The Myokine Irisin Is Released in Response to Saturated Fatty Acids and Promotes Pancreatic B-Cell Survival and Insulin Secretion

Diabetes Volume 66, November 2017 2849

The Myokine Irisin Is Released in Response to Saturated Fatty Acids and Promotes Pancreatic b-Cell Survival and Insulin Secretion

Annalisa Natalicchio,1 Nicola Marrano,1 Giuseppina Biondi,1 Rosaria Spagnuolo,1 Rossella Labarbuta,1 Immacolata Porreca,2 Angelo Cignarelli,1 Marco Bugliani,3 Piero Marchetti,3 Sebastio Perrini,1 Luigi Laviola,1 and Francesco Giorgino1

Diabetes 2017;66:2849–2856 | https://doi.org/10.2337/db17-0002

This study explored the role of irisin as a new pancreatic correlate with markers of insulin resistance in subjects with- b-cell secretagogue and survival factor and its potential out diabetes, while they are reduced in overt type 2 diabetes role in the communication between skeletal muscle and (2,3). The biological functions of irisin include effects on pancreatic b-cells under lipotoxic conditions. Recombi- multiple tissues (1,4–7). Irisin also improves glucose toler- nant irisin stimulated insulin biosynthesis and glucose- ance and insulin sensitivity and increases energy expendi- stimulated insulin secretion (GSIS) in a PKA-dependent ture in both obese and diabetic mice (8). Irisin increases STUDIES ISLET – manner and prevented saturated fatty acid induced apo- proliferation of insulin-secreting INS-1E cells and protects b ptosis in human and rat pancreatic -cells, as well as in them from high glucose–induced apoptosis (9). human and murine pancreatic islets, via AKT/BCL2 signal- If chronically in excess, saturated free fatty acids (FFAs) ing. Treatment of myotubes with 0.5 mmol/L palmitate for can reduce insulin biosynthesis (10) and secretion (11) and 4 h, but not with oleate, promoted an increase in irisin promote b-cell apoptosis (12,13). Excessively high plasma release in the culture medium. Moreover, increased serum levels of FFAs, particularly long-chain saturated FFAs, also levels of irisin were observed in mice fed with a high-fat diet. Mouse serum rich in irisin and the conditioned medium foster insulin resistance (14). Palmitate can directly impair from myotubes exposed to palmitate for 4 h signiﬁcantly insulin signaling in skeletal muscle cells (15) and alter the reduced apoptosis of murine pancreatic islets and insulin- expression of some myokines (16). secreting INS-1E cells, respectively, and this was abrogated Skeletal muscle, the major site of insulin-stimulated glucose ﬁ in the presence of an irisin-neutralizing antibody. Finally, disposal (17), has recently been identi ed as a secretory organ in vivo administration of irisin improved GSIS and increased able to release myokines. It is thus plausible that this organ b-cell proliferation. In conclusion, irisin can promote might interact with the endocrine pancreas by releasing myo- b-cell survival and enhance GSIS and may thus participate kines that adjust insulin secretion to the actual insulin need in the communication between skeletal muscle and b-cells for appropriate peripheral glucose utilization (18,19). In this under conditions of excess saturated fatty acids. study, we have investigated the effects of irisin on pancreatic b-cells and the release of this myokine by skeletal muscle cells.

Irisin is a newly discovered muscle-derived hormone, pro- RESEARCH DESIGN AND METHODS duced by cleavage of the membrane protein ﬁbronectin type Animals III domain–containing protein 5 (FNDC5) and proposed to Animal experimentations were conducted after approval by bridge exercise with metabolic homeostasis (1). Irisin levels the Ethics Committee (CESA) of Istituto di Ricerche Genetiche

1Section of Internal Medicine, Endocrinology, Andrology and Metabolic Diseases, This article contains Supplementary Data online at http://diabetes Department of Emergency and Organ Transplantation, University of Bari Aldo .diabetesjournals.org/lookup/suppl/doi:10.2337/db17-0002/-/DC1. Moro, Bari, Italy A.N. and N.M. contributed equally to the work. 2Istituto di Ricerche Genetiche “Gaetano Salvatore” (IRGS), Biogem, Ariano Irpino, © 2017 by the American Diabetes Association. Readers may use this article as Italy long as the work is properly cited, the use is educational and not for proﬁt, and the 3Endocrinology and Metabolism of Transplantation, Azienda Ospedaliero Universi- work is not altered. More information is available at http://www.diabetesjournals taria Pisana, Pisa, Italy .org/content/license. Corresponding author: Francesco Giorgino, [email protected]. Received 1 January 2017 and accepted 12 July 2017. 2850 Irisin Effects on Pancreatic b-Cells Diabetes Volume 66, November 2017

Figure 1—Effects of irisin on palmitate-induced apoptosis in pancreatic b-cells. Rat (INS-1E) and human (1.1B4) b-cells and murine and human islets were treated or not with recombinant irisin (10–100 nmol/L) for 24 h and then incubated with or without 0.5 mmol/L palmitate for 24 h. A: Apoptosis was measured by assessing caspase-3 cleavage in INS-1E insulin-secreting cells (n = 11 independent experiments). Total caspase-3 is also shown. Densitometric analysis of the related bands was expressed as relative optical density, corrected using b-actin as a loading control and normalized against untreated control. B: Apoptosis was evaluated by measuring cytoplasmic oligonucleosomes with an ELISA assay (data expressed as percentage of untreated control) in INS-1E cells (n = 7 independent experiments), murine islets (n = 3 independent experiments), 1.1B4 human pancreatic cells (n = 9 independent experiments), and human islets (n = 3 independent experiments). C: Apoptosis was evaluated by a TUNEL assay. Left: Representative confocal images of murine islets (islets from n = 3 different isolations for each condition) treated as described above and subsequently immunostained for insulin (green) and apoptosis by TUNEL (red). Nuclear staining was performed with TO-PRO-3 (blue). Right: TUNEL-positive cells normalized for total cell number. D: AKT phosphorylation was measured by immunoblotting and quantiﬁed by densitometry (n = 6 independent experiments) in INS-1E cells treated with 100 nmol/L recombinant irisin for different times. Densitometric analysis of the related bands was expressed as relative optical density, corrected using total ΑΚΤ as a loading control and normalized against untreated control. E: BCL2 and BAX protein content was measured by immunoblotting and quantiﬁed by densitometry in INS-1E cells. The BCL2-to-BAX ratio was also calculated (n = 8 independent experiments). F and G: INS-1E cells were treated or not with diabetes.diabetesjournals.org Natalicchio and Associates 2851

“Gaetano Salvatore” (IRGS), Biogem, Italy (internal ID 0907) TMR red (Roche Biochemicals, Indianapolis, IN) and by in accordance with the Guide for the Care and Use of Lab- assessing caspase-3 cleavage. Proliferation was assessed with oratory Animals (National Institutes of Health publication the BrdU ELISA kit (colorimetric) (Abcam, Cambridge, U.K.). no. 85-23, revised 1985) and regulations of Italy and the EU. Irisin and Insulin Assays CD-1 and C57BL/6 mice were purchased from Charles River Irisin concentrations were measured using an irisin-competitive Laboratories (Calco, Lecco, Italy). The high-fat diet (HFD) (60% ELISA kit (AdipoGen SA, Liestal, Switzerland). Insulin levels of energy from fat) was from Mucedola (Settimo Milanese, were measured by specific ELISA (Mercodia AB, Sylveniusgatan, Milan, Italy) (formulation given in Supplementary Table 1). Uppsala, Sweden). Glucose-stimulated insulin secretion In Vivo Irisin Administration (GSIS) was performed as previously described (13). More Eighteen male 6-week-old C57BL/6 mice were randomized details are given in Supplementary Data. intothreegroups:6micewereimmediatelysacrificed; the remaining mice were intraperitoneally injected daily with irisin Statistical Analyses t (0.5 mg/gbodywt)orvehicle(PBS)for14days.Intraperi- Data were analyzed by the Student test or ANOVA, as 6 toneal glucose tolerance tests were carried out, and then appropriate, and are presented as mean SD. Statistical fi P , animals were sacrificed.Pancreas,serum,andplasmawere signi cance was set at value 0.05. collected. RESULTS Pancreatic Islets and Cell Culture Mouse and human islets were isolated and cultured as Irisin Prevents Palmitate-Induced b-Cell Apoptosis previously described (20,21). More details on cell lines and and Induces Insulin Secretion culture conditions are given in Supplementary Data. INS-1E cells treated with 0.5 mmol/L palmitate for 24 h Palmitate and oleate were obtained from Sigma-Aldrich showed increased caspase-3 cleavage, as expected. However, (St. Louis, MO) and prepared as previously reported (13). pretreatment with recombinant irisin for 24 h reduced this All controls were treated with the same volume of 10% response in a dose-dependent manner. A similar effect was fatty acid–free BSA solution without palmitate or oleate. observed in human pancreatic 1.1B4 cells and in murine A B fl Cells were preincubated with recombinant irisin and human pancreatic islets (Fig. 1 and ). Immuno uo- fi b (AdipoGen SA, Liestal, Switzerland) for the indicated rescence of murine islets con rmed that -cells showed doses and times. The chemical inhibitors AKT Inhibitor increased apoptosis in response to palmitate and that this C VIII (2.5 mmol/L) or H-89 (5 mmol/L) (Calbiochem; Merck was counteracted by irisin (Fig. 1 ). Since treatment of Millipore, Darmstadt, Germany)wereadded30minor INS-1E cells with irisin increased AKT phosphorylation (Fig. D 60 min before irisin. 1 ), which is known to upregulate B-cell lymphoma 2 (BCL2) expression (23), we investigated whether irisin pre- Immunoblotting vented palmitate-induced apoptosis through a mechanism Cells lysates were obtained and analyzed by immunoblot- involving BCL2 and BCL2-associated X protein (BAX)—key ting as previously described (13). A list of the antibodies apoptosis regulatory proteins. While palmitate increased BAX used is shown in Supplementary Table 2. expression and, consequently, reduced the BCL2-to-BAX ra- Gene Expression Analysis by Quantitative RT-PCR tio, pretreatment with irisin increased BCL2 expression and RNA isolation, cDNA synthesis, and mRNA quantitation restored the BCL2-to-BAX ratio (Fig. 1E). When INS-1E cells were carried out as previously described (13). A list of the were pretreated with the AKT Inhibitor VIII, irisin-mediated primer sequences used is shown in Supplementary Table 3. AKT phosphorylation was abrogated, and augmentation of BCL2 protein expression and inhibition of palmitate- Immunofluorescence and Immunohistochemistry F G fl induced apoptosis were prevented (Fig. 1 and ). Immuno uorescence and immunohistochemistry were car- Exposure of human and murine islets and rat INS-1E cells ried out as previously described (13,22). More details are to irisin also increased insulin (Ins) mRNA levels and insulin given in Supplementary Data. content and augmented GSIS (Fig. 2A–K). These responses Apoptosis and BrdU Assays were dependent on PKA activation, since they were abro- Apoptosis was measured by using the Cell Death Detection gated by the PKA inhibitor H-89, as was the irisin-induced ELISAPLUS Kit and In Situ Cell Death Detection Kit and CREB phosphorylation at 25 mmol/L glucose (Fig. 2H–L).

100 nmol/L recombinant irisin for 24 h in the presence or absence of the AKT Inhibitor VIII (2.5 mmol/L) for 60 min and then incubated with or without 0.5 mmol/L palmitate for 24 h. F: AKT phosphorylation (P-AKT) and BCL2 protein content were measured by immunoblotting and quantiﬁed by densitometry (n = 8 independent experiments). Densitometric analysis of the related bands was expressed as relative optical density, corrected using b-actin as loading controls and normalized against untreated control. G: Cell death was evaluated by measuring cytoplasmic oligonucleosomes with an ELISA assay and expressed as percentage of untreated control. Data are expressed as mean 6 SD. *P < 0.05 vs. no palmitate; †P < 0.05 vs. palmitate; ‡P < 0.05 vs. basal; #P < 0.05 vs. control without AKT Inhibitor VIII. Palm, palmitate. 2852 Irisin Effects on Pancreatic b-Cells Diabetes Volume 66, November 2017

Figure 2—Effects of irisin on insulin biosynthesis and GSIS. A–C: Human islets were incubated with 100 nmol/L recombinant irisin for 60 min. A: Insulin content was evaluated by an ELISA assay (n = 3 independent experiments). B: GSIS was measured after 60 min at 2.8 mmol/L glucose (white bars = basal secretion) followed by 60 min at 16.8 mmol/L glucose (dark bars = stimulated secretion) (n = 3 independent experiments). Secretion is normalized to protein concentration and is expressed as a percentage of untreated control. C: Fold increase of 16.8 mmol/L vs. 2.8 mmol/L GSIS. D–G: Murine islets were incubated with 100 nmol/L recombinant irisin for 60 min. D: Ins gene expression was evaluated by quantitative RT-PCR analysis and normalized to 18S gene expression (n = 4 independent experiments). E: Insulin content was evaluated by immunoﬂuorescence. Left: Representative confocal images of murine islets (islets from n = 3 different isolations for each condition) immunostained for insulin (green). Nuclear staining was performed with TO-PRO-3 (blue). The intensity of the insulin signal was measured using the ImageJ software and normalized against islets area (right). F: GSIS was measured after 60 min at 3 mmol/L glucose (white bars = basal secretion) followed by 60 min at 25 mmol/L glucose (dark bars = stimulated secretion) (n = 3 independent experiments). Secretion is normalized to protein concentration and is expressed as a percentage of untreated control. G: Fold increase of 25 mmol/L vs. 3 mmol/L GSIS. H–K: INS-1E cells were treated with or without 5 mmol/L H-89 for 30 min before incubation with 100 nmol/L recombinant irisin for 60 min. H: Ins gene expression was evaluated by quantitative RT-PCR analysis normalized to Gusb gene expression (n = 5 independent experiments). I: Insulin content was evaluated by an ELISA assay (n = 9 independent experiments). J: GSIS was measured after 60 min at 3 mmol/L glucose (white bars = basal diabetes.diabetesjournals.org Natalicchio and Associates 2853

Moreover, irisin increased proliferation of INS-1E cells and enhanced 90 min after intraperitoneal glucose injec- human pancreatic 1.1B4 cells (Supplementary Fig. 1). tion, and this resulted in a faster decrease of glucose levels (Fig. 4B). In vivo dosing of irisin significantly aug- Irisin Is Secreted in Response to Saturated FFAs mented the b-cell area and slightly reduced the a-cell area and Promotes b-Cell Survival within pancreatic islets, thus increasing the b-cell–to–a-cell Treatment of rat myotubes with palmitate for 4 h resulted ratio (Fig. 4C). Moreover, insulin content and b-cell pro- in increased FNDC5 mRNA levels and protein content and liferation in the islets were markedly increased after irisin caused a threefold higher release of irisin in the culture administration (Fig. 4C and D). medium compared with untreated myotubes. By contrast, in myotubes treated with palmitate for 24 h, FNDC5 mRNA levels and protein content were significantly reduced and DISCUSSION irisin release in the culture medium was unchanged (Fig. Here, we demonstrate for the first time that recombinant 3A–C). Validation of the ELISA kit and irisin antibody is irisin protects human and rodent b-cells and islets from shown in Supplementary Fig. 2. Oleate, a monounsaturated palmitate-induced apoptosis through a mechanism involv- FFA, did not stimulate irisin release (Supplementary Fig. 3). ing AKT/BCL2 signaling (Fig. 1). Similarly, in the human Human myotubes showed a similar regulation of irisin re- umbilical vein endothelial cells and INS-1E cells irisin par- lease by palmitate (Fig. 3E), even though FNDC5 mRNA tially suppressed high glucose–induced apoptosis by affect- levels were unchanged (Fig. 3D). Furthermore, treatment ing BCL2, BAX, and caspase expression (7,9). In vivo with a small interfering RNA of FNDC5 prevented the administration of irisin also improves GSIS and increases b palmitate-induced increase of FNDC5 protein expression, insulin content and -cell mass and proliferation (Fig. 4). while irisin release was unaffected (Supplementary Fig. 4). Furthermore, irisin promotes insulin biosynthesis and se- fi Plasma irisin levels were also measured in vivo in CD-1 cretion in a PKA-dependent manner (Fig. 2). The bene cial b mice fed with a standard diet (SD) or HFD for 63 days. The effects of irisin on -cells appear to be glucose dependent HFD caused a rapid increase of blood irisin concentrations, and comparable with those of glucagon-like peptide 1 and which persisted at later times (Fig. 3F). Interestingly, mu- its analogs (24). Like glucagon-like peptide 1, irisin induces rine pancreatic islets cultured in the presence of serum cyclic AMP generation (data not shown) and activates PKA fi from mice fed with an HFD containing high levels of irisin and AKT, suggesting the existence of a speci c receptor that fi were protected from palmitate-induced apoptosis; the pro- has not been yet identi ed. tective effect was attenuated in the presence of a neutral- Short-term treatment of rat myotubes with palmitate izing antibody against irisin (Fig. 3G), which inhibited the results in increased FNDC5 mRNA levels and protein irisin-induced increase of UCP-1 expression in 3T3-L1 adi- content and raises irisin levels in the culture medium threefold compared with control (Fig. 3A–C). The mech- pocytes (Supplementary Fig. 5). Moreover, culture medium anism by which saturated FFAs induce irisin release from was collected from L6 myotubes exposed to 0.5 mmol/L skeletal muscle cells is still unclear and may involve the palmitate for 4 h (PM4h) or not (PM0h), and the effects activation of FNDC5 cleavage by specificprotease(s).In on b-cell apoptosis were evaluated. Samples of L6 myotube- human myotubes, higher irisin release in the culture free culture medium with (CTR2) or without (CTR1) pal- medium in response to palmitate occurred in the ab- mitate were used as additional controls. CTR2 increased sence of changes in FNDC5 mRNA levels (Fig. 3D and E). INS-1E apoptosis, as expected, while apoptosis was sig- Furthermore, in rat myotubes, FNDC5 knockdown pre- nificantly reduced when INS-1E cells were exposed to vented the palmitate-induced increase of FNDC5 protein PM4h, and this effect was abrogated by the neutralizing H expression without affecting irisin release (Supplemen- antibody directed against irisin (Fig. 3 ). tary Fig. 4). Altogether, these results suggest independent In Vivo Irisin Administration Improves Insulin Secretion regulation of FNDC5 expression and cleavage by FFAs. Iri- and Glucose Tolerance and Increases the Pancreatic sin release from myotubes is also influencedbythetypeof b-Cell Mass FFA, since it is not stimulated by oleate, a monounsaturated Intraperitoneal injection of irisin for 14 days in C57BL/6 FFA (Supplementary Fig. 3). mice resulted in a 1.5-fold increase in irisin plasma levels Mice receiving an HFD and gaining weight display an compared with vehicle (Fig. 4A). Furthermore, in irisin- early increase in plasma irisin levels that persists thereafter compared with vehicle-treated mice, insulin secretion was (Fig. 3F). This is in line with the observed positive correlation

secretion) followed by 60 min at 25 mmol/L glucose (dark bars = stimulated secretion) (n = 18 independent experiments). Secretion is normalized to protein concentration and is expressed as a percentage of untreated control. K:Foldincreaseof25mmol/Lvs.3mmol/LGSIS.L:INS-1E cells were cultured in medium at 25 mmol/L glucose and were treated with or without 5 mmol/L H-89 for 30 min before incubation with 100 nmol/L recombinant irisin for 10 min. CREB phosphorylation was measured by immunoblotting and quantiﬁed by densitometry. Densito- metric analysis of the related bands was expressed as relative optical density, corrected using total CREB as loading control and normalized against untreated control (n = 7 independent experiments). Data are expressed as mean 6 SD. §P < 0.05 vs. 2.8 or 3 mmol/L glucose; *P < 0.05 vs. basal; †P < 0.05 vs. irisin. 2854 Irisin Effects on Pancreatic b-Cells Diabetes Volume 66, November 2017

Figure 3—Effects of saturated FFAs on FNDC5 expression and irisin release by skeletal muscle cells and effects of secreted irisin on b-cell survival. L6 rat skeletal muscle cells were exposed to 0.5 mmol/L palmitate for 4 h or 24 h. A: Fndc5 gene expression was evaluated by quantitative RT-PCR analysis and normalized to Gusb gene expression (n = 6 independent experiments). B: FNDC5 protein expression was evaluated by immunoblotting and quantiﬁed by densitometry (n = 8 independent experiments). Densitometric analysis of the related bands was expressed as relative optical density, corrected using b-actin as loading control and normalized against the untreated control. C: Irisin secretion in the culture medium was measured by ELISA assay and was expressed as percentage of untreated control (n = 3 independent experiments). The irisin concentration in the conditioned medium of L6 cells treated with 0.5 mmol/L palmitate for 4 h was 2.1 nmol/L. D and E: Human skeletal muscle cells were exposed to 0.5 mmol/L palmitate for 4 h. D: FNDC5 gene expression was evaluated by quantitative RT-PCR analysis and normalized to 18S gene expression (n = 6 independent experiments). E: Irisin secretion in culture medium was measured by ELISA assay and was expressed as percentage of untreated control (n = 6 independent experiments). F: From weaning at the age of 5 weeks onward, CD-1 mice received an SD; then, they were randomized to either an HFD or the SD for an additional 63 days. Blood samples were collected from the tail vein. Plasma irisin levels in mice fed with HFD (n =15)orSD(n = 15) during the observation were measured by ELISA assay and were expressed as fold over baseline (left) or as nmol/L (right) (data from days 0, 1, 2, 4, 21, 28, and 35 in the HFD group). G: Murine islets were cultured for 24 h with or without serum from mouse fed with HFD as described above (collected on days 1, 2, 4, 21, 28, and 35) (n = 6 independent experiments), in presence or absence of a neutralizing antibody directed against irisin, and then incubated with or without 0.5 mmol/L palmitate for 24 h. Apoptosis was evaluated by measuring cytoplasmic oligonucleosomes with an ELISA assay (data expressed as percentage of untreated control) diabetes.diabetesjournals.org Natalicchio and Associates 2855

Figure 4—Effects of in vivo administration of irisin on insulin secretion, glucose tolerance, pancreatic islet structure, and b-cell proliferation in mice. Eighteen male 6-week-old C57BL/6 mice were randomized into three groups: 6 mice were immediately sacriﬁced, 6 mice were daily injected intraperitoneally with irisin (0.5 mg/g body wt) for 14 days, and 6 mice were treated with vehicle (PBS). A: Plasma irisin levels were measured by ELISA assay in mice injected with irisin (n = 6) or vehicle (n = 6) for 14 days. B: On day 14, an intraperitoneal glucose tolerance test was performed: mice were intraperitoneally injected with 20% glucose (2 g glucose/kg body wt), and serum and plasma were collected immediately before and 45 min and 90 min after glucose injection. Left: Serum insulin concentrations were evaluated by an ELISA assay. Right: Blood glucose levels were measured using a commercial glucometer. Mice injected with irisin, n = 6; mice injected with vehicle, n =6.C,left: Representative confocal images of pancreas sections (3 different areas from each section were analyzed) from mice treated for 14 days with irisin (0.5 mg/g body wt) (n = 6) or vehicle (n = 6). Sections were immunostained for insulin (green) and glucagon (red). Nuclear staining was with TO-PRO-3 (blue). Right: b-anda-Cell areas were measured as insulin- or glucagon-positive cell areas and normalized against total islet area. The ratio of b-cell to a-cell area was also calculated. Intensity of the insulin signal was measured using the ImageJ software and normalized against total islet area. D, left: Representative confocal images of pancreas sections (3 different areas from each section were analyzed) from mice treated as described above. Images were captured at 363 oil objective showing insulin (green), nuclear Ki-67 expression (red; arrows), and nuclear staining with TO-PRO-3 (blue). Right: Ki67/insulin double-positive cells normalized for total insulin-positive cells. Data are expressedas mean 6 SD. *P < 0.05 vs. vehicle. Results were similar in mice sacriﬁced at time 0 and vehicle-treated mice.

between serum irisin levels and BMI, adiposity markers, and irisin is estimated to be released by skeletal muscle (1). The HOMA of insulin resistance index in humans (3). Adipose availability of tissue-speciﬁc FNDC5 knockout mice will help tissue also secretes irisin (25); however, ;72% of circulating toaddressthisissue.

(n = 6 independent experiments). H: Samples of conditioned medium were obtained by culturing L6 rat myotubes for 0 h (PM0h) or 4 h (PM4h) with 0.5 mmol/L palmitate. Culture medium not exposed to L6 cells with (CTR2) or without (CTR1) 0.5 mmol/L palmitate was used as additional control. The conditioned medium was collected, transferred into sixwell plates containing INS-1E cells, in presence or absence of a neutralizing antibody directed against irisin, and left for 24 h. Cell death was evaluated by measuring cytoplasmic oligonucleosomes with an ELISA assay (data expressed as percentage of CTR1) (left) or by assessing caspase-3 cleavage (right) (n = 4 independent experiments). Total caspase-3 is also shown. Densitometric analysis of the related bands was expressed as relative optical density, corrected using b-actin as a loading control and normalized against CTR1. Data are expressed as mean 6 SD. *P < 0.05 vs. basal; #P < 0.05 vs. baseline; †P < 0.05 vs. SD; §P < 0.05 vs. palmitate; @P < 0.05 vs. no neutralizing antibody; aP < 0.05 vs. CTR1 and PM0h; bP < 0.05 vs. CTR2. Ab, neutralizing antibody against irisin; Palm, palmitate. 2856 Irisin Effects on Pancreatic b-Cells Diabetes Volume 66, November 2017

We could not target the actions of irisin on pancreatic 3. Park KH, Zaichenko L, Brinkoetter M, et al. Circulating irisin in relation to insulin b-cells in vivo because of lack of identification of the irisin resistance and the metabolic syndrome. J Clin Endocrinol Metab 2013;98:4899– receptor and difficulty in neutralizing circulating irisin in 4907 mice. However, the serum from HFD-fed mice, containing 4. Colaianni G, Cuscito C, Mongelli T, Oranger A, Mori G, Brunetti G, Colucci S, Cinti S, Grano M. Irisin enhances osteoblast differentiation in vitro. Int J Endocrinol high levels of irisin, protected murine pancreatic islets 2014; 2014 against palmitate-induced apoptosis, and this was largely 5. Moon H-S, Dincer F, Mantzoros CS. Pharmacological concentrations of irisin abrogated in the presence of a neutralizing antibody against increase cell proliferation without influencing markers of neurite outgrowth and G irisin (Fig. 3 ). This is in line with the observation that synaptogenesis in mouse H19-7 hippocampal cell lines. Metabolism 2013;62:1131–1136 myotube-derived irisin in the conditioned medium also pre- 6. Park MJ, Kim DI, Choi JH, Heo YR, Park SH. New role of irisin in hepatocytes: vented apoptosis of b-cells (PM4h) (Fig. 3H). However, in The protective effect of hepatic steatosis in vitro. Cell Signal 2015;27:1831–1839 contrast to recombinant irisin, PM4h did not enhance GSIS 7. Song H, Wu F, Zhang Y, et al. Irisin promotes human umbilical vein endothelial (Supplementary Fig. 6). Since irisin elicits distinct biological cell proliferation through the ERK signaling pathway and partly suppresses high effects at different doses (1,4–7), it is possible that higher glucose-induced apoptosis. PLoS One 2014;9:e110273 irisin concentrations than those in PM4h are required to 8. Zhang Y, Li R, Meng Y, et al. Irisin stimulates browning of white adipocytes enhance GSIS. Alternatively, other myokines in PM4h could through mitogen-activated protein kinase p38 MAP kinase and ERK MAP kinase signaling. Diabetes 2014;63:514–525 counteract the action of irisin on insulin secretion. 9. Liu S, Du F, Li X, et al. Effects and underlying mechanisms of irisin on the In conclusion, the myokine irisin has antiapoptotic b b b proliferation and apoptosis of pancreatic cells. PLoS One 2017;12:e0175498 actions on pancreatic -cells and stimulates -cell prolifer- 10. Poitout V, Hagman D, Stein R, Artner I, Robertson RP, Harmon JS. Regulation of ation, insulin biosynthesis, and secretion. Under conditions the insulin gene by glucose and fatty acids. J Nutr 2006;136:873–876 of excess saturated FFAs, irisin release from skeletal muscle 11. Giacca A, Xiao C, Oprescu AI, Carpentier AC, Lewis GF. Lipid-induced pan- may be promoted as an adaptive response, signaling to the creatic b-cell dysfunction: focus on in vivo studies. Am J Physiol Endocrinol Metab endocrine pancreas to compensate for the increased insulin 2011;300:E255–E262 resistance and initial deterioration of glucose tolerance. 12. Natalicchio A, Tortosa F, Labarbuta R, et al. Erratum to: the p66Shc redox adaptor protein is induced by saturated fatty acids and mediates lipotoxicity-induced apoptosis in pancreatic beta cells. Diabetologia 2015;58:2682 Acknowledgments. The authors thank Dr. Paola Pontrelli, Dr. Chiara Divella, 13. Natalicchio A, Labarbuta R, Tortosa F, et al. Exendin-4 protects pancreatic beta and Dr. Michele Lauriero (Nephrology, Dialysis and Transplantation Unit, Department cells from palmitate-induced apoptosis by interfering with GPR40 and the MKK4/7 – of Emergency and Organ Transplantation, University of Bari Aldo Moro, Bari, Italy) for stress kinase signalling pathway. Diabetologia 2013;56:2456 2466 technical assistance in immunohistochemistry experiments and confocal images 14. Boden G. Role of fatty acids in the pathogenesis of insulin resistance and – acquisition. NIDDM. Diabetes 1997;46:3 10 Funding. This work is supported by a grant from Ager - Agroalimentare e ricerca 15. Yu C, Chen Y, Cline GW, et al. Mechanism by which fatty acids inhibit insulin (Claims of Olive oil to iMProvE The market ValuE of the product [COMPETITIVE]) to activation of insulin receptor substrate-1 (IRS-1)-associated phosphatidylinositol – F.G. 3-kinase activity in muscle. J Biol Chem 2002;277:50230 50236 Duality of Interest. A.N., N.M., and F.G. are named inventors of a pending 16. Yang M, Wei D, Mo C, et al. Saturated fatty acid palmitate-induced insulin patent application related to the work described. F.G. has received grant support resistance is accompanied with myotube loss and the impaired expression of health fi from Takeda, Eli Lilly, and LifeScan. L.L. is a consultant for and has received lecture bene t myokine genes in C2C12 myotubes. Lipids Health Dis 2013;12:104 fees from AstraZeneca, Eli Lilly, Boehringer Ingelheim, Takeda, Sanofi, Novo Nordisk, 17. Baron AD, Brechtel G, Wallace P, Edelman SV. Rates and tissue sites of Roche, Medtronic, and Jannsen. F.G. is a consultant for and has received lecture fees non-insulin- and insulin-mediated glucose uptake in humans. Am J Physiol 1988; – from AstraZeneca, Eli Lilly, Novo Nordisk, Roche, Sanofi, and Takeda. No other 255:E769 E774 potential conflicts of interest relevant to this article were reported. 18. Pedersen BK, Steensberg A, Fischer C, et al. Searching for the exercise factor: – Author Contributions. A.N., N.M., and F.G. designed the study. G.B. and is IL-6 a candidate? J Muscle Res Cell Motil 2003;24:113 119 b R.S. acquired data. N.M., G.B., R.L., and I.P. performed experiments and data 19. Maedler K, Schumann DM, Sauter N, et al. Low concentration of interleukin-1 b analysis. A.N., N.M., and G.B. performed data analysis and interpretation. G.B., R.S., induces FLICE-inhibitory protein-mediated -cell proliferation in human pancreatic – R.L., A.C., and M.B. participated in interpretation and discussion of data. A.N., N.M., islets. Diabetes 2006;55:2713 2722 and F.G. wrote the manuscript. P.M., S.P., and L.L. critically reviewed the manuscript 20. Li D-S, Yuan Y-H, Tu H-J, Liang Q-L, Dai L-J. A protocol for islet isolation from – for intellectual content. All the authors approved the final version of the manuscript. mouse pancreas. Nat Protoc 2009;4:1649 1652 F.G. is the guarantor of this work and, as such, had full access to all the data in the 21. Lupi R, Del Guerra S, Fierabracci V, et al. Lipotoxicity in human pancreatic islets – study and takes responsibility for the integrity of the data and the accuracy of the and the protective effect of metformin. Diabetes 2002;51(Suppl. 1):S134 S137 data analysis. 22. Takeda Y, Fujita Y, Honjo J, et al. Reduction of both beta cell death and alpha Prior Presentation. Parts of this study were presented in abstract form at the cell proliferation by dipeptidyl peptidase-4 inhibition in a streptozotocin-induced – 76th Scientific Sessions of the American Diabetes Association, New Orleans, LA, 10– model of diabetes in mice. Diabetologia 2012;55:404 412 14 June 2016, and at the 52nd Annual Meeting of the European Association for the 23. Pugazhenthi S, Nesterova A, Sable C, et al. Akt/protein kinase B up-regulates Study of Diabetes, Munich, Germany, 12–16 September 2016. Bcl-2 expression through cAMP-response element-binding protein. J Biol Chem 2000;275:10761–10766 References 24. Tudurí E, López M, Diéguez C, Nadal A, Nogueiras R. Glucagon-like peptide 1. Boström P, Wu J, Jedrychowski MP, et al. A PGC1-a-dependent myokine that 1 analogs and their effects on pancreatic islets. Trends Endocrinol Metab 2016;27: drives brown-fat-like development of white fat and thermogenesis. Nature 2012;481: 304–318 463–468 25. Moreno-Navarrete JM, Ortega F, Serrano M, et al. Irisin is expressed and 2. Liu JJ, Wong MD, Toy WC, et al. Lower circulating irisin is associated with produced by human muscle and adipose tissue in association with obesity and type 2 diabetes mellitus. J Diabetes Complications 2013;27:365–369 insulin resistance. J Clin Endocrinol Metab 2013;98:E769–E778