Beneficial effects of parenteral GLP-1 delivery by cell therapy in insulin-deficient streptozotocin diabetic mice.

Parenteral delivery of long-acting glucagon-like peptide-1 (GLP-1) mimetics has received much attention as a therapeutic option for diabetes. However, cell therapy-based GLP-1 treatments may provide a more physiological regulation of blood glucose. The present study assessed the effects of chronic GLP-1 delivery by cell therapy, using the GLP-1-secreting GLUTag cell line, in normoglycemic and streptozotocin-induced diabetic mice. GLUTag cell aggregates were transplanted into the subscapular region of mice. Over 30 days, cellular transplantation gave rise to encapsulated and well-vascularized growths, which contained immunoreactive GLP-1. Cell implantation was well tolerated and had no appreciable metabolic effects in normal mice. However, transplantation significantly (P<0.001) countered excessive food and fluid intake in diabetic mice and maintained normal body weight. Circulating glucose (P<0.01) and glucagon (P<0.05) were significantly reduced and plasma insulin and GLP-1 dramatically increased. This was associated with significantly (P<0.01) improved glucose tolerance in diabetic mice. Histological examination of the pancreata of these mice revealed elevations (P<0.001) in islet and ß-cell area, with reduced (P<0.001) α-cell area. Increased ß-cell mass reflected the enhanced proliferation relative to apoptosis. These studies emphasize the potential of chronic GLP-1 delivery by cell therapy as a potential therapeutic option for diabetes.

Daily administration of the GIP-R antagonist (Pro3)GIP in streptozotocin-induced diabetes suggests that insulin-dependent mechanisms are critical to anti-obesity-diabetes actions of (Pro3)GIP.

AIM: Glucose-dependent insulinotropic polypeptide-receptor (GIP-R) antagonism using (Pro3)GIP improves glucose tolerance and ameliorates insulin resistance and abnormalities of islet structure and function in a commonly used model of obesity-diabetes, namely ob/ob mice. The effect of GIP-R antagonism in a streptozotocin (STZ)-induced model of insulin deficiency has not been evaluated. The present study has investigated the effects of daily administration of (Pro(3))GIP to STZ-treated mice. METHODS: Swiss TO mice received once-daily injection of (Pro3)GIP (25 nmol/kg body weight) or saline 4 days prior to and 16 days after injection of STZ, and effects on metabolic parameters and islet architecture were assessed. RESULTS: (Pro3)GIP treatment had no significant effect on hyperphagia or body weight loss. However, hyperglycaemia and glycated haemoglobin were worsened, glucose tolerance further decreased and insulin sensitivity was impaired by (Pro3)GIP. These effects were observed on an STZ-induced background characterized by severe reductions of circulating insulin, beta-cell mass and pancreatic insulin stores. CONCLUSIONS: These data indicate that the beneficial actions of the GIP-R antagonist, (Pro3)GIP, in obesity-diabetes appear to be largely mediated through insulin-dependent mechanisms that merit further investigation.

Function of a long-term, GLP-1-treated, insulin-secreting cell line is improved by preventing DPP IV-mediated degradation of GLP-1.

Glucagon-like peptide-1 (GLP-1) is an important insulinotropic hormone with potential in the treatment of type 2 diabetes. However, the short biological half-life of the peptide after cleavage by dipeptidylpeptidase IV (DPP IV) is a major limitation. Inhibition of DPP IV activity and the development of resistant GLP-1 analogues is the subject of ongoing research. In this study, we determined cell growth, insulin content, insulin accumulation and insulin secretory function of a insulin-secreting cell line cultured for 3 days with either GLP-1, GLP-1 plus the DPP IV inhibitor diprotin A (DPA) or stable N-acetyl-GLP-1. Native GLP-1 was rapidly degraded by DPP IV during culture with accumulation of the inactive metabolite GLP-1(9-36)amide. Inclusion of DPA or use of the DPP IV-resistant analogue, N-acetyl-GLP-1, improved cellular function compared to exposure to GLP-1 alone. Most notably, basal and accumulated insulin secretion was enhanced, and glucose responsiveness was improved. However, prolonged GLP-1 treatment resulted in GLP-1 receptor desensitization regardless of DPP IV status. The results indicate that prevention of DPP IV action is necessary for beneficial effects of GLP-1 on pancreatic beta cells and that prolonged exposure to GLP-1(9-36)amide may be detrimental to insulin secretory function. These observations also support the ongoing development of DPP-IV-resistant forms of GLP-1, such as N-acetyl-GLP-1.

Comparison of the secretory properties of four insulin-secreting cell lines.

The insulin-secretory responsiveness of four popular and widely used insulin-secreting cells lines (RINm5F, HIT-T15, INS-1 and BRIN-BD11 cells) to a range of stimuli including glucose, amino acids, neurotransmitters, peptide hormones and sulphonylureas was studied. Differences were seen in the pattern of responsiveness of the cell lines to the various modulators of insulin release. While these studies revealed that INS-1 cells had the highest insulin content, only BRIN-BD11 cells exhibited a significant step-wise insulin secretory response to increasing glucose concentrations. BRIN-BD11 cells also showed pronounced insulin responses to leucine, KIC, L-arginine, L-alanine and palmitic acid. All the cell lines tested gave significant responses to the neurotransmitters carbachol and glibenclamide with increased insulin release. A comparison was made between the functional characteristics of the various cell lines with those of freshly isolated rat islets. This illustrated the general value of each cell line as a model for studies of insulin secretion. Electrofusion-derived BRIN-BD11 cells appeared to closely mimic the glucose sensitivity and overall secretory performance of normal rat islets.

Effects of cytotoxic agents on functional integrity and antioxidant enzymes in clonal beta-cells.

Effects of cytotoxic agents and hydrogen peroxide were examined using pancreatic BRIN-BD11 cells and the parental insulinoma RINm5F cell line. Cell viability was determined using the MTT colorimetric assay and the TUNEL assay was used to assess apoptosis and acridine orange assay was used to determine levels of apoptosis versus necrosis. RT-PCR studies were employed to investigate the effects of the toxins on the expression of antioxidative enzymes, superoxide dismutase (SOD), glutathionine peroxidase (GPX) and catalase (CAT). Streptozotocin, hydrogen peroxide, alloxan and ninhydrin exerted time- and concentration-dependent toxic effects on BRIN-BD11 and RINm5F cells. RT-PCR showed that 90 minutes exposure of BRIN-BD11 cells or RINm5F cells to 5 mM ninhydrin down regulates SOD, GPX and CAT antioxidative enzymes. Glutathionine peroxidase gene expression was also down regulated in both types of cell by hydrogen peroxide. There were no significant differences in antioxidant gene expression after exposure to the other toxins under the conditions employed. TUNEL assay revealed that streptozotocin (8 mM) and hydrogen peroxide (125 microM) had no significant effect on the number of cells undergoing apoptosis. However after exposure to ninhydrin (5 mM) almost 100% of the non-viable BRIN-BD11 cells and around 50% of the RINm5F cells were dying by apoptosis. With the BRIN-BD11 cells there was around a 30% increase in the number of apoptotic cells compared with 50% in the RINm5F cells after exposure to alloxan (16 mM). The results indicate multiple effects of cytotoxic agents on functional integrity and antioxidant enzyme gene expression in clonal beta-cells.

Functional examination of microencapsulated bioengineered insulin-secreting beta-cells.

Clonal insulin-secreting BRIN-BD11 cells engineered by electrofusion were encapsulated inside natrium alginate beads and cultured in RPMI 1640 culture media. Acute insulin secretory responses to glucose and amino acids were compared between microencapsulated cells and non-encapsulated cells maintained in monolayer culture. Encapsulated cells exhibited a 1.5-fold, 2.9-fold and 4.2-fold increase (P< 0.001) in insulin release in response to 16.7 mmol/l glucose, 10 mmol/l L-arginine and 10 mmol/l L-alanine respectively. Insulin output by non-encapsulated cells was approximately 30% greater but the relative magnitudes of responses were similar. This is the first study to demonstrate the stability of cellular engineered insulin-secreting cells encapsulated in alginate beads, illustrating the utility of this approach for cellular engineering and potential transplantation in diabetes.

Comparative functional study of clonal insulin-secreting cells cultured in five commercially available tissue culture media.

The electrofusion-derived rat insulin-secreting cell line BRIN-BD11 was cultured in five different commercially available media to determine the optimum medium for the in vitro maintenance of such clonal cell lines. Cells were cultured in RPMI-1640, DMEM, McCOY'S, F-12K, or MEM culture medium supplemented with 10% (v/v) fetal bovine serum and antibiotics (100 U/ml penicillin and 0.1 g/L streptomycin). Insulin secretion studies performed after 10 days revealed RPMI-1640 to be the best performing medium in terms of insulin secretory responsiveness to a range of stimuli including glucose, L-alanine, L-arginine, carbachol, and glibenclamide. Insulin release was significantly decreased (p < 0.01 to p < 0.05) in all other media compared to RPMI-1640. Only the cells cultured in RPMI-1640 and DMEM showed a significant glucose-induced insulin secretory response (p < 0.01 and p < 0.05). McCOY'S gave the next best result followed by F-12K and MEM. After the 10-day culture period, the highest insulin content was found in cells cultured in RPMI-1640 and DMEM with significantly lower levels of insulin in cells cultured in McCOY'S, F-12K, and MEM (p < 0.01 to p < 0.001). RPMI-1640 was used for further studies to investigate the effects of 5.6-16.7 mmol/L glucose in culture on the secretory responsiveness of BRIN-BD11 cells. Significant responses to a number of nonglucidic secretagogues were seen following culture at 5.6 and 16.7 mmol/L glucose, although responsiveness was less than after culture with 11.1 mmol/L glucose. At 16.7 mmol/L glucose culture, glucose-stimulated insulin release was abolished.

Culture and function of electrofusion-derived clonal insulin-secreting cells immobilized on solid and macroporous microcarrier beads.

In view of the advantages of the bulk production of clonal pancreatic beta cells, an investigation was made of the growth and insulin secretory functions of an electrofusion-derived cell line (BRIN-BD11) immobilized on a solid microcarrier, cytodex-1 or a macroporous microcarrier, cultispher-G. For comparison, similar tests were performed using BRIN-BD11 cells present in single cell suspensions or allowed to form pseudoislets. Similar growth profiles were recorded for each microcarrier with densities of 4.4 x 10(5) +/- 0.3 cells/ml and 4.2 x 10(5) +/- 0.2 cells/ml achieved using cytodex-1 and cultispher-G, respectively. Cell viability began to decline on day 5 of culture. Insulin concentration in the culture medium reached a peak of 26 +/- 2.0 ng/ml and 24 +/- 2.2 ng/ml for cells grown on cytodex-1 and cultispher-G, respectively. Cells grown on both types of microcarrier showed a significant 1.5-1.8-fold acute insulin-secretory response to 16.7 mmol/l glucose. L-alanine (10 mmol/l) and L-arginine (10 mmol/l) also induced significant 3 4 fold increases of insulin release. BRIN-BD11 cells immobilized on cytodex-1 or cultispher-G out-performed single cell suspensions and pseudoislets in terms of insulin-secretory responses to glucose and amino acids. A 1.3-fold, 2.2-fold and 1.7-fold stimulation of insulin secretion was observed for glucose, L-alanine and L-arginine respectively in single cell suspensions. Corresponding increases for pseudoislets were 1.6-1.8-fold for L-alanine and L-arginine, with no significant response to glucose alone. These data indicate the utility of micro-carriers for the production of functioning clonal beta cells.

Production and characterization of specific antibodies for evaluation of glycated insulin in plasma and biological tissues.

Previous studies have shown that glycation of insulin occurs in pancreatic beta-cells under conditions of hyperglycaemia and that the site of glycation is the N-terminal Phe(1) of the insulin B-chain. To enable evaluation of glycated insulin in diabetes, specific antibodies were raised in rabbits and guinea-pigs by using two synthetic peptides (A: Phe-Val-Asn-Gln-His-Leu-Cys-Tyr, and B: Phe-Val-Asn-Gln-His-Leu-Tyr-Lys) modified by N-terminal glycation and corresponding closely to the N-terminal sequence of the glycated human insulin B-chain. For immunization, the glycated peptides were conjugated either to keyhole limpet haemocyanin or ovalbumin using glutaraldehyde, m-maleimidobenzoyl-N-hydroxysuccinimide ester or 1-ethyl-3-(3-dimethylamino propyl) carbodiimide hydrochloride. Antibody titration curves, obtained using I(125)-tyrosylated tracer prepared from glycated peptide A, revealed high-titre antisera in five groups of animals immunized for 8-28 weeks. The highest titres were observed in rabbits and guinea-pigs immunized with peptide B coupled to ovalbumin using glutaraldehyde. Under radioimmunoassay conditions, these antisera exhibited effective dose (median) (ED(50)) values for glycated insulin of 0.3-15 ng/ml and 0.9-2.5 ng/ml respectively, with negligible cross-reactivity against insulin or other islet peptides. The degree of cross-reaction with glycated proinsulin was approximately 50%. Glycated insulin in plasma of control and hydrocortisone-treated diabetic rats measured using rabbit 3 antiserum (1:10 000 dilution; sensitivity <19 pg/ml) was 0. 08+/-0.01 and 1.5+/-0.6 ng/ml (P<0.01), corresponding to 4 and 16% of total circulating insulin concentration respectively. Immunocytochemistry studies of the pancreas of streptozotocin-treated diabetic rats using a 1:1000 dilution of guinea-pig 2 antiserum revealed clusters of fluorescent positively stained cells in islets. These studies document the successful production of polyclonal antisera specific for glycated insulin and their usefulness in radioimmunoassays and immunocytochemistry. The demonstration of glycated insulin in plasma and islets of animal models of diabetes supports the view that glycation of insulin is involved in the pathogenesis of this disease.

Drug-induced desensitization of insulinotropic actions of sulfonylureas.

K(ATP)-channel-dependent and K(ATP)-channel-independent insulin-releasing actions of the sulfonylurea, tolbutamide, were examined in the clonal BRIN-BD11 cell line. Tolbutamide stimulated insulin release at both nonstimulatory (1.1 mM) and stimulatory (16. 7 mM) glucose. Under depolarizing conditions (16.7 mM glucose plus 30 mM KCl) tolbutamide evoked a stepwise K(ATP) channel-independent insulinotropic response. Culture (18 h) with tolbutamide or the guanidine derivative BTS 67 582 (100 microM) markedly reduced (P < 0. 001) subsequent responsiveness to acute challenge with tolbutamide, glibenclamide, and BTS 67 582 but not the imidazoline drug, efaroxan. Conversely, 18 h culture with efaroxan reduced (P < 0.001) subsequent insulinotropic effects of efaroxan but not that of tolbutamide, glibenclamide, or BTS 67 582. Culture (18 h) with tolbutamide reduced the K(ATP) channel-independent actions of both tolbutamide and glibenclamide. Whereas culture with efaroxan exerted no effect on the K(ATP) channel-independent actions of sulfonylureas, BTS 67 582 abolished the response of tolbutamide and inhibited that of glibenclamide. These data demonstrate that prolonged exposure to tolbutamide desensitizes both K(ATP)-channel-dependent and -independent insulin-secretory actions of sulfonylureas, indicating synergistic pathways mediated by common sulfonylurea binding site(s).