Prolonged insulin independence after islet allotransplants in recipients with type 1 diabetes.

We sought to determine the long-term outcomes in type 1 diabetic recipients of intraportal alloislet transplants on a modified immunosuppressive protocol. Six recipients with hypoglycemia unawareness received one to two islet infusions. Induction therapy was with antithymocyte globulin (ATG) plus etanercept for tumor necrosis factor-alpha blockade. Recipients received cyclosporine and everolimus for maintenance immunosuppression for the first year posttransplant, with mycophenolic acid or mycophenolate mofetil subsequently substituted for everolimus. Recipients have been followed for 1173 +/- 270 days since their last infusion for islet graft function (insulin independence, hemoglobin A(1c) levels and C-peptide production) and for adverse events associated with the study protocol. Of the six recipients, five were insulin-independent at 1 year, and four continue to be insulin-independent at a mean of 3.4 +/- 0.4 years posttransplant. None of the six recipients experienced recurrence of severe hypoglycemia. Measured glomerular filtration rate decreased from 110.5 +/- 21.2 mL/min/1.73 m(2) pretransplant to 82.6 +/-19.1 mL/min/1.73 m(2) at 1 year posttransplant. In conclusion, islet transplants restored insulin independence for a mean of >3 years in four of six recipients treated with ATG and etanercept induction therapy and with cyclosporine and, initially, everolimus for maintenance. Our results suggest this immunosuppressive protocol may allow long-term graft survival.

Modulation of immune responses after portal venous injection of antigen.

BACKGROUND: How the localization of antigen to the liver, through means such as oral ingestion, induces tolerance is poorly understood. METHODS: To elucidate potential mechanisms we used an adoptive transfer system wherein ova-specific T cells were infused into a syngeneic host, and antigen-specific T-cell responses after delivery of soluble antigen into the liver were monitored. RESULTS: After infusion of antigen into the portal vein, the frequency of antigen-specific T cells in lymph nodes draining the liver was lower than the frequency in peripheral lymph nodes. These findings were the reverse of what is typically observed after subcutaneous injection of antigen with adjuvant. Infusion of antigen with adjuvant into the portal vein did not alter this pattern of antigen-specific T-cell localization; however, an increased frequency of T cells, compared with antigen alone, was observed in peripheral lymph nodes and spleen. After exposure to antigen via the portal vein, T cells isolated from lymph nodes draining the liver and challenged with antigen in vitro exhibited a diminished proliferative response compared with T cells isolated from nondraining lymph nodes. This hyporesponsiveness was not observed when the antigen was administered with adjuvant. CONCLUSIONS: Our findings suggest that the influence of the liver on immune responses might reflect two processes: (1) loss of antigen-specific T cells after primary antigen injection, and (2) hyporesponsiveness on reexposure to antigen. These mechanisms may contribute to the prevention of undesirable immune responses to foods and enteric bacteria in the gastrointestinal tract. Furthermore, these results underscore the importance of minimizing inflammation in circumstances such as islet transplantation, if endogenous mechanisms of tolerance induction are to be maximized.

Insulin down-regulates the inducible nitric oxide synthase pathway: nitric oxide as cause and effect of diabetes?

Evidence in this paper indicates that insulin can down-regulate the inducible nitric oxide synthase (iNOS) pathway in vivo. The iNOS pathway is up-regulated in diabetes-prone rats and mice and is associated with an autoimmune process. However, the results presented here indicate that macrophage nitric oxide (NO) production and iNOS mRNA expression are also elevated in rats or mice made diabetic by streptozotocin injection in which there is no primary autoimmune component. Insulin administration reduces NO production in autoimmune-prone and streptozotocin-induced diabetic rodents. Finally, insulin decreases macrophage NO production in normal hosts. These results indicate that the autoimmune paradigm is inadequate to explain increased NO in diabetes. As a potential mechanism to explain insulin-mediated regulation of NO production, TGF-1 may be involved because 1) macrophages from diabetic mice produce less TGF-beta1 than macrophages from normal hosts; 2) the circulating TGF-beta1 level is lower in diabetic mice; and 3) insulin administration increases circulating TGF-beta1 in normal mice. Together, these results provide evidence that increased NO in diabetes is not only a cause but also an effect of beta-cell destruction and results in part from a heretofore unrecognized immunomodulatory activity of insulin.

Nitric oxide mediates early dysfunction of rat and mouse islets after transplantation.

Evidence presented in this paper indicates that nitric oxide (NO), generated by a nonspecific "wound"-type of inflammation, is an important mediator of the early dysfunction of transplanted islets in rodents. Although allogeneic islets stimulate NO production to a greater degree than syngeneic islets, the amounts of NO produced after either are significantly elevated above baseline. Inhibition of NO production by N(G)-monomethyl-L-arginine (NMA), markedly decreases the time needed to restore euglycemia after intraportal transplantation of syngeneic islets in diabetic rats. The dose of NMA used was not observably toxic, with no significant changes in blood pressure, hepatic artery blood flow, serum hepatic enzyme levels, or in weight compared with control animals. In rat recipients of intraportal syngeneic transplants, evidence that NO is produced at the site of implantation includes (1) an early and transient increase in posttransplant hepatic vein nitrate levels (pretransplant, 90 microM; 24 hr, 230 microM; 48 hr, 250 microM; 72 hr, 170 microM; and 96 hr, 140 microM), (2) concurrent appearance of inducible NO synthase mRNA in liver extracts, and (3) immunohistochemical localization of inducible NO synthase within the transplanted islets. Suppression of NO production or inhibition of NO activity is a potential strategy to increase the early function and engraftment transplanted islets in the clinical setting.

Expression of intrahepatic inducible nitric oxide synthetase mRNA correlates with production of nitric oxide during intraportal isogeneic and allogeneic rat islet transplantation.

Intrahepatic NO production is related to the islet mass transplanted. Nitric oxide production is higher in recipients of allogeneic rather than syngeneic islets. In addition, in allogeneic recipients a possible second peak of NO production was observed at 120 hours corresponding to the time of cellular rejection of the islet grafts (P = .22 vs 96 hours). Finally, the time to rejection of Wistar rat donor islets transplanted into Lewis rat diabetic recipients treated with NMA was not affected. However, inhibiting NO production in the minimal islet transplant model decreased the time to islet function, it does not affect the time to clinical rejection in recipients of a high number of allogeneic islet, which functions immediately. High-level NO has been shown to inhibit T-cell activation in vitro, and thus decreasing the levels by administrating NMA may accentuate the rejection response, canceling out the beneficial effect that might otherwise have occurred on islet function. Further experiments are required to clarify these issues.

The role of nitric oxide in IL-1 beta-mediated dysfunction of rodent islets of Langerhans. Implications for the function of intrahepatic islet grafts.

Products of inflammation, such as interleukin-1 beta (IL-1 beta) and nitric oxide (NO), may impair early function of pancreatic islet grafts. In in vitro studies, freshly isolated rat islets of Langerhans cultured for 24 hr (10 islets/well) in the presence of 20 IU/ml of IL-1 beta released 57% less insulin (mean +/- S.E. of 151 +/- 61 microU) on the average than control (385 +/- 89 microU) cultures (n = 9, P = 0.08). Nitrite levels in the medium (indirect measure of NO) after islets were cultured for a 24-hr period were nearly 3-fold greater in IL-1 beta-exposed islets than control islet cultures (5.8 +/- 1.0 microM vs. 2.2 +/- 0.3 microM, P = 0.03). Production of nitrite by islet cells in the presence of IL-1 beta was inhibited in cultures also containing 2 mM L-NG-monomethyl-Arginine (L-NMMA) (3.4 +/- 0.4 microM, n = 9, P = 0.09 vs. control). When islets were maintained for 1 hr in 30 mg/dl glucose followed by 300 mg/dl for 1 hr, insulin release (stimulated) increased 6-fold (from 7 +/- 2 to 45 +/- 11 microU) in control cultures but only 3-fold (from 4 +/- 2 to 12 +/- 4 microU) in IL-1 beta-exposed cultures (n = 9, P = .01). Addition of 2 mM L-NMMA to islet cultures in the presence of IL-1 beta (20 IU/ml) (n = 9) restored insulin release to normal (from 6 +/- 2 to 38 +/- 9 microU, P > or = 0.6), suggesting that NO mediates the inhibitory effect of IL-1 beta on beta-cell function. In in vivo studies, rats with streptozotocin-induced diabetes (blood glucose > 400 mg/dl) received minimal (750 hand-picked islets) intraportal beta cell mass isografts with (n = 5) or without (n = 9) treatment with 100 mg/7 days of L-NMMA from 3 days before transplantation to 4 days after transplantation (POD -3 to +4). L-NMMA-treated rats became euglycemic (glucose < 200 mg/dl) earlier than nontreated rats (mean +/- SD of 6.4 +/- 2.5 vs. 16.7 +/- 4.7 days posttransplant, P = 0.001). These findings support the hypothesis that NO is a mediator of beta cell dysfunction after intraportal transplantation of freshly isolated islets of Langerhans.