Interferon- (cid:103) Increases the Sensitivity of Islets of Langerhans for Inducible Nitric-oxide Synthase Expression Induced by Interleukin 1*

The purpose of this study was to evaluate the effects of interferon- (cid:103) (IFN- (cid:103) ) alone and in combination with interleukin 1 (cid:98) (IL-1 (cid:98) ) on inducible nitric-oxide synthase (iNOS) mRNA and protein expression, nitrite produc- tion, and insulin secretion by islets of Langerhans. Treatment of rat islets with IL-1 (cid:98) results in a concentra-tion-dependent increase in the production of nitrite that is maximal at 5 units/ml. Individually, 0.1 unit/ml IL-1 (cid:98) or 150 units/ml rat IFN- (cid:103) do not stimulate iNOS expression or nitrite production by rat islets; however, in com- bination, these cytokines induce the expression of iNOS and the production of nitrite to levels similar in magni- tude to the individual effects of 5 units/ml IL-1 (cid:98) . The islet (cid:98) -cell, selectively destroyed during insulin-depend-ent diabetes mellitus, appears to be one islet cellular source of iNOS as 150 units/ml rat IFN- (cid:103) and 0.1 unit/ml IL-1 (cid:98) induced similar effects in primary (cid:98) -cells purified by fluorescence-activated cell sorting and in the rat in- sulinoma cell line, RINm5F. iNOS expression and nitrite production by rat islets in response to 150 units/ml rat IFN- (cid:103) and 0.1 unit/ml IL-1 (cid:98) (cid:98) to determine if human islets respond in a similar manner. In this study we show that IL-1 (cid:98) , at a concentration as low as 1 unit/ml (5.7 p M ), is able to stimulate high levels of nitrite production by human islets in the presence of human IFN- (cid:103) (750 units/ml). We also show that in the presence of 75 units/ml human IFN- (cid:103) , as little as 10 units/ml IL-1 (cid:98) is required to induce a 2-fold increase in the level of nitrite production. These results indicate that IFN- (cid:103) reduces the concentration of IL-1 (cid:98) required to stimulate iNOS expression by human islets in a manner similar to IFN- (cid:103) ’s effects on rat islets.

Insulin-dependent diabetes mellitus is an autoimmune disease characterized by the selective destruction of insulin secreting ␤-cells found in islets of Langerhans. Many lines of evidence support a role for the involvement of cytokines as effector molecules that participate in the development of diabetes. Mandrup-Poulsen et al. (1) first showed that treatment of isolated rat islets with conditioned media derived from activated mononuclear cells results in a potent inhibition of insulin secretion followed by islet destruction. The active component of this conditioned media was determined to be the cytokine IL-1 1 (2). IL-1-induced inhibition of insulin secretion is both timeand concentration-dependent and requires mRNA transcription and new protein synthesis (3). Recently, IL-1-induced inhibition of insulin secretion has been attributed to the expression of iNOS and increased production of nitric oxide by ␤-cells (4,5). Southern et al. (6) first demonstrated that treatment of rat islets with IL-1␤ and tumor necrosis factor results in an inhibition of insulin secretion that is attenuated by the nitric-oxide synthase inhibitor nitro-L-arginine methyl ester. We and others (7)(8)(9) have shown that IL-1␤-induced inhibition of insulin secretion and IL-1␤-induced nitrite production by rat islets are completely prevented by N G -monomethyl-L-arginine (NMMA) and aminoguanidine (AG). The expression of iNOS by rat islets has been demonstrated at the level of mRNA and protein (10 -12). Immunohistochemical colocalization of iNOS and insulin demonstrates that IL-1␤ selectively induces the expression of iNOS by ␤-cells (12). The inhibitory and destructive effects of IL-1␤ on islet function and viability are mediated, in part, by the ability of nitric oxide to target and inhibit the enzymatic activity of mitochondrial enzymes, including aconitase and the electron transport chain complexes I and II (7,13,14). Treatment of rat islets for 18 h with IL-1␤ results in an 80% inhibition of aconitase activity that is prevented by NMMA (7,13). IL-1␤ has also been shown to reduce islet cellular levels of ATP and to inhibit glucose oxidation in a nitric oxide-dependent manner (15).
Autoimmune diabetes is associated with a local inflammatory reaction (insulitis) in and around pancreatic islets. In an activated state, T-lymphocytes and macrophages, primary cellular components of islet insulitis, release high levels of IL-1 and IFN-␥, respectively. Although the effects of IL-1 on islet function have been examined in detail, few studies have investigated the effects of IFN-␥ on ␤-cell function and viability. Many lines of evidence support a role for IFN-␥ in the devel-opment of autoimmune diabetes. This evidence includes 1) IFN-␥ mRNA expression in islets correlates with the development of insulitis and diabetes in the nonobese diabetic mouse (16); 2) transgenic mice expressing IFN-␥ under control of the insulin promoter develop insulitis and diabetes (17); and 3) monoclonal antiserum specific for IFN-␥ attenuates the development of diabetes in the nonobese diabetic mouse (18,19). In this study we have examined the effects of IFN-␥ alone and in combination with IL-1␤ on iNOS expression, nitrite formation, and islet function and viability. Alone, IFN-␥ does not modulate islet function or viability; however, IFN-␥ increases the sensitivity of rat islets to IL-1␤ by stabilizing IL-1-induced iNOS mRNA expression resulting in the increased production of nitric oxide. The increased sensitivity of rat islets for IL-1 results in the inhibition of ␤-cell function and islet destruction at concentrations of IL-1␤ that alone have no effect on islet viability or function. Islet Isolation and Culture-Islets were isolated from male Sprague Dawley rats by collagenase digestion as described previously (20). Following isolation, islets were cultured overnight in complete CMRL-1066 (CMRL-1066 containing 2 mM L-glutamine, 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, and 100 g/ml streptomycin) under an atmosphere of 95% air and 5% CO 2 at 37°C. Human islets were incubated for 48 h at 37°C in complete CMRL-1066 before the initiation of experiments. Prior to each experiment, islets were washed 3 times in complete CMRL-1066, counted, and then cultured for an additional 3 h at 37°C. Experiments were then initiated by the addition of cytokines and iNOS inhibitors, followed by culture for the indicated times.
For experiments using RINm5F cells, cells were removed from growth flasks by treatment with 0.05% trypsin, 0.02% EDTA at 37°C. Cells were washed 2 times with complete CMRL-1066, plated at a concentration of 200,000 cells/200 l of complete CMRL-1066 in 96-well microtiter plates, and then cultured at 37°C for 2-3 h prior to cytokine treatment. Experiments were initiated by the addition of cytokines, and the RINm5F cells were then incubated for the indicated times at 37°C.
Islet Dispersion and Macrophage Depletion-Isolated rat islets were dispersed into individual cells by treatment with trypsin (1.0 mg/ml) in Ca 2ϩ -and Mg 2ϩ -free Hanks' solution at 37°C for 3 min as stated previously (21). The dispersed islet cells were counted and then immediately aliquoted into 24-well microtiter plates (100,000 cells/well in 400 l of complete CMRL-1066) and cultured for 1 h at 37°C. For macrophage depletion studies, islets were cultured for 7 days at 24°C followed by 2 days at 37°C prior to cell dispersion (24). Where indicated, islet cells were pretreated for 30 min with IRAP; cytokines were then added, and the islet cells were incubated for 24 h.
Purification of ␣and ␤-Cells by Fluorescence-activated Cell Sorting (FACS)-Islets isolated from 10 rats were cultured overnight (ϳ1200 islets/3 ml) in complete CMRL-1066 media under an atmosphere of 95% air and 5% CO 2 at 37°C. Islets were then dispersed into individual cells as stated above. Dispersed islet cells were incubated for 60 min at 37°C in complete CMRL-1066 prior to cell sorting. Islet cells were purified as described previously (11,13,22) using a FACSTAR ϩ flow cytometer (Becton Dickinson, San Jose, CA). The cells were illuminated at 488 nm, and emission was monitored at 515-535 nm. The sorting process yielded a 95% population of ␤-cells and an 80 -85% population of ␣-cells.
Insulin Secretion-Islets (220/ml of complete CMRL-1066) were cultured for 40 h with the indicated concentrations of IL-1␤, IFN-␥, and aminoguanidine (AG). The islets were then isolated and washed 3 times in Krebs-Ringer bicarbonate buffer (KRB, 25 mM Hepes, 115 mM NaCl, 24 mM NaHCO 3 , 5 mM KCl, 1 mM MgCl 2 , 2.5 mM CaCl 2 , and 0.1% bovine serum albumin, pH 7.4) containing 3 mM D-glucose. Groups of 20 islets were counted into 10-ϫ 75-mm borosilicate test tubes and preincubated for 30 min at 37°C with shaking in 200 l of the same buffer. The preincubation buffer was removed, and glucose-stimulated insulin secretion was initiated by the addition of 200 l of KRB containing either 3 or 20 mM D-glucose. Islets were then incubated at 37°C for 30 min, the incubation buffer was removed, and insulin content was determined by radioimmunoassay (23).
Islet Viability-Islets (25/500 l of complete CMRL-1066) were cultured for 96 h in 24-well microtiter plates with the indicated concentrations of IL-1␤, IFN-␥, and aminoguanidine. Islet degeneration was determined in a double-blind manner by phase-contrast microscopic analysis. Islet degeneration is characterized by the loss of islet integrity, disintegration, and partial dispersion of islets as described previously (14,24,25).
Western Blot Analysis-RINm5F cells (400,000/400 l of complete CMRL-1066), cultured in 24-well microtiter plates with the indicated concentrations of IL-1␤ and IFN-␥ for 24 h at 37°C, were washed 3 times with 0.1 M phosphate-buffered saline (PBS), pH 7.4, followed by the addition of 25 l of sodium dodecyl sulfate (SDS) sample mix (0.25 M Tris-HCl, 20% ␤-mercaptoethanol, and 4% SDS). The lysed cells were then transferred to 1.5-ml microcentrifuge tubes, and the individual wells of the microtiter plate were rinsed with 15 l of distilled H 2 O which was then added to the corresponding lysed samples. The samples were boiled for 4 min followed by the addition of 4 l of loading dye (0.05% bromphenol blue in 80% glycerol). Rat islets (150/400 l of complete CMRL-1066) were cultured for 40 h with the indicated concentrations of IL-1␤ and IFN-␥ at 37°C under an atmosphere of 95% air and 5% CO 2 . The islets were isolated by centrifugation (6,000 ϫ g, 3 min) and washed 3 times with 0.1 M PBS. Islets were lysed by the addition of 25 l of SDS sample mix and 15 l of distilled H 2 O, boiled for 4 min, followed by the addition of 4 l loading dye. Proteins were separated by SDS-gel electrophoresis using standard conditions (26) and transferred to Nitrocell nitrocellulose membranes (Pharmacia Biotech Inc.) under semi-dry transfer conditions. Blots were blocked overnight in TBST (20 mM Tris, 500 mM NaCl, and 0.1% Tween 20, pH 7.5) containing 5% nonfat dry milk. Blots were washed one time with TBST and then incubated for 1.5 h at room temperature with rabbit antimouse iNOS (1:2000 dilution) in TBST containing 1% nonfat dry milk. Following incubations with the primary antisera, blots were washed 4 times with TBST (5 min/wash) and then incubated for 1 h at room temperature with horseradish peroxidase-conjugated donkey anti-rabbit secondary antibody at a dilution of 1:7000. The blots were washed 3 times in TBST and once in 0.1 M PBS at room temperature. Detection of rat iNOS was by enhanced chemiluminescence according to manufacturer's specifications (Amersham Corp.).
Northern Blot Analysis-RINm5F cells (10 ϫ 10 6 cells/3 ml complete CMRL-1066) were cultured for 6 and 12 h at 37°C with the indicated concentrations of IL-1␤ and IFN-␥. For the mRNA stability experiments, RINm5F cells (10 ϫ 10 6 cells/3 ml of complete CMRL-1066) were cultured for 6 h in the presence of the indicated concentrations of IL-1 and IFN-␥. Actinomycin D (1 M) was then added, and the cells were cultured for an additional 6 h. After culture, the cells were washed 3 times with 0.1 M PBS, pH 7.4, and total RNA was isolated using the RNeasy kit (Qiagen, Inc., Chatsworth, CA). Total cellular RNA (10 -20 g) was denatured and fractionated by gel electrophoresis using a 1.0% agarose gel containing 2.2 M formaldehyde. RNA was transferred by capillary action in 20 ϫ SSC (3 M NaCl, 0.3 M sodium citrate, pH 7.0) to Duralon UV nylon membranes (Stratagene, La Jolla, CA), and the membranes were hybridized to a 32 P-labeled probe specific for rat iNOS or cyclophilin (27). The cDNA probe was radiolabeled with [␣-32 P]dCTP by random priming using the Prime-a-Gene nick translation system from Promega (Madison, WI). iNOS cDNA probe corresponds to bases 509 -1415 of the rat iNOS coding region. 28 S RNA band or cyclophilin was used as an internal control for RNA loading. Hybridization and autoradiography were performed as described previously (28).
Densitometry and Image Analysis-Autoradiograms were scanned into NIH Image version 1.59 using a COHU high performance CCD camera (Brookfield, WI). Densities were determined using NIH Image version 1.59 software. PhosphorImaging analysis of RINm5F cell mRNA stability experiments was performed using a Molecular Dynamics PhosphorImager and Molecular Dynamics ImageQuant Software Version 3.3 (Molecular Dynamics, Inc.). For Western blot data, autoradiograms were scanned into NIH Image and then imported into Canvas 3.5 (Deneba Software, Miami, Fl) for the preparation of figures. Nitrite Determination-Nitrite production was determined by mixing 50 l of culture medium with 50 l of Griess reagent (29). The absorbance at 540 nm was measured, and nitrite concentrations were calculated from a sodium nitrite standard curve.
Statistics-Statistical comparisons were made between groups using a one-way analysis of variance. Significant differences between treatment groups compared with untreated controls (p Ͻ 0.05) were evaluated using a Scheffe's F test posthoc analysis.

Rat IFN-␥ Reduces the Concentration of IL-1␤ Required to Stimulate Rat Islet iNOS Expression by 10-Fold-To determine
if IFN-␥ modulates iNOS expression, the effects of rat IFN-␥, alone and in combination with IL-1␤, on nitrite production and iNOS expression by isolated rat islets were examined. Incubation of rat islets for 40 h with IL-1␤ results in a concentrationdependent increase in the production of nitrite (Fig. 1A). IL-1␤ induces the first detectable increase in nitrite production by rat islets at 0.5 units/ml IL-1␤ with maximal nitrite production observed at 1 and 5 units/ml IL-1␤ (data not shown for 1 unit/ml IL-1␤). Alone, rat IFN-␥ (concentrations from 1.5 to 150 units/ml) does not stimulate nitrite formation by rat islets; however, in the presence of 0.1 unit/ml IL-1␤ (which alone does not stimulate nitrite formation), rat IFN-␥ induces the concen-tration-dependent production of nitrite by rat islets. The levels of nitrite produced in response to 0.1 unit/ml IL-1␤ and 150 units/ml rat IFN-␥ are similar in magnitude to the levels produced by rat islets treated with 5 units/ml IL-1␤. Rat IFN-␥ also slightly increases (ϳ8%) the production of nitrite by rat islets stimulated with 5 units/ml IL-1␤ as compared with 5 units/ml IL-1␤ alone.
The effects of rat IFN-␥ and IL-1␤ on iNOS expression from the same islets used in Fig. 1A are shown in Fig. 1B. Alone, 0.1 unit/ml IL-1␤ or 150 units/ml rat IFN-␥ do not induce the expression of iNOS by rat islets; however, a combination of 0.1 unit/ml IL-1␤ and 150 units/ml rat IFN-␥ stimulates the expression of iNOS to levels similar in magnitude to the expression of iNOS induced by 5 units/ml IL-1␤. Also, the combination of 5 units/ml IL-1␤ and 150 units/ml rat IFN-␥ induce iNOS expression to levels that exceed those induced by the treatment of rat islets with 5 units/ml IL-1␤. This effect is consistent with the ability of rat IFN-␥ to increase the level of nitrite produced by rat islets in response to maximal concentrations of IL-1␤.
These results indicate that rat IFN-␥, in combination with IL-1␤ at concentrations that alone do not induce iNOS expression, stimulate the expression of iNOS by rat islets to levels that are similar to the individual effects of maximal concentrations of IL-1␤. For convenience, we have defined 0.1 unit/ml IL-1␤ as submaximal and 1 and 5 units/ml IL-1␤ as maximal concentrations of IL-1␤.

The Combination of Rat IFN-␥ and Submaximal Concentrations of IL-1␤ Inhibit Glucose-stimulated Insulin Secretion and Induce Islet Degeneration in a Nitric
Oxide-dependent Manner-Our previous studies have shown that nitric oxide mediates the inhibitory effects of IL-1␤ on glucose-stimulated insulin secretion (8). The effects of rat IFN-␥ and IL-1␤ on insulin secretion by rat islets were examined to determine if nitric oxide production, stimulated by submaximal concentrations of IL-1␤ in the presence of rat IFN-␥, is associated with an inhibition of insulin secretion. Treatment of rat islets with a maximal concentration of IL-1␤ results in a complete inhibition of glucose-stimulated insulin secretion (Table I). We have previously shown that IL-1␤-induced inhibition of insulin secretion is prevented by the NOS inhibitors NMMA and AG and that iNOS inhibitors do not modulate glucose-stimulated insulin secretion in the absence of cytokines (8,9). Incubation of islets for 40 h with a combination of rat IFN-␥ and a submaximal  concentration of IL-1␤ also results in a complete inhibition of insulin secretion that is prevented by AG (Table I). Individually, submaximal IL-1␤ or rat IFN-␥ do not inhibit glucosestimulated insulin secretion. The lack of an inhibitory effect of submaximal IL-1␤ or rat IFN-␥ on insulin secretion is consistent with the inability of these cytokines to stimulate nitrite production by rat islets. These findings indicate that treatment of rat islets with a submaximal concentration of IL-1␤, in the presence of rat IFN-␥, results in an inhibition of insulin secretion that is mediated by the production of nitric oxide. The effects of rat IFN-␥, alone and in combination with IL-1␤, on islet viability are also shown in Table I. Incubation of islets for 96 h with a maximal concentration of IL-1␤ results in the complete degeneration of islets. Islet degeneration is characterized by loss of islet integrity and islet dispersion (14,24,25). The destructive effects of IL-1␤ on islet viability are completely prevented by the iNOS inhibitor aminoguanidine (AG), indicating that nitric oxide participates in IL-1␤-induced islet degeneration. Alone, rat IFN-␥ (150 units/ml) or a submaximal concentration of IL-1␤ do not induce islet degeneration; however, in combination these cytokines stimulate islet degeneration to levels similar to the individual effects of maximal concentrations of IL-1␤ alone. The destructive effects of submaximal concentrations of IL-1␤, in combination with rat IFN-␥, are completely prevented by AG. These findings indicate that islet degeneration stimulated by rat IFN-␥ and submaximal concentrations of IL-1␤ is mediated by the production of nitric oxide.

Effects of Rat IFN-␥ and IL-1␤ on iNOS Expression and Nitrite Formation by FACS Purified ␣and ␤-Cells-Islets
contain a heterogeneous population of both endocrine and nonendocrine cells, of which the insulin-secreting ␤-cell is selectively destroyed during the development of autoimmune diabetes. To determine if rat IFN-␥ increases the sensitivity of ␤-cells for IL-1␤-induced iNOS expression, we have examined the effects of rat IFN-␥ alone, and in combination with IL-1␤, on nitrite production and iNOS expression by primary rat ␣and ␤-cells purified by FACS. As shown in Fig. 2, treatment of primary ␤-cells with submaximal IL-1␤ in the presence of rat IFN-␥ stimulates the production of nitrite to levels similar in magnitude to the effects of maximal concentrations of IL-1␤ alone. Also, nitrite production by primary ␤-cells incubated with maximal concentrations of IL-1␤, or the combination of maximal IL-1␤ and rat IFN-␥, are virtually identical.
The effects of rat IFN-␥ and IL-1␤ on iNOS expression correlate with the effects of these cytokines on nitrite production by primary ␤-cells. As shown in Fig. 2B, submaximal concentrations of IL-1␤ or rat IFN-␥ do not stimulate the expression of iNOS by primary ␤-cells. However, in combination, these cytokines stimulate the expression of iNOS to levels that are slightly higher than the effects of maximal concentrations of IL-1␤ on iNOS expression by primary ␤-cells. Also shown in Fig. 2 are the effects of IL-1␤ and rat IFN-␥ on nitrite formation and iNOS expression by primary rat ␣-cells. Individually or in combination, IL-1␤ and rat IFN-␥ do not stimulate the production of nitrite or the expression of iNOS by primary ␣-cells. These experiments demonstrate that the combination of submaximal concentrations of IL-1␤ in the presence of rat IFN-␥ stimulates the expression of iNOS by primary ␤-cells, suggesting that the ␤-cell is one islet cellular source of iNOS under these conditions.

Time-dependent Effects of Rat IFN-␥ and IL-1␤ on Nitrite Formation and iNOS mRNA Expression and Stability-We
have examined the time-dependent production of nitrite and iNOS mRNA accumulation using RINm5F cells. RINm5F cells represent a homogeneous population of ␤-cells that respond to IL-1 and IFN-␥ in a manner similar to the effects of these cytokines on iNOS expression by intact islets. As shown in Fig.  3A, a maximal concentration of IL-1␤ stimulates the time-dependent production of nitrite that is first apparent at 6 h, then progresses linearly from 6 to 24 h, with little increase in the level of nitrite from 24 to 48 h. Individually, rat IFN-␥ or submaximal concentrations of IL-1␤ do not stimulate the production of nitrite by RINm5F cells at any time point examined; however, the combination of these cytokines induces the timedependent production of nitrite by RINm5F cells that is similar to the effects of maximal concentrations of IL-1␤ alone. Nitrite production induced by the combination of submaximal concentrations of IL-1␤ in the presence of rat IFN-␥ is first detected 6 h after the addition of cytokines (control, 2.9 Ϯ 0.6 pmol/2000 cells versus IL-1␤ ϩ IFN-␥, 4.3 Ϯ 0.6 pmol/2000 cells) and increases linearly from 6 to 24 h. The rate by which IL-1␤ alone or the combination of IL-1␤ and rat IFN-␥ stimulate nitrite formation by RINm5F cells was determined by linear regression of nitrite data from 6 to 24-h time points shown in Fig. 3A. The rate of nitrite formation induced by the combination of submaximal IL-1␤ and rat IFN-␥ is reduced compared with the individual effects of maximal IL-1␤ (2.5 pmol of nitrite/h versus 3.6 pmol nitrite/h, respectively). Also, the maximal level of nitrite produced in response to submaximal IL-1␤ and rat IFN-␥ is ϳ20 -30% less than that induced by maximal IL-1␤ alone.
Although nitrite production by RINm5F cells in response to submaximal concentrations of IL-1␤ in combination with rat IFN-␥ is similar to the effects of maximal IL-1␤ (in terms of the time dependence), the effects of these two conditions on iNOS mRNA accumulation are different. As shown in Fig. 3B, iNOS mRNA accumulation in response to a maximal concentration of IL-1␤ is 2-fold higher than the effects of submaximal concentrations of IL-1␤ and rat IFN-␥ following a 6-h incubation. However, following a 12-h incubation, the levels of iNOS mRNA that accumulate in response to submaximal concentrations of IL-1␤ and rat IFN-␥ are nearly identical to the levels observed following a 6-h exposure, whereas maximal IL-1␤induced iNOS mRNA accumulation is reduced to near background levels.
The persistence of iNOS mRNA accumulation following a 12-h exposure of RINm5F cells with submaximal concentrations of IL-1␤ in combination with IFN-␥ compared with maximal concentrations of IL-1␤ alone (Fig. 3B) suggests that IFN-␥ may stabilize IL-1-induced iNOS mRNA. To examine this question, an analysis of iNOS mRNA stability using the transcriptional inhibitor actinomycin D was performed. In Fig.  3C, RINm5F cells were incubated for 6 h in the presence of IFN-␥ and maximal or submaximal concentrations of IL-1␤ or with maximal concentrations of IL-1␤ alone. Actinomycin D was then added, and the RINm5F cells were cultured for 6 additional h. As shown in Fig. 3C, ϳ70% IL-1␤-induced iNOS mRNA is degraded in the 6-h incubation following the addition of actinomycin D; however, only ϳ30% iNOS mRNA is degraded in the presence of maximal concentrations of IL-1 in combination with IFN-␥, and only ϳ40% iNOS mRNA is degraded in the presence of submaximal concentrations of IL-1 in combination with IFN-␥. These data suggest a role for IFN-␥ in the stabilization of IL-1␤-induced iNOS mRNA that results in the persistence of iNOS mRNA accumulation after a 12-h exposure to these cytokines (Fig. 3B). Whereas maximal IL-1␤induced iNOS mRNA accumulation is ϳ2-fold greater than for submaximal IL-1␤ in the presence of INF-␥, the increase in iNOS mRNA stability afforded by IFN-␥ ultimately results in similar levels of iNOS protein expression (data not shown) and nitrite production under both conditions.
Rat IFN-␥-induced iNOS Expression by Islet Cells Requires the Endogenous Release of IL-1-Islets contain resident macrophages that are known to express and release IL-1. We have previously shown that endogenous release of IL-1 within islets results in an inhibition of insulin secretion that is mediated by ␤-cell expression of iNOS (12). To further examine the extent to which the presence of IFN-␥ is able to increase the sensitivity of islets to IL-1, islets were dispersed into individual cells and treated with varying concentrations of IFN-␥ or with a maximal concentration of IL-1␤ alone (Fig. 4). Islet dispersion involves the treatment of intact islets with trypsin, an experimental manipulation that results in the destruction of 2-3% islet cells (based on trypan blue exclusion, data not shown). As shown in Fig. 4A, treatment of dispersed islet cells with rat IFN-␥ stimulates nitrite formation and iNOS expression (inset) in a concentration-dependent manner. The interleukin-1 receptor antagonist protein (IRAP), which competes with IL-1 for receptor binding (33), completely prevents IFN-␥-induced nitrite production and iNOS expression by dispersed islet cells. These findings indicate that sufficient levels of IL-1␤ are released during dispersion to stimulate iNOS expression in the presence of IFN-␥.
The cellular source of the IL-1 released during islet cell dispersion is believed to be the resident islet macrophage. To provide evidence for the intra-islet macrophage as a source of IL-1, the effects of macrophage depletion on IFN-␥-induced iNOS expression and nitrite formation by dispersed islet cells were examined. Macrophage depletion was accomplished by culturing intact islets for 7 days at 24°C. This culture condition has previously been shown to deplete over 95% of the islet lymphoid population (24). As shown in Fig. 4B, IFN-␥ no longer stimulates iNOS expression (inset) or nitrite production by islet were cultured for 6, 12, 24, and 48 h with indicated concentrations of IL-1␤ and rat IFN-␥. Nitrite formation was measured on culture supernatant as described under "Experimental Procedures." B, RINm5F cells (10 ϫ 10 6 cells/3 ml of complete CMRL-1066) were cultured with IL-1␤ and rat IFN-␥ for the indicated times. Total RNA was isolated and probed for iNOS by Northern analysis as stated under "Experimental Procedures." 28 S RNA band is as indicated and is used to control for RNA loading. C, iNOS mRNA stability was analyzed by Northern blot analysis of total RNA isolated from RINm5F cells incubated for 6 h with the indicated concentrations of IL-1␤, and IL-1␤ plus IFN-␥, followed by the addition of 1 M actinomycin D and continued culture for 6 h. PhosphorImaging analysis was used to quantitate the levels of iNOS mRNA, and the values have been corrected for RNA loading using cyclophilin as a control. iNOS mRNA accumulation was set at 100% following the 6-h exposure prior to the addition of actinomycin D, and the level of iNOS mRNA accumulation in the presence of actinomycin D is expressed as a % of this level. Results for nitrite measurements are average Ϯ S.E. of four individual experiments. Northern analysis of iNOS mRNA is representative of three individual experiments, and iNOS mRNA stability is the average of two individual experiments. cells dispersed from macrophage-depleted islets. The culture conditions used for macrophage depletion do not damage islet endocrine cells as maximal concentrations of IL-1 induce the expression of iNOS and the production of nitrite. These findings provide evidence for the resident islet macrophage as the cellular source of IL-1 and indicate that IFN-␥ increases the sensitivity of islet cells for iNOS expression stimulated by the endogenous release of IL-1.
Effects of Human IFN-␥ and IL-1␤ on Nitrite Formation by Human Islets of Langerhans-We have also evaluated the effects of human IFN-␥ and IL-1␤ on nitrite production by human islets of Langerhans. We and others (30,31) have shown previously that the minimal combination of cytokines required to stimulate iNOS expression and nitrite formation by human islets is IFN-␥ and IL-1␤. Concentrations of IL-1␤ and human IFN-␥ that have been used to stimulate iNOS expression and nitrite production by human islets are 50 -75 and 750 -1000 units/ml, respectively (30 -32). Because rat IFN-␥ is effective at reducing the amount of IL-1␤ required to stimulate nitrite formation by rat islets, the effects of human IFN-␥ on IL-1␤induced nitrite formation by human islets were examined. In the presence of 750 units/ml human IFN-␥, IL-1␤ induces a concentration-dependent increase in nitrite formation (Fig. 5) by human islets that is first detected at 1 unit/ml IL-1␤ and increases linearly to 50 units/ml IL-1␤. Nitrite production stimulated by 75 units/ml IL-1␤ and 750 units/ml human IFN-␥ is slightly lower than the level of nitrite produced by human islets treated with 50 units/ml IL-1␤ and 750 units/ml human IFN-␥. In the presence of 75 units/ml human IFN-␥ and IL-1␤ at concentrations from 1 to 75 units/ml, lower levels of nitrite are produced as compared with 750 units/ml human IFN-␥; however, 10 units/ml IL-1␤ and 75 units/ml human IFN-␥ induce a 2-fold increase in nitrite production as compared with untreated human islets. Although human IFN-␥ reduces the concentration of IL-1 required to stimulate iNOS expression by human islets, the concentrations of human IFN-␥ used in these studies may be higher than those present in and around islets during the development of diabetes.

DISCUSSION
In this report we have examined the effects of IFN-␥, alone and in combination with IL-1, on iNOS expression and nitrite production by both rat and human islets of Langerhans. We demonstrate that rat IFN-␥ increases the sensitivity of rat islets for IL-1-induced iNOS expression by 10-fold. Alone, concentrations of IL-1␤ as low as 0.1 unit/ml (0.57 pM) or 150 units/ml rat IFN-␥ do not induce iNOS expression or nitrite production by rat islets, RINm5F cells, or primary ␤-cells purified by FACS; however, in combination, these cytokines induce the expression of iNOS and the production of nitrite to similar levels induced by maximal concentrations of IL-1␤. Also, IL-1␤, alone or in combination with rat IFN-␥, does not stimulate nitrite formation or iNOS expression by primary ␣-cells, the other major endocrine cell type found in islets. These results are consistent with previous studies that have identified the ␤-cell as the islet cellular source of iNOS in response to IL-1␤ (12, 13) and provide the first direct evidence that IFN-␥ modulates the function and viability of primary ␤-cells by increasing the sensitivity of ␤-cells for IL-1␤-induced iNOS expression. The combination of IL-1␤ and rat IFN-␥ also results in inhibition of insulin secretion and islet destruction that are prevented by the iNOS inhibitor AG. These findings indicate that nitric oxide participates in the inhibitory and destructive effects of submaximal concentrations of IL-1␤ plus rat IFN-␥ on insulin secretion and islet destruction.
The mechanism by which IFN-␥ increases the sensitivity of ␤-cells for iNOS expression and nitrite production in response to IL-1␤ appears to be associated with an increase in the stability of iNOS mRNA. Consistent with previous studies (34), maximal concentrations of IL-1␤ induce an 8-fold increase in the accumulation of iNOS mRNA following a 6-h exposure; however, iNOS mRNA accumulation is reduced to near background levels following a 12-h incubation. In contrast, nearly equivalent levels of iNOS mRNA accumulate in RINm5F cells treated for 6 or 12 h with submaximal concentrations of IL-1␤ in combination with rat IFN-␥. Stability studies indicate that iNOS mRNA induced by IL-1␤ and rat IFN-␥ is approximately 2-fold more stable than the individual effects of IL-1␤. These findings indicate that IFN-␥ may induce the expression of factors, and/or activate factors that stabilize iNOS mRNA, thus preventing iNOS mRNA degradation or that IFN-␥ may inhibit the activity of factors that are required for down-regulating iNOS mRNA expression, and/or the degradation of iNOS mRNA. These possibilities are currently under investigation.
It is clear from studies with rat islets that IFN-␥ reduces the concentration of IL-1␤ required to stimulate iNOS expression by ␤-cells; however, it is important to evaluate the effectiveness of human IFN-␥ on iNOS expression by human islets treated with maximal and submaximal concentrations of IL-1␤ to determine if human islets respond in a similar manner. In this study we show that IL-1␤, at a concentration as low as 1 unit/ml (5.7 pM), is able to stimulate high levels of nitrite production by human islets in the presence of human IFN-␥ (750 units/ml). We also show that in the presence of 75 units/ml human IFN-␥, as little as 10 units/ml IL-1␤ is required to induce a 2-fold increase in the level of nitrite production. These results indicate that IFN-␥ reduces the concentration of IL-1␤ required to stimulate iNOS expression by human islets in a manner similar to IFN-␥'s effects on rat islets.
We have also evaluated the effects of endogenous IL-1 release on iNOS expression and nitrite production by rat islets. Macrophages are believed to play a primary role in the development of autoimmune diabetes. Islets contain approximately 10 -15 resident macrophages. Macrophage depletion, by silica treatment or feeding a diet deficient in essential fatty acids, prevents the natural occurrence of diabetes in the Bio Breeding rat and prevents the development of diabetes induced by multiple injections of streptozotocin in CD-1 mice (35)(36)(37). We have previously shown that treatment of rat islets with tumor necrosis factor ϩ lipopolysaccharide, conditions known to activate macrophages, results in the release of IL-1 within islets and that IL-1 subsequently stimulates iNOS expression and nitric oxide production by ␤-cells resulting in the inhibition of ␤-cell function (12). These studies have led to the suggestion that activation of resident islet macrophages may represent a triggering event associated with the initiation of autoimmune diabetes (12,38). We show that IFN-␥, in the absence of exogenously added IL-1, induces the expression of iNOS and the production of nitrite by islets physically dispersed into individual cells by trypsin treatment. Under these conditions, iNOS expression and nitrite production are prevented by IRAP, indicating that during the dispersion process low levels of IL-1 are released from cells that may be damaged during this procedure. The cellular source of IL-1 in the dispersed islet cells appears to be resident macrophages. Macrophage depletion of islets prior to islet dispersion completely prevents IFN-␥-induced nitrite formation and iNOS expression. Cellular mechanisms associated with macrophage release of IL-1 are incompletely characterized; however, macrophage death, by either apoptosis or necrosis, is known to result in the release of IL-1 (39,40). Macrophage damage and IL-1 release during islet dispersion is consistent with these previous studies showing IL-1 release following macrophage death.
The release of IL-1, in islets following cellular damage, represents a novel mechanism that may be associated with the initiation of autoimmune diabetes. Environmental toxins or viral infections have been proposed to target and destroy ␤-cells during the initiation of autoimmune diabetes (41,42). In contrast, our data support a potential role for the resident islet macrophage as a target for such an event. Viral infection of macrophages, or macrophage death stimulated by chemical toxins, in islets under immune surveillance or peri-insulitis (conditions associated with the migration but not infiltration of T-lymphocytes into islets), could result in the release of IL-1 in the local environment of the islet. In the presence of IFN-␥, the release of low levels of IL-1 would be sufficient to stimulate iNOS expression and nitric oxide production by ␤-cells. Nitric oxide production by ␤-cells would then result in the inhibition of function and the destruction of ␤-cells, resulting in the release of autoantigens and the initiation of a T-cell-mediated immune response. In addition, tissue damage associated with acute and chronic inflammation and injury may be mediated by a similar mechanism. Low levels of IL-1 released in the presence of IFN-␥ could result in high levels of iNOS expression and nitric oxide production by target cells, leading to target cell damage. Under these conditions, macrophage production of nitric oxide may participate in target tissue damage; however, as our studies indicate, cytokine release by activated resident macrophages and cytokine-induced iNOS expression by target cells may be the more important mechanism associated with tissue damage. If our hypothesis is correct, human macrophage production of nitric oxide is not required for target cell damage. This interpretation is consistent with the difficulties in demonstrating human macrophage production of nitric oxide. In conclusion, our studies show that the T-cell cytokine, IFN-␥, directly inhibits the function and viability of islets by reducing the concentration of IL-1 required to stimulate iNOS expression and nitric oxide production by ␤-cells. These findings support an effector role for IFN-␥, in concert with IL-1, in mediating the initial destruction of ␤-cells during the development of autoimmune diabetes.