Evidence That β Cell Death in the Nonobese Diabetic Mouse Is Fas Independent

Helen E. Thomas, Rima Darwiche, John A. Corbett and Thomas W. H. Kay
J Immunol August 1, 1999, 163 (3) 1562-1569;

Abstract

Recent studies suggest that Fas expression on pancreatic β cells may be important in the development of autoimmune diabetes in the nonobese diabetic (NOD) mouse. To address this, pancreatic islets from NOD mice were analyzed by flow cytometry to directly identify which cells express Fas and Fas ligand (FasL) ex vivo and after in vitro culture with cytokines. Fas expression was not detected on β cells isolated from young (35 days) NOD mice. In vitro, incubation of NOD mouse islets with both IL-1 and IFN-γ was required to achieve sufficient Fas expression and sensitivity for islets to be susceptible to lysis by soluble FasL. In islets isolated from older (≥125 days) NOD mice, Fas expression was detected on a limited number of β cells (1–5%). FasL was not detected on β cells from either NOD or Fas-deficient MRLlpr/lpr islets. Also, both NOD and MRLlpr/lpr islets were equally susceptible to cytokine-induced cell death. This eliminates the possibility that cytokine-treated murine islet cells commit “suicide” due to simultaneous expression of Fas and FasL. Last, we show that NO is not required for cytokine-induced Fas expression and Fas-mediated apoptosis of islet cells. These findings indicate that β cells can be killed by Fas-dependent cytotoxicity; however, our results raise further doubts about the clinical significance of Fas-mediated β cell destruction because few Fas-positive cells were isolated immediately before the development of diabetes.

Fas (CD95/APO-1) is a member of the TNFR family that is able to signal apoptosis via a conserved intracellular death domain. Fas is constitutively expressed on a number of mouse tissues including thymocytes. Its expression can be up-regulated by IFN-γ or combinations of IFN-γ and TNF-α or IL-1 in various cells lines. The ligand for Fas (FasL)3 is sufficient for induction of apoptosis of Fas-expressing cells and is more effective as a membrane-bound protein although it is functional in soluble form. The Fas-FasL interaction leads to activation of an intracellular signaling pathway that results in cell death by caspase-dependent mechanisms (reviewed in Refs. 1 and 2).

Type 1 insulin-dependent diabetes mellitus (IDDM) is a T cell-mediated disease in which recent evidence, mostly from the nonobese diabetic (NOD) mouse model, suggests that pancreatic β cells are destroyed by apoptotic mechanisms (3, 4, 5). It has been suggested that Fas may play a role in β cell apoptosis in IDDM both in humans and in the NOD mouse. Much of the evidence for this comes from in vitro studies with isolated human or mouse islets of Langerhans. IL-1 has been shown to up-regulate Fas expression on β cells, resulting in apoptosis after the addition of agonist anti-Fas Ab (6, 7, 8). IL-1 in combination with IFN-γ has also been shown to up-regulate inducible NO synthase (iNOS) expression in rodent and human β cells, leading to NO-dependent islet dysfunction (9, 10). One study suggested that NO is able to regulate Fas expression on human β cells, and not on other cell types within the islet (11), providing a potential mechanism for selective β cell destruction in IDDM.

The spontaneous mouse mutant lpr (lymphoproliferation) lacks functional Fas expression due to a mutation in the gene (12). When this mutation is crossed onto a NOD genetic background (NODlpr/lpr), the mice do not develop diabetes or insulitis, either spontaneously or in T cell adoptive transfer experiments, suggesting that Fas expression is required for disease (13, 14). However, recent reports suggest that Fas plays only a minor role in β cell destruction, as Fas-deficient islet grafts are not protected from diabetogenic T cells (15, 16). Immunohistochemical staining of human pancreas sections revealed Fas expression in islets of diabetic but not normal subjects, with FasL expression on infiltrating T cells, suggesting a mechanism for β cell destruction (11). In mice, Chervonsky et al. found that Fas expression on islet cells of old NOD mice (≥12 wk) was up-regulated after transfer of a diabetogenic T cell clone; however, Fas expression in the spontaneous NOD mouse model was not examined (13). Also, Fas expression was found on β cells from syngeneic islets grafted into diabetic NOD mouse recipients, and this correlated with expression of inflammatory cytokines (17). However, the data in all of these studies using two-color immunohistochemical staining are difficult to interpret, especially when looking for Fas, FasL, and apoptosis in infiltrated islets where staining on the infiltrating cells may interfere with the β cell staining. Flow cytometry is a more powerful technique for these studies, where specific cell types can be easily identified.

Although a large body of in vitro evidence suggests that Fas may participate in β cell destruction, it remains to be determined whether Fas is involved in the natural progression of diabetes in the NOD mouse, which has different mechanisms of disease from adoptive transfer models. Furthermore, the lpr mutation has multiple effects such as lymphadenopathy, constitutive up-regulation of FasL, and dysregulation of T cell populations (18, 19), which makes it difficult to conclude that Fas is required for the development of diabetes in NOD mice. Additionally, some reagents that have been used to examine FasL expression have been shown to be nonspecific (20), making it difficult to confirm that Fas-FasL participates in β cell destruction during the development of diabetes.

In this study, we have used techniques to directly address some of these problems with the literature. We have defined Fas expression on β cells from NOD mice of different ages by flow cytometry, where β cells can easily be distinguished from other islet cells and infiltrating cells. We present evidence indicating that up-regulation of Fas expression on the majority of NOD mouse β cells does not correlate with the onset of autoimmune diabetes, although expression on a few cells suggests that some Fas-mediated lysis may occur. We have also used MRLlpr/lpr islets to demonstrate that killing of β cells by Fas and FasL coexpression is not important in cytokine-induced damage.

Materials and Methods

Mice and cytokines

NOD/Lt and MRLlpr/lpr mice were housed in the animal facility at the Walter and Eliza Hall Institute of Medical Research. The prevalence of diabetes of NOD/Lt mice at our institution is 70–75% of females and 15% of males by 300 days.

Recombinant murine IFN-γ (used at 100 U/ml) was obtained from Genentech (South San Francisco, CA), and recombinant human IL-1β (10–100 U/ml) from Genzyme (Cambridge, MA). Soluble human FasL (100 ng/ml) was kindly provided by Dr. J. Tschopp (University of Lausanne, Lausanne, Switzerland). NG-monomethyl-l-arginine (NMMA) (3 mM) and sodium nitroprusside (SNP) (100 μM) were purchased from Sigma (St. Louis, MO). Z-VAD-fmk (1 μg/ml) was provided by Dr. G. Vairo (Walter and Eliza Hall Institute)

Islet isolation

Islets of Langerhans were isolated from mice by pancreas digestion with collagenase P (Boehringer Mannheim, Mannheim, Germany) followed by a BSA density gradient (First Link, U.K.), as previously described (21, 22). Islets were hand picked and cultured at 37°C in 5% CO2 in CMRL Medium-1066 (Life Technologies, Gaithersburg, MD) supplemented with 10% FCS and antibiotics.

Flow cytometry

For flow cytometric analysis, islets were either stained on the day of isolation or cultured with cytokines in DMEM containing 2.5 mM glucose, 10% FCS, and antibiotics. This glucose concentration was used to allow for detection of β cells on the basis of high autofluorescence due to intracellular flavin adenine dinucleotide (23). There is no evidence that expression of cell-surface molecules is altered by incubation of islets in low glucose, and we have used the islet endocrine cell marker A2B5 (24) on islets incubated in normal (10 mM) glucose to show this (not shown). We have examined the cell types present in the autofluorescent populations of islets analyzed by flow cytometry. Autofluorescent cells were sorted (FACStar+; Becton Dickinson, Mountain View, CA), fixed onto slides by cytospin, and stained by direct immunofluorescence. Eighty to 90% of the autofluorescent cells stained positive for insulin, and all were positive for A2B5, suggesting that the cells are all of islet endocrine nature (not shown).

Islets for staining were dispersed into single cells with 0.2% trypsin (Calbiochem, La Jolla, CA), 10 mM EDTA in HBSS, and allowed to recover in culture medium for 1 h before staining with mAb. Islet cell suspensions were stained using standard procedures. Cells were finally washed and resuspended in balanced salt solution with 2% FCS and 1 μg/ml propidium iodide to stain dead cells. Analysis was performed on a FACScan (Becton Dickinson). Hamster anti-mouse Fas (Jo2; PharMingen, San Diego, CA) was used for analysis of surface Fas expression, with hamster anti-human B anti-human 2 as an isotype control Ab for background staining. This was followed by biotinylated anti-hamster Ig (PharMingen) and PE-conjugated streptavidin (Caltag Laboratories, Burlingame, CA).

Apoptosis assay

Quantitation of apoptosis was determined according to the method of Nicoletti et al. (25). After 3–5 days in culture with cytokines, islets were dispersed into a single-cell suspension with trypsin, followed by recovery in complete medium at 37°C for 30 min. The cells were then resuspended in a hypotonic fluorochrome solution (50 μg/ml propidium iodide, 0.1% sodium citrate, 0.1% Triton X-100) and kept overnight in the dark at 4°C. Cells were then analyzed on the flow cytometer. This assay measures fragmented nuclei, and therefore greater than one fragment can be derived from one apoptotic cell. The level of induced apoptosis over spontaneous was determined by the formula 100 × (% number of apoptotic cells with cytokine − % number of background apoptotic cells)/(100 − % number of background apoptotic cells). This formula allows results from experiments with different levels of basal apoptosis to be combined. Data from at least three individual experiments were pooled. All experiments were conducted on islets isolated from ≤40-day-old NOD mice.

Nitrite determination

Nitrite was detected in the cultures by mixing 50 μl supernatant with 50 μl Griess reagent (26). Absorbances were read at 540 nm, and nitrite concentration was calculated off a standard curve with sodium nitrite.

Results

Fas is up-regulated on primary murine β cells following exposure to cytokines

To confirm that Fas expression can be induced by cytokines on murine β cells, islets isolated from young (≤40 days) NOD mice were cultured in vitro with combinations of the cytokines IFN-γ and IL-1β. Fas expression was then examined on β cells by flow cytometry. Alone, IL-1β (10 U/ml) stimulates low levels of Fas expression on NOD β cells (Fig. 1⇓A), and this is enhanced by the addition of 100 U/ml IFN-γ (Fig. 1⇓B). IFN-γ alone did not significantly up-regulate β cell expression of Fas (Fig. 1⇓A) when compared with either untreated islets or islets isolated from MRLlpr/lpr mice, which are deficient in Fas (Fig. 1⇓). Fas expression was induced by IL-1 and IL-1 plus IFN-γ after a 24-h incubation with cytokines and was not further increased with longer incubations (not shown). These data are consistent with the findings of Yamada et al., which showed Fas up-regulation on murine islets by IL-1α (6). The finding that cytokines up-regulate Fas is not unique to NOD β cells. We have also analyzed Fas expression after cytokine treatment of islets isolated from C57BL/6 mice with the same results (not shown).

FIGURE 1.
FIGURE 1.

Fas is induced on β cells in vitro by IL-1 and IFN-γ. Islets isolated from 40-day-old NOD mice and MRLlpr/lpr mice were treated in culture for 24 h with IL-1 and IFN-γ either (A) alone or (B) in combination. Islet cell suspensions were then stained with hamster anti-mouse Fas followed by biotinylated anti-hamster Ig and PE-streptavidin and were analyzed by flow cytometry. β cells were identified and electronically gated by autofluorescence.

Murine islets undergo Fas-mediated apoptosis

To demonstrate that Fas expression on β cells is functional, islets were incubated with cytokines plus soluble FasL (sFasL) (27), and apoptosis was measured by propidium iodide staining of fragmented nuclei. Concentrations of sFasL between 1 and 1000 ng/ml were examined, and sFasL-induced killing was maximal at 100 ng/ml (not shown). This concentration was used for all subsequent experiments. IFN-γ (100 U/ml) and IL-1β (10 U/ml) induced a 2- to 4-fold increase in DNA fragmentation of islet cells over background after a 4-day culture (Fig. 2⇓, A and B). This cell death was attenuated by the iNOS inhibitor NMMA (3 mM), indicating that cell death induced by cytokines (in the absence of sFasL) is due to the increased production of NO (Fig. 2⇓). In the presence of IFN-γ and IL-1β (to up-regulate Fas) and sFasL, the level of cell death increased 8- to 10-fold over background, demonstrating that Fas expression induced by cytokines is functional. When NMMA was added with cytokines and sFasL, cytokine-induced NO-mediated cell death was inhibited, but FasL-mediated death was still observed (Fig. 2⇓A). Alone, sFasL did not induce islet cell death compared with untreated islets, indicating the absence of functional Fas expressed constitutively on the islet cells of young NOD mice. FasL-mediated death was inhibited by the caspase inhibitor Z-VAD-fmk, demonstrating that the islet cells are killed by a caspase-dependent mechanism. Fas-deficient MRLlpr/lpr islets were used as a control for Fas-independent cytokine-induced cell death (Fig. 2⇓C). These islets were found to be as susceptible to NO-dependent cell death as NOD islets, but this apoptosis did not increase upon addition of sFasL to the lpr islets, as expected.

FIGURE 2.
FIGURE 2.

Islets are susceptible to Fas-mediated lysis in vitro. Islets were isolated from 35- to 45-day-old NOD mice and counted into petri dishes with 100 islets/ml. Cytokines were added as indicated and the islets were left undisturbed for 4 days. Percentage DNA fragmentation was analyzed by flow cytometry. A, Mean and SD of data from at least three independent experiments. B, NOD and C, MRLlpr/lpr islets treated with cytokines as indicated for 4 days followed by DNA fragmentation analysis. Data obtained in one representative experiment is shown.

Fas-induced death of islet cells is independent of NO

Cytokines induce the up-regulation of iNOS and the production of NO by β cells in vitro, and it has been demonstrated that NO, either induced by cytokines or exogenously supplied by donor compounds, is able to prime human β cells for Fas-mediated killing by up-regulation of Fas (11). The mechanism by which NO modulates Fas expression by human islets is unknown. We have examined whether NO could regulate murine β cell expression of Fas and directly modulate Fas-sFasL-mediated islet cell killing by analyzing the effects of endogenously produced or exogenously added NO on both Fas expression and FasL-mediated islet cell death. To confirm that cytokines stimulate NO production by NOD mouse islets, we show in Fig. 3⇓ that incubation with IL-1 plus IFN-γ for 24 h results in the increased production of nitrite and nitrite production is attenuated by NMMA. To determine whether NO participates in cytokine regulation of Fas expression, the effect of NMMA on IL-1 plus IFN-γ-induced Fas expression was examined by flow cytometry.

FIGURE 3.
FIGURE 3.

Nitrite production after in vitro culture of islets with cytokines. Culture medium from islets treated with cytokines was sampled after 48 h for the presence of nitrite. Mean and SD are shown of data from at least three independent experiments.

Inhibiting cytokine-induced NO production does not inhibit Fas expression by NOD mouse islets after 24-h incubation (Fig. 4⇓A). Also, the exogenous addition of NO using the NO donor compound SNP does not induce Fas expression on β cells (Fig. 4⇓B), despite detection of 222 ± 98.7 pmol nitrite in the culture supernatant after 24-h incubation with SNP. These results indicate that NO does not participate in up-regulation of Fas. Thus, while NO is able to cause damage to the islets, it does not appear to be involved in Fas-mediated destruction of NOD islets.

FIGURE 4.
FIGURE 4.

Nitric oxide has no effect on Fas expression on β cells. NOD mouse islets were treated in vitro for 24 h as indicated. Cell suspensions were then stained with hamster anti-mouse Fas, biotinylated anti-hamster Ig, and PE-streptavidin, followed by analysis by flow cytometry. β cells were electronically gated. A, Fas expression on β cells after inhibition of iNOS with 3 mM NMMA. B, Fas expression on β cells after treatment of islets with the NO donor compound SNP. Mean and SDs from at least three independent experiments are shown. C, DNA fragmentation analysis after incubation with the NO donor compound SNP. One representative experiment is shown.

SNP was unable to induce death of islet cells on its own or in the presence of sFasL and IFN-γ (after 4-day culture), again showing that NO is unable to affect Fas expression (Fig. 4⇑C). The fact that a high level of exogenous NO is unable to induce cell death suggests that NO production by islet cells themselves is required for damage to occur.

FasL is not detectable on β cells of NOD mice

Several groups have reported expression of FasL on human and mouse islets, proposing direct self-killing as a mechanism for β cell destruction (8, 28). These studies have been conducted using immunohistochemical techniques with antisera that may not be specific for FasL (20). We have analyzed FasL expression on β cells by flow cytometry, which is a more sensitive method, using MFL-3 (hamster anti-FasL Ab), which has been shown to be specific for FasL (29). Islets were isolated from NOD mice of between 40 and 120 days of age and stained for FasL expression. As a positive control, islets from rat insulin promoter (RIP)-FasL transgenic animals were stained at the same time (30). While a high level of FasL staining was consistently detected on RIP-FasL transgenic β cells, indicating that the process of islet isolation does not prevent FasL detection, we did not observe FasL staining on β cells isolated from nontransgenic littermates or from either young or old NOD mice (Fig. 5⇓). FasL was also not detected by immunohistochemical analysis (L. Stephens and P. Augstein, unpublished observations). In addition, we were unable to detect FasL staining after incubation of islets with the cytokines IFN-γ and IL-1 (not shown).

FIGURE 5.
FIGURE 5.

FasL is not detectable on normal β cells. Islet cells were stained on the day of isolation with MFL-3 (hamster anti-mouse FasL) followed by biotinylated anti-hamster Ig and PE-streptavidin. Staining on β cells is shown. A, FasL expression on β cells from RIP-FasL transgenic mice or nontransgenic littermates. B, FasL staining on β cells from 45- and 105-day-old NOD mice.

Although we did not detect FasL expression on NOD β cells, we cannot rule out the possibility that low levels, undetectable by flow cytometry, are expressed, particularly as detection of FasL on activated T cells is difficult (not shown). To further explore this possibility, we analyzed cytokine-induced cell death in islets deficient in Fas expression. MRLlpr/lpr and NOD islets were equally susceptible to cytokine-mediated cell death (Fig. 2⇑C). If FasL were constitutively expressed on β cells and Fas were induced by cytokines, the level of cell death in the normal NOD islets with cytokines alone would be expected to be higher than that seen in the MRLlpr/lpr islets. This is not the case. Thus direct autolysis of β cells by Fas/FasL cannot occur in NOD mouse islets. We have also repeated this experiment with NODlpr/lpr islets with the same result (not shown). As expected, cell death of lpr islets was not enhanced by the addition of sFasL (Fig. 2⇑C).

Relevance of Fas-mediated apoptosis of β cells in vivo

These results and those of others have clearly demonstrated that Fas expression is induced on murine β cells by cytokines in vitro. However, the relevance of Fas expression to the induction of autoimmune diabetes in the NOD mouse is unclear. Evidence for the role of Fas in IDDM comes primarily from NODlpr/lpr mice, which do not develop insulitis or diabetes (13, 14, 15). In these NODlpr/lpr experiments, the importance of Fas expression on β cells and Fas-mediated β cell death is unclear. In NOD mice, IFN-γ and IL-1 are expressed within the islet during the development of autoimmune diabetes. To examine the role of cytokines and Fas during the progression of islet insulitis and diabetes development, we assessed Fas expression on β cells of NOD mice at various stages of disease.

A group of 60 female NOD mice was set aside to monitor diabetes incidence and Fas expression on islets. Five mice per age group were killed for islet isolation at 120, 125, 130, and 135 days of age. Blood glucose measurements (range 5.2–9.3 mmol/L) were determined before islet isolation from individual mice. Although these mice were nondiabetic, some had elevated blood glucose measurements that suggests they were close to diabetes, and many other mice in the cohort were also progressing to diabetes. Additionally, the number of islets isolated from each mouse was determined (between 17 and 370 per mouse), and the mice with low islet numbers may have been prediabetic, undergoing β cell destruction and islet loss. Islets were stained for Fas expression, and levels were compared with an isotype control Ab, Fas staining on Fas-deficient MRLlpr/lpr islets, and on islets from 35-, 70-, and 90-day-old NOD mice. A high level of Fas expression was detected on infiltrating T cells. Even though there were relatively few infiltrating cells in the young islets, the level of Fas staining on T cells did not change as the mice became older and developed more severe insulitis (Fig. 6⇓). T cell Fas staining was used as an internal positive control and shows that Fas expression is maintained during the islet isolation procedure.

FIGURE 6.
FIGURE 6.

Fas expression on NOD islet cells. Islets were isolated from NOD mice and cell suspensions were stained on the day of isolation with hamster anti-mouse Fas, biotinylated anti-hamster Ig, and PE-streptavidin followed by analysis of cell populations by flow cytometry. A, Fas expression on infiltrating lymphocytes from NOD mice of 47, 92, and 120 days of age. B, Dot plot and histogram profiles showing Fas expression of cells from (i) MRLlpr/lpr mice and NOD mice of (ii) 35 and (iii) 130 days old. The β cells are identified by their high autofluorescence (shown on the x-axis of profiles on the left). Infiltrating cells can be seen in the older NOD mice as the strongly Fas positive/weak autofluorescence population. Histograms of the autofluorescent β cell populations (solid lines) are overlaid with that of isotype control Ab (dotted lines). C, Diabetes incidence of female NOD/Lt mice between 100 and 140 days of age. Mice were monitored for diabetes by urine glucose measurement with a follow-up blood glucose determination if required. Mice were considered diabetic with a blood glucose over 15 mmol/L.

Dot plots and histograms representative of Fas staining on islets isolated from MRLlpr/lpr, 35- and 130-day-old NOD mice are shown in Fig. 6⇑B. Fas expression was only observed on 1–5% of the β cells from mice of ≥125 days of age, and the level of staining was not as high as that seen after in vitro culture with cytokines. This is more apparent when the histograms of β cell Fas staining are overlaid with that of isotype control staining (Fig. 6⇑B). Diabetes incidence in the mice not killed for islet isolation (n = 47) during the experiment was 75% at 135 days of age (Fig. 6⇑C). This indicates that we analyzed Fas expression on islets during a period when many of the mice are developing diabetes. Islets from mice 70–90 days of age were also studied with similar results (not shown). Therefore, unless Fas-expressing β cells are very short-lived and not detected in this experiment, it is unlikely that a large number of β cells express Fas before the onset of autoimmune diabetes, raising doubt whether Fas-mediated lysis of β cells plays an important role in β cell destruction in NOD mice.

Discussion

The techniques available for studying expression of cell death-inducing molecules and apoptosis of pancreatic β cells have generally been restricted to immunohistochemical analysis, which can be difficult to interpret due to the presence of many cell types within NOD islets on histological sections. We have avoided this problem by using alternative techniques that enable us to identify specific cell types within the islet expressing Fas and FasL. We have found that while Fas is expressed at high levels by infiltrating lymphocytes, only a limited number β cells from prediabetic NOD mice express Fas at low levels. It has been suggested that nondestructive insulitis progresses throughout the life of the NOD mouse, but that β cell destruction is rapid and occurs immediately before the onset of diabetes (31). If this were the case and Fas-mediated lysis were a mechanism for β cell death, we would expect to see large numbers of Fas-positive cells in mice close to the time that diabetes occurs, which we did not observe.

Alternative mechanisms of β cell destruction include soluble factors such as cytokines and NO, as well as perforin released by CTL. Perforin and FasL are believed to be the major mechanisms of killing by CTL. It is known that perforin plays a role in the effector phase of β cell destruction as diabetes is reduced and delayed in perforin knockout mice in the presence of infiltrate similar to that of wild-type mice (32). However, it is likely that a combination of factors is important for diabetes to occur in the NOD mouse. We have shown that class I MHC expression on β cells is required for insulitis, also implicating CD8+ T cells in initiation of disease (33, 34).

Our data suggest that the β cells we isolated from NOD mice may not have been exposed to high levels of IL-1 in vivo because levels of Fas expression achieved in vitro by addition of IL-1 and IFN-γ were not observed on cells from infiltrated islets. IFN-γ is present in the infiltrated islet (35) at sufficient levels to uniformly up-regulate class I MHC, for example. We have shown that NOD mice of increasing age and degree of insulitis have gradually increasing levels of class I MHC expression on β cells and that mice with β cells unresponsive to IFN-γ have no such class I MHC up-regulation, despite the presence of insulitis (22). This shows that the rise in class I MHC seen on β cells is due to local IFN-γ. Both resident and infiltrating macrophages are thought to produce IL-1 (36, 37, 38). As both IL-1 and IFN-γ are required for in vitro up-regulation of functional Fas, either IL-1 concentrations may be limiting or factors which inhibit IL-1-iduced Fas expression may be present within the islet. It is also possible that β cells in close proximity to macrophages have up-regulated Fas expression, and that these cells may be rapidly destroyed by FasL-bearing T cells. This could explain the low numbers of Fas-positive β cells observed in NOD islets.

We have demonstrated that β cell “suicide” due to β cell expression of both Fas and FasL does not occur. β cells did not appear to express FasL while mice expressing FasL transgenically displayed high levels of FasL. These results suggested that FasL expression was maintained throughout the islet isolation procedure and we were not missing FasL expression on normal NOD β cells due to technical problems. We cannot rule out the possibility that FasL is expressed at levels undetectable by flow cytometry, although our data show that such levels are not functional. In the absence of soluble FasL, IL-1 and IFN-γ kill MRLlpr/lpr islets to the same extent as normal NOD islets in a NO-dependent fashion. Thus, it is unlikely that FasL is constitutively expressed on β cells and the cells are lysed due to up-regulation of Fas by cytokines, as has been suggested by others (8). FasL is most likely expressed on activated T cells within the infiltrate.

IFN-γ is required for Fas-mediated lysis of islets in vitro. IL-1 and sFasL do not cause islet cell death in vitro, even though IL-1 is able to up-regulate low levels of Fas expression on β cells. IFN-γ on its own has little effect on Fas expression, but in combination with IL-1 leads to enhanced Fas up-regulation in vitro. It is possible that the high levels of Fas induced by IL-1 plus IFN-γ are required for activated cell death. Humans with a mutation in the Fas gene have the rare disease autoimmune lymphoproliferative syndrome, despite having one normal copy of Fas. It has been suggested that the mutant gene acts as a dominant negative receptor (39, 40); however, it is also possible that insufficient levels of Fas are expressed with only one functional copy of the gene. Although no phenotype is observed in mice heterozygous for the lpr mutation, decreased killing of cells from these mice has been observed in vitro (41). Our data support the possibility that high levels of Fas are required for killing. In addition to its effect on Fas expression, IFN-γ is known to regulate caspase-1 expression (42, 43), and may also activate other intracellular cell death signaling molecules. Our data show that caspases are important mediators of β cell destruction, as inhibition of caspases with Z-VAD-fmk prevented cytokine-induced apoptosis due to Fas-FasL interaction.

Stassi et al. reported that functional Fas expression on human β cells is induced by NO (11). This has also been shown in other cell types including vascular smooth muscle cells (44), and NO has been shown to regulate expression of several other genes such as IL-8 (45) and superoxide dismutase (46) by various mechanisms. However NO has also been shown to inhibit Fas-mediated apoptosis by suppressing caspase-3 activation in a cyclic GMP-independent manner (47, 48, 49), demonstrating the wide range of effects NO has on cells, possibly depending on its concentration (50). We found that NO had no effect on either Fas expression or Fas-mediated apoptosis of islet cells from NOD mice. The differences between our result in mice and that of Stassi et al. in human islets suggests a species difference in the effects of cytokines and cell death-inducing agents on islets. We also found that while cytokine-induced NO production was toxic to islets in vitro, exogenously supplied NO was not, even at higher concentrations than those produced by islet cells. β cells produce NO in response to cytokines (37). Also, transgenic mice expressing iNOS in their β cells develop insulin-dependent diabetes (51). Therefore, it is likely that NO production by the islet cells themselves is required for toxicity.

We have also presented evidence that Fas is not constitutively expressed on β cells of young NOD mice before the onset of insulitis and that as insulitis progresses only a few β cells express Fas. The lack of Fas staining on β cells from young mice suggests that Fas-mediated lysis of β cells is probably not involved in initiation of disease. The patchy, low level expression on β cells from mice with insulitis implies that Fas plays only a minor role in the effector phase of disease. Fas-deficient NODlpr/lpr fetal pancreas grafts are unable to reverse diabetes in NOD mice as they undergo T cell-mediated destruction (15). These grafts appeared slightly less damaged than wild-type NOD grafts, suggesting that there may be some Fas-mediated β cell destruction in autoimmune diabetes. Therefore, it is possible that the few Fas positive β cells within the islets of old NOD mice are playing a role in β cell destruction, in addition to perforin-dependent lysis, which has been shown in knockout animals to be required for the effector phase of autoimmune diabetes. It is likely that other mechanisms of β cell destruction exist, including Fas, TNFR1 (52), and nitric oxide-mediated damage. The effects of making any one of these deficient may be difficult to observe in the presence of perforin.

The role of Fas-mediated destruction of β cells is very topical, with many seemingly inconsistent findings being published. Of these findings, it is clear that Fas-deficient NODlpr/lpr mice to not develop pathology (13, 14, 15). However, using NODlpr/lpr pancreas grafts, it has been suggested that this absence of disease may be due to other immune defects in the lpr mouse (15, 16). The published data on Fas expression by β cells (11, 13, 17) are consistent with our flow cytometry data and reveal at best only a limited number of Fas-positive β cells. Analysis of FasL expression by β cells (8) was performed using antisera of doubtful specificity, and there is little functional evidence for β cell FasL expression. As these studies were performed with human islets, it is possible that the data differ from ours due to species differences; however, there is no convincing demonstration in the literature of FasL expression by β cells.

In conclusion, we have shown that there is a limited amount of Fas expression in β cells from NOD mice. FasL expression on β cells is undetectable both by flow cytometry and functional analysis using islets from MRLlpr/lpr mice. This is contrary to other studies using two-color immunohistochemistry and co-localization on serial sections as methods for Fas and FasL detection. While we do not yet have a complete understanding of the role of Fas and FasL in the development of IDDM, more recent data, including ours, makes a major role of Fas-FasL interactions in β cell death uncertain. Our inability to find Fas expression on the majority of β cells does not equivocally rule out a role for Fas in β cell death in NOD mice. This will be directly tested by several groups making transgenic NOD mice with FasL-insensitive β cells.

Acknowledgments

We thank Dr. Janette Allison for providing RIP-FasL transgenic mice and NODlpr/lpr mice, as well as for many helpful discussions. We also thank Dr. Leigh A. Stephens and Dr. Petra Augstein for discussion about unpublished observations and critical reading of the manuscript.

Footnotes

  • ↵1 This work was supported by the National Health and Medical Research Council of Australia (Regkey 973002), a Diabetes Interdisciplinary Research Program and Career Development Awards (to J.A.C. and T.W.H.K.) from the Juvenile Diabetes Foundation International, and National Institutes of Health Grants DK-52194 and AI 44458.

  • ↵2 Address correspondence and reprint requests to Dr. Thomas W. H. Kay, Autoimmunity and Transplantation Division, Walter and Eliza Hall Institute, P.O. Royal Melbourne Hospital, Victoria 3050, Australia. E-mail address: kay{at}wehi.edu.au

  • ↵3 Abbreviations used in this paper: FasL, Fas ligand; NOD, nonobese diabetic; IDDM, insulin-dependent diabetes mellitus; iNOS, inducible NO synthase; SNP, sodium nitroprusside; NMMA, NG-monomethyl-l-arginine; sFasL, soluble FasL; RIP, rat insulin promoter.

  • Received March 4, 1999.
  • Accepted May 18, 1999.

References

  1. ↵
    Nagata, S., P. Golstein. 1995. The Fas death factor. Science 267: 1449
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Ashkenazi, A., V. M. Dixit. 1998. Death receptors: signaling and modulation. Science 281: 1305
    OpenUrlAbstract/FREE Full Text
  3. ↵
    O’Brien, B. A., B. V. Harmon, D. P. Cameron, D. J. Allan. 1997. Apoptosis is the mode of β-cell death responsible for the development of IDDM in the nonobese diabetic (NOD) mouse. Diabetes 46: 750
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Kurrer, M. O., S. V. Pakala, H. L. Hanson, J. D. Katz. 1997. β cell apoptosis in T cell-mediated autoimmune diabetes. Proc. Natl. Acad. Sci. USA 94: 213
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Augstein, P., A. G. Elefanty, J. Allison, L. C. Harrison. 1998. Apoptosis and β-cell destruction in pancreatic islets on NOD mice with spontaneous and cyclophosphamide-accelerated diabetes. Diabetologia 41: 1381
    OpenUrlCrossRefPubMed
  6. ↵
    Yamada, K., N. Takane-Gyotoku, X. Yuan, F. Ichikawa, C. Inada, K. Nonaka. 1996. Mouse islet cell lysis mediated by interleukin-1-induced Fas. Diabetologia 39: 1306
    OpenUrlCrossRefPubMed
  7. ↵
    Stassi, G., M. Todaro, P. Richiusa, M. Giordano, A. Mattina, M. S. Sbriglia, A. Lo Monte, G. Buscemi, A. Galluzzo, C. Giordano. 1995. Expression of apoptosis-inducing CD95 (Fas/Apo-1) on human β-cells sorted by flow-cytometry and cultured in vitro. Transpl. Proc. 27: 3271
    OpenUrlPubMed
  8. ↵
    Loweth, A. C., G. T. Williams, R. F. James, J. H. Scarpello, N. G. Morgan. 1998. Human islets of Langerhans express Fas ligand and undergo apoptosis in response to interleukin-1β and Fas ligation. Diabetes 47: 727
    OpenUrlAbstract
  9. ↵
    Corbett, J. A., M. A. Sweetland, J. L. Wang, J. Lancaster, Jr, M. L. McDaniel. 1993. Nitric oxide mediates cytokine-induced inhibition of insulin secretion by human islets of Langerhans. Proc. Natl. Acad. Sci. USA 90: 1731
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Eizirik, D. L., M. Flodstrom, A. E. Karlsen, N. Welsh. 1996. The harmony of the spheres: inducible nitric oxide synthase and related genes in pancreatic β cells. Diabetologia 39: 875
    OpenUrlPubMed
  11. ↵
    Stassi, G., R. D. Maria, G. Trucco, W. Rudert, R. Testi, A. Galluzzo, C. Giordano, M. Trucco. 1997. Nitric oxide primes pancreatic beta cells for Fas-mediated destruction in insulin-dependent diabetes mellitus. J. Exp. Med. 186: 1193
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Cohen, P. L., R. A. Eisenberg. 1991. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9: 243
    OpenUrlCrossRefPubMed
  13. ↵
    Chervonsky, A. V., Y. Wang, F. S. Wong, I. Visintin, R. A. Flavell, C. Janeway, Jr, L. A. Matis. 1997. The role of Fas in autoimmune diabetes. Cell 89: 17
    OpenUrlCrossRefPubMed
  14. ↵
    Itoh, N., A. Imagawa, T. Hanafusa, M. Waguri, K. Yamamoto, H. Iwahashi, M. Moriwaki, H. Nakajima, J. Miyagawa, M. Namba, S. Makino, S. Nagata, N. Kono, Y. Matsuzawa. 1997. Requirement of Fas for the development of autoimmune diabetes in nonobese diabetic mice. J. Exp. Med. 186: 613
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Allison, J., A. Strasser. 1998. Mechanisms of β cell death in diabetes: a minor role for CD95. Proc. Natl. Acad. Sci. USA 95: 13818
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Kim, Y.-H., S. Kim, K.-A. Kim, H. Yagita, N. Kayagaki, K.-W. Kim, M.-S. Lee. 1999. Apoptosis of pancreatic β cells detected in accelerated diabetes of NOD mice: no role of Fas-Fas ligand interaction in autoimmune diabetes. Eur. J. Immunol. 29: 455
    OpenUrlCrossRefPubMed
  17. ↵
    Suarez-Pinzon, W., O. Sorensen, R. C. Bleackley, J. F. Elliott, R. V. Rajotte, A. Rabinovitch. 1999. β-cell destruction in NOD mice correlates with Fas (CD95) expression on β-cells and proinflammatory cytokine expression in islets. Diabetes 48: 21
    OpenUrlAbstract
  18. ↵
    Benoist, C., D. Mathis. 1997. Cell death mediators in autoimmune diabetes: no shortage of suspects. Cell 89: 1
    OpenUrlCrossRefPubMed
  19. ↵
    Chu, J. L., P. Ramos, A. Rosendorff, J. Nikolic-Zugic, E. Lacy, A. Matsuzawa, K. B. Elkon. 1995. Massive upregulation of the Fas ligand in lpr and gld mice: implications for Fas regulation and the graft-versus-host disease-like wasting syndrome. J. Exp. Med. 181: 393
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Stokes, T. A., M. Rymaszewski, P. L. Arscott, S. H. Wang, J. D. Bretz, J. Barton, J. R. J. Baker. 1998. Constitutive expression of FasL in thyrocytes. Science 279: 2015
    OpenUrlFREE Full Text
  21. ↵
    Lake, S. P., J. Anderson, J. Chamberlain, S. J. Gardner, P. R. Bell, R. F. James. 1987. Bovine serum albumin density gradient isolation of rat pancreatic islets. Transplantation 43: 805
    OpenUrlCrossRefPubMed
  22. ↵
    Thomas, H. E., J. L. Parker, R. D. Schreiber, T. W. Kay. 1998. IFN-γ action on pancreatic β cells causes class I MHC upregulation but not diabetes. J. Clin. Invest. 102: 1249
    OpenUrlCrossRefPubMed
  23. ↵
    Pipeleers, D. G., P. A. in’t Veld, M. Van de Winkel, E. Maes, F. C. Schuit, W. Gepts. 1985. A new in vitro model for the study of pancreatic A and B cells. Endocrinology 117: 806
    OpenUrlCrossRefPubMed
  24. ↵
    Eisenbarth, G. S., F. S. Walsh, M. Nirenberg. 1979. Monoclonal antibody to a plasma membrane antigen of neurons. Proc. Natl. Acad. Sci. USA 76: 4913
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Nicoletti, I., G. Migliorati, M. C. Pagliacci, F. Grignani, C. Riccardi. 1991. A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 139: 271
    OpenUrlCrossRefPubMed
  26. ↵
    Green, L. C., D. A. Wagner, J. Glogowski, P. L. Skipper, J. S. Wishnok, S. R. Tannenbaum. 1982. Analysis of nitrate, nitrite, and [15N]nitrate in biological fluids. Anal. Biochem. 126: 131
    OpenUrlCrossRefPubMed
  27. ↵
    Schneider, P., J. L. Bodmer, N. Holler, C. Mattmann, P. Scuderi, A. Terskikh, M. C. Peitsch, J. Tschopp. 1997. Characterization of Fas (Apo-1, CD95)-Fas ligand interaction. J. Biol. Chem. 272: 18827
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Signore, A., A. Annovazzi, E. Procaccini, P. E. Beales, J. Spencer, R. Testi, G. Ruberti. 1997. CD95 and CD95-ligand expression in endocrine pancreas of NOD, NOR and BALB/c mice. Diabetologia 40: 1476
    OpenUrlCrossRefPubMed
  29. ↵
    Kayagaki, N., N. Yamaguchi, F. Nagao, S. Matsuo, H. Maeda, K. Okumura, H. Yagita. 1997. Polymorphism of murine Fas ligand that affects the biological activity. Proc. Natl. Acad. Sci. USA 94: 3914
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Allison, J., H. M. Georgiou, A. Strasser, D. L. Vaux. 1997. Transgenic expression of CD95 ligand on islet β cells induces a granulocytic infiltration but does not confer immune privilege upon islet allografts. Proc. Natl. Acad. Sci. USA 94: 3943
    OpenUrlAbstract/FREE Full Text
  31. ↵
    Gazda, L. S., B. Charlton, K. J. Lafferty. 1997. Diabetes results from a late change in the autoimmune response of NOD mice. J. Autoimmun. 10: 261
    OpenUrlCrossRefPubMed
  32. ↵
    Kagi, D., B. Odermatt, P. Seiler, R. M. Zinkernagel, T. W. Mak, H. Hengartner. 1997. Reduced incidence and delayed onset of diabetes in perforin-deficient nonobese diabetic mice. J. Exp. Med. 186: 989
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Kay, T. W., J. L. Parker, L. A. Stephens, H. E. Thomas, J. Allison. 1996. RIP-β2-microglobulin transgene expression restores insulitis, but not diabetes, in β2-microglobulin null nonobese diabetic mice. J. Immunol. 157: 3688
    OpenUrlAbstract
  34. ↵
    Kay, T., H. L. Chaplin, J. L. Parker, L. A. Stephens, H. E. Thomas. 1997. CD4(+) and CD8(+) T lymphocytes: clarification of their pathogenic roles in diabetes in the NOD mouse. Res. Immunol. 148: 320
    OpenUrlCrossRefPubMed
  35. ↵
    Rabinovitch, A., W. L. Suarez-Pinzon, O. Sorensen, R. C. Bleackley, R. F. Power. 1995. IFN-γ gene expression in pancreatic islet-infiltrating mononuclear cells correlates with autoimmune diabetes in nonobese diabetic mice. J. Immunol. 154: 4874
    OpenUrlAbstract
  36. ↵
    Welsh, M., N. Welsh, K. Bendtzen, J. Mares, E. Strandell, C. Oberg, S. Sandler. 1995. Comparison of mRNA contents of interleukin-1β and nitric oxide synthase in pancreatic islets isolated from female and male nonobese diabetic mice. Diabetologia 38: 153
    OpenUrlCrossRefPubMed
  37. ↵
    Corbett, J. A., M. L. McDaniel. 1995. Intraislet release of interleukin 1 inhibits β cell function by inducing β cell expression of inducible nitric oxide synthase. J. Exp. Med. 181: 559
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Mandrup-Poulsen, T.. 1996. The role of interleukin-1 in the pathogenesis of IDDM. Diabetologia 39: 1005
    OpenUrlCrossRefPubMed
  39. ↵
    Fisher, G. H., F. J. Rosenberg, S. E. Straus, J. K. Dale, L. A. Middleton, A. Y. Lin, W. Strober, M. J. Lenardo, J. M. Puck. 1995. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81: 935
    OpenUrlCrossRefPubMed
  40. ↵
    Vaishnaw, A. K., J. R. Orlinick, J. Chu, P. H. Krammer, M. V. Chao, K. B. Elkon. 1999. The molecular basis for apoptotic defects in patients with CD95 (Fas/Apo-1) mutations. J. Clin. Invest. 103: 355
    OpenUrlCrossRefPubMed
  41. ↵
    Rathmell, J. C., M. P. Cooke, W. Y. Ho, J. Grein, S. E. Townsend, M. M. Davis, C. C. Goodnow. 1995. CD95 (Fas)-dependent elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature 376: 181
    OpenUrlCrossRefPubMed
  42. ↵
    Kumar, A., M. Commane, T. W. Flickinger, C. M. Horvath, G. R. Stark. 1997. Defective TNF-α-induced apoptosis in STAT1-null cells due to low constitutive levels of caspases. Science 278: 1630
    OpenUrlAbstract/FREE Full Text
  43. ↵
    Chin, Y. E., M. Kitagawa, K. Kuida, R. A. Flavell, X. Y. Fu. 1997. Activation of the STAT signaling pathway can cause expression of caspase 1 and apoptosis. Mol. Cell. Biol. 17: 5328
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Fukuo, K., S. Hata, T. Suhara, T. Nakahashi, Y. Shinto, Y. Tsujimoto, S. Morimoto, T. Ogihara. 1996. Nitric oxide induces upregulation of Fas and apoptosis in vascular smooth muscle. Hypertension 27: 823
    OpenUrlAbstract/FREE Full Text
  45. ↵
    Andrew, P. J., H. Harant, I. J. Lindley. 1995. Nitric oxide regulates IL-8 expression in melanoma cells at the transcriptional level. Biochem. Biophys. Res. Commun. 214: 949
    OpenUrlCrossRefPubMed
  46. ↵
    Sano, H., M. Hirai, H. Saito, I. Nakashima, K. I. Isobe. 1997. A nitric oxide-releasing reagent, S-nitroso-N-acetylpenicillamine, enhances the expression of superoxide dismutases mRNA in the murine macrophage cell line RAW264-7. Immunology 92: 118
    OpenUrlCrossRefPubMed
  47. ↵
    Dimmeler, S., J. Haendeler, A. Sause, A. M. Zeiher. 1998. Nitric oxide inhibits APO-1/Fas-mediated cell death. Cell Growth Differ. 9: 415
    OpenUrlAbstract
  48. ↵
    Kim, Y. M., R. V. Talanian, T. R. Billiar. 1997. Nitric oxide inhibits apoptosis by preventing increases in caspase-3-like activity via two distinct mechanisms. J. Biol. Chem. 272: 31138
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Mannick, J. B., X. Q. Miao, J. S. Stamler. 1997. Nitric oxide inhibits Fas-induced apoptosis. J. Biol. Chem. 272: 24125
    OpenUrlAbstract/FREE Full Text
  50. ↵
    Kolb, H., V. Kolb-Bachofen. 1998. Nitric oxide in autoimmune disease: cytotoxic or regulatory mediator?. Immunol. Today 19: 556
    OpenUrlCrossRefPubMed
  51. ↵
    Takamura, T., I. Kato, N. Kimura, T. Nakazawa, H. Yonekura, S. Takasawa, H. Okamoto. 1998. Transgenic mice overexpressing type 2 nitric-oxide synthase in pancreatic β cells develop insulin-dependent diabetes without insulitis. J. Biol. Chem. 273: 2493
    OpenUrlAbstract/FREE Full Text
  52. ↵
    Kagi, D., A. Ho, B. Odermatt, A. Zakavian, P. S. Ohashi, T. W. Mak. 1999. TNF-receptor 1-deficient β cell toxicity as an effector pathway in autoimmune diabetes. J. Immunol. 162: 4598
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

The Journal of Immunology: 163 (3)
The Journal of Immunology
Vol. 163, Issue 3
1 Aug 1999
Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section