Search

Aims/hypothesis. Type I (insulin-dependent) diabetes results mainly from T-cell-mediated autoimmune destruction of pancreatic beta cells. Cytotoxic T lymphocytes destroy target cells via a perforin-based or Fas-based mechanism. Our previous study indicated that the Fas-Fas ligand (FasL) pathway is required for the development of autoimmune diabetes in the NOD mouse. We now investigated whether or not the Fas-FasL system is involved in the beta-cell destruction in human Type I diabetes. Methods. We immunohistochemically analysed pancreas biopsy specimens of 13 recent-onset patients. Results. Pancreatic islets were identified but showed various degrees of reduction in beta-cell volume in all patients. Out of 13 patients 6 had insulitis. In these 6 patients Fas was expressed in both the islets and infiltrating cells but not in either cell type in the 7 other patients without insulitis. Double immunostaining showed that Fas was positive in 92.2 to 97.7 % of beta cells but only in 17.6 to 46.7 % of alpha cells in Fas-positive, insulin-remaining islets. We found FasL was expressed exclusively in islet-infiltrating cells in patients with insulitis. Double immunostaining revealed that the most prevalent phenotype of FasL-positive cells was CD8, which was followed by macrophages and CD4. Conclusion/interpretation. The interaction between Fas on beta cells and FasL on infiltrating cells might trigger selective apoptotic beta-cell death in inflamed islets, leading to immune-mediated Type I diabetes. [Diabetologia (1999) 42: 1332–1340]

Aims/hypothesis. To determine whether the clinical heterogeneity observed in the development of Type I (insulin-dependent) diabetes mellitus correlates with immunohistochemical differences observed at diagnosis. Methods. Patients (n = 17) with recent-onset diabetes clinically considered to be insulin dependent (Type I), underwent pancreatic biopsy for immunohistological analysis. These patients were divided into two groups based on the presence or absence of islet immunological abnormalities (insulitis or hyperexpression of MHC class I antigens or both). The patients were also HLA typed and tested for islet cell antibodies and antibodies to glutamic acid decarboxylase (GAD-Ab). All patients were followed monthly for 2 years and their fasting plasma glucose, haemoglobin A1C and daily insulin doses were recorded. The clinical course of patients with islet immunological abnormalities was compared with that of patients without those abnormalities. Results. Patients with and without islet immunological abnormalities did not differ with regard to HLA type or islet cell antibodies. Antibodies to glutamic acid decarboxylase correlated with the presence of insulitis and MHC class I hyperexpression. These local immunological abnormalities were also associated with higher haemoglobin A1C values (p < 0.05) and a trend towards greater insulin requirements. Further, patients with the islet abnormalities had higher fasting plasma glucose concentrations 2 years after the biopsy than at the time of the biopsy (p < 0.05). Conclusion/interpretation. The heterogeneous clinical course observed following diagnosis in patients with Type I diabetes correlates with islet immunological abnormalities. Insulitis and hyperexpression of MHC class I correlate with deteriorating glycaemic control. [Diabetologia (1999) 42: 574–578]

Cytokines are thought to contribute to the induction of pancreatic beta-cell destruction in insulin-dependent diabetes mellitus. The molecular mechanisms that underlie beta-cell death were investigated by studying cytokine-induced cell death in beta-cell lines. A combination of three cytokines (interleukin-1 beta, tumour necrosis factor-alpha, and interferon-gamma) induced apoptotic cell death in the mouse pancreatic beta-cell line beta TC1, as judged from the appearance of cells with hypodiploid nuclei and oligonucleosomal DNA fragmentation. The same treatment also induced apoptosis in the mouse pancreatic alpha-cell line alpha TC1 and the NOD/Lt mouse beta-cell line NIT-1, although to a lesser extent than in beta TC1 cells. The abundance of endogenous Bcl-2 in beta TC1 cells was lower than that in the other two cell lines. Overexpression of human Bcl-2 in beta TC1 cells partially protected them from cytokine-induced cell death. These results suggest that apoptosis may be responsible, at least in part, for cytokine-induced beta-cell destruction and that Bcl-2 prevents apoptosis in pancreatic islet cells.

Viral infection is assumed to trigger or exacerbate autoimmune responses against pancreatic beta cells leading to the development of insulin-dependent diabetes mellitus (IDDM). We therefore examined by polymerase chain reaction the presence of two candidate viruses, cytomegalovirus and Epstein-Barr virus, in IDDM pancreases. Pancreas tissues were obtained by biopsy under laparoscopy from 16 recent-onset IDDM patients: age 17-53 years; disease duration 0-7 months; six had flu-like symptoms before onset. Frozen sections were made and subjected to DNA amplification. DNA samples were prepared from the frozen sections and polymerase chain reaction was performed using primers specific to cytomegalovirus, Epstein-Barr virus and control gene for HLA-DP. Cytomegalovirus- and Epstein-Barr virus-infected cells were used for positive control. Southern blot analysis could detect cytomegalovirus DNA from as few as 2 x 10(-1) cytomegalovirus-infected cells and Epstein-Barr virus DNA from two Epstein-Barr virus-infected cells. This highly sensitive analysis, however, could not detect cytomegalovirus or Epstein-Barr virus genomes in pancreases of recent-onset IDDM. A single copy human gene (HLA-DP) was amplified from all IDDM pancreases indicating that DNA amplification was performed without inhibition. We conclude that cytomegalovirus or Epstein-Barr virus genomes are unlikely to exist in pancreas biopsy specimens of recent-onset IDDM patients.