We have previously reported that transplantation (Tx) of prevascularized donor islets as composite islet-kidneys (IK) reverses diabetic hyperglycemia in miniature swine. In order to test the potential clinical applicability of this strategy, we have extended it to a fully allogeneic nonhuman primate model. IKs were prepared in baboons by isolating islets from 50% to 70% partial pancreatectomies and injecting them under the autologous renal capsule, allowing vascularization before allogeneic Tx. Baboons with diabetes induced by stereptozotocin or total pancreatectomy, received composite IKs (n = 3) or free islets under the renal capsule or intraportally (n = 3), across fully allogeneic barriers with an immunosuppressive regimen consisting of ATG followed by MMF and tacrolimus. FBS of two of IK recipients decreased immediately after Tx and no insulin therapy was required throughout the experimental period (225 and 301 days). In contrast, all recipients of allogeneic free islets showed unstable FBS levels and required insulin within 2 months. We conclude that in addition to maintaining creatinine in the normal range, fully allogeneic IKs from single primate donors can achieve glucose regulation without insulin therapy, while free islets do not. These results support the feasibility of composite allogeneic IK Tx as a potential cure for end-stage diabetic nephropathy.

We have previously shown that free fatty acids (FFA) impair hepatic insulin extraction in vivo and thus generate hyperinsulinemia, a suspected risk factor for atherosclerosis and cancer. Hepatic insulin extraction is a receptor-mediated event, which is initiated by hepatocyte insulin binding. In the present study, we investigated the effect of FFA on insulin binding in freshly isolated rat hepatocytes maintained at 10 mM glucose. Hepatocyte insulin binding decreased after 1 h exposure to oleate in a concentration-dependent manner reaching a maximum (35-40%) at 125 microM. Inhibition of FFA oxidation by >90% with the carnitine palmitoyltransferase I (CPT-I) inhibitor methylpalmoxirate (MP, 30 microM) did not prevent the effect of oleate. However, when hepatocytes were treated with the PKC inhibitor bisindolylmaleimide (BIM, 1 microM) the effect of oleate was abolished. Subcellular fractionation and immunoblotting of specific PKC isoforms revealed that oleate-induced hepatic PKC-delta membrane translocation, but did not translocate-epsilon, -theta, -alpha, -betaI and -betaII. These results indicate that PKC-delta activation mediated the FFA-induced decrease in hepatocyte insulin binding under our conditions, and thus provides a mechanistic basis for FFA-induced hyperinsulinemia.

BACKGROUND:To follow up on previously successful transplantation of encapsulated islets in mice, the present study was performed in rats to determine the effects of several factors, including alginate composition and concentration of cross-linking agent and capsule size on the effectiveness of encapsulated islets. METHODS:Highly purified alginate of either high guluronic acid or high mannuronic acid (M) with low endotoxin content was used. Regular-size (0.8-1.1 mm) or small microcapsules (0.5-0.7 mm) were produced by cross-linking with BaCl2 without additional poly-L-lysine coating and were transplanted into abdominal cavity of normoglycemic (empty capsules) or streptozotocin induced diabetic Lewis rats (islet containing capsules). RESULTS:Empty regular-size capsules made of different alginate compositions had similar biocompatibility and stability results. Compared with empty capsules, regular-size capsules made of high-M alginate containing syngeneic islets had inferior stability indicated with lower fractional volume retrieved. Islet-containing smaller-size microcapsules made of high-M alginate were more stable and had less cellular attachment compared with the regular-size capsules, although the normoglycemic period was comparable between two groups of rats receiving transplants with smaller-size microcapsules (48+/-8 days, n=8) or regular-size capsules (59+/-11 days, n=4) in allogeneic experiments. In syngeneic experiments, all of the rats (n=4) maintained normoglycemia up to 210 days after transplantation. CONCLUSION:These results indicate that regular-size alginate capsules do less well in rats than in our previous experiments with mice. Smaller capsules made of alginate cross-linked with barium appear to provide better stability and may be a useful strategy for use in larger recipients.

P1194 Aims: We have previously demonstrated the cure of surgically induced diabetes mellitus (DM) by the transplantation of prevascularized, composite islet-kidneys (IK) into MGH-miniature swine, using 5,000 islet equivalents (IE)/kg recipient weight. In contrast, the same number of IE did not cure DM if administered directly into the portal vein or beneath the renal capsule, across the same genetic barrier. Clinical studies have also required injection of much higher numbers of IE (25,000 IE/kg) for cure of DM. Methods: Donors were subjected to 70% partial pancreatectomy; islets were isolated and cultured for 3 days. Autograft islets were then injected under the right kidney capsule and allowed to develop vasculature over the next 6-8 weeks before allogeneic IK transplantation (Tx). Recipients of IKs were totally pancreatectomized 3-5 days prior to IK Tx and received a 12-day course of FK-506, starting on the day IK Tx. Of 8 MGH swine recipients, 3 received 2,000 IE/kg, 3 received 3,500 IE/kg and 2 received 5,000 IE/kg across a two-haplotype, fully MHC-mismatched barrier. Graft function was assessed by daily blood glucose and creatinine (Cr) level, biopsies and immunologic assays. Results: All 3 recipients of 2,000 IE/kg maintained normal renal function, however, never achieved normoglycemia. All of these recipients completed the 12-day course of FK-506. The first recipient of 3,500 IE/kg required exogenous insulin for the first 2 weeks following IK Tx and then maintained blood glucose levels of 120-180 mg/dl without exogenous insulin. CML and MLR at day 30 and beyond showed donor specific unresponsiveness. The second recipient in this group immediately became normoglycemic, however, was sacrificed on POD 2 due to an unexplained neurologic event. The third recipient, like the first, received exogenous insulin for the first three weeks (by 1 week after cessation of FK506) and has not required insulin since then. This animal is currently post op day 50 and gaining weight. Both recipients of 5,000 IE/kg IKs had normal renal function and maintained normoglycemia. No insulin was required. Conclusions: These data indicate that 3,500 IE/kg is the minimal effective dose of vascularized islets capable of reversing surgically induced DM as part of a tolerance protocol of renal transplantation. This number is much lower than the number of IEs required to cure DM using intraportal injection of free islets, suggesting the clinical potential of this technology for treatment of type I DM with renal failure.

Recently, islet transplantation in patients with type 1 diabetes has had greater success than in the past, but the important question of whether the kinetics of islet secretion are able to accommodate the metabolic demands of special conditions such as exercise remains unanswered. Syngeneic rat islets (4,000 islet equivalents/rat) were transplanted into the liver, kidney, and peritoneal cavity (encapsulated or nonencapsulated) of rats with streptozocin-induced diabetes. Normoglycemic transplanted rats and age-matched controls were subjected to 30 min of moderate exercise on a treadmill 5 weeks after transplantation. Although control rats maintained near normoglycemia during and after exercise, the rats with islet transplants had significantly lower blood glucose levels. For the rats with islets in the liver, increased C-peptide levels were found at 30 min (790 +/- 125 and 1,450 +/- 250 pmol/l at 0 and 30 min, respectively; P < 0.01), whereas a decrease was found in controls and in rats with islets transplanted into the peritoneal cavity or under the kidney capsule. Moreover, increased glucagon levels were found after exercise in the rats with islets transplanted into the liver (62 +/- 6, 165 +/- 29, 155 +/- 27, and 97 +/- 13 pg/ml at 0, 30, 60, and 90 min, respectively; P < 0.05), whereas no changes in glucagon levels were observed in controls. In conclusion, moderate exercise caused hypoglycemia in rats with islet transplants in different sites including liver, kidney, and peritoneal cavity. C-peptide and glucagon responses to exercise were very different in rats with transplanted islets compared with controls. This islet dysfunction led to exercise-induced hypoglycemia.

Recent advances in islet cell transplantation have led to insulin independence in a majority of islet transplant recipients. However, there exists a need to overcome the shortage of donor tissue and the necessity for lifelong immunosuppression. Preclinical studies in large animal models are necessary to evaluate the safety and efficacy of alternative approaches for clinical islet transplantation. The nonhuman primate serves as an appropriate animal model for such investigations; however, a major impediment in performing such preclinical research has been the difficulty in isolating islets of sufficient quantity and quality. The current study describes a simple and cost-effective method to isolate nonhuman primate islets to support preclinical islet transplantation research. The results of islet isolations from 54 cynomolgus monkeys and 4 baboons are reported. The pancreas was infused with Liberase HI and subjected to static digestion. The digested tissue was shaken, filtered through a mesh screen, applied to a discontinuous gradient, and centrifuged in much the same manner as with conventional rodent islet isolations. Islets were collected from the two interfaces, washed, and transplanted. Following purification, cynomolgus monkey islet isolation yields were 50,100 ± 3120 IE total or 8760 ± 420 IE/g pancreas with the percent purity and viability of 90.8 ± 0.9 and 90.7 ± 0.7, respectively. Total insulin content of the isolated islets was 405 ± 53 μg insulin with DNA content being and 976 ± 117 μg DNA, corresponding to a ratio of 0.57 μg insulin/μg DNA. STZ-induced diabetes was reversed in both mouse and nonhuman primate recipients, which possessed significant levels of c-peptide following transplantation and well-granulated islet grafts. The technique yields sufficient numbers of pure and viable islets to support preclinical research to develop improved strategies to prevent the immune destruction of the transplanted islet graft.

JTT-501, a new insulin sensitizer, improves peripheral glucose uptake in insulin-resistant animals such as KK-Ay mice and Zucker fatty rats. However, the effect of JTT-501 on hepatic glucose metabolism has not been addressed. To investigate this effect, experiments were performed on 6 alloxan-diabetic dogs. Three experiments were conducted for each dog: the treatment experiment, which followed a 10-day oral treatment with JTT-501 30 mg · kg−1 · d−1, and 2 control experiments 2 weeks before and 2 weeks after the treatment experiment. A hyperinsulinemic-hyperglycemic clamp was performed with the tracer dilution method (intraportal insulin infusion rate, 18 pmol · kg−1 · min−1). Arterial hyperglycemia (10 mmol/L) was maintained by adjusting the peripheral glucose infusion rate. After a 45-minute basal period (period I), portal glucose infusion (22.2 μmol · kg−1min−1) was administered for 120 minutes (period II). This was followed by a 90-minutes recovery period (period III). JTT-501 increased insulin-stimulated glucose utilization (P < .05) and enhanced insulin-mediated suppression of glucose production (P < .05) in periods I and III. Net hepatic glucose balance (NHGB) determined by the arterial-venous (A-V) difference method was increased by JTT-501 in period II (P < .01). We conclude that JTT-501 enhances both hepatic and peripheral insulin sensitivity and therefore may have important therapeutic effects in type 2 diabetes.

To determine whether glucagon-like peptide (GLP)-1 increases insulin sensitivity in addition to stimulating insulin secretion, we studied totally depancreatized dogs to eliminate GLP-1's incretin effect. Somatostatin was infused (0.8 microg x kg(-1) x min(-1)) to inhibit extrapancreatic glucagon in dogs, and basal glucagon was restored by intraportal infusion (0.65 ng x kg(-1) x min(-1)). To simulate the residual intraportal insulin secretion in type 2 diabetes, basal intraportal insulin infusion was given to obtain plasma glucose concentrations of approximately 10 mmol/l. Glucose was clamped at this level for the remainder of the experiment, which included peripheral insulin infusion (high dose, 5.4 pmol x kg(-1) x min(-1), or low dose, 0.75 pmol x kg(-1) x min(-1)) with or without GLP-1(7-36) amide (1.5 pmol x kg(-1) x min(-1)). Glucose production and utilization were measured with 3-[3H]glucose, using radiolabeled glucose infusates. In 12 paired experiments with six dogs at the high insulin dose, GLP-1 infusion resulted in higher glucose requirements than saline (60.9+/-11.0 vs. 43.6+/-8.3 micromol x kg(-1) x min(-1), P< 0.001), because of greater glucose utilization (72.6+/-11.0 vs. 56.8+/-9.7 micromol x kg(-1) x min(-1), P<0.001), whereas the suppression of glucose production was not affected by GLP-1. Free fatty acids (FFAs) were significantly lower with GLP-1 than saline (375.3+/-103.0 vs. 524.4+/-101.1 micromol/l, P<0.01), as was glycerol (77.9+/-17.5 vs. 125.6+/-51.8 micromol/l, P<0.05). GLP-1 receptor gene expression was found using reverse transcriptase-polymerase chain reaction of poly(A)-selected RNA in muscle and adipose tissue, but not in liver. Low levels of GLP-1 receptor gene expression were also found in adipose tissue using Northern blotting. In 10 paired experiments with five dogs at the low insulin dose, GLP-1 infusion did not affect glucose utilization or FFA and glycerol suppression when compared with saline, suggesting that GLP-1's effect on insulin action was dependent on the insulin dose. In conclusion, in depancreatized dogs, GLP-1 potentiates insulin-stimulated glucose utilization, an effect that might be contributed in part by GLP-1 potentiation of insulin's antilipolytic action.

Hyperinsulinemia is a common finding in obesity and results from insulin hypersecretion and impaired hepatic insulin extraction. In vitro studies have shown that free fatty acids (FFAs), which are often elevated in obesity, can impair insulin binding and degradation in isolated rat hepatocytes. To investigate whether FFAs impair hepatic insulin extraction (E(H)) in vivo, either saline (SAL) or 10% Intralipid (0.03 ml x kg(-1) x min(-1)) plus heparin (0.44 U x kg(-1) x min(-1)) (IH) was infused into normal dogs to elevate FFA levels. Insulin was infused intraportally at 18 pmol x kg(-1) x min(-1) for 150 min (period A, high insulin dose), and then at 2.4 pmol x kg(-1) x min(-1) for another 150 min (period B, low insulin dose). After the low portal insulin dose, additional insulin was infused peripherally at 8.4 pmol x kg(-1) x min(-1) for 120 min (period C) to assess the clearance of insulin from the peripheral plasma. In 16 paired experiments, FFA levels were 1,085 +/- 167, 1,491 +/- 240, 1,159 +/- 221 micromol/l (IH) and 221 +/- 44, 329 +/- 72, 176 +/- 44 micromol/l (SAL) in periods A, B, and C, respectively. Peripheral insulin levels were greater with IH (P < 0.001) than with SAL in all periods (1,620 +/- 114, 126 +/- 12, 1,050 +/- 72 pmol/l for IH vs. 1,344 +/- 168, 96 +/- 4.2, 882 +/- 60 pmol/l for SAL). Glucose clearance was impaired by IH in all periods (P < 0.05), whereas glucose production was slightly increased by IH during period B. Peripheral insulin clearance (Cl) and E(H) were calculated from the insulin infusion rate and insulin concentration data in each period by taking into account the nonlinearity of insulin kinetics. Cl was lower (P < 0.01) with IH (9.6 +/- 0.6, 12.0 +/- 0.9, 10.2 +/- 0.6 ml x kg(-1) x min(-1)) than with SAL (11.2 +/- 1, 13.6 +/- 0.7, 11.9 +/- 0.9 ml x kg(-1) x min(-1)) in periods A, B, and C. E(H) was also lower (P < 0.05) with IH (25 +/- 4, 40 +/- 5, 32 +/- 5%) than with SAL (30 +/- 2.8, 47 +/- 3, 38 +/- 3%). We conclude that FFAs can impair hepatic insulin extraction in vivo at high and low insulin levels, an effect that may contribute to the peripheral hyperinsulinemia of obesity.

Prolonged exposure of pancreatic islets to free fatty acids (FFAs) inhibits glucose-stimulated insulin secretion (GSIS) in vitro. However, FFA inhibition of GSIS has not been clearly demonstrated in vivo. We examined the in vivo effect of prolonged elevation of plasma FFAs on GSIS using a two-step hyperglycemic clamp in rats treated with a 48-h intravenous infusion of either 20% Intralipid plus heparin (INT) (5 microl/min plus heparin, 0.1 U/min; n = 8), oleate (OLE) (1.3 microEq/min; n = 6), saline (SAL) (n = 6), or bovine serum albumin (BSA) (vehicle for OLE; n = 5). Because there was no difference in any of the parameters between BSA and SAL rats, these groups were combined as control rats (CONT) (n = 11). At the end of the 48-h OLE/INT/CONT infusions, after an overnight fast, plasma glucose was clamped for 2 h at 13 mmol/l and for another 2 h at 22 mmol/l. Preclamp plasma FFAs were elevated twofold (P < 0.01) versus CONT with both INT and OLE (NS, INT vs. OLE). Preclamp glucose, insulin, and C-peptide levels were higher in INT than in CONT rats (P < 0.05), suggesting insulin resistance, but they were not different in OLE and CONT rats. The insulin and C-peptide responses to the rise in plasma glucose from basal to 13 mmol/l were lower in OLE (336 +/- 72 pmol/l and 1.2 +/- 0.1 nmol/l, P < 0.01 and P < 0.05, respectively) than in CONT (552 +/- 54 and 1.9 +/- 0.1) rats, but they were not different between CONT and INT rats (648 +/- 150 and 2.0 +/- 0.4). The insulin and C-peptide responses to the rise in plasma glucose from 13 to 22 mmol/l were lower in both INT (1,188 +/- 204 pmol/l and 3.0 +/- 0.3 nmol/l, P < 0.01 and P < 0.001) and OLE (432 +/- 60 and 1.7 +/- 0.2, P < 0.001 vs. CONT or INT) rats than in CONT rats (1,662 +/- 174 and 5.0 +/- 0.6). In summary, 1) both INT and OLE decreased GSIS in vivo in rats, and 2) the impairing effect of INT on GSIS was less than that of OLE, which might be due to the different type of fatty acid (mostly polyunsaturated in INT versus monounsaturated as OLE) and/or to differential effects of INT and OLE on insulin sensitivity. In conclusion, prolonged elevation of plasma FFAs can desensitize the insulin secretory response to glucose in vivo, thus inducing a beta-cell defect that is similar to that found in type 2 diabetes.