The radical nitric oxide (NO) is a possible mediator of pancreatic beta-cell damage in insulin-dependent diabetes mellitus (IDDM). NO is produced by the enzyme nitric oxide synthase (NOS), in a reaction where arginine is the main substrate. There are different isoforms of NOS, but in the context of immune mediated beta-cell damage the inducible form of NOS (iNOS) is the most relevant. The beta-cell iNOS is similar and encoded by the same gene on chromosome 17 as the iNOS expressed in macrophages and other nucleated cells. iNOS activation depends on gene transcription and de novo enzyme synthesis, and NO seems to induce a negative feedback on iNOS expression. While iNOS mRNA is induced by interleukin-1 beta (IL-1 beta) alone in rodent insulin-producing cells, a combination of two (IL-1 beta + interferon gamma) (IFN-gamma) or three (IL-1 beta + IFN gamma + tumour necrosis factor alpha) cytokines is required for iNOS activation in human pancreatic islets. The promoter region of the murine iNOS gene has at least 25 binding sites for different transcription factors, and the nuclear transcription factor kappa B is necessary for cytokine-induced iNOS transcription in both rodent and human pancreatic islets. The nature of other transcription factors relevant for iNOS regulation in these cells remains to be determined. Induction of iNOS is paralleled by induction of several other cytokine-dependent genes in beta cells, including argininosuccinate synthetase, cyclooxygenase and manganese superoxide dismutase. Some of these genes may contribute to beta-cell damage, while others are probably involved in beta-cell defence and/or repair. Regulation of iNOS and other related genes in beta cells is complex, and differs in several aspects from that observed in macrophages. There are also important differences in iNOS regulation between rodent and human pancreatic islets. A detailed knowledge of the molecular regulation of these genes in beta cells may be instrumental in the development of new approaches to prevent beta-cell destruction in early IDDM.

Most autoimmune diabetes occurs in those without a diabetic relative, but few cases are identifiable prospectively. To model general population prediction, 491 consecutive newly diabetic children from all of Sweden were tested for autoantibodies to glutamate decarboxylase (GAD65ab), insulin (IAA), and islet cells (ICA), and for HLA-DQ genotypes by PCR; 415 matched control children were tested in parallel. GAD65ab sensitivity/specificity was 70/96%, versus 84/96% for ICA, 56/97% for IAA, 93/93% (any positive), 39/99.7% (all positive), and 41/99.7% (GAD65ab plus IAA). The latter's 25% predictive value was not improved by requiring concomitant high-risk HLA genotypes. GAD65ab were associated with DQA1 *0501/Bi *0201 (DQ2; P = 0.007) but not DQA1*0301/Bl*0302 (DQ8), and IAA with DQA1*0301/Bl*0302 (DQ8; P = 0.03) but not DQA1*0501/B1*0201 (DQ2). GAD65ab were more prevalent in females than males (79 vs. 63%; P < 0.0001) but did not vary with onset age nor season. Combining the three antibody assays yielded sufficient sensitivity for screening. GADab were relatively sensitive/specific for diabetes, but even with HLA marker combinations yielded predictive values insufficient for early immunointervention in the low-prevalence general population. (J. Clin. Invest. 1995. 95:1505-1511

Glutamic acid decarboxylase (GAD; glutamate decarboxylase, L-glutamate 1-carboxy-lyase, EC 4.1.1.15), which catalyzes formation of gamma-aminobutyric acid from L-glutamic acid, is detectable in different isoforms with distinct electrophoretic and kinetic characteristics. GAD has also been implicated as an autoantigen in the vastly differing autoimmune disease stiff-man syndrome and insulin-dependent diabetes mellitus. Despite the differing GAD isoforms, only one type of GAD cDNA (GAD-1), localized to a syntenic region of chromosome 2, has been isolated from rat, mouse, and cat. Using sequence information from GAD-1 to screen a human pancreatic islet cDNA library, we describe the isolation of an additional GAD cDNA (GAD-2), which was mapped to the short arm of human chromosome 10. Genomic Southern blotting with GAD-2 demonstrated a hybridization pattern different from that detected by GAD-1. GAD-2 recognizes a 5.6-kilobase transcript in both islets and brain, in contrast to GAD-1, which detects a 3.7-kilobase transcript in brain only. The deduced 585-amino acid sequence coded for by GAD-2 shows less than 65% identity to previously published, highly conserved GAD-1 brain sequences, which show greater than 96% deduced amino acid sequence homology among the three species. The function of this additional islet GAD isoform and its importance as an autoantigen in insulin-dependent diabetes remain to be determined.

Interleukin-1␤ (IL-1␤) is cytotoxic to rat pancreatic ␤-cells by inhibiting glucose oxidation, causing DNA damage and inducing apoptosis. Nitric oxide (NO) is a necessary but not sufficient mediator of these effects. IL-1␤ induced kinase activity toward Elk-1, activation transcription factor 2, c-Jun, and heat shock protein 25 in rat islets. By Western blotting with phosphospecific antibodies and by immunocomplex kinase assay, IL-1␤ was shown to activate extracellular signal-regulated kinase (ERK) 1/2 and p38 mitogen-activated protein kinase (p38) in islets and rat insulinoma cells. Specific ERK1/2 and p38 inhibitors individually reduced but in combination blocked IL-1␤-mediated islet NO synthesis, and reverse transcription-polymerase chain reaction of inducible NO synthase mRNA showed that ERK1/2 and p38 controlled IL-1␤-induced islet inducible NO synthase expression at the transcriptional level. Hyperosmolarity caused phosphorylation of Elk-1, activation transcription factor 2, and heat shock protein 25 and activation of ERK1/2 and p38 in islets comparable to that induced by IL-1␤ but did not lead to NO synthesis. Inhibition of p38 but not of ERK1/2 attenuated IL-1␤-mediated inhibition of glucose-stimulated insulin release. We conclude that ERK1/2 and p38 activation is necessary but not sufficient for IL-1␤-mediated ␤-cell NO synthesis and that p38 is involved in signaling of NO-independent effects of IL-1␤ in ␤-cells.

Nitric oxide (NO) and reactive oxygen species (ROS) are crucial elements in cytokine-mediated ␤-cell destruction. In insulin-producing RINm5F cells, overexpression of cytoprotective enzymes provides significant protection against the synergistic toxicity of NO and ROS. We therefore examined whether overexpression of catalase (Cat), glutathione peroxidase (Gpx), and Cu/Zn superoxide dismutase (SOD) can provide protection for bioengineered RINm5F cells against cytokine-mediated toxicity. A 72-h exposure of RINm5F control cells to interleukin-1␤ (IL-1␤) alone or a combination of IL-1␤, tumor necrosis factor-␣, and ␥-interferon resulted in a time-and concentration-dependent decrease of cell viability in the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) cytotoxicity assay. Although IL-1␤ alone caused only a moderate reduction of viability in the range of 25%, the cytokine mixture induced a significant loss of viability of >75%. This increased toxicity of the cytokine mixture compared with that of IL-1␤ alone could be explained by a higher rate of NO generation within the early 24-48 h incubation period that would favor the toxic synergism of NO and oxygen free radicals. Overexpression of Cat, Gpx, and Cu/Zn SOD protected against toxicity of the cytokine mixture but not against that of IL-1␤ alone. The reduction of cytokine-mediated toxicity was evident also because of an increased proliferation rate and a drastic decrease in the cell death rate. The improved antioxidant defense status did not prevent the activation of iNOS after cytokine exposure. However, RINm5F cells overexpressing cytoprotective enzymes showed a significantly lower level of ROS-damaged protein residues. Thus, protection through Cat, Gpx, and Cu/Zn SOD overexpression was apparently because of an inactivation of ROS generated in the signal cascades of the cytokines. Overexpression of cytoprotective enzymes thus represents a feasible strategy to protect insulin-producing cells against cytokine-mediated cytotoxicity.

Insulin-dependent diabetes mellitus (IDDM) is associated with autoreactivity against GAD but the diagnostic sensitivity (positivity in disease) and specificity (negativity in health) of isoform-specific GAD antibodies have yet to be defined in assay systems suitable for screening large number of samples. One set of IDDM patient (n = 10) and control (n = 50) standard sera were used to develop quantitative antibody assays with in vitro synthesized recombinant 35S-methionine-labelled GAD65 and GAD67, respectively, and protein A-Sepharose to separate free from antibody-bound ligand. Binding levels were not normally distributed (p < 0.0001) and therefore, the diagnostic accuracy of GAD antibodies was analysed by the ROC plots in population-based, consecutively-diagnosed, recent onset, 0-14 year-old patients (n = 105), and matched, healthy control subjects (n = 157). The ROC plots showed that the diagnostic sensitivity of GAD65 antibodies was 77% and the specificity 92% compared with 8% and 98%, respectively for GAD67 antibodies. In the IDDM sera, GAD65 and GAD67 antibodies were concordant in 7% (6 of 81) and GAD65 antibodies and ICA in 89% (72 of 81) without a correlation between the autoantibody levels. Autoantibodies to recombinant human islet GAD65 are specific and sensitive markers for childhood IDDM in this immunoassay with in vitro synthesized 35S-methionine-labelled recombinant GAD.

A model of the pathogenesis of insulin-dependent diabetes mellitus, i.e. the initial phase of beta-cell destruction, is proposed: in a cascade-like fashion efficient antigen presentation, unbalanced cytokine, secretion and poor beta-cell defence result in beta-cell destruction by toxic free radicals (O2- and nitric oxide) produced by the beta cells themselves. This entire process is under polygenetic control.

The radical nitric oxide (NO) is a possible mediator of pancreatic beta-cell damage in insulin-dependent diabetes mellitus (IDDM). NO is produced by the enzyme nitric oxide synthase (NOS), in a reaction where arginine is the main substrate. There are different isoforms of NOS, but in the context of immune mediated beta-cell damage the inducible form of NOS (iNOS) is the most relevant. The beta-cell iNOS is similar and encoded by the same gene on chromosome 17 as the iNOS expressed in macrophages and other nucleated cells. iNOS activation depends on gene transcription and de novo enzyme synthesis, and NO seems to induce a negative feedback on iNOS expression. While iNOS mRNA is induced by interleukin-1 beta (IL-1 beta) alone in rodent insulin-producing cells, a combination of two (IL-1 beta + interferon gamma) (IFN-gamma) or three (IL-1 beta + IFN gamma + tumour necrosis factor alpha) cytokines is required for iNOS activation in human pancreatic islets. The promoter region of the murine iNOS gene has at least 25 binding sites for different transcription factors, and the nuclear transcription factor kappa B is necessary for cytokine-induced iNOS transcription in both rodent and human pancreatic islets. The nature of other transcription factors relevant for iNOS regulation in these cells remains to be determined. Induction of iNOS is paralleled by induction of several other cytokine-dependent genes in beta cells, including argininosuccinate synthetase, cyclooxygenase and manganese superoxide dismutase. Some of these genes may contribute to beta-cell damage, while others are probably involved in beta-cell defence and/or repair. Regulation of iNOS and other related genes in beta cells is complex, and differs in several aspects from that observed in macrophages. There are also important differences in iNOS regulation between rodent and human pancreatic islets. A detailed knowledge of the molecular regulation of these genes in beta cells may be instrumental in the development of new approaches to prevent beta-cell destruction in early IDDM.