HEAT SHOCK INHIBITS CYTOKINE-INDUCED iNOS EXPRESSION

heat shock prevents iNOS expression and nitrite production by rat and human islets by inhibiting cytokine-induced activation of the transcriptional regulator NF-kB. In addition, evidence is presented that implicates NO as the stimulus for IL-1-induced hsp 70 expression by rat islets. Materials and Methods Materials RINm5F cells were from the Washington University Tissue Culture Support Center (St. Louis, MO). Human islets were obtained from the Islet Isolation Core Facility at Washington University School of Medicine and the Diabetes Research Institute at the University of Miami (Miami, FL). [a-32P]Deoxycytidine triphosphate and enhanced chemiluminescence reagents were purchased from Amersham Pharmacia Biotech (Arlington Heights, IL). NF-kB consensus oligonucleotide and rabbit anti-IkB-a antiserum were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Horseradish peroxidase-conjugated donkey antirabbit IgG was obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Rabbit antiserum specific for the C-terminal 27 amino acids of mouse macrophage iNOS was a gift from Dr. Thomas Misko (G.D. Searle & Co., St. Louis, MO). iNOS and cyclophilin cDNAs were gifts from Dr. Charles Rodi (Monsanto Corporate Research, St. Louis, MO) and Dr. Steve Carroll (Department of Pathology, University of Alabama-Birmingham), respectively. Collagenase type XI was purchased from Sigma Chemical Co. (St. Louis, MO). RPMI 1640, CMRL1066, penicillin, streptomycin, and human recombinant IFN-g were from Gibco BRL (Grand Island, NY). FCS was obtained from Hyclone Laboratories, Inc. (Logan, UT). Human recombinant IL-1b was obtained from Cistron Biotechnology (Pine Brook, NJ). NMMA was purchased from Calbiochem (San Diego, CA). All other reagents were from commercially available sources.

Islet isolation and culture Islets were isolated from 250 –300 g male Sprague-Dawley rats (Harlan Sprague Dawley, Inc., Indianapolis, IN) by collagenase digestion as described previously (21). After isolation, islets were cultured overnight in complete CMRL-1066 media (CMRL-1066 media supplemented with 2 mm l-glutamine, 10% heat-inactivated FCS serum, 100 U/ml penicillin, and 100 mg/ml streptomycin) under an atmosphere of 95% air and 5% CO2 at 37 C.

Heat shock The heat-shock response was induced in RINm5F cells and isolated rat and human islets by culturing for 30, 60, or 90 min at 42 C, respectively. The islets and cells were then incubated at 37 C for an additional 3 h to allow for recovery from heat shock. Unless otherwise indicated, experiments were initiated by the addition of 1 U/ml IL-1 for RINm5F cells and rat islets or 75 U/ml IL-1 1 750 U/ml IFN-g for human islets.

Nitrite formation Nitrite formation was measured by mixing 50 ml culture media with 50 ml Griess reagent (22). The absorbance was measured at 540 nm, and nitrite concentrations were calculated from a sodium nitrite standard curve.

5051

Islet dispersion Isolated islets (;1000/condition) were dispersed into individual cells by treatment with trypsin (1 mg/ml) in Ca12- and Mg12-free Hanks solution at 37 C for 3 min as described previously (12). The dispersed islets were counted, plated (1 3 106 cells/ml of complete CMRL-1066), and allowed to incubate overnight before initiation of experiments.

Aconitase activity Heat shocked and control RINm5F cells (5 3 106 cells/2 ml complete CMRL-1066) were incubated in the presence or absence of 1 U/ml IL-1b and 0.5 mm NMMA for 18 h. Dispersed rat islets (1 3 106 cells/2 ml complete CMRL-1066) were incubated for 1 h with the NO donor compound, sodium (Z)-1(N, N-diethylamino) diazen-1-ium-1,2-diolate (DEA-NO), in the presence or absence of actinomycin D, or incubated for 1 h with DEA-NO, then washed three times with complete CMRL1066 to remove the DEA-NO, and cultured for an additional 5 h in the presence or absence of actinomycin D. After culture, the cells were isolated by centrifugation, and mitochondrial aconitase activity was measured at 340 nm as described previously (6, 12).

Western blot analysis Following cytokine treatment, islets were washed three times in 0.1 m PBS, pH 7.4. The islet pellet was dissolved in 20 ml SDS sample mix (0.25 m Tris-HCl, 20% b-mercaptoethanol, and 4% SDS) and 10 ml H2O and boiled for 4 min. Proteins (10 mg/lane) were separated on 10% SDS-polyacrylamide gels by the method of Laemmli (24) and transferred to nitrocellulose membranes under semidry transfer conditions. Blots were blocked in TBST (20 mm Tris, 500 mm NaCl, and 0.1% Tween 20, pH 7.5) containing 5% nonfat dry milk overnight at 4 C and then incubated for 1.5 h at room temperature with rabbit antihuman IkB-a (1:5000 dilution), mouse antirat hsp 70 (1:2000 dilution), rabbit antimouse iNOS (1:2000 dilution), or rabbit antihuman iNOS (1:800 dilution) in TBST containing 1% nonfat dry milk as indicated. The blots were then washed three times with TBST for 5 min per wash, followed by incubation with horseradish peroxidase-conjugated donkey antirabbit or donkey antimouse secondary antibody (1:5000 dilution) for 40 min. The blots were washed three times with TBST followed by target protein detection by enhanced chemiluminescence (Amersham Pharmacia Biotech).

Northern blot analysis Control or heat-shocked islets (1000 islets/condition) were incubated for 6 h in the presence or absence of 1 U/ml IL-1b at 37 C. The islets were isolated by centrifugation, washed three times with 0.1 m PBS, and total RNA was isolated using the RNeasy kit (Qiagen, Chatsworth, CA). Total RNA (5–10 mg) was denatured and fractionated by gel electrophoresis using a 1.0% agarose gel containing 2.2 m formaldehyde. RNA was transferred by capillary action to a Duralon UV nylon membrane (Stratagene, La Jolla, CA), and the membrane was then hybridized to a 32 P-labeled DNA probe for rat iNOS or cyclophilin. The probe was radiolabeled with [a-32P]deoxycytidine triphosphate by random priming using the Prime-a-Gene nick translation system from Promega Corp. (Madison, WI). iNOS DNA corresponds to bases 509-1415 of rat iNOS coding region. Cyclophilin was used as an internal control for RNA loading. Hybridization and autoradiography were performed as described previously (25, 26).

Nuclear extraction and gel shift analysis Glucose-stimulated insulin secretion Following cytokine stimulation, islets were washed three times with Krebs-Ringer bicarbonate buffer (25 mm HEPES, 115 mm NaCl, 24 mm NaHCO3, 5 mm KCl, 1 mm MgCl2, and 2.5 mm CaCl2, pH 7.4) containing 3 mm d-glucose and 0.1% BSA. Groups of 20 islets were counted into 10 mm 3 75 mm borosilicate tubes and preincubated for 30 min in 200 ml of the same buffer at 37 C with shaking. The preincubation solution was removed, and glucose-stimulated insulin secretion was initiated by the addition of 200 ml Krebs-Ringer bicarbonate buffer containing either 3 mm or 20 mm d-glucose. Islets were incubated for 30 min, incubation media was removed, and insulin content was determined by RIA (23).

Nuclear proteins were extracted from 1 3 106 dispersed islet cells using previously described methods (27). Protein concentrations were determined by the Pierce bicinchonic acid-microassay (Pierce Chemical Co., Rockford, IL). Double-stranded synthetic oligonucleotide probe for NF-kB was end labeled using [g-32P]ATP and T4 polynucleotide kinase (Promega Corp.). Binding reactions contained 10 mg nuclear extracts, 0.5–1 ng oligonucleotide probe, 2 mg/ml poly(dI-dC) (Boehringer Mannheim, Indianapolis, IN) in a buffer containing 20 mm HEPES, pH 7.8, 50 mm NaCl, 1 mm DTT, 1 mm EDTA, and 5% glycerol and incubated for 20 min at 30 C. DNA and protein complexes were resolved on 4% Tris-glycine nondenaturing polyacrylamide gels using a Tris-glycine

5052


buffer system (0.05 m Tris-HCl, pH 8.3, 0.38 m glycine, 2 mm EDTA, 28). After electrophoresis, gels were dried and exposed to photographic film.

Statistical analysis densitometry and image analysis Statistical comparisons were made between groups using a one-way ANOVA. Significant differences between groups were determined by Scheffe´’s F test poct hoc analysis. Autoradiograms were scanned using a COHU high-performance charged coupled device camera (Brookfield, WI), and densities were determined using NIH Image version 1.59 software.

Results IL-1 and DEA-NO stimulate hsp 70 expression by rat islets

Previous studies showed that IL-1 stimulates hsp 70 expression by rat islets (19, 20). Consistent with these previous studies, we show that treatment of rat islets for 18 h with 1 U/ml IL-1 results in the expression of hsp 70, as determined by Western blot analysis (Fig. 1a). The iNOS inhibitor NMMA completely prevents IL-1-induced nitrite production (data not shown) and hsp 70 expression by rat islets. These results indicate that IL-1-induced hsp 70 expression requires the endogenous production of NO by rat islets. To confirm that NO stimulates hsp 70, rat islets were treated with or without the NO donor compound DEA-NO for 1 h. As shown in Fig. 1b, DEA-NO (500 mm) stimulates the expression of hsp 70 by rat islets. We also compared the effects of IL-1 and heat shock on hsp 70 expression by rat islets (Fig. 1a). Islets subjected to a 60-min incubation at 42 C (heat shock) express 5.6 6 1.6-fold higher levels of hsp 70 than the levels stimulated by IL-1 (quantitated by densitometry), and neither IL-1 nor NMMA further modulate hsp 70 expression following heat shock (6.6 6 1.3 and 6.4 6 2.3-fold increases above IL-1-induced hsp 70 expression, respectively).

FIG. 1. Effects of heat shock, IL-1, and DEA-NO on hsp 70 expression by rat islets. a, Control or heat-shocked rat islets (100 islets/400 ml complete CMRL-1066) were cultured for 18 h with 1 U/ml IL-1 and 0.5 mM NMMA. b, Rat islets were cultured for 1 h with or without 500 mM DEA-NO. Islets were then isolated, and hsp 70 expression was determined by Western blot analysis as in Materials and Methods. Results are representative of three individual experiments.

Endo • 1998 Vol 139 • No 12

Effects of heat shock on cytokine-induced iNOS messenger RNA (mRNA) and protein expression and nitrite production by rat islets

Because hsp 70 overexpression prevents NO-mediated RINm5F cell death (18), and IL-1-induced islet damage is due to the expression of iNOS and the increased production of NO by b-cells (1, 29, 30), we evaluated the effects of heat shock on IL-1-induced iNOS expression. Treatment of rat islets for 6 h with IL-1b results in the accumulation of iNOS mRNA, as determined by Northern blot analysis (Fig. 2a). However, IL-1 fails to stimulate iNOS mRNA accumulation in rat islets pretreated for 60 min at 42 C before cytokine stimulation. Similar to the inhibitory effects of heat shock on

FIG. 2. Effects of heat shock on IL-1-induced iNOS expression by rat islets. a, Control or heat-shocked rat islets (1000/condition) were treated for 6 h with or without 1 U/ml IL-1. Total islet RNA was isolated, and iNOS mRNA accumulation was determined by Northern analysis. Cyclophilin mRNA accumulation is shown as a control for RNA loading. Control and heat-shocked and rat islets (100/condition) were treated for 18 h with 1 U/ml IL-1, islets were isolated, and iNOS protein expression determined by Western blot analysis (b); nitrite released into culture supernatant was determined by Griess assay (c). Results are representative of three individual experiments (a, b), or average 6 SEM of three individual experiments (c). Significantly different from untreated control: *, P , 0.05.


IL-1-induced iNOS mRNA accumulation, islets pretreated for 60 min at 42 C before IL-1 stimulation fail to express iNOS at the protein level (Fig. 2b) or to produce nitrite (Fig. 2c). The inhibitory effects of heat shock on IL-1-induced iNOS expression do not appear to be associated with heat-shockinduced inhibition of protein synthesis. IL-1-induced iNOS mRNA expression does not require de novo protein synthesis (28, 30), and [35S]methionine incorporation is nearly identical in control and rat islets treated for 60 min at 42 C before IL-1 stimulation (data not shown). These findings indicate that heat shock prevents cytokine-induced iNOS expression and nitrite production by rat islets at sites upstream of IL1-induced iNOS mRNA expression. Heat shock inhibits cytokine-induced IkB degradation and NF-kB nuclear translocation by rat islets

NF-kB activation and nuclear localization is one signaling event associated with IL-1-induced iNOS expression by rat islets (28, 31, 32). Because heat shock inhibits IL-1-induced iNOS expression by rat islets, the effects of heat shock on NF-kB activation was evaluated by gel shift analysis. Treatment of rat islets for 30 min with IL-1b stimulates the nuclear localization of NF-kB as shown by the reduced mobility of the NF-kB and DNA probe complex (Fig. 3a). However, NF-kB nuclear localization is attenuated in islets pretreated for 60 min at 42 C before IL-1 stimulation. The specificity of the NF-kB/DNA interaction was confirmed by inhibition of complex formation in the presence of 100-fold excess unla-

FIG. 3. Effects of heat shock on IL-1-induced NF-kB activation and IkB degradation in rat islets. a, Control and heat-shocked rat islets (1000/condition) were treated for 30 min with or without 1 U/ml IL-1. Islets were isolated, and NF-kB nuclear localization was examined by gel shift analysis. b, Control and heat-shocked islets (100/condition) were treated with or without 1 U/ml IL-1 for 30 min. Islets were isolated, lysed in SDS sample mix, and immunoreactive IkB was detected by Western blot analysis. Results are representative of three independent experiments.

5053

beled NF-kB oligonucleotide DNA (data not shown). Also, inclusion of antiserum specific for NF-kB (p65) in the binding reaction further retards the migration of the NF-kB and DNA complex (supershift; data not shown). These findings indicate that one mechanism by which heat shock prevents IL1-induced iNOS expression by rat islets is by the inhibition of IL-1-induced NF-kB nuclear localization. In unstimulated cells, NF-kB is found in the cytoplasm as an inactive heterodimer of p50/p65 subunits bound to the inhibitory protein, IkB. Upon stimulation, IkB is phosphorylated on specific serine residues that target IkB for degradation (33). To determine whether heat shock inhibits NF-kB nuclear localization by preventing IL-1-induced IkB degradation, the effects of a 60-min incubation at 42 C on IL1-induced IkB degradation by rat islets were examined. Treatment of rat islets for 30 min with IL-1b results in complete degradation of IkB (Fig. 3b). However, IL-1-induced IkB degradation is attenuated in rat islets subjected to heat shock before cytokine stimulation. These findings suggest that heat shock inhibits cytokine-induced NF-kB nuclear localization by a mechanism that involves the inhibition of IkB degradation. Protective effects of heat shock on IL-1-induced inhibition of mitochondrial aconitase activity and glucose-stimulated insulin secretion

The mechanism by which IL-1 inhibits glucose-stimulated insulin secretion involves the expression of iNOS by b-cells followed by NO-mediated inhibition of b-cell oxidative metabolism (1, 29, 30). Therefore, if heat shock prevents IL- 1-induced iNOS expression, it should also prevent the damaging effects of IL-1 on aconitase activity and insulin secretion by rat islets. Treatment of rat insulinoma RINm5F cells for 18 h with 1 U/ml IL-1 results in a potent inhibition of mitochondrial aconitase activity that is completely prevented by coincubation with NMMA (Fig. 4a). However, IL- 1 fails to inhibit aconitase activity of RINm5F cells preincubated under heat-shock conditions (30 min at 42 C: conditions that stimulate hsp 70 expression and prevent IL- 1-induced iNOS expression; data not shown) before incubation with IL-1. Heat shock also prevents IL-1-induced inhibition of glucose-stimulated insulin secretion by rat islets. Treatment of rat islets for 18 h with IL-1 results in a complete inhibition of glucose-stimulated insulin secretion (Fig. 4b). IL-1-induced inhibition of insulin secretion is prevented if islets are pretreated for 60 min at 42 C before cytokine stimulation. However, heat-shocked islets release higher levels of insulin in response to both 3 mm and 20 mm glucose as compared with non– heat-shocked islets. Although heat shock increases insulin secretion, the fraction of insulin secreted in response to 20 mm glucose as compared with 3 mm glucose is nearly identical whether or not the islets were heat shocked (Fig. 4b, percent of control, hatched bars). The mechanism by which heat shock increases the levels of insulin secretion in response to basal and stimulatory concentrations of glucose is unknown. It is possible that heat shock stimulates islet damage resulting in leakage of insulin from islets; however, this interpretation is not likely, because heat shock does not in-

5054


Endo • 1998 Vol 139 • No 12

FIG. 5. Effects of heat shock on IL-1 1 IFN-g-induced iNOS expression and NF-kB nuclear localization in human islets. a, Control and heat-shocked human islets (120 islets/condition) were treated for 40 h with 75 U/ml IL-1 and 750 U/ml IFN-g. Islets were isolated, and iNOS expression was examined by Western blot analysis. b, Control and heat-shocked human islets (1000/condition) were treated for 30 min with IL-1 1 IFN-g. Islets were isolated, and NF-kB nuclear localization was determined by gel shift analysis. Results are representative of three experiments from three independent human islet isolations.

FIG. 4. Heat shock prevents IL-1-induced inhibition of mitochondrial aconitase activity and insulin secretion. a, Control and heat-shocked RINm5F cells were treated for 18 h with 1 U/ml IL-1, 0.5 mM NMMA, or both. Cells were isolated, and mitochondrial aconitase activity was determined as in Materials and Methods. b, Control and heat-shocked islets were treated for 18 h with or without 1 U/ml IL-1. Islets were isolated, and glucose-stimulated insulin secretion was examined. Also shown is percent of control, where glucose-induced insulin secretion (20 mM glucose-3 mM glucose) by control islets was set at 100%. Results for both a and b are average 6 SEM of three independent experiments. Significantly different from untreated control (a), and untreated control incubated with 20 mM glucose (b): *, P , 0.05; significantly different from untreated control incubated with 3 mM glucose (b) D P , 0.05.

hibit mitochondrial aconitase activity (Fig. 4a) or induce morphological damage to islets (as determined by phase contrast microscopic analysis; data not shown). Effects of heat shock on cytokine-induced NF-kB activation and iNOS expression by human islets

Previous studies have shown that IL-1 and IFN-g are the minimal combination of cytokines required to stimulate iNOS expression by human islets (34). Treatment of human islets for 40 h with IL-1 1 IFN-g results in iNOS expression as determined by Western blot analysis (Fig. 5a). IL-1 1 IFN-g-induced iNOS expression is markedly reduced if human islets are pretreated for 90 min at 42 C before cytokine stimulation. The mechanism by which heat shock attenuates IL-1 1 IFN-g-induced iNOS expression appears to be asso-

ciated with the inhibition of cytokine-induced NF-kB activation. Incubation of human islets for 30 min with IL-1b 1 IFN-g results in NF-kB nuclear localization as evidenced by the reduced mobility of a DNA probe containing the consensus sequence for NF-kB binding (Fig. 5b). However, IL-1 1 IFN-g fails to stimulate NF-kB nuclear localization in human islets heat shocked for 90 min at 42 C before cytokine stimulation. These findings, which are similar to the effects of heat shock on IL-1-induced iNOS expression by rat islets, indicate that heat shock prevents IL-1 1 IFN-g-induced iNOS expression by human islets by a mechanism that is associated with the inhibition of NF-kB nuclear localization. NO is a stimulus for recovery protein expression by rat islets

Treatment of rat islets for 18 h with IL-1 results in a complete inhibition of insulin secretion that correlates with a potent inhibition of mitochondrial aconitase activity (15). We have previously shown that the addition of NMMA to islets pretreated for 18 h with IL-1b, followed by an 8-h incubation in the presence of both IL-1b and NMMA, results in complete recovery of both insulin secretion and aconitase activity (12, 15). Islet recovery from IL-1-induced damage appears to be an active process, because the transcriptional inhibitor actinomycin D, prevents the recovery of both insulin secretion and aconitase activity (12, 15). However, it is not clear if IL-1, NO (produced following IL-1-induced iNOS expression), or both stimulate the expression of potential recovery proteins. Data shown in Fig. 1 suggests that NO mediates IL-1-induced


hsp 70 expression by rat islets. Therefore, we examined the effects of actinomycin D on recovery of aconitase activity in islets treated with NO. In these experiments, islets were either treated with DEA-NO for 1 h, washed, and aconitase activity determined, or the islets were treated with DEA-NO for 1 h, washed three times to remove the NO donor compound, and cultured for an additional 5 h in the absence of DEA-NO. The islets were then isolated and aconitase activity was determined. Treatment of rat islets for 1 h with the NO donor compound DEA-NO results in an approximate 65% inhibition of mitochondrial aconitase activity (Fig. 6, solid bars). Islets completely recover aconitase activity to control levels if the islets are washed to remove the NO donor compound and incubated for an additional 5 h in the absence of DEA-NO (Fig. 6, hatched bars). Incubation of rat islets with the transcriptional inhibitor actinomycin D during the 5-h recovery period (in the absence of DEA-NO) prevents islet recovery of mitochondrial aconitase activity. The inhibitory effects of actinomycin D on the recovery of aconitase activity are not due to an inhibition of constitutive mRNA expression, because actinomycin D does not inhibit islet aconitase activity during a 5-h incubation (Fig. 6). These findings suggest that NO directly inhibits islet mitochondrial aconitase activity, and that NO also stimulates the expression of factors that participate in the recovery from NO-mediated damage. Discussion

IL-1-induced inhibition of insulin secretion and islet destruction is characterized by the expression of iNOS by b-cells, followed by NO-mediated inhibition of islet oxidative metabolism (1, 29, 30). Although Margulis et al. (35) first showed that liposomal delivery of hsp 70 into rat islets protects against IL-1-induced inhibition of islet function, the mechanism by which hsp 70 protects islets is unknown.

FIG. 6. Effects of actinomycin D on recovery of islet aconitase activity. Rat islets (1000/condition) were treated with or without 500 mM DEA-NO for 1 h and then isolated, and aconitase activity was determined (solid bars). To examine recovery from NO-induced inhibition of aconitase activity, islets were treated for 1 h in presence or absence of 500 mM DEA-NO, washed three times with complete CMRL-1066 tissue culture media, and then allowed to recover for 5 h in presence or absence of 1 mM actinomycin D (hatched bars). Following recovery, islets were isolated, and aconitase activity was determined. Results are average 6 SEM of three independent experiments. Significantly different from untreated control: *, P , 0.05.

5055

Others have shown that increased expression of hsp 70 in islets or RINm5F cells, either by heat shock or overexpression, prevents NO-induced islet cell and RINm5F cell lysis, respectively (16, 18). In the current study, we examined the effects of heat shock on IL-1-induced inhibition of islet metabolic function and insulin secretion. Rat islets, pretreated under heat-shock conditions (60 min incubation at 42 C), are resistant to IL-1-induced inhibition of mitochondrial aconitase activity and insulin secretion. The protective actions of heat shock on cellular metabolism are associated with an inability of islets to express iNOS or produce NO in response to IL-1. Activation of the transcriptional regulator NF-kB is required for IL-1-induced iNOS expression by rat islets (28, 32). We show that heat shock prevents IL-1-induced IkB degradation and NF-kB nuclear localization in rat islets. Heat shock also inhibits cytokine-induced iNOS expression by human islets of Langerhans. Human islets preincubated under heat-shock conditions (90 min at 42 C) do not express iNOS in response to IL1 1 IFN-g. Similar to rat islets, heat shock prevents IL-1 1 IFN-g-induced NF-kB activation and nuclear localization. These findings indicate that heat shock provides functional protection against the damaging effects of IL-1 by preventing iNOS expression by rat islets. In addition, these studies provide the first evidence that heat shock prevents cytokine-induced expression of human iNOS (human islets), by inhibiting NF-kB nuclear localization. Although our studies are not capable of discerning whether the protective actions of heat shock are mediated by increased hsp 70 expression or additional biochemical changes associated with heat shock (17), human islets constitutively express high levels of hsp 70 (36), and hsp 70 expression does not appear to be further altered by heat shock or IL-1 1 IFN-g treatment (data not shown). The findings that human islets expressing constitutively high levels of hsp 70 (in the absence of heat shock; data not shown) still express iNOS in response to IL-1 1 IFN-g, whereas heat shock prevents IL-1 1 IFN-g-induced NF-kB nuclear localization and iNOS expression by human islets, suggest that the inhibitory effects of heat shock on cytokine-induced iNOS expression may be independent of hsp 70. The heatshock response has been shown to prevent cytokine-induced iNOS expression by rat astrocytes, rat hepatocytes, and murine lung epithelial cells (37–39). In addition, Kim et al. (40) showed that hsp 70 expression protects hepatocytes from TNF-induced apoptosis. The protective actions of the heatshock response are associated with the inhibition of NF-kB activation and nuclear localization. Heat shock may prevent NF-kB activation by either inhibiting IkB degradation (37), stimulating IkB expression (41), or both. Although our studies do not discriminate between these two possibilities, the net result of each event is the sequestering of NF-kB in the cytoplasm and the inhibition of NF-kB-dependent mRNA transcription. A putative heat-shock element in the IkB promoter region has been identified, and the heat-shock response has been shown to stimulate IkB expression and inhibit TNF-induced IkB degradation in A549 lung adenocarcinoma cells (41). The heat-shock response may also inhibit cytokine signaling at sites upstream of NF-kB activa-

5056


tion. Gabai et al. (42) showed that heat shock prevents the activation of stress-activated Jun N-terminal kinase (JNK). In addition, IL-1 stimulates JNK activation in RINm5F cells (43); however, the effects of heat shock on IL-1-induced JNK activation and the role of JNK in IL-1-induced iNOS expression by RINm5F cells have not been examined. We are currently examining the potential role of JNK in IL-1-induced iNOS expression, and whether heat shock prevents JNK activation in rat islets. Islets also have the ability to recover from NO-mediated damage, and this recovery process is dependent on de novo mRNA transcription; however, the stimulus that induces the expression of potential recovery factors has yet to be identified (12, 15). We show that a 1-h incubation with the NO donor compound DEA-NO results in the inhibition of islet mitochondrial aconitase activity. Islets completely recover normal aconitase activity if DEA-NO is removed by washing, and the islets are allowed to recover for an additional 5 h. The transcriptional inhibitor actinomycin D inhibits the recovery of islet aconitase activity. These results indicate that NO stimulates the expression of factors that participate in islet recovery from NO-mediated damage. We also show that IL-1-induced hsp 70 expression is inhibited by NMMA, and that DEA-NO stimulates hsp 70 expression by rat islets. These findings suggest that hsp 70 may be one potential protein that is expressed in response to NO and that participates in islet recovery from NO mediated islet damage. Acknowledgments We thank Jessica Gorman and Colleen Kelly for expert technical assistance, Dr. Camillo Ricordi and his laboratory at the University of Miami Diabetes Research Institute (Miami, FL) and the Islet Isolation Core Facility at Washington University School of Medicine (St. Louis, MO) for providing human islets, and Dr. David Wink for providing DEA-NO. We also thank the Diabetes Research and Training Center at Washington University School of Medicine (St. Louis, MO) for performing insulin RIAs.

References 1. Mandrup-Poulsen T 1996 The role of interleukin-1 in the pathogenesis of IDDM. Diabetologia 39:1005–1029 2. Rabinovitch A 1993 Roles of cytokines in IDDM pathogenesis and islet b-cell destruction. Diabetes Rev 1:215–240 3. Southern C, Schulster D, Green IC 1990 Inhibition of insulin secretion by interleukin-1b and tumor necrosis factor-a via an L-arginine-dependent nitric oxide generation mechanism. FEBS Lett 276:42– 44 4. Eizirik DL, Cagliero E, Bjorklund A, Welsh N 1993 Interleukin-1-induced expression of nitric oxide synthase in insulin-producing cells is preceded by c-fos induction and depends on gene transcription and protein synthesis. FEBS Lett 317:62– 66 5. Corbett JA, Kwon G, Misko TP, Rodi CP, McDaniel ML 1994 Tyrosine kinase involvement in IL-1b-induced expression of iNOS by b-cells purified from islets of Langerhans. Am J Physiol 267:C48 –C54 6. Corbett JA, Wang JL, Sweetland MA, Lancaster Jr JR, McDaniel ML 1992 IL-1b induces the formation of nitric oxide by b-cells from rodent islets of Langerhans. J Clin Invest 90:2384 –2391 7. Corbett JA, McDaniel ML 1995 Intraislet release of IL-1 inhibits b-cell function by inducing b-cell expression of iNOS. J Exp Med 181:559 –568 8. Corbett JA, McDaniel ML 1996 Selective inhibition of iNOS by aminoguanidine. Methods Enzymol 268:398 – 408 9. Eizirik DL, Sandler S, Welsh N, Cetkovic-Cvrlje M, Nieman A, Geller DA, Pipeleers DG, Bendtzen K, Hellerstrom C 1994 Cytokines suppress human islet function irrespective of their effects on nitric oxide generation. J Clin Invest 93:1968 –1974 10. Corbett JA, Sweetland MA, Wang JL, Lancaster Jr JR, McDaniel ML 1993 Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans. Proc Natl Acad Sci USA 90:1731–1735

Endo • 1998 Vol 139 • No 12

11. Welsh N, Eizirik DL, Bendtzen K, Sandler S 1991 Interleukin-1b-induced nitric oxide production in isolated rat pancreatic islets requires gene transcription and may lead to inhibition of Krebs cycle enzyme aconitase. Endocrinology 129:3167–3173 12. Scarim AL, Heitmeier MR, Corbett JA 1997 Irreversible inhibition of metabolic function and islet destruction after a 36-hour exposure to interleukin-1b. Endocrinology 138:5301–5307 13. Palmer JP, Helqvist S, Spines GA, Molvig J, Mandrup-Poulsen T, Andersen HU, Nerup J 1989 Interactions of b-cell activity and IL-1 concentration and exposure time in isolated rat islets of Langerhans. Diabetes 38:1211–1216 14. Comens PG, Wolf BA, Unanue ER, Lacy PE, McDaniel ML 1987 Interleukin-1 is a potent modulator of insulin secretion from isolated rat islets of Langerhans. Diabetes 36:963–970 15. Corbett JA, McDaniel ML 1994 Reversibility of interleukin-1b-induced islet destruction and dysfunction by inhibition of nitric oxide synthase. Biochem J 299:719 –724 16. Bellmann K, Wenz A, Radons J, Burkart V, Kleemann R, Kolb H 1995 Heat shock induces resistance in rat pancreatic islet cells against nitric oxide, oxygen radicals and streptozotocin toxicity in vitro. J Clin Invest 95:2840 –2845 17. Welch WJ 1992 Mammalian stress response: cell physiology, structure/function of stress proteins, and implications for medicine and disease. Physiol Rev 72:1063–1081 18. Bellmann K, Jaattela M, Wissing D, Burkart V, Kolb H 1996 Heat shock protein hsp70 overexpression confers resistance against nitric oxide. FEBS Lett 391:185–188 19. Helqvist S, Hansen BS, Johannesen J, Andersen HU, Nielsen JH, Nerup J 1989 Interleukin 1 induces new protein formation in isolated rat islets of Langerhans. Acta Endocrinol (Copenh) 121:135–140 20. Helqvist S, Polla BS, Johannesen J, Nerup J 1991 Heat shock protein induction in rat pancreatic islets by recombinant human interleukin-1b. Diabetologia 34:150 –156 21. Lacy PE, Kostianovsky M 1967 Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes 16:35–39 22. Green LC, Wagner DA, Glogowski J, Skipper PL, Wishnok JS, Tannenbaum SR 1982 Analysis of nitrate, nitrite, and [15N] nitrate in biological fluids. Anal Biochem 126:131–138 23. Wright PH, Makulu DR, Vichick D, Sussman KE 1971 Insulin immunoassay by back-titration. Diabetes 20:33– 45 24. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680 – 685 25. Heitmeier MR, Scarim AL, Corbett JA 1997 Interferon-g increases the sensitivity of islets of Langerhans for inducible nitric-oxide synthase expression induced by interleukin-1. J Biol Chem 272:13697–13704 26. Burd PR, Rogers HW, Gordon JR, Martin CA, Jayaraman S, Wilson SD, Dvorak AM, Galli SJ, Dorf ME 1989 Interleukin-3-dependent and -independent mast cells stimulated with IgE and antigen express multiple cytokines. J Exp Med 170:245–257 27. Greenlund AC, Farrar MA, Bibiano BL, Schreiber RD 1994 Ligand-induced IFN-g receptor tyrosine phosphorylation couples the receptor to its signal transduction system (p91). EMBO J 13:1591–1600 28. Kwon G, Corbett JA, Rodi CP, Sullivan P, McDaniel ML 1995 IL-1b-induced nitric oxide synthase expression by rat pancreatic b-cells: evidence for the involvement of nuclear factor kB in signaling mechanism. Endocrinology 136:4790 – 4795 29. Eizirik DL, Flodstrom M, Karlesn AE, Welsh N 1996 The harmony of the spheres: inducible nitric oxide synthase and related genes in pancreatic beta cells. Diabetologia 39:875– 890 30. Corbett JA, Kwon G, Hill JR, McDaniel ML 1995 Nitric oxide mediates cytokine-induced islet destruction. In: Marshall SM, Home PD, Rizza RA (eds) The Diabetes Annual 9. Elsevier Science B. V., Amsterdam, The Netherlands, pp 265–294 31. Flodstrom M, Welsh N, Eizirik DL 1996 Cytokines activate the nuclear factor kB (NF-kB) and induce nitric oxide production by human pancreatic islets. FEBS Lett 385:4 – 6 32. Bedoya FJ, Flodstrom M, Eizirik DL 1995 Pyrrolidine dithiocarbamate prevents IL-1-induced nitric oxide synthase mRNA, but not superoxide dismutase mRNA in insulin producing cells. Biochem Biophys Res Commun 210:816 – 822 33. Baeuerle PA, Henkel T 1994 Function and activation of NF-kB in the immune system. Annu Rev Immunol 12:141–179 34. Corbett JA, Kwon G, Marino M, Rodi CP, Sullivan PM, McDaniel ML 1996 Tyrosine kinase inhibitors prevent cytokine-induced expression of iNOS and COX-2 by human islets of Langerhans. Am J Physiol 39:C1581–C1587 35. Margulis BA, Sandler S, Eizirik DL, Welsh N, Welsh M 1991 Liposomal delivery of purified heat shock protein hsp 70 into rat pancreatic islets as protection against interleukin 1b-induced impaired b-cell function. Diabetes 40:1418 –1422 36. Welsh N, Margulis B, Borg LA, Wiklund HJ, Saldeen J, Flodstrom M, Mello MA, Andersson A, Pipeleers DG, Hellerstrom C, Eizirik DL 1995 Differences


37. 38. 39. 40.

in the expression of heat-shock proteins and antioxidant enzymes between human and rodent pancreatic islets: implications for the pathogenesis of insulin-dependent diabetes mellitus. Mol Med 1:806 – 820 Wong HR, Ryan M, Wispe JR 1997 Stress response decreases NF-kB nuclear translocation and increases IkB expression in A549 cells. J Clin Invest 99:2423–2428 deVera ME, Kim YM, Wong HR, Wang Q, Billiar TR, Geller DA 1996 Heat shock response inhibits cytokine-inducible nitric oxide synthase expression in rat hepatocytes. Hepatology 24:1238 –1245 Feinstein DL, Galea E, Aquino DA, Li GC, Xu H, Reis DJ 1996 Heat shock protein 70 suppresses astroglial-inducible nitric-oxide synthase expression by decreasing NFkappaB activation. J Biol Chem 271:17224 –17232 Kim YM, deVera ME, Watkins SC, Billiar TR 1997 Nitric oxide protects

5057

cultured rat hepatocytes from tumor necrosis factor-a-induced apoptosis by inducing heat shock protein 70 expression. J Biol Chem 272:1402–1411 41. Wong HR, Ryan M, Wispe JR 1997 Stress response decreases NF-kB nuclear translocation and increases I-kBa expression in A549 cells. J Clin Invest 99:2423–2428 42. Gabai VL, Meriin AB, Mosser DD, Caron AW, Rits S, Shifrin VI, Sherman MY 1997 Hsp 70 prevents activation of stress kinases. J Biol Chem 272:18033–18037 43. Welsh N 1996 Interleukin-1b-induced ceramide and diacylglycerol generation may lead to activation of the c-Jun NH2-terminal kinase and the transcription factor ATF2 in the insulin-producing cell line RINm5F. J Biol Chem 271:8307– 8312