Inhibition of Y1 receptor signaling improves islet transplant outcome

Abstract

Failure to secrete sufficient quantities of insulin is a pathological feature of type-1 and type-2 diabetes, and also reduces the success of islet cell transplantation. Here we demonstrate that Y1 receptor signaling inhibits insulin release in β-cells, and show that this can be pharmacologically exploited to boost insulin secretion. Transplanting islets with Y1 receptor deficiency accelerates the normalization of hyperglycemia in chemically induced diabetic recipient mice, which can also be achieved by short-term pharmacological blockade of Y1 receptors in transplanted mouse and human islets. Furthermore, treatment of non-obese diabetic mice with a Y1 receptor antagonist delays the onset of diabetes. Mechanistically, Y1 receptor signaling inhibits the production of cAMP in islets, which via CREB mediated pathways results in the down-regulation of several key enzymes in glycolysis and ATP production. Thus, manipulating Y1 receptor signaling in β-cells offers a unique therapeutic opportunity for correcting insulin deficiency as it occurs in the pathological state of type-1 diabetes as well as during islet transplantation.Islet transplantation is considered one of the potential treatments for T1DM but limited islet survival and their impaired function pose limitations to this approach. Here Loh et al. show that the Y1 receptor is expressed in β- cells and inhibition of its signalling, both genetic and pharmacological, improves mouse and human islet function.

Fig. 1

Improved glycemic control in diabetic mice transplanted with Y1 receptor-deficient islets. a Photomicrograph showing immune fluorescence staining for insulin (green), PYY (red) and DAPI (blue) in islets from human, wild type, and PYY knockout mice. Scale bar, 25 μm. b Y1 receptor mRNA expression in pancreatic islets isolated from lean or obese mice determined by qRT-PCR (n = 4). c Pancreatic islets from C57BL/6JAusb mice were isolated and cultured. Insulin secretion was determined in response to 2 and 11 mM glucose in the presence of PYY and/or Y1 receptor antagonist BIBO3304 (n = 3). d Perifusion experiment to record insulin release over a 50 min period using islets from C57BL/6JAusb mice stimulated with 11 mM glucose and treated with PYY or Y1 antagonist BIBO3304. (n = 4–5). e Results expressed as area under the curve. f Alloxan-induced diabetic mice were transplanted with an optimal number (300) of WT islets (n = 8) or a minimal number (60) from WT islets (n = 15) or Y1^−/− islets (n = 13) and blood glucose levels were monitored for 20 days. g Results expressed as area under the curve. h–k Diabetic mice receiving 60 WT (n = 4) or Y1^−/− (n = 4) islets were fasted overnight and i.v. glucose tolerance tests (1 g/kg body weight) were performed at day 5 post-transplant. Blood glucose levels and insulin production during glucose tolerance tests were monitored. Results are expressed over the time course and as area under the curve. l, m 7-week-old C57BL/6JAusb mice were treated with placebo (n = 6) or with 0.5 μM of the Y1 receptor specific antagonist BIBO3304 (n = 6), 2 h prior to i.p. glucose injection (1 g/kg body weight) and serum insulin levels in response to glucose administration at indicated time points were determined. Results are also expressed as area under the curve. Data are shown as mean ± s.e.m. *P < 0.05, **P < 0.01, calculated by t-test (b, c) or two-way ANOVA analysis

Fig. 2

Improved glycemic control in diabetic mice transplanted with WT islets treated with a Y1 receptor antagonist. a Alloxan-induced diabetic mice were transplanted with an optimal number of WT islets (300) (n = 6) or 60 WT islets (n = 12) and treated daily with either 0.5 μM BIBO3304 (n = 6) or placebo (n = 6) for 9 days, respectively, after which BIBO3304 treatment was stopped but blood glucose levels continuously monitored till day 60 post-transplant. b Results for the first 9 days expressed as area under the curve. c, d, g, h Diabetic mice were transplanted with 60 WT islets and subsequently orally treated with 0.5 μM BIBO3304 or placebo and i.v. glucose tolerance tests (1 g/kg body weight) (n = 4 per group) were performed at day 5 post-transplant. Blood glucose levels and insulin production were monitored. Results are also expressed as area under the curve. e Representative photomicrographs of islet transplant grafts from placebo and BIBO3304-treated mice showing immunofluorescent staining for insulin (red), Ki67 (green) and nucleus counterstained with DAPI (blue). Arrows indicate Ki67-positive β-cells. f Quantification of Ki67-positive β-cells in grafts of placebo or BIBO3304-treated mice (n = 3). i, j Diabetic mice were transplanted with 60 WT islets (n = 10 per group) and half of the mice were treated with 0.5 μM BIBO3304 from day 1, and the other half were treated with placebo. Mice originally receiving placebo were then treated with 0.5 μM BIBO3304 from day 9 for 10 days, after which treatment was discontinued. Blood glucose levels were monitored till day 60 after which survival nephrectomy was performed. k Diabetic mice were transplanted with 60 WT islets (n = 10) and treated with 0.5 μM BIBO3304 for 9 days after which treatment was discontinued. Blood glucose levels were monitored till day 260 after which survival nephrectomy was performed. Data are shown as mean ± s.e.m. *P < 0.05, **P < 0.01, calculated by t-test (b, d, f, h) or two-way ANOVA analysis

Fig. 3

Improved human islet transplant outcomes due to Y1 receptor antagonism. a Human islets from three independent donors were isolated and cultured for 48 h. Insulin secretion from human islets was determined in response to 2.8 and 28 mM glucose with or without 1 μM BIBO3304 (n = 3; triplicate; three independent donors). b, c Alloxan-induced NOD RAG1^−/− diabetic mice were transplanted with human islets (IEQ 1400) and subsequently orally treated with or without 0.5 μM BIBO3304 and blood glucose levels were monitored for 15 days after which survival nephrectomy was performed (n = 7 per group; two independent donors). Results are also expressed as area under the curve. d, e Human islet recipient mice treated with placebo or 0.5 μM BIBO3304 for 7 days (n = 4) were fasted for 6 h and i.p. glucose tolerance tests (1 g/kg body weight) were performed. Results were also expressed as area under the curve. f Serum insulin levels in human islet recipient mice were determined at day 15 post-transplantation (n = 7). g Alloxan-induced diabetic mice were transplanted with a full MHC mismatch (BALB/c islet  −> C57BL/6) optimal number of allogeneic islets (300) (n = 6) or 60 allogeneic islets (n = 14) and treated daily with either 0.5 μM BIBO3304 (n = 6) or placebo (n = 8). h 6-week-old female NOD mice were treated with placebo or 0.5 μM BIBO3304 daily and glucose levels monitored weekly (n = 16 per group). i When the blood glucose levels in BIBO3304-treated mice eventually also reached 12 mM, a 10-fold higher dose of BIBO3304 was given in an attempt to reverse the hyperglycemia (n = 3). j, k, l Glucose tolerance tests were performed in BIBO3304-treated and placebo-treated NOD mice at 11, 15, and 19 weeks of age, respectively (n = 6 per group). m Results were also expressed as area under the curve. n, o Insulitis was scored on islets from NOD mice either treated with 0.5 μM BIBO3304 or placebo at 12 or 22 weeks of age (n = 6, n = 5), respectively. Data are means ± s.e.m. *P < 0.05, **P < 0.01; ***P < 0.001 calculated by t-test (a, c, e, f) or two-way ANOVA analysis

Fig. 4

Y1 receptor signaling regulates β-cell function via a cAMP-CREB-dependent pathway. a Pancreas weight of BIBO3304 vs. placebo-treated WT mice (n = 3). b Total islet number of BIBO3304 compared to placebo-treated WT mice (n = 3). c Representative photomicrographs of islets stained for insulin from BIBO3304 and placebo-treated WT mice (n = 3). d Total islet area of BIBO3304 compared to placebo-treated WT mice (n = 3). e Quantification of Ki67-positive β-cells in islets from BIBO3304 or placebo-treated WT mice. f cAMP production of WT islets in response to 2 and 20 mM glucose co-stimulation with GLP-1 receptor agonist, Exenatide (100 nM), in the presence or absence of PYY (50 nM), or BIBO3304 (1 μM) (n = 5). g cAMP production of Y1^−/− islets in response to 2 and 20 mM glucose co-stimulation with GLP-1 receptor agonist, Exenatide (100 nM), in the presence or absence of PYY (50 nM), or BIBO3304 (1 μM) (n = 5). h, i Pancreatic islets from WT mice treated with placebo or BIBO3304 and Y1^−/− mice were isolated and subjected to SDS–PAGE and western blot analysis using anti-p-CREB and α-tubulin antibodies (n = 5–7). j, k Representative fluorescence micrographs of pancreata from WT mice treated with placebo or BIBO3304 and Y1^−/− mice showing staining for insulin (red), phospho-CREB (green) and nuclear counterstained with DAPI (blue), scale bar = 20 μm. l, m Pancreatic islets from Y1^−/− and WT mice treated with placebo or BIBO3304 were isolated and genes involved in insulin secretory pathways including glucokinase, Tpi1, pyruvate carboxylase, Mdh2, ATP citrate Lyase, and Hadh were determined using quantitative RT-PCR. Cyclophillin A and Rpl19 were used as a housekeeping gene (n = 3). Data are means ± s.e.m. *P < 0.05, **P < 0.01, calculated by t-test (c, e, f) or two-way ANOVA analysis