Autoimmune Insulitis Research Articles

Beta-cells, oxidative stress, lysosomal stability, and apoptotic/necrotic cell death.

Reactive oxygen intermediates (ROI) may be involved in the destruction of pancreatic beta-cells during the development of insulin-dependent diabetes mellitus (IDDM). To investigate the possible role of lysosomes in this process, normal mouse beta-cells were cultured as monolayers at D-glucose concentrations of 1.6 (pronounced crinophagy), 11 or 28 mM (minimal crinophagy), subjected to a low level of oxidative stress and returned to standard culture conditions. Some cultures were exposed to desferrioxamine (Des) before the oxidative stress. As a result of such stress, many of the cells' lysosomes ruptured with consequent apoptosis or necrosis. Cells kept at 1.6 mM glucose were rich in secretory granules, showed crinophagy/autophagy, were very sensitive to oxidative stress, and had the least stable lysosomes. Cells kept at 28 mM glucose did not show crinophagy, contained fewer secretory granules, were less sensitive to oxidative stress, and had more stable lysosomes. Des-treated cells behaved almost as cells not exposed to oxidative stress at all. The findings suggest that iron may occur together with zinc within the secretory granules and that it sensitizes crinophagic lysosomes to oxidative stress. The stress that was applied in this study may be comparable to what occurs within the vicinity of activated macrophages during autoimmune insulitis.

A shared TCR CDR3 sequence in NOD mouse autoimmune diabetes.

T cells involved in autoimmune diseases have been characterized by the genetic elements used to construct their autoimmune TCR. In the present study, we sequenced the alpha and beta chains of the TCR expressed by a CD4(+) T cell clone, C9, functional in NOD mouse diabetes. Clone C9 can adoptively transfer diabetes or, when attenuated, C9 can be used to vaccinate NOD mice against diabetes. Clone C9 recognizes a peptide epitope (p277) of the 60 kDa heat shock protein (hsp60) molecule. We now report that the C9 TCR beta chain features a CDR3 peptide sequence that is prevalent among NOD mice. This CDR3 element is detectable by 2 weeks of age in the thymus, and later in the spleen and in the autoimmune insulitis. Thus, a TCR CDR3beta sequence appears to be a common idiotope associated with mouse diabetes.

Differential Impact of T Cell Repertoire Diversity in Diabetes-Prone or -Resistant IL-10 Transgenic Mice

Expression of IL-10 transgene (tg) in pancreatic β cells failed to induce autoimmune insulitis and diabetes in (BALB/c × NOD)F1 mice. However, IL-10-expressing tg littermates from backcrosses (N2 and N3) with NOD mice became diabetic at 5 to 10 weeks of age in an MHC-dependent manner. In this study, we tested the possibility that enhancement in frequency of islet antigen (Ag)-specific T cells overrides the protective effects of a diabetes-resistant genetic background and promotes diabetes in IL-10 tg (BALB/c × NOD)F1 mice. For this test, we introduced the IL-10 transgene into tg BDC2.5 mice expressing the islet Ag-specific Vβ4 T cell repertoire by breeding Ins-IL-10+/BALB/c mice with BDC2.5 mice. The progeny (Ins-IL-10+/BALB/c × BDC2.5+)F1 mice doubly tg for IL-10 and Vβ4 (BDC2.5) T cell repertoire, developed diabetes at 10 to 18 weeks of age with a much more aggressive T cell infiltrate in the pancreatic islets than in single tg mice. Surprisingly, these diabetic mice were free from acute pancreatitis but had apoptotic β cells in the islet infiltrate. Conversely, mice tg for Vβ4 (BDC2.5) T cell repertoire but not IL-10 had no diabetes and no apoptotic β cells in the islet infiltrate. Therefore, an increase in the frequency of islet-specific T cells apparently overcomes the protection from diabetes by a resistant genetic background. Interestingly, N2 backcross mice doubly tg for Vβ4 (BDC2.5) T cell repertoire and IL-10, compared to N2 backcross mice tg for IL-10 only, eventually became diabetic but with a delayed onset and reduced incidence of disease. These findings demonstrate that, along with IL-10, an increase in frequency of islet antigen-specific T cells (a) overrides the protective effect of genetic resistance to autoimmune diabetes in F1 mice and (b) delays the onset of an otherwise accelerated diabetes in (Ins-IL-10+/NOD)N2 backcross mice.

Minute oxidative stress is sufficient to induce apoptotic death of NIT‐1 insulinoma cells

When cultured NIT-1 cells were subjected to a low level of oxidative stress (30 microM hydrogen peroxide for 15 min at 37 degrees C) several of their lysosomes ruptured, as demonstrated by intravital staining with the lysosomotropic weak base acridine orange. Such rupture is due to intralysosomal, iron-catalyzed oxidative reactions, since it was largely prevented by previous endocytotic uptake of desferrioxamine. The resultant limited leakage of lysosomal hydrolytic enzymes into the cytosol could be important for an apoptotic-type degradation/fragmentation process within initially intact plasma membranes. In contrast, extensive lysosomal rupture leads to necrosis. The development of the damage process was followed by light- and electron microscopy; and by the TUNEL-reaction. As a result of the applied oxidative stress, which is comparable to that expected to occur within the microenvironment surrounding activated macrophages under oxidative burst (e.g. during autoimmune insulitis), about 90% of the cells eventually died due to post-apoptotic secondary necrosis. The few surviving cells phagocytosed the debris from their fragmented neighbours and began to divide about 24 h after the insult. Thus the sensitivity to oxidative stress varies, perhaps as a consequence of varying amounts of intralysosomal redox-active iron, as we have found to be the case in several other cellular systems. Since the NIT-1 cells are highly differentiated, and in many ways like beta cells, we consider our result to be of value for the understanding of beta-cell death during the development of insulin-dependent (Type I) diabetes mellitus (IDDM).

Glutamic acid decarboxylase antibodies are more frequent than islet cell antibodies in islet transplanted IDDM patients and persist or occur despite immunosuppression.

Pancreatic islet grafts transplanted into patients with autoimmune diabetes are potentially threatened by two immune responses, allograft rejection and the recurrence of autoimmune insulitis. In the present study we investigated the humoral autoimmune response directed to islet autoantigens by studying islet cell antibodies and glutamic acid decarboxylase (GAD 65) antibodies in twenty-one insulin-dependent diabetes-mellitus (IDDM) patients undergoing intraportal islet allotransplantation. Islet transplantation was performed according to the following recipient categories: Islet after kidney transplantation (n=10), simultaneous islet and kidney transplantation (n=6) and islet transplant alone (n=5). GAD 65 antibodies were detected in a radioligand GAD 65 antibody assay using recombinant, in vitro translated, human 35S-methionin labelled GAD 65 as tracer. Islet cell antibodies were determined by indirect immunofluorescence technique on human pancreas. In six out of twenty-one patients we observed GAD 65 antibodies before islet transplantation and the GAD 65 antibodies persisted despite immunosuppression. In contrast only two subjects were concordantly islet cell antibody positive and the titre decreased post transplantation. In addition we observed occurrence of GAD 65 antibodies in five subjects that were shown to be antibody negative before islet transplantation with three of them subsequently becoming positive for islet cell antibodies. The remaining ten patients were GAD 65 antibody and islet cell antibody negative before islet transplantation and remained negative thereafter. Interestingly none of the patients was exclusively positive for islet cell antibodies without being positive for GAD 65 antibodies. In summary we have demonstrated in twenty-one islet grafted individuals that humoral autoimmunity to islet antigens can persist or occur despite immunosuppression. Islet cell antibodies appear to be less frequent (5 out of 21, 23%) compared to GAD 65 antibodies (11 out of 21, 52%) suggesting that they are more affected by immunosuppressive therapy. We conclude that GAD 65 antibodies are a useful tool to further evaluate a possible link between persistent autoimmunity and early or late graft failure after islet transplantation.

Lentivirus-mediated transduction of islet grafts with interleukin 4 results in sustained gene expression and protection from insulitis.

Autoimmune destruction of islets in the pancreas leads to the development of insulin-dependent diabetes mellitus (IDDM). Replacement of insulin-producing tissue by transplantation of islets provides a cure to disease but requires immunosuppression or a means of controlling anti-graft immune responses. To promote islet survival we have utilized a local approach by expressing immunoregulatory molecules in islet grafts. The results presented here show that the human immunodeficiency virus (HIV)-based lentiviral vector is capable of stably transducing whole islets. Foreign reporter gene expression was observed both in vitro and in vivo 30 days after transplantation. Grafts containing insulin-positive beta-islet cells expressing foreign protein indicate that transduction does not interfere with glucose regulation. The absence of inflammatory infiltrates in grafts suggests that transduction does not activate the immune system. When islets transduced with an HIV vector expressing IL-4 were transplanted into diabetes-prone mice, animals were protected from autoimmune insulitis and islet destruction. As demonstrated by proliferative and cytokine analysis, protection was consistent with a switching of islet-antigen-specific T cell responses toward a Th2 phenotype. These results suggest that HIV-based lentivirus vectors can efficiently transduce islet cells with genes encoding potentially therapeutic molecules, for possibly managing diabetes.

Interferon-gamma plays a role in pancreatic islet-cell destruction of reovirus type 2-induced diabetes-like syndrome in DBA/1 suckling mice.

Reovirus type 2 (Reo-2) infection in DBA/1 suckling mice causes insulitis, which leads to pancreatic islet-cell destruction, resulting in a diabetes-like syndrome. T-helper (Th) 1 cytokines are thought to play a key role in islet inflammation in insulin-dependent diabetes mellitus. We examined this hypothesis in the Reo-2-induced diabetes-like syndrome. We used reverse transcriptase polymerase chain reaction (PCR) and quantitative PCR techniques to examine mRNA expression of interferon (IFN)-gamma (Th1 type cytokine), and interleukin (IL)-4 (Th2 type cytokine) in splenic cells. We observed that in Reo-2 infected mice the level of IFN-gamma expression increases with the development of insulitis, whereas expression of message for IL-4 is minimal to detectable with the immuno-inflammatory process 10 days after infection. The treatment of monoclonal antibody (mAb) against mouse IFN-gamma during the expansion phase of insulitis (5-9 days after infection) inhibited the development of insulitis and the elevation of blood glucose concentrations in a dose dependent manner. Furthermore altered CD4+/CD8+ cell ratio compared with uninfected mice in the splenic cells by the infection was recovered to the ratio of uninfected mice by the treatment of mAb against mouse IFN-gamma, suggesting normalization of T cell balance in immune system. These results suggest that Reo-2-triggered autoimmune insulitis may be mediated by Th1 lymphocytes and IFN-gamma may play a role in islet inflammation leading to islet cell destruction.

Transfer of human serum IgG to nonobese diabetic Igmu null mice reveals a role for autoantibodies in the loss of secretory function of exocrine tissues in Sjögren's syndrome.

The NOD (nonobese diabetic) mouse has been studied as an animal model for autoimmune insulin-dependent diabetes and Sjögren's syndrome. NOD.Igmu null mice, which lack functional B lymphocytes, develop progressive histopathologic lesions of the submandibular and lachrymal glands similar to NOD mice, but in the absence of autoimmune insulitis and diabetes. Despite the focal appearance of T cells in salivary and lachrymal tissues, NOD.Igmu null mice fail to lose secretory function as determined by stimulation of the muscarinic/cholinergic receptor by the agonist pilocarpine, suggesting a role for B cell autoantibodies in mediating exocrine dryness. Infusion of purified serum IgG or F(ab')2 fragments from parental NOD mice or human primary Sjögren's syndrome patients, but not serum IgG from healthy controls, alters stimulated saliva production, an observation consistent with antibody binding to neural receptors. Furthermore, human patient IgG fractions competitively inhibited the binding of the muscarinic receptor agonist, [3H]quinuclidinyl benzilate, to salivary gland membranes. This autoantibody activity is lost after preadsorption with intact salivary cells. These findings indicate that autoantibodies play an important part in the functional impairment of secretory processes seen in connection with the autoimmune exocrinopathy of Sjögren's syndrome.

Amelioration of diabetes in nonobese diabetic mice with advanced disease by linomide-induced immunoregulation combined with Reg protein treatment.

Oral linomide, (quinoline-3-carboxamide), has been shown to prevent autoimmune insulitis, islet destruction, and diabetes in NOD mice treated at an early stage of the disease, but confers only partial protection in animals with advanced disease. Reg protein, the gene product of a complementary DNA isolated from a regenerating rat islet library, has been previously shown to induce expansion of beta-cell mass in pancreatectomized rats. To determine the effect of treatment combining immunomodulation and Reg protein on advanced autoimmune diabetes, we treated female NOD mice with oral linomide and i.p. Reg protein injections. In 14-week-old animals with less severe disease (glucose tolerant), treatment with each agent alone resulted in amelioration of diabetes, as did treatment with Reg alone in 5-week-old prediabetic mice. In 14-week-old animals with more severe disease (glucose intolerant), only treatment with the combination of both agents, but not that with each separately, resulted in amelioration of diabetes. Our study suggests that treatment aimed at abrogation of autoimmunity combined with expansion of beta-cell mass constitutes a potential therapeutic approach for treatment of insulin-dependent diabetes mellitus.

Impact of Dietary Fat on Th1/Th2 Cytokine Gene Expression in the Pancreas and Gut of Diabetes-prone BB Rats

The effect of dietary n-3 or n-6 polyunsaturated fatty acids on the development of autoimmune insulitis was analysed in diabetes-prone BB rats. Litter-matched groups of rats received a standard open formula NIH-07 (National Institutes of Health, NIH) diet enriched with 10% fish oil, 10% flaxseed oil or with 10% palm oil plus 2% cholesterol during the period of insulitis onset (50–70 days of age). Analysis of cytokine gene expression in pancreatic RNA revealed an increase of IFN-γ and a decrease of IL-10 mRNA with onset of insulitis. When compared to unsupplemented NIH, none of the three fat-enriched diets depressed the rise of IFN-γ gene expression or the influx of leukocytes into islets. However, all of the fat-enriched diets led to significantly higher IL-10 mRNA levels. Although a specific anti-inflammatory effect of fish oil was not seen in the pancreas, a clear shift of the Th1/Th2 cytokine mRNA ratio towards Th2 was seen in the gut-associated immune system. We conclude that diets high in fat support IL-10 without suppressing IFN-γ gene expression in islet inflammation. A special anti-inflammatory effect of fish oil was not seen in pancreatic lesions of BB rats, although there was strong modulation of the IFN-γ/IL-10 mRNA ratio in the gut associated immune system.

Stage-specific degeneration of germ cells in the seminiferous tubules of non-obese diabetic mice.

The cause of fertility problems in insulin-dependent diabetes is largely unknown. To evaluate the role of autoimmunity-associated phenomena in the testis as a possible cause of the derangement in spermatogenesis, the stage-specific apoptosis of germ cells in the insulitis phase of pre-diabetes was quantified in the testes of non-obese diabetic (NOD) mice. The seminiferous epithelium of normal BALB/c and NOD mice contained cells positive for in-situ end-labelling (ISEL) of DNA. ISEL-positive germ cells formed clusters in the seminiferous epithelium of the NOD mice in marked contrast to the seminiferous epithelium of the BALB/c mice, which contained only individual cells positive for ISEL. ISEL-positive cells were present in the basal and luminal compartments of the epithelium. Ultrastructural analysis and demonstration of externalized phosphatidyl serine confirmed that the cells were undergoing apoptosis. The ultrastructurally apoptotic cells included spermatogonia, spermatocytes and spermatids. In cytological squash preparations of segments of seminiferous tubules from NOD mice aged 17-20 weeks, the number of ISEL-positive cells/mm tubule was significantly lower in segments at stages I-II of the seminiferous epithelial wave but higher at stages III-IV in comparison to BALB/c mice. The numbers of ISEL-positive cells/mm tubule in the other stages were similar in the two strains of mice. Analysis of 32P-3'-end labelled DNA from the testes showed that the BALB/c mice had relatively more DNA fragmentation than did the NOD mice. These data suggest that autoimmune insulitis in the NOD mice is associated with increased amounts and abnormal stage distribution of apoptosis in the seminiferous epithelium, resulting in derangement of spermatogenesis.

Patterns of hemopoietic reconstitution in nonobese diabetic mice: dichotomy of allogeneic resistance versus competitive advantage of disease-resistant marrow.

Complete replacement of the immune system via allogeneic bone marrow transplantation is sufficient to prevent diabetes in the nonobese diabetic (NOD) mouse model. In the present study we examined whether mixed allogeneic reconstitution would be sufficient to interrupt the autoimmune process with respect to occurrence of overt diabetes, as well as preexisting autoimmune insulitis. NOD mice were lethally irradiated and reconstituted with a mixture of NOD and B10.BR marrow. A relative resistance to allogeneic bone marrow engraftment was noted in NOD recipients of the mixed bone marrow inoculum, compared with disease-resistant controls. Moreover, unlike disease-resistant controls, all animals that initially repopulated as mixed donor/host chimeras became predominantly allogeneic by 4 mo, suggesting a competitive advantage for long term engraftment for disease-resistant marrow. All but one mouse in the group that engrafted with allogeneic marrow remained free of diabetes for the entire follow-up period (n = 22). Moreover, in all animals examined, virtually all islets were free of insulitis. In contrast, 74% of NOD mice that received similar conditioning and failed to engraft with donor marrow developed acute diabetes and intra-islet insulitis was present in all animals examined. These data suggest that NOD mice exhibit a relative resistance to engraftment compared with disease-resistant recipients. Conversely, animals that initially repopulated as a mixture of syngeneic and donor marrow become converted to virtually all donor by 4 mo. These data provide additional support that a defective stem cell is responsible for autoimmune diabetes in this experimental model.

Active stage of autoimmune diabetes is associated with the expression of a novel cytokine, IGIF, which is located near Idd2.

Recently, interferon-gamma-inducing-factor (IGIF) has been described as a novel monokine that is a more potent interferon-gamma (IFN-gamma) inducer than IL-12. By cloning IGIF from affected tissue and studying IGIF gene expression, we describe for the first time a close association of this cytokine with an autoimmune disease. The non-obese diabetic (NOD) mouse spontaneously develops autoimmune insulitis and diabetes which can be accelerated and synchronized by a single injection of cyclophosphamide. IGIF mRNA was demonstrated by reverse transcriptase PCR in NOD mouse pancreas during early stages of insulitis. Levels of IGIF mRNA increased rapidly after cyclophosphamide treatment and preceded a rise in IFN-gamma mRNA, and subsequently diabetes. Interestingly, these kinetics mimick that of IL-12p40 mRNA, resulting in a close correlation of individual mRNA levels. Cloning of the IGIF cDNA from pancreas RNA followed by sequencing revealed identity with the IGIF sequence cloned from Kupffer cells and in vivo preactivated macrophages. When extending our study to macrophages of the spleen we observed that NOD mouse macrophages responded to cyclophosphamide with IGIF gene expression while macrophages from Balb/c mice treated in parallel did not. The IGIF gene position is located within the Idd2 interval on mouse chromosome 9 and therefore it is a candidate for the Idd2 susceptible gene. We conclude that IGIF expression is abnormally regulated in autoimmune NOD mice and closely associated with diabetes development.

Dendritic cells in the autoimmune insulitis in NOD mouse models of diabetes.

Dendritic cells (DC) are thought to play important roles as antigen presenting cells in the autoimmune insulitis preceding insulin dependent diabetes mellitus (IDDM). Not only are DC the first cells appearing in the pancreatic inflammatory process in animal models of diabetes (NOD mouse1.2 and BB rat3.4), DC are also abundantly present in progressed human insulitis5. Although insulitis in the NOD mouse starts around 3 weeks, overt diabetes will not develop until an age of 20 weeks. The lag between occult insulitis and diabetes, which can also be observed in human IDDM, was always considered to reflect the time needed for accumulation of a sufficient number of destructive T cells in the islets.

Histologic findings in islets of whole pancreas allografts: lack of evidence for recurrent cell-mediated diabetes mellitus.

Cell-mediated cytotoxicity directed toward pancreatic islets plays a significant role in the pathogenesis of type 1 diabetes mellitus. Histologically, autoimmune insulitis appears as mononuclear infiltrates that precede actual beta cell loss. Experimental studies by Bartlett et al. on the BB rat model suggest that whole pancreas transplants are not susceptible to recurrent autoimmune diabetes. To test this hypothesis on clinical grounds, we evaluated the islets in pancreatectomies and percutaneous needle biopsies from 55 whole pancreas allografts. In this material, islet inflammation was never seen independently of allograft rejection, but was always associated with inflammation of the surrounding acinar structures. The degree of islet inflammation and the occasional presence of islet necrosis correlated well with the degree of rejection and only occurred in the higher grades of the latter (grades III-V). We did not observe isolated insulitis as a histological indication of selective cell-mediated cytotoxicity against beta cells.

Successful treatment of diabetes in nod mice with advanced disease by islet isografts following immunoregulation with Linomide (quinoline-3-carboxamide)

We have recently documented that oral Linomide (quinoline-3-carboxamide) prevents autoimmune insulitis, islet destruction, and diabetes in NOD mice treated at an early stage (5 wk of age) of the disease. In this report, we show that treatment of female NOD mice with advanced disease (age 23–24 wk) by syngeneic islet transplantation and oral Linomide administration results in prevention of graft insulitis and diabetes in the Linomide group up to 40 wk (diabetes at 40 wk: isograft recipients with Linomide n = 0 of 6; isograft recipients alone n = 5 of 6; p < 0.0001). The extent of protection from glucose intolerance by the combination of transplantation with Linomide was superior to that of Linomide alone [blood glucose (mean ±SD) 60 min post-i.p. injection of 1 g/kg body weight glucose: Linomide plus isograft 6.7 ±1 mmol/L; Linomide alone 18.7 ±6.3 mmol/L; p < 0.00011. Thus, Linomide should be considered a potential immunoregulatory modality in patients undergoing pancreatic islet or whole organ transplantation.

Successful treatment of diabetes in NOD mice with advanced disease by islet isografts following immunoregulation with Linomide (quinoline-3-carboxamide).

We have recently documented that oral Linomide (quinoline-3-carboxamide) prevents autoimmune insulitis, islet destruction, and diabetes in NOD mice treated at an early stage (5 wk of age) of the disease. In this report, we show that treatment of female NOD mice with advanced disease (age 23-24 wk) by syngeneic islet transplantation and oral Linomide administration results in prevention of graft insulitis and diabetes in the Linomide group up to 40 wk (diabetes at 40 wk: isograft recipients with Linomide n = 0 of 6; isograft recipients alone n = 5 of 6; p < 0.0001). The extent of protection from glucose intolerance by the combination of transplantation with Linomide was superior to that of Linomide alone [blood glucose (mean +/- SD) 60 min post-i.p. injection of 1 g/kg body weight glucose: Linomide plus isograft 6.7 +/- 1 mmol/L; Linomide alone 18.7 +/- 6.3 mmol/L; p < 0.0001]. Thus, Linomide should be considered a potential immunoregulatory modality in patients undergoing pancreatic islet or whole organ transplantation.

Mouse islet cell lysis mediated by interleukin-1-induced Fas.

This study was conducted to investigate the possible involvement of Fas in beta-cell death in insulitis of Type 1 (insulin-dependent) diabetes mellitus. Although primary cultured Balb/c mouse islet cells did not express Fas mRNA, 4-12 hours of treatment with 10(2)-10(3) U/l of mouse interleukin-1 alpha (IL-1 alpha) induced the expression of Fas mRNA. Surface Fas expression was detected by immunofluorescence flow cytometry using a non-cytolytic anti-Fas monoclonal antibody after 6 or 12 h of incubation with 10(3) U/l of IL-1 alpha. Primary islet cells were resistant to an agonistic anti-Fas monoclonal antibody. However, 12 h pretreatment with IL-1 alpha sensitized islet cells to its cytolytic effect. Significant cell death was observed 24 h after the addition of anti-Fas, and progressively increased until 72 h, when specific 51Cr release was 72 +/- 6%. Agarose gel electrophoresis of DNA extracted from cells exposed to IL-1 alpha and agonistic anti-Fas showed internucleosomal DNA fragmentation, a hallmark of apoptotic cell death. Since the Fas antibody showed no cross-reactive activity of tumour necrosis factor (TNF), the cytotoxic effect was not mediated by TNF receptors. A protein synthesis inhibitor cycloheximide augmented Fas-mediated islet cell death. The Fas-mediated killing of islet cells was not L-arginine-dependent, or blocked by N(G)-monomethyl-L-arginine. beta-TC1 cells also expressed Fas mRNA when exposed to IL-1 alpha or IL-1 alpha plus interferon-gamma. These observations suggest that Fas-mediated apoptosis may be a mechanism of islet cell death in autoimmune insulitis.

Differential expression of ICAM-1 and LFA-1 versus L-selectin and VCAM-1 in autoimmune insulitis of NOD mice and association with both Th1- and Th2-type infiltrates.

The infiltration of pancreatic islets by mononuclear cells is the hallmark of the development of insulin dependent diabetes mellitus (IDDM) in the NOD mouse, an animal model for human IDDM. The aim, of this study was to correlate adhesion molecule expression with the degree of islet infiltration and to compare Th1- and Th2-driven islet inflammation. Cryostat sections of NOD mouse pancreata before and after diabetes development were analysed by semiquantitative immunohistochemistry. NOD mouse islets did not show the expression of ICAM-1, LFA-1, L-selectin and VCAM-1 prior to infiltration by mononuclear cells. Furthermore, islets with early stage insulitis (grade 1, periinsular location of small infiltrates) still were devoid of adhesion molecule expression. ICAM-1 and LFA-1 were first demonstrable in islets with strong periinsular infiltrates (insulitis grade 2) while L-selectin and VCAM-1 were only seen in islets with mild or strong intraislet infiltration (grade 3-4). Adhesion molecules were demonstrable in areas of macrophage and T-lymphocyte infiltrates but not in adjacent endocrine islet tissue. Islets of all infiltration stages contained Th2 lymphocytes (positive for IL-4). Substantial numbers of Th1 cells (positive for IFN-gamma, TNF-alpha, IL-2 and/or IL-2 receptor) were observed only after acceleration of diabetes development by a single injection of cyclophosphamide (250 mg/kg i.p.). Interestingly, the adhesion molecule expression pattern in islets with "Th1' versus "Th2 insulitis' was not different. In conclusion, the expression of adhesion molecules in islets during the development of autoimmune diabetes does not precede mononuclear infiltration but probably occurs in response to the activation of initial small infiltrates. ICAM-1 and LFA-1 expression is seen prior to L-selectin and VCAM-1. However, adhesion molecule expression during Th1 versus Th2 cell infiltration is very similar, suggesting similar adhesion molecule requirements of the two Th subsets.

Antibodies to glutamic acid decarboxylase in the prediction of insulin dependency

Antibodies to glutamic acid decarboxylase (anti-GAD) predict the progression of adults masquerading as NIDDM to insulin dependency and predict the eventual occurrence of IDDM in healthy pregnant women in Finland. Almost 80% of prediabetic and newly diagnosed IDDM cases are positive for anti-GAD. However, approximately 20% of these groups do not have a humoral response to GAD so it cannot be claimed that anti-GAD is the exclusive autoimmune phenomenon. Nevertheless, 94% of children with newly diagnosed IDDM that we studied had an autoimmune response to either GAD, ICA or IAA, singly or in combination. The anti-GAD assay also has a substantial role in the diagnosis and classification of diabetes presenting in adult life since a proportion of adults who present with apparent NIDDM actually have a slowly evolving autoimmune insulitis, a condition we have called latent autoimmune diabetes in adults (LADA). It appears likely that anti-GAD will be predictive for IDDM in both first degree relatives and the general population. As a result of the cost and relative ease of performance, it will provide a practical alternative to ICA, particularly in population screening. Comparisons of testing for anti-GAD and ICA as predictors of IDDM using large population groups are now in progress in our laboratory.