We herein report cytotoxicity of advanced glycation end-products (AGEs) on pancreatic beta cells. AGEs stimulated reactive oxygen species (ROS) generation but did not arrest proliferation of the INS-1 cell line. Pancreatic beta cell lines or primary cultured islets possess a receptor for AGE (RAGE), and its expression increased after AGE treatment. TUNEL staining and FACS analysis using annexin V/PI antibodies showed that apoptosis increased in INS-1 cells or primary cultured islets when incubated with BSA conjugated with glyceraldehyde (AGE2) or glucoaldehyde (AGE3), compared with those conjugated with glucose (AGE1). Reaction of INS-1 cells to Ki67, which is a cellular marker for proliferation, was also increased after AGE treatment. The ability of primary cultured islets to secrete insulin was retained even after AGE treatment under either low or high glucose conditions. The antiserum against RAGE partially prevented AGE-induced cellular events. Treatment of beta cells with the antioxidant metallothionein results in a significant reduction in pathologic changes. AGEs might be able to induce apoptosis as well as proliferation of pancreatic beta cell lines or primary cultured islets. Moreover, antibody array showed that RAD51 and RAD52 were significantly decreased in AGE2-treated INS-1 cells. AGEs might inhibit homologous DNA recombination for repairing DNA of INS-1 cells damaged by ROS generation. It might be suggested that treatment of AGEs resulted in ROS production and apoptosis through their receptor on pancreatic beta cells. AGEs might deteriorate function of pancreatic beta cells in patients with long-term hyperglycemia.

Studies of animals with spontaneous autoimmune diabetes have revealed that autoreactive T cells that mediate islet beta cell destruction can be manipulated by the administration of Th(2) cytokines. Using gene delivery to express the targeted protein, we can overcome the need for frequent administration of cytokines on account of their short half-lives. In this study, the effect of hTGFbeta gene delivery was evaluated both in vitro and in vivo using an adenovirus vector (Ad) constructed with an hTGFbeta cDNA. In vitro transfection assays of the construct in HepG2, beta cell lines, and islets showed good expression levels of hTGFbeta and activation of smad3. Ad-hTGFbeta enhanced differentiation and proliferation in the beta cell line or islets without causing apoptosis. Of interest, Ad-hTGFbeta transduction in CD4(+)CD25(-) T cells resulted in a significant enhanced expression of CD25 and a regulatory T cell-specific transcription factor, Foxp3. To evaluate in vivo efficacy, Ad-hTGFbeta was intravenously injected into 7-week-old NOD mice and compared to the transduction using the vector only. The Ad-hTGFbeta group had persistent gene expression for longer than 5 weeks, and high TGFbeta serum level was secreted. There was no difference in the degree of insulitis between the Ad-hTGFbeta group and controls. Although we found favorable in vitro results, such as decrease in islet apoptosis, enhanced proliferation and differentiation, and increase in the level of CD4(+)CD25(+) regulatory T cells, there was no difference in reduction of the development of T1D between controls and Ad-hTGFbeta-injected mice. Nevertheless, if we find the appropriate mode and timing of TGFbeta gene transduction, Ad-hTGFbeta gene therapy might be useful in therapeutic cytokine delivery for the treatment of T1D.