Glycoprotein 130 receptor signaling mediates α-cell dysfunction in a rodent model of type 2 diabetes

(1)

Article

Reference

Glycoprotein 130 receptor signaling mediates α-cell dysfunction in a rodent model of type 2 diabetes

CHOW, Samuel Z, et al.

Abstract

Dysregulated glucagon secretion accompanies islet inflammation in type 2 diabetes. We recently discovered that interleukin (IL)-6 stimulates glucagon secretion from human and rodent islets. IL-6 family cytokines require the glycoprotein 130 (gp130) receptor to signal. In this study, we elucidated the effects of α-cell gp130 receptor signaling on glycemic control in type 2 diabetes. IL-6 family cytokines were elevated in islets in rodent models of this disease.

gp130 receptor activation increased STAT3 phosphorylation in primary α-cells and stimulated glucagon secretion. Pancreatic α-cell gp130 knockout (αgp130KO) mice showed no differences in glycemic control, α-cell function, or α-cell mass. However, when subjected to streptozotocin plus high-fat diet to induce islet inflammation and pathophysiology modeling type 2 diabetes, αgp130KO mice had reduced fasting glycemia, improved glucose tolerance, reduced fasting insulin, and improved α-cell function. Hyperinsulinemic-euglycemic clamps revealed no differences in insulin sensitivity. We conclude that in a setting of islet inflammation and pathophysiology modeling [...]

CHOW, Samuel Z, et al . Glycoprotein 130 receptor signaling mediates α-cell dysfunction in a rodent model of type 2 diabetes. Diabetes , 2014, vol. 63, no. 9, p. 2984-95

DOI : 10.2337/db13-1121 PMID : 24812426

Available at:

http://archive-ouverte.unige.ch/unige:55460

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

Samuel Z. Chow,¹Madeleine Speck,¹Piriya Yoganathan,¹Dominika Nackiewicz,¹Ann Maria Hansen,² Mette Ladefoged,²Björn Rabe,³Stefan Rose-John,³Peter J. Voshol,⁴Francis C. Lynn,¹

Pedro L. Herrera,⁵Werner Müller,⁶Helga Ellingsgaard,⁷and Jan A. Ehses¹

Glycoprotein 130 Receptor Signaling Mediates a -Cell

Dysfunction in a Rodent Model of Type 2 Diabetes

Diabetes 2014;63:2984–2995 | DOI: 10.2337/db13-1121

Dysregulated glucagon secretion accompanies islet in- ﬂammation in type 2 diabetes. We recently discovered that interleukin (IL)-6 stimulates glucagon secretion from human and rodent islets. IL-6 family cytokines require the glycoprotein 130 (gp130) receptor to signal.

In this study, we elucidated the effects ofa-cell gp130 receptor signaling on glycemic control in type 2 diabetes. IL-6 family cytokines were elevated in islets in rodent models of this disease. gp130 receptor activation increased STAT3 phosphorylation in primary a-cells and stimulated glucagon secretion. Pancreatic a-cell gp130 knockout (agp130KO) mice showed no differences in glycemic control, a-cell function, or a-cell mass. However, when subjected to streptozotocin plus high-fat diet to induce islet inﬂammation and pathophysiology modeling type 2 diabetes,agp130KO mice had reduced fasting glycemia, improved glucose tolerance, reduced fasting insulin, and improved a-cell function.

Hyperinsulinemic-euglycemic clamps revealed no differences in insulin sensitivity. We conclude that in a setting of islet inﬂammation and pathophysiology modeling type 2 diabetes, activation ofa-cell gp130 receptor signaling has deleterious effects on a-cell function, pro- moting hyperglycemia. Antagonism of a-cell gp130

receptor signaling may be useful for the treatment of type 2 diabetes.

Islet inflammation (1,2) and pancreatica-cell dysfunction (3–6) contribute to hyperglycemia in patients with type 2 diabetes. Pancreatic islets from patients with type 2 diabetes are infiltrated with macrophages (7,8), express elevated proinflammatory cytokines (9,10), and express features of fibrosis (11), consistent with reports from animals and primates with this disease (7,12–18). The detrimental effects of inflammation on islet b-cell function were recently confirmed, when the interleukin (IL)-1 receptor antagonist reduced hyperglycemia and improved b-cell insulin secretion in patients with type 2 diabetes (1).

Pancreatica-cell dysfunction is detectable in the early stages of type 2 diabetes, where the inverse relationship between pulsatile insulin and glucagon secretion is lost (19). This dysregulation is associated with relative hyperglucagonemia, which contributes to the development of fasting hyperglycemia associated with frank type 2 diabetes (6). Indeed, inhibition of glucagon action by various means reduces hyperglycemia in rodent models of this disease (5,6).

1Department of Surgery, Faculty of Medicine, University of British Columbia, Child and Family Research Institute, Vancouver, British Columbia, Canada

2Diabetes Research Unit, Novo Nordisk A/S, Måløv, Denmark

3Institute of Biochemistry, Medical Faculty, Christian Albrechts University of Kiel, Kiel, Germany

4Institute of Metabolic Science, University of Cambridge Metabolic Research Laboratories and National Institute for Health Research Biomedical Research Centre, Addenbrooke’s Hospital, Cambridge, U.K.

5Department of Genetic Medicine and Development, Faculty of Medicine, Univer- sity of Geneva, Geneva, Switzerland

6Faculty of Life Sciences, University of Manchester, Manchester, U.K.

7The Centre of Inﬂammation and Metabolism and the Centre for Physical Activity Research, Department of Infectious Diseases, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark

Corresponding author: Jan A. Ehses, [email protected].

Received 19 July 2013 and accepted 7 April 2014.

This article contains Supplementary Data online at http://diabetes .diabetesjournals.org/lookup/suppl/doi:10.2337/db13-1121/-/DC1.

S.Z.C. and M.S. contributed equally to this work.

ISLETSTUDIES

(3)

We recently identified IL-6 as a key cytokine elevated during islet inflammation in rodents with type 2 diabetes (7,20), with effects on a-cell glucagon secretion, GLP-1 secretion, and survival (21,22). IL-6 is part of a family of IL-6–type cytokines that all share the same common signal transducer for signaling, the glycoprotein 130 (gp130) (23). In support of a role for IL-6 cytokines in human disease, a recent study of global gene expression in human islets identified a group of coexpressed genes associated with type 2 diabetes. Within this group of genes, expression of two IL-6 family cytokines, IL-6 and IL-11, were both increased in islets from patients with type 2 diabetes: IL-6 was increased 2.75-fold, while IL-11 mRNA was increased 1.61-fold (24).

Despite these advances, there remain significant gaps in our knowledge regarding how the gp130 receptor regulatesa-cell function. In our previous reports, we were unable to distinguish systemic IL-6 effects from tissue- specific IL-6 cytokine-mediated effects on the a-cell. In addition, we did not evaluate how IL-6 family cytokines regulate glycemia via actions on glucagon or GLP-1 secretion in a setting of islet inflammation and decreasedb-cell mass modeling type 2 diabetes (21,22). Thus we generated a-cell gp130 receptor knockout (agp130KO) mice and investigated the role of the a-cell in a nongenetic rodent model of type 2 diabetes.

RESEARCH DESIGN AND METHODS Animals

Male rodents were used for all experiments. Goto- Kakizaki (GK) rats (Taconic), Wistar rats (Taconic), BKS.Cg-m^+/+Lepr^db/BomTac (db/db) mice (Taconic), C57BL/6J mice (CDM), and Gt(ROSA)26Sortm4(ACTB-tdTomato,-EGFP)Luo/J (mT/mG) mice (The Jackson Laboratory) were from com- mercial sources. Gp130^ﬂ^/^ﬂand Gcg-Cre mice were provided by W. Müller (25) and P. Herrera (26), respectively. Gcg- Cre mice were bred with mT/mG mice to produce Gcg-Cre mT/mG mice. All rodents were housed at the Child and Family Research Institute (CFRI) animal facility, maintained on a 12-h light/dark cycle, and fed a normal chow diet (13 kcal% fat). All procedures were approved by the University of British Columbia Committee on Animal Care or the Principles of Laboratory Care (Denmark).

Streptozotocin, High-Fat Diet, and Streptozotocin/

High-Fat Diet Mouse Models

Streptozotocin (STZ; Sigma-Aldrich) was prepared in acetate buffer and administered (25 mg/kg) via intraperitoneal injections for 5 consecutive days at 7–9 weeks of age. Con- trol mice were administered an intraperitoneal injection of acetate buffer. Some mice were put on high-fat diet (HFD;

58 kcal% fat with sucrose; Research Diets) 3 weeks following intraperitoneal injection of acetate buffer or STZ.

Islet Isolation and Immunocytochemistry

Mouse and rat islets were isolated and cultured as described (7,27,28). For pSTAT3 experiments, islets were dispersed and plated on 804G extracellular matrix–coated

8-chamber slides (Nunc) (29). Islet cells were serum starved followed by treatment with IL-6, hyper-IL (HIL)- 6 (30), leukemia inhibitory factor (LIF), or IL-27 (100 ng/mL was used for all cytokine treatments unless otherwise stated). Cells were paraformaldehydeﬁxed, permeabilized with methanol, blocked with 5% goat serum (Invitrogen), and incubated with rabbit anti-pSTAT3 (1:100; Cell Signal- ing Technology) and mouse anti-glucagon (1:1,000; Sigma- Aldrich). Slides were incubated with Alexa 488 donkey anti-rabbit (1:250; Jackson ImmunoResearch) and Alexa 594 donkey anti-mouse (1:450; Jackson ImmunoResearch) and mounted using Vectashield with DAPI (Vector Labora- tories). Imaging was acquired with a BX61 microscope and Leica SP5 II confocal imaging system and quantiﬁed using Image-Pro Analyzer. An average of 1,1306 261 glucagon- positivea-cells were counted per experimental condition.

Pancreatic a-cell apoptosis was assessed by TUNEL staining as described (21) and imaged using a BX61 microscope. An average of 3,164 6 722 glucagon-positive a-cells were counted per experimental condition.

Flow Cytometry

Isolated islets were dispersed and incubated with Fc block (1:100; eBioscience). Islet cells were stained with CD45- eFluor 450 (1:250; clone 30-F11; eBioscience), CD11b-PE (1:1,200; clone M1/70; eBioscience), Ly-6c allophycocya- nin (1:1,200; clone HK1.4; eBioscience), and the viability dye eFluor 506 (1:1,000; eBioscience). Unstained, single stains, andﬂuorescence minus one controls were used for setting gates and compensation with the help of the CFRI FACS core facility using a BD LSR II instrument (BD Biosciences). Cells were gated on live cells and CD45⁺cells followed by CD11b⁺Ly6c⁺ cells. Islets were pooled from two mice per treatment per time point.

Secretion Assays

Islets were plated on 804G extracellular matrix–coated plates for secretion assays. Aprotinin (250 kallikrein inhibitor units/mL; Sigma-Aldrich) and dipeptidyl peptidase- 4 inhibitor (50mmol/L; Millipore) were added to each well, and islets were treated with IL-6 or HIL-6. Conditioned media were collected, and 70% acid ethanol was added to islets for total hormone content. Glucagon and GLP-1 were measured by radioimmunoassay (Millipore). Glucose- stimulated insulin secretion from islets was performed as previously published (7). For total pancreatic insulin, glucagon, and GLP-1 content, hormones were extracted using 70% acid ethanol and determined using Luminex technology (Millipore). Total pancreatic protein concentration was determined using the Pierce BCA Protein Assay kit (Thermo Scientiﬁc). Ex vivo secretion of IL-6 and LIF was determined by Luminex, and soluble IL-6 receptor (sIL-6R) secretion was assayed by ELISA (31).

Physiological Measurements

For blood sampling, aprotinin (250 kallikrein inhibitor units/mL plasma; Sigma-Aldrich) and dipeptidyl peptidase-4 inhibitor (50 mmol/L; Millipore) were added to the

(4)

collection tubes. Mice were fasted overnight and injected intraperitoneally with 1.5 g/kg body weight glucose for glucose tolerance tests (intraperitoneal glucose tolerance test [IPGTT]). For insulin tolerance tests (ITTs), mice were fasted 2 h and injected intraperitoneally with 1 unit/kg insulin (Novolin GE, Novo Nordisk). Plasma insulin and glucagon were measured using ELISA (ALPCO and Mercodia, respectively), total GLP-1 was measured using Meso Scale Discovery technology, and plasma IL-6 was measured using Luminex technology (Millipore).

Hyperinsulinemic-Euglycemic Clamps

Hyperinsulinemic-euglycemic clamps were performed as previously described (32–34). Mice were fasted overnight and anesthetized with acepromazine, midazolam, and fen- tanyl, with an initial dose by intraperitoneal injection and top-up doses by subcutaneous injection. Once fully immo- bilized from anesthesia, the tail vein was cannulated, and a 1-h basal infusion of³H-D-glucose (1.2mCi/h) was started to determine steady-state tracer. Duplicate blood samples were taken from the saphenous vein at the end of the basal period to measure fasting blood glucose levels, for deter- mination of endogenous glucose production, and for insulin measurements. Hyperinsulinemia was induced with a constant infusion of insulin (3.5 mU/kg/min). To main- tain euglycemia (;5.0 mmol/L), a variable infusion of 12.5%D-glucose was started simultaneously. Once euglycemia had been achieved, steady state was maintained for 60 min, after which triplicate blood samples were taken for further analysis. Brieﬂy, insulin concentrations were measured by ELISA, and plasma samples were counted using a 1450 MicroBeta TriLux LSC and Luminescence Counter (PerkinElmer) after extraction by trichloroacetic acid pre- cipitation. Whole-body glucose use (mmol kg²¹min²¹) was determined as the ratio of the speciﬁc activity of glucose to the rate of³H-D-glucose appearance. Subsequently, endogenous glucose production (mmol kg²¹ min²¹) could be calculated as the difference between whole-body glucose uptake and exogenous glucose infusion.

Immunohistochemistry

Antigen retrieval was performed on sectioned tissues using 10 mmol/L citrate buffer (Fisher Scientiﬁc) followed by blocking with 1% goat serum and donkey/goat anti- mouse IgG (1:30; eBioscience). Sections were incubated with guinea pig anti-insulin (1:1,000; Millipore) and mouse anti-glucagon (1:2,000) followed by incubation with Alexa 488 donkey anti–guinea pig (1:250; Jackson ImmunoResearch) and Alexa 594 donkey anti-mouse (1:450). Sections were mounted using Vectashield with DAPI and imaged on a BX61 microscope.

Pancreatic Gcg-Cre X mT/mG and mT/mG cryosections were paraformaldehyde ﬁxed and equilibrated through a sucrose gradient. Tissue was frozen in optimum cutting temperature compound (Tissue-Tek). Sections were permeabilized with 0.1% Triton X-100 for 30 min and then blocked in 5% goat serum and 1% BSA (Fisher Scientiﬁc).

Sections were incubated in 1% BSA and 0.1% Triton

X-100 with mouse anti-glucagon (1:2,000) followed by incubation with Alexa 647 goat anti-mouse (1:250;

Abcam) and mounted with Vectashield. Images were taken on a Leica SP5 II confocal imaging system.

For pancreaticb- anda-cell mass analysis, all islets in three sections (spaced 200mm apart) were analyzed using Image-Pro Analyzer. Insulin and glucagon-positive area was quantiﬁed using the area/count function. Endocrine cell mass was calculated as previously described (21).

Gene Expression Analysis

Total RNA was extracted from islets using the MN NucleoSpin RNA II kit (Macherey-Nagel), RNA was re- versed transcribed and quantitative PCR was performed using PrimeTime primers/probes (IDT) in a ViiA 7 real- time PCR system (ABI) as previously described (35). Dif- ferential expression was determined by the 2^2DDCT method (36) withRplp0and18S as reference genes.

Statistical Analysis

Data are expressed as means6SEM/SD with the number of individual experiments presented in the figure legends. All data were analyzed using the nonlinear regression analysis program Prism (GraphPad), and significance was tested using Studentttest or ANOVA with post hoc tests for multiple comparison analysis. Significance was set atP,0.05.

RESULTS

IL-6 Family Cytokines Are Elevated in Pancreatic Islets From Rodent Models of Type 2 Diabetes

We analyzed islet IL-6 family cytokine expression in two rodent models, the GK rat anddb/dbmouse. Both the GK rat and thedb/dbmouse displayed elevated blood glucose and declining insulin levels at 12 and 15 weeks of age, respectively, suggestive ofb-cell failure (Fig. 1A,B,D, and E). Isolated islets from 8–12-week-old GK rats had signif- icantly increased levels ofIl6,Lif,Clcf1, andOsmmRNA relative to Wistar controls (Fig. 1C); similar effects were observed in islets from 15-week-olddb/dbmice (compared with Fig. 1CandF). Interestingly, both IL-6 family cytokines and glycemia were more elevated in db/db mice relative to GK rats. Thus elevated IL-6 family cytokine expression accompanies hyperglycemia and declining insulin levels in rodent models of type 2 diabetes, and the magnitude of islet IL-6 cytokine expression may contribute to the prevailing hyperglycemia in these models.

gp130 Receptor Activation Stimulates STAT3

Phosphorylation and Glucagon Secretion Froma-Cells To determine if gp130 receptor signaling is activated in a-cells following IL-6 treatment, dispersed islets were stimulated with IL-6 or HIL-6 (IL-6 fused to the sIL-6R [37]; HIL-6 can stimulate all gp130 receptors in the ab- sence of an IL-6R) and immunostained for pSTAT3. IL-6 induced phosphorylation of STAT3 in a-cells within 30 min to a similar degree as HIL-6 (Fig. 2A). Next we assessed the effect of different IL-6 doses and time of exposure on glucagon secretion in islets. IL-6 increased glucagon secretion in a concentration- and time-dependent

(5)

manner, with no signiﬁcant differences in glucagon content (Fig. 2B and C). Consistent with our pSTAT3 data, IL-6 stimulated glucagon secretion to a similar degree as HIL-6 (Fig. 2D and E), suggesting that a-cell gp130 receptor signaling is maximally activated by IL-6 and sup- porting our previous observation that pancreatic a-cells express the IL-6 transmembrane receptor (21). To ensure that IL-6 did not induce a-cell death, causing glucagon release into the incubation medium, islets were treated with IL-6 and apoptosis visualized by TUNEL staining. IL-6 protected from cell death induced by a mixture of cytokines (IL-1b, TNFa, and IFNg) (Fig. 2F), consistent with our previous observations under nutrient stress (21). Col- lectively, these data indicate that gp130 signaling within a-cells leads to phosphorylation of STAT3 and stimulation of glucagon secretion.

Generation and Validation ofagp130KO Mice

To assess Cre recombinase activity withina-cells, Gcg-Cre mice were crossed with mT/mG reporter mice. Pancreatic

sections revealed expression of EGFP in a proportion of a-cells stained with glucagon, demonstrating active Cre recombinase in 44.1 6 1.5% of a-cells (Fig. 3Aand B), similar to previous reports using Gcg-Cre mice (38).

Next we assessed pSTAT3 levels in response to gp130 receptor activation in ﬂ/ﬂ controls and agp130KO islets.

Stimulation of pSTAT3 with IL-6 and LIF was signiﬁcantly decreased withina-cells ofagp130KO islets (Fig. 3C). IL-27 failed to induce pSTAT3 within control islets, suggesting lack of IL-27 receptors ona-cells. Finally, we assessed glucagon secretion in response to gp130 receptor activation in agp130KO islets. IL-6 stimulated glucagon secretion was reduced by;50% inagp130KO islets (Fig. 3D). There was no effect of IL-6 on GLP-1 secretion following 48-h exposure of mouse islets (Fig. 3E). Thusagp130KO islets have Cre- mediated recombination occurring in a proportion ofa-cells and reduced gp130 receptor signaling, coinciding with a 50%

reduction in gp130 receptor–mediated glucagon secretion.

To rule out anya-cell or intestinal L-cell developmental abnormalities in agp130KO mice, we assessed several

Figure 1—IL-6 family cytokine expression is increased in rodents with type 2 diabetes. Nonfasted glycemia (A) and insulin levels (B) in Wistar and GK rats. IL-6 family cytokine expression levels in islets from 8–12-week-old Wistar and GK rats (C). Expression levels were assessed by quantitative real-time PCR, normalized to the housekeeping geneRplp0, and expressed as fold control (Wistar). Nonfasted glycemia (D) and insulin levels (E) in 4–15-week-olddb/dbmice. IL-6 family cytokine expression levels in islets from 4-, 8-, and 15-week-old db/dbmice (F). Expression levels were assessed by quantitative real-time PCR, normalized to the housekeeping gene Rplp0, and expressed as fold control (4 weeks db/db). Data represent mean 6 SEM from 3–4 Wistar and 3–4 GK rats (A–C). Data represent mean6 SEM from 5–20 db/dbmice (D–E) and mean 6SD from islets pooled from $3 mice/age (F). White bars represent Wistar/

4-week db/db, gray bars represent 8-week db/db, and black bars represent GK or 15-week db/db. ★P < 0.05, ★★P < 0.01,

★★★P<0.001 as tested by Studentttest. nd, not detected.

(6)

physiological parameters in agp130KO mice.agp130KO mice displayed no difference in body weight, glucose tolerance, glucose-stimulated insulin secretion, or insulin sensitivity at 16–18 weeks of age (Fig. 3F–I). Fasting GLP-1 levels were unchanged, and pancreatic a-cell function was normal, as assessed by glucose-mediated glucagon suppression during a GTT and hypoglycemia-mediated glucagon secretion during an ITT (Fig. 3JandK). Lastly,b-cell and a-cell mass were similar between genotypes (Fig. 3L andM). Thusagp130 receptor signaling does not appear to affecta-cell development or function under normal chow- fed conditions.

Pancreatica-Cell Dysfunction and Islet Inﬂammation in the STZ/HFD Mouse Model of Type 2 Diabetes To investigate the role of the a-cell gp130 receptor in a mouse model of human type 2 diabetes, we generated a nongenetic model displaying features of type 2 diabetes.

Mice were administered STZ, HFD, or STZ in combination with HFD as depicted in Fig. 4A. STZ alone caused glucose intolerance, tended to reduce insulin secretion in response to glucose in vivo, had minimal effects on glucose- mediated glucagon suppression and hypoglycemia- induced glucagon secretion in vivo, and tended to reduce b-cell mass and increase a-cell mass (Fig. 4B–L). HFD

Figure 2—IL-6 and HIL-6 activatea-cell STAT3 signaling and stimulate glucagon secretion from islets. STAT3 activation in primary mouse a-cells following 30-min cytokine treatment (A). Representative image indicates staining for glucagon (red), pSTAT3 (green), and DAPI (blue). Glucagon secretion (B) and glucagon content (C) following IL-6 treatment of mouse islets at the indicated doses and times.

Glucagon secretion (D) and glucagon content (E) from IL-6– and HIL-6–treated mouse islets following 48 h. TUNEL-positive a-cells were analyzed following 24-h exposure to IL-6, cytokine mix (200 pg/mL IL-1b, 1 ng/mL TNFa, 5 ng/mL IFNg), or cytokine mix plus IL-6 (F). Data represent mean6SEM fromn= 3 mice performed in three independent experiments (A),n= 3–6 mice performed in two independent experiments (BandC),n= 6 mice performed in three independent experiments (DandE), andn= 4 mice in two independent experiments (F).★P<0.05;★★P<0.01;★★★P<0.001 vs. untreated control; #P<0.05 vs. cytokine mix as tested by ANOVA with Dunnett (A–E) or Newman-Kuels posttest (F). Ctrl, control; cyt, cytokine; CM, cytokine mix.

(7)

alone increased body weight, caused glucose intolerance, increased insulin secretion in response to glucose in vivo, impaired glucose-mediated glucagon suppression, had no effect on hypoglycemia-induced glucagon secretion in vivo, and had minimal effects onb- anda-cell mass (Fig. 4B–L).

STZ in combination with HFD caused increased body weight, fed hyperglycemia, glucose intolerance, and marked a-cell dysfunction (Fig. 4B–H). In addition,b-cell mass was reduced by ;50%, in parallel with impaired glucose- stimulated insulin secretion from islets ex vivo and reduced

Figure 3—agp130KO mice exhibit partial loss ofa-cell gp130 receptor function and normal glucose homeostasis anda-cell function under chow-fed conditions. Gcg-Cre mice were bred with mT/mG mice to assess activity ofa-cell Cre recombinase (AandB). Images indicate nonrecombinant cells (tdTomato), recombinant cells (EGFP), and glucagon-positive cells (purple). EGFP was expressed in 44.161.5% of a-cells (B). STAT3 activation in primary mousea-cells fromfl/flandagp130KO mice following 30-min cytokine treatment (C). Glucagon (D) and GLP-1 (E) secretion fromfl/flandagp130KO mouse islets following IL-6 treatment. Basal glucagon secretion was 2.2660.54 forfl/fl and 2.3560.53 foragp130KO as percentage glucagon content. Basal GLP-1 secretion was 1.9760.39 forfl/fland 1.3160.17 for agp130KO as percentage GLP-1 content. Body weight offl/flandagp130KO mice at 16–18 weeks of age (F). IPGTT (G) (1.5 g/kg), insulin levels during IPGTT (H), and ITT (I) (1 unit/kg). Fasting GLP-1 levels (J) and glucagon levels during IPGTT and ITT (K). Pancreaticb- and a-cell mass (LandM). Gray bars representfl/flcontrol, and light red bars representagp130KO. Scale bars represent 50mm. Data represent n= 3–5 mice per genotype (AandB),n= 3–4 mice per genotype performed in three independent experiments (C),n= 6 mice per genotype performed in two independent experiments (DandE), and 3–12 mice per genotype (F–M).★P<0.05;★★P<0.01 as tested by Studentt test. Ctrl, control;fl/fl, gp130fl/fl; KO,agp130KO.

(8)

Figure 4—Pancreatica-cell dysfunction and islet inﬂammation in STZ-, HFD-, and STZ/HFD-treated mice. Schematic of chow, STZ (25 mg/kg), HFD, and STZ/HFD treatment groups (A) with weekly fed blood glucose monitoring (B). IPGTT (C) (1.5 g/kg glucose), ITT (D) (1 unit/kg insulin), and body weight (E) of chow, STZ, HFD, and STZ/HFD mice. Plasma insulin (F) and glucagon (GandH) during IPGTT and ITT, and fed insulin (I) after 9 weeks. Pancreaticb- anda-cell mass (JandK) in treated mice with representative images (L). Glucose-stimulated insulin secretion (M) and insulin content (N) of islets ex vivo after 9 weeks. Pancreatic islet inﬁltrating monocytes (CD11b⁺Ly6c⁺cells) in chow and STZ-treated mice (O). Expression levels of IL-6 family cytokines post-STZ injection and during HFD relative to chow mice (P). IL-6, sIL-6R, and LIF protein secretion from 100 islets/well isolated 3 weeks post-STZ (Q). Islet-derived IL-6 (R) from 100 islets/well and

(9)

islet insulin content (Fig. 4J–N).a-Cell mass was also sig- niﬁcantly increased in STZ/HFD mice (Fig. 4K). Thus treatment with STZ plus HFD resulted in a mouse with impaireda- andb-cell function and hyperglycemia.

To characterize islet inflammation post-STZ, islet infiltrating monocytes (CD11b⁺Ly6c⁺cells) were analyzed byflow cytometry. The number of islet monocytes tended to increase at weeks 0.5 and 1 post-STZ and were signifi- cantly increased at week 2 post-STZ (Fig. 4O). To determine if islet IL-6 family cytokine expression was increased, we analyzed islet gene expression at multiple times post-STZ and during subsequent HFD feeding (Fig.

4P). IL-27mRNA was increased;1 week post-STZ,Osm mRNA was increased 2 weeks post-STZ, andIL-6andLif mRNA were increased 3 weeks post-STZ. Other IL-6 family cytokines were unchanged (Fig. 4P and not shown).

Consistent with these mRNA data, IL-6 and LIF protein secretion were increased in islets isolated 3 weeks post- STZ, with no change in sIL-6R secretion (Fig. 4Q). Islet IL-6 secretion was not signiﬁcantly increased at week 9 in our treatment groups (Fig. 4R), indicating that IL-6 cytokines are only transiently increased following STZ and not further increased by HFD. However, plasma IL-6 levels were signiﬁcantly increased in our STZ/HFD group at the end of our study (Fig. 4S). Taken together, these data show that STZ/HFD-treated mice have many hall- mark characteristics of type 2 diabetes, including pancreatic a-cell dysfunction and increased expression of pancreatic islet IL-6 family cytokines.

Partial Knockout of thea-Cell gp130 Receptor Protects From Hyperglycemia Following STZ/HFD To determine the consequence ofa-cell gp130 signaling during islet inﬂammation and in a pathophysiological setting modeling type 2 diabetes, we subjected agp130KO mice to STZ alone or STZ followed by HFD feeding.

agp130KO mice displayed no difference in body weight following STZ or STZ/HFD (Fig. 5A). When subjected to STZ alone,agp130KO mice showed mildly improved glucose tolerance, with no differences in insulin sensitivity, insulin secretion, or a-cell function in vivo (Fig. 5C–I).

However, following STZ/HFD, agp130KO mice had decreased fasting glycemia, improved glucose tolerance, and decreased fasting insulin and showed improvements in glucose modulated glucagon secretion in vivo (Fig. 5B–

I). This was despite decreased fasting GLP-1 levels (Fig.

5E). ITTs indicated no difference in insulin sensitivity in agp130KO mice (Fig. 5F). No differences in pancreatic glucagon, insulin, or GLP-1 content (Fig. 5J–L) were

observed between genotypes, along with no differences in b- and a-cell mass (Fig. 5M–O). STZ/HFD-treated Gcg-Cre mice did not show any glucose intolerance compared with wild-type mice excluding any phenotypic effect of Cre recombinase expression ina-cells (Supplementary Fig. 1). Further, agp130KO mice on HFD alone showed no phenotypic differences in glucose tolerance ora- and b-cell function (Supplementary Fig. 2).

Finally, to determine if changes in insulin sensitivity contributed to the protective phenotype in STZ/HFD- treated mice, we performed hyperinsulinemic-euglycemic clamps. There were no differences in glucose infusion rate, whole-body insulin-stimulated glucose uptake, or hepatic insulin sensitivity between genotypes (Fig. 6). In- terestingly, hepatic glucose production under fasting conditions tended to be reduced inagp130KO mice (Fig. 6D), consistent with reduced fasting glycemia in these mice.

Taken together, these data show that gp130 receptor signaling contributes to a-cell dysfunction in a nongenetic model of type 2 diabetes, and partial inhibition of gp130 receptor signaling improves glucose tolerance and protects from hyperglycemia through improved a-cell function.

DISCUSSION

Normally, glucagon secretion is increased during fasting and suppressed as blood glucose rises following a meal in a counterregulatory fashion to insulin. However, during type 2 diabetes, glucagon secretion is inappropriately elevated, contributing to hyperglycemia (5,6). Pioneering studies by Müller et al. (39) and elegant clamp studies by Shah et al. (3) have demonstrated that lack of glucagon suppression contributes to postprandial hyperglycemia in people with type 2 diabetes. The main mechanism underlying this hyperglucagonemia is thought to beb-cell dysfunction and a decrease in b-cell–derived secretory products that are known to have inhibitory effects on a-cell secretion, including insulin, g-aminobutyric acid, and zinc ions (40). All of these factors inhibit glucagon secretion, and their secretion is likely decreased in frank type 2 diabetes. Our data demonstrate that islet inﬂam- mation also contributes toa-cell dysfunction, with a central role for gp130 receptor signaling in this process and with implications for type 2 diabetes.

agp130KO mice were protected from glucose intolerance following STZ alone, but not following HFD alone.

This was consistent with STZ treatment causing a transient increase in IL-6 and LIF secretion from islets, while

systemic IL-6 (S) during fed after 9 weeks. White bars/circles represent chow control, horizontally striped bars/white squares represent STZ, vertically striped bars/white triangles represent HFD, and black bars/triangles represent STZ/HFD. Scale bars represent 50mm. Data represent mean6SEM fromn= 5 mice (B–I),n= 3–11 mice (J–L) per experimental group,n= 3–5 mice per experimental group (MandN),n= 4–5 mice per time point (O),n= 5–6 mice per time point (P),n= 4–5 mice for each experimental group (Q–S).★P<0.05;★★P<0.01;

★★★P<0.001;★★★★P<0.0001 as tested by two-way ANOVA with Bonferroni post test (B–D), ANOVA with Dunnett post test (E–H), or Studentttest (M–Q) compared with chow controls. Ctrl, control; ns, not signiﬁcant.

(10)

HFD alone did not increase islet IL-6 family cytokine expression. One caveat in our study is that isolating islets induces IL-6 mRNA expression and secretion (41). It is therefore likely that islet IL-6 is elevated for a more prolonged period of time than depicted in Fig. 4P. Interest- ingly, islet IL-6 cytokine mRNA levels correlated very well

with macrophage inﬁltration following STZ. We recently found that depletion of islet macrophages reduced islet amyloid polypeptide–induced islet IL-6 mRNA expression (42). Whether islet macrophages are the cellular source of IL-6 family cytokines causinga-cell dysfunction should be addressed in future studies.

Figure 5—agp130KO mice display improved glucose homeostasis anda-cell function following STZ/HFD. Body weight (A) and fasting glycemia (B) offl/flandagp130KO mice at 16–18 weeks of age following STZ or STZ/HFD. IPGTT (C) (1.5 g/kg) with quantification of area under curve above basal (D), fasting GLP-1 (E), and ITT (F) (1 unit/kg) with corresponding plasma insulin (G) and glucagon levels (H–I).

Insulin (J), glucagon (K), and GLP-1 (L) pancreatic content offl/flandagp130KO mice following STZ/HFD. Pancreaticb- anda-cell mass (M–N), and representative islet images (O) of STZ/HFDfl/flandagp130KO mice. Representative image indicates glucagon (red), insulin (green), and nuclei (blue; DAPI). Scale bars represent 50mm. Data represent mean6SEM fromn= 4–6 mice per genotype treated with STZ (A–I),n= 22–23 mice per genotype treated with STZ/HFD (A–D),n= 4 mice per genotype (E),n= 4–8 mice per genotype treated with STZ/HFD (F),n= 11–14 mice per genotype treated with STZ/HFD (G–I),n= 4–8 mice per genotype (J–N) from 1–4 independent cohorts of mice. Black border white bars/triangles represent STZfl/fl, red border white bars/triangles represent STZagp130KO, black border gray bars/circles represent STZ/HFDfl/fl, and red border red bars/squares represent STZ/HFDagp130KO.★P<0.05;★★P<0.01 as tested by Studentttest. InC,★P<0.05 and #P<0.05 as tested by Studentttest compared withfl/flcontrols. AUC, area under the curve; KO,agp130KO; ns, not significant.

(11)

Interestingly, the combination of STZ plus HFD did not prolong or increase islet IL-6 family cytokine expression, but it did result in further impairments in a-cell function compared with STZ or HFD treatments alone.

One possible explanation for this synergism is that HFD mediatesa-cell dysfunction by causing a-cell insulin resistance and that STZ exacerbates these effects by de- creasing insulin and inducing inﬂammation. This might explain the synergistic effect of STZ plus HFD on glucose- mediated glucagon suppression in vivo during an IPGTT.

With respect to ITT responses, hypoglycemia-induced glucagon secretion is strongly driven by the parasympathetic nervous system (43,44). Thus STZ acts synergistically with HFD to impair this response. Regardless of the mechanism of action, these hypotheses are consistent with effects of gp130 receptor signaling on a-cell function and glycemic control being more pronounced in the STZ/HFD model, compared with STZ or HFD treatments alone. Thus gp130 receptor signaling acts as a modiﬁer of a-cell function in a synergistic manner with HFD-induced effects ona-cell function.

Improved a-cell function in agp130KO mice treated with STZ/HFD resulted in reduced fasting hyperglycemia, likely due to reduced hepatic glucose output. Reduced glycemia was likely responsible for the decreased fasting insulin observed in agp130KO mice. We ruled out any contribution of increased insulin sensitivity inagp130KO mice treated with STZ/HFD by performing hyperinsulinemic- euglycemic clamps. However, reduced fasting glycemia may have improvedb-cell function in agp130KO mice, likely by reducing the effects of glucotoxicity on theb-cell. This

sequence of events may also help explain how a transient increase in islet IL-6 cytokines can have long-lasting effects on glycemic control.

Two independent groups recently investigated the effect of gp130 receptor cytokines on thea-cell. McGuinness and colleagues found that IL-6²^/²mice had a blunted glucagon response to endotoxin that was restored by replace- ment of IL-6 (45). They went on to show that IL-6 ampliﬁes adrenergic-dependent glucagon secretion from mouse islets (46). In a separate study, Fernández-Millán et al.

showed thata-cell mass expansion and hyperglucagonemia during suckling in rats is partly IL-6–dependent, contributing to hyperglycemia postweaning (47). These studies, our previous work (21), and data presented here support the notion that IL-6 cytokines are modulators of a-cell glucagon secretion. Taking theseﬁndings into con- sideration, we propose that gp130 receptor actions are context-dependent and that IL-6 cytokines modulate a-cell glucagon secretion to allow for increased glucose output during normal physiology (e.g., during acute in- fection or postexercise), while also driving excessive a-cell glucagon secretion and dysfunction during type 2 diabetes.

Exposure of human islets to IL-6 for 4 days or enriched human pancreatic a-cells to IL-6 enhances GLP-1 secretion in addition to glucagon secretion, perhaps via upreg- ulation of pro(hormone)-convertase 1/3 (22). Mouse islets exposed to IL-6 for 2 days did not have elevated GLP-1 secretion. Dedifferentiation of a-cells to a pre- a-cell state has been shown to induce GLP-1 secretion (48). In addition, prolonged culture of human islets in

Figure 6—No difference in insulin sensitivity inagp130KO mice following STZ/HFD. Hyperinsulinemic-euglycemic clamps were performed after STZ plus 6 weeks of HFD feeding inﬂ/ﬂandagp130KO littermates. Blood glucose during clamping (A) and glucose infusion rate (B).

Glucose turnover (C), suppression of hepatic glucose production (D), and insulin levels (E) during clamp. Data represent mean6SEM from n= 7ﬂ/ﬂandn= 6agp130KO mice from two independent cohorts.Pvalue was determined using Studentttest. GIR, glucose infusion rate;

KO,agp130KO.

(12)

vitro results in increased numbers of glucagon-positive cells due to conversion ofb-cells toa-cells (49). We sur- mise that signals causing a-cell dedifferentiation or conversion of b-cells to a-cells act together with IL-6 cytokines to stimulate GLP-1 secretion from a-cells. In the future, the mechanisms underlying IL-6–stimulated a-cell GLP-1 secretion will need to be determined and their role tested in vivo.

Despite decreased systemic GLP-1 levels inagp130KO mice (likely due to receptor deletion in L cells), activation ofa-cell gp130 receptor signaling in a setting of reduced b-cell mass and HFD-induced insulin resistance had detrimental effects on normal a-cell function and glycemic control. This may suggest that effects of gp130 receptor cytokines ona-cell function are more important for glycemic control than effects on L-cell GLP-1 secretion in type 2 diabetes. While it remains to be determined what the mechanisms are that facilitate IL-6 cytokine actions on glucagon secretion, insight into how IL-6/gp130 inﬂu- encesa-cell secretory processes and/ora-cell differentia- tion may allow us to identify druggable targets that modify the actions of the a-cell and could thereby be used for the treatment of type 2 diabetes.

Acknowledgments.The authors thank G. Soukhatcheva, M. Komba, and L. Xu (CFRI) for technical assistance provided by the CFRI islet isolation core facility and CFRIﬂow cytometry core and Bruce Verchere (University of British Columbia) for critical review of the manuscript.

Funding.This work was supported by funding from the CFRI (J.A.E.), the University of British Columbia (J.A.E.), the Canadian Institutes of Health Research (PCN-110793 and PNI-120292 to J.A.E.), the Canadian Diabetes Association (OG-3-13-4097-JE), the Deutsche Forschungsgemeinschaft (Bonn, Germany;

SFB877; project A1 to S.R.-J.), and the Cluster of Excellence“Inﬂammation at Interfaces”(to S.R.-J.). S.Z.C. is supported by a Canadian Institutes of Health Research Transplantation Training Program Scholarship. H.E. is supported by the Danish Council for Independent Research and the Novo Nordisk Foundation. D.N.

is supported by a University of British Columbia Transplantation Training Program and Vanier Canada Graduate Scholarship. J.A.E. has salary support from a CFRI scientist award and a Canadian Diabetes Association scholar award.

Duality of Interest.No potential conﬂicts of interest relevant to this article were reported.

Author Contributions.S.Z.C. designed and performed experiments, analyzed data, and wrote the manuscript. M.S. designed and performed experiments and analyzed data. P.Y., D.N., A.M.H., and M.L. designed and performed experiments. B.R. and S.R.-J. performed experiments and provided reagents. P.J.V. and H.E. designed experiments. F.C.L., P.L.H., and W.M. provided reagents and mice.

J.A.E. supervised the study, designed and performed experiments, analyzed data, and wrote the manuscript. All authors edited the manuscript. J.A.E. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

References

1. Larsen CM, Faulenbach M, Vaag A, et al. Interleukin-1-receptor antagonist in type 2 diabetes mellitus. N Engl J Med 2007;356:1517–1526

2. Donath MY, Shoelson SE. Type 2 diabetes as an inﬂammatory disease. Nat Rev Immunol 2011;11:98–107

3. Shah P, Vella A, Basu A, Basu R, Schwenk WF, Rizza RA. Lack of suppression of glucagon contributes to postprandial hyperglycemia in subjects with type 2 diabetes mellitus. J Clin Endocrinol Metab 2000;85:4053–4059

4. Baron AD, Schaeffer L, Shragg P, Kolterman OG. Role of hyperglucagonemia in maintenance of increased rates of hepatic glucose output in type II diabetics.

Diabetes 1987;36:274–283

5. Dunning BE, Gerich JE. The role of alpha-cell dysregulation in fasting and postprandial hyperglycemia in type 2 diabetes and therapeutic implications.

Endocr Rev 2007;28:253–283

6. Unger RH, Cherrington AD. Glucagonocentric restructuring of diabetes:

a pathophysiologic and therapeutic makeover. J Clin Invest 2012;122:4–12 7. Ehses JA, Perren A, Eppler E, et al. Increased number of islet-associated macrophages in type 2 diabetes. Diabetes 2007;56:2356–2370

8. Richardson SJ, Willcox A, Bone AJ, Foulis AK, Morgan NG. Islet-associated macrophages in type 2 diabetes. Diabetologia 2009;52:1686–1688

9. Böni-Schnetzler M, Thorne J, Parnaud G, et al. Increased interleukin (IL)- 1beta messenger ribonucleic acid expression in beta -cells of individuals with type 2 diabetes and regulation of IL-1beta in human islets by glucose and au- tostimulation. J Clin Endocrinol Metab 2008;93:4065–4074

10. Igoillo-Esteve M, Marselli L, Cunha DA, et al. Palmitate induces a pro- inﬂammatory response in human pancreatic islets that mimics CCL2 expression by beta cells in type 2 diabetes. Diabetologia 2010;53:1395–1405 11. Lee E, Ryu GR, Ko SH, et al. Antioxidant treatment may protect pancreatic beta cells through the attenuation of isletﬁbrosis in an animal model of type 2 diabetes. Biochem Biophys Res Commun 2011;414:397–402

12. Homo-Delarche F, Calderari S, Irminger JC, et al. Islet inﬂammation and ﬁbrosis in a spontaneous model of type 2 diabetes, the GK rat. Diabetes 2006;55:

1625–1633

13. Eguchi K, Manabe I, Oishi-Tanaka Y, et al. Saturated fatty acid and TLR signaling linkbcell dysfunction and islet inﬂammation. Cell Metab 2012;15:518– 533

14. Jones HB, Nugent D, Jenkins R. Variation in characteristics of islets of Langerhans in insulin-resistant, diabetic and non-diabetic-rat strains. Int J Exp Pathol 2010;91:288–301

15. Masters SL, Dunne A, Subramanian SL, et al. Activation of the NLRP3 in- ﬂammasome by islet amyloid polypeptide provides a mechanism for enhanced IL-1bin type 2 diabetes. Nat Immunol 2010;11:897–904

16. Youm YH, Adijiang A, Vandanmagsar B, Burk D, Ravussin A, Dixit VD.

Elimination of the NLRP3-ASC inﬂammasome protects against chronic obesity- induced pancreatic damage. Endocrinology 2011;152:4039–4045

17. Chentouf M, Dubois G, Jahannaut C, et al. Excessive food intake, obesity and inﬂammation process in Zucker fa/fa rat pancreatic islets. PLoS ONE 2011;6:

e22954

18. Nicol LE, Grant WF, Comstock SM, et al. Pancreatic inﬂammation and increased islet macrophages in insulin-resistant juvenile primates [published correction appears in J Endocrinol 2013;218:X1]. J Endocrinol 2013;217:207– 213

19. Rohrer S, Menge BA, Grüber L, et al. Impaired crosstalk between pulsatile insulin and glucagon secretion in prediabetic individuals. J Clin Endocrinol Metab 2012;97:E791–E795

20. Ehses JA, Lacraz G, Giroix MH, et al. IL-1 antagonism reduces hyperglycemia and tissue inﬂammation in the type 2 diabetic GK rat. Proc Natl Acad Sci U S A 2009;106:13998–14003

21. Ellingsgaard H, Ehses JA, Hammar EB, et al. Interleukin-6 regulates pancreatic alpha-cell mass expansion. Proc Natl Acad Sci U S A 2008;105:13163– 13168

22. Ellingsgaard H, Hauselmann I, Schuler B, et al. Interleukin-6 enhances insulin secretion by increasing glucagon-like peptide-1 secretion from L cells and alpha cells. Nat Med 2011;17:1481–1489

23. Scheller J, Chalaris A, Schmidt-Arras D, Rose-John S. The pro- and anti- inﬂammatory properties of the cytokine interleukin-6. Biochim Biophys Acta 2011;1813:878–888

24. Mahdi T, Hänzelmann S, Salehi A, et al. Secreted frizzled-related protein 4 reduces insulin secretion and is overexpressed in type 2 diabetes. Cell Metab 2012;16:625–633

(13)

25. Betz UA, Bloch W, van den Broek M, et al. Postnatally induced inactivation of gp130 in mice results in neurological, cardiac, hematopoietic, immunological, hepatic, and pulmonary defects. J Exp Med 1998;188:1955–1965

26. Herrera PL. Adult insulin- and glucagon-producing cells differentiate from two independent cell lineages. Development 2000;127:2317–2322

27. Westwell-Roper C, Dai DL, Soukhatcheva G, et al. IL-1 blockade attenuates islet amyloid polypeptide-induced proinﬂammatory cytokine release and pancreatic islet graft dysfunction. J Immunol 2011;187:2755–2765

28. Ladefoged M, Buschard K, Hansen AM. Increased expression of toll-like receptor 4 and inﬂammatory cytokines, interleukin-6 in particular, in islets from a mouse model of obesity and type 2 diabetes. APMIS 2013;121:531–538 29. Ribaux P, Ehses JA, Lin-Marq N, et al. Induction of CXCL1 by extracellular matrix and autocrine enhancement by interleukin-1 in rat pancreatic beta-cells.

Endocrinology 2007;148:5582–5590

30. Peters M, Blinn G, Solem F, Fischer M, Meyer zum Büschenfelde KH, Rose- John S. In vivo and in vitro activities of the gp130-stimulating designer cytokine Hyper-IL-6. J Immunol 1998;161:3575–3581

31. Atreya R, Mudter J, Finotto S, et al. Blockade of interleukin 6 trans signaling suppresses T-cell resistance against apoptosis in chronic intestinal inﬂammation:

evidence in crohn disease and experimental colitis in vivo. Nat Med 2000;6:583–588 32. Voshol PJ, Jong MC, Dahlmans VE, et al. In muscle-speciﬁc lipoprotein lipase-overexpressing mice, muscle triglyceride content is increased without inhibition of insulin-stimulated whole-body and muscle-speciﬁc glucose uptake.

Diabetes 2001;50:2585–2590

33. Huynh FK, Levi J, Denroche HC, et al. Disruption of hepatic leptin signaling protects mice from age- and diet-related glucose intolerance. Diabetes 2010;59:

3032–3040

34. Levi J, Gray SL, Speck M, et al. Acute disruption of leptin signaling in vivo leads to increased insulin levels and insulin resistance. Endocrinology 2011;152:

3385–3395

35. Ehses JA, Meier DT, Wueest S, et al. Toll-like receptor 2-deﬁcient mice are protected from insulin resistance and beta cell dysfunction induced by a high-fat diet. Diabetologia 2010;53:1795–1806

36. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;

25:402–408

37. Jones SA, Scheller J, Rose-John S. Therapeutic strategies for the clinical blockade of IL-6/gp130 signaling. J Clin Invest 2011;121:3375–3383 38. Lu J, Herrera PL, Carreira C, et al. Alpha cell-speciﬁc Men1 ablation triggers the transdifferentiation of glucagon-expressing cells and insulinoma development. Gastroenterology 2010;138:1954–1965

39. Müller WA, Faloona GR, Aguilar-Parada E, Unger RH. Abnormal alpha-cell function in diabetes. Response to carbohydrate and protein ingestion. N Engl J Med 1970;283:109–115

40. Kawamori D, Welters HJ, Kulkarni RN. Molecular pathways underlying the pathogenesis of pancreatic alpha-cell dysfunction. Adv Exp Med Biol 2010;654:

421–445

41. Johansson U, Olsson A, Gabrielsson S, Nilsson B, Korsgren O. Inﬂammatory mediators expressed in human islets of Langerhans: implications for islet transplantation. Biochem Biophys Res Commun 2003;308:474–479

42. Westwell-Roper CY, Ehses JA, Verchere CB. Resident macrophages mediate islet amyloid polypeptide-induced islet IL-1bproduction and beta cell dysfunction. Diabetes. 12 November 2013 [Epub ahead of print]

43. Patel DG. Role of parasympathetic nervous system in glucagon response to insulin-induced hypoglycemia in normal and diabetic rats. Metabolism 1984;33:

1123–1127

44. Lamy CM, Sanno H, Labouèbe G, et al. Hypoglycemia-Activated GLUT2 Neurons of the Nucleus Tractus Solitarius Stimulate Vagal Activity and Glucagon Secretion. Cell Metab 2014;19:527–538

45. Tweedell A, Mulligan KX, Martel JE, Chueh FY, Santomango T, McGuinness OP. Metabolic response to endotoxin in vivo in the conscious mouse: role of interleukin-6. Metabolism 2011;60:92–98

46. Lundblad TM, Coldren A, Brissova M, Chukwuma V, Ansari T, McGuinness OP. Interleukin-6 ampliﬁes adrenergic-dependent glucagon secretion. Diabetes 2012;61:A477

47. Fernández-Millán E, de Toro-Martín J, Lizárraga-Mollinedo E, Escrivá F, Álvarez C. Role of endogenous IL-6 in the neonatal expansion and functionality of Wistar rat pancreatic alpha cells. Diabetologia 2013;56:1098–1107

48. Habener JF, Stanojevic V. Alpha cells come of age. Trends Endocrinol Metab 2013;24:153–163

49. Spijker HS, Ravelli RB, Mommaas-Kienhuis AM, et al. Conversion of mature humanb-cells into glucagon-producinga-cells. Diabetes 2013;62:2471–2480