Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Inhibition of DPP-4 reduces acute mortality after myocardial infarction with restoration of autophagic response in type 2 diabetic rats

Inhibition of DPP-4 reduces acute mortality after myocardial infarction with restoration of... Background: Type 2 diabetes mellitus ( T2DM) worsens the outcome after myocardial infarction (MI). Here, we hypothesized that inhibition of dipeptidyl peptidase-4 (DPP-4) improves survival after MI in T2DM by modifying autophagy in the non-infarcted region of the heart. Methods and results: Under baseline conditions, there was no significant difference between levels of myocardial autophagy marker proteins in OLETF, a rat model of T2DM, and in LETO, a non-diabetic control. However, in contrast to the response in LETO, LC3-II protein and LC3-positive autophagosomes in the non-infarcted region of the myo- cardium were not increased after MI in OLETF. The altered autophagic response in OLETF was associated with lack of AMPK/ULK-1 activation, attenuated response of Akt/mTOR/S6 signaling and increased Beclin-1–Bcl-2 interaction after MI. Treatment with vildagliptin (10 mg/kg/day s.c.), a DPP-4 inhibitor, suppressed Beclin-1–Bcl-2 interaction and increased both LC3-II protein level and autophagosomes in the non-infarcted region in OLETF, though it did not normalize AMPK/ULK-1 or mTOR/S6 signaling. Plasma insulin level, but not glucose level, was significantly reduced by vildagliptin at the dose used in this study. Survival rate at 48 h after MI was significantly lower in OLETF than in LETO (32 vs. 82%), despite similar infarct sizes. Vildagliptin improved the survival rate in OLETF to 80%, the benefit of which was abrogated by chloroquine, an autophagy inhibitor. Conclusions: The results indicate that vildagliptin reduces T2DM-induced increase in post-MI acute mortality pos- sibly by restoring the autophagic response through attenuation of Bcl-2-Beclin-1 interaction. Keywords: Type 2 diabetes, Autophagy, DPP-4 inhibitor, Myocardial infarction, Mortality of the myocardium, often called “diabetic cardiomyo- Background pathy” [3]. Augmentation of contractile function in the Diabetes mellitus (DM) is associated with poor outcome non-infarcted region is crucial in acute compensation for after acute myocardial infarction (MI) even in the era of the lost contraction in the infarct region. u Th s, pre-exist - reperfusion therapy [1, 2]. The poor outcome after MI in patients with DM has been explained by extensive ather ing contractile dysfunction in patients with DM, if any, potentially compromises such a compensatory response osclerotic lesions, increased myocardial susceptibility to after MI and increases mortality. Recently, we confirmed ischemia/reperfusion injury, and contractile dysfunction DM-induced increase in acute mortality after MI in an animal model of obese type 2 DM (T2DM), Otsuka- *Correspondence: [email protected] 1 Long-Evans-Tokushima Fatty rats (OLETF). The mortal - Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, South-1, West-16, Chuo-ku, ity rate at 48 h after MI was significantly higher in OLETF Sapporo 060-8543, Japan due to lethal heart failure than in non-diabetic control Full list of author information is available at the end of the article © 2015 Murase et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 2 of 16 rats, Long-Evans-Tokushima-Otsuka rats (LETO), while To examine the hypotheses, we used OLETF as a model infarct sizes were similar in OLETF and LETO [4]. Inter- of T2DM in the present study as in previous studies estingly, OLETF had preserved ventricular contractility [4, 33, 34]. Results of the experiments showed that a with mildly impaired relaxation under baseline condi- DPP-4 inhibitor, vildagliptin, significantly improved tions [4]. The findings indicate that the altered response survival after MI in OLETF and that the protective of the non-infarcted myocardium, rather than base- effect of vildagliptin was closely associated with resto - line ventricular contractility, contributes to the T2DM- ration of the autophagic response in the non-infarcted induced increase in acute mortality after MI. myocardium. Autophagy is a cellular process of lysosome-medi- ated degradation of cytoplasmic components or dam- Methods aged organelles in response to cellular stress [5–13]. It Animals and experimental protocol has been suggested that autophagy has roles in critical The present study was conducted in strict accordance adaptive mechanisms in the heart under hemodynamic with the Guideline for the Care and Use of Labora- stress conditions such as pressure overload [6] and loss tory Animals published by the US National Institute of of a contractile region by MI [5, 7–11]. In the case of MI, Health (NIH publication No. 85-23, revised 1996) and autophagic activity was augmented in the non-infarcted was approved by the Animal Use Committee of Sap- remote area and border area but not in the infarcted poro Medical University. Protocols of the experiments myocardium after MI, and the autophagic activity pro- are summarized in Fig.  1. Male LETO and OLETF at gressively increased in the remote area during a 3-week ages of 25–30  weeks were used in all experiments. period after MI [8, 9]. The increase in autophagic activity LETO were pretreated with saline, vildagliptin (10  mg/ has protective effects against remodeling and dysfunc - kg/day), a DPP-4 inhibitor, or exenatide (10  μg/kg/day), tion of the ventricle after MI [8, 9]. However, autophagy a GLP-1 receptor agonist, for 2  weeks, or with chloro- has been shown to be impaired by DM in non-cardiac quine (10 mg/kg/day), an autophagy inhibitor [9, 12, 35], [14] and cardiac tissues [15–20], though advanced gly- for 1 week. Vildagliptin was kindly provided by Novartis cogen endproducts reportedly activate autophagy in Pharma AG (Basel, Switzerland). The pharmacological cardiomyocytes [21]. Whether the increase in myocar- agents were administered via osmotic minipumps (Alzet, dial autophagic activity after MI is impaired by DM and Cupertino, CA, USA), not via drinking water contain- whether such an impairment, if any, is treatable by phar- ing the agents, because the amount of water rats drink macological agents have not been clarified. per day is not consistent. The dose of vildagliptin was We hypothesized that autophagic response of the selected on the basis of a result in a previous report [36], myocardium to MI is impaired by T2DM and that inhi- and we confirmed that this dose of vildagliptin signifi - bition of dipeptidyl peptidase-4 (DPP-4) would attenu- cantly increased the serum level of GLP-1 (see “Results”). ate the T2DM-induced increase in post-MI mortality OLETF also received saline, vildagliptin, exenatide or by restoring the autophagic response. The rationale chloroquine as did LETO, and an additional group of for the hypotheses is four-fold. First, a significant asso - OLETF received both vildagliptin and chloroquine. In ciation of preserved ventricular function and activa- separate groups of rats, blood samples were collected via tion of autophagy has been demonstrated for different the carotid artery under anesthesia 12  h after fasting to types of cardiac stress in non-diabetic animals [5–11]. examine the effects of these pharmacological agents on Second, DM impairs intracellular signaling mecha- metabolic parameters. nisms relevant to autophagy, including PI3K-Akt sign- aling, in the myocardium [3]. Third, activation of the Oral glucose tolerance test glucagon-like peptide-1 (GLP-1) receptor or treatment An oral glucose tolerance test (OGTT) was performed in with a DDP-4 inhibitor triggers AMP-activated protein LETO treated with a vehicle and OLETF treated with a kinase (AMPK) signaling [22, 23, 24], which facilitates vehicle or vildagliptin (10  mg/kg/day) for 2  weeks. After autophagy [5, 9, 15, 18, 25]. In fact, a GLP-1 analog, fasting for 12  h, rats were administered glucose (2  g/kg liraglutide, has been shown to promote autophagy in body weight) by gavage, and blood glucose and insulin non-cardiac cells [26]. Fourth, it has been reported levels before and after glucose administration were meas- that the activity of circulating DPP-4 is associated with ured by using a Glutest-mint (Sanwa Kagaku Kenkyusho, left ventricular dysfunction in patients [27, 28]. Con- Nagoya, Japan) and a rat insulin RIA kit (Linco Research versely, inhibition of DPP-4 and use of a GLP-1 analog Inc, St. Charles, MO, USA), respectively. Blood for GLP-1 prevented cardiomyopathy, improved cardiac function assay using a GLP-1 (Active) ELISA kit (Millipore) was and post-MI survival rate, and attenuated ventricular collected before glucose administration in sampling tubes remodeling in DM and non-DM animals [27, 29–32]. containing a DPP-4 inhibitor. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 3 of 16 Fig. 1 Experimental protocols. Experimental protocols for survival study (a) and cardiac tissue sampling (b). Vilda vildagliptin, Exe exenatide, CQ chloroquine, LV left ventricle. Echocardiography Cardiac tissue sampling after MI Echocardiography was performed before induction of MI Since the mortality rate at 24–48  h after MI was high as previously reported [4]. in OLETF at ages of 25–30  weeks [4], myocardial tis- sue sampling for biochemical analyses and immunohis- Induction of MI and mortality monitoring tochemistry was performed at 12  h after MI. Rats were Rats were prepared for induction of MI as in our previous anesthetized and ventilated, and blood pressure and study [4]. In brief, rats were anesthetized with sodium heart rate were monitored by a catheter placed in the pentobarbital (40  mg/kg, i.p.), and the level of anesthe- carotid artery. The chest was re-opened and the hearts sia was continuously monitored during the experiment were excised and immediately immersed in ice-cold and an additional dose of pentobarbital was administered saline. The myocardium in the non-infarcted region was when necessary. Rats were then intubated and ventilated quickly excised in the saline, frozen in liquid nitrogen, with a rodent respirator (model 683, Harvard Apparatus, and stored at −80°C until use for biochemical and histo- South Natick, MA, USA). After left thoracotomy, a mar- logical analyses. ginal branch of the left coronary artery was permanently ligated by using a 5–0 silk thread to induce MI. We used Immunohistochemistry a permanent occlusion model of MI to avoid the possi- Frozen heart tissues were embedded in OCT compound bility that pharmacological pretreatments modify infarct (Tissue-Tek) and snap-frozen in liquid nitrogen. After the size and induce an inter-group difference in mechanical tissues had been sectioned at 8  μm in thickness with a stress on the non-infarcted region. The surgical wounds cryostat at −20°C, the sections were incubated with rab- were repaired and the rats were returned to their cages. bit polyclonal anti-LC3 antibody (MBL, PM036, 1:250) in All rats were allowed ad-lib access to water but restricted PBS containing 1% BSA and 0.3% triton X-100 overnight from food for 12 h. Survival rate of rats was determined at 4°C. The samples were then incubated with an Alexa at 24 and 48  h after MI. Rats that had survived at 48  h Fluor 488 anti-rabbit IgG antibody (Invitrogen) for 1  h after MI were euthanized by a pentobarbital overdose at room temperature. After nuclei had been stained with and heart tissue was excised and fixed in 10% formalde - Hoechst33342 (Dojindo, Kumamoto, Japan), samples hyde for infarct size analysis. were mounted on slides for image analysis. Fluorescence Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 4 of 16 images were obtained using a FLoid Cell Imaging Sta- CA, USA). First-strand cDNA was synthesized using a tion (Life Technologies). The number of LC3 dots was SuperScript VILO cDNA Synthesis Kit (Life Technolo- counted and analyzed in 40 randomly selected fields from gies). DNA amplification was performed in StepOne five hearts in each group. (Life Technologies). Analyses of B-type natriuretic pep- tide (BNP) and β-actin mRNA levels were performed by Immunoblotting using Taqman gene expression assays (Rn00676450_g1 Frozen tissue samples were homogenized in ice-cold Nppb and Rn00667869_m1 Actb, respectively). For buffer (CelLytic MT Cell Lysis Reagent) including pro- p62 and 18S, we used Power SYBR PCR Master Mix tease and phosphatase inhibitor cocktails (Nacalai Tesque, (Applied Biosystems, Inc) and the following oligonucle- Inc., Kyoto, Japan). The homogenate was centrifuged at otide primers: for rat p62, 5′-ATCAGCCTCTGGTGG 15,000g for 15 min at 4°C to obtain the supernatant. Equal GAGAT-3′ and 5′-CCCATCCACAGGTGAACTCC-3′; amounts of protein were analyzed by immunoblot assays for rat 18S, 5′-CGGACAGGATTGACAGATTG-3′ using specific antibodies (see Additional file  1: Table S1). and 5′-CAAATCGCTCCACCAACTAA-3′. All assays Intensities of individual bands were quantified by using were performed in duplicate and by the standard curve Image J software (National Institutes of Health). method using serial cDNA dilution. Beclin‑1–Bcl‑2 interaction Statistical analyses Frozen myocardial tissue samples were homogenized in Data are presented as mean ± SEM. Differences between ice-cold buffer containing 20 mM Tris (pH 7.4), 137 mM treatment groups were assessed by one-way analysis of NaCl, 10% glycerol, 0.3% CHAPS, and 2 mM EDTA sup- variance (ANOVA) followed by the Student–Newman– plemented with protease and phosphatase inhibitor cock- Keuls post hoc test for multiple comparisons. Differences tails. The homogenates were then centrifuged at 15,000g in time course between two groups in OGTT were ana- for 15  min to obtain supernatants. After quantification lyzed by 2-way ANOVA for repeated measures followed of protein concentration, the lysates were incubated for by the Student–Newman–Keuls post hoc test for multi- 30  min with 40  μl of protein A magnetic beads (New ple comparisons. Survival rates after MI were compared England Biolads, Ipswich, MA, USA) to remove endog- by Kaplan–Meier curves and log-rank statistics. For all enous IgG. Equal amounts (2,000  μg) of lysates were tests, p < 0.05 was considered statistically significant. incubated with either 6  μg of rabbit anti-Beclin-1 anti- body or normal rabbit IgG overnight at 4°C. The mixture Results was then incubated with 50 μl of fresh beads for 1 h. The Metabolic profiles beads were washed three times with PBS containing pro- Data for metabolic profiles of LETO and OLETF and tease inhibitor cocktail and re-suspended in SDS sample the effects of vildagliptin, exenatide and chloroquine on loading buffer followed by denaturation. Immunoprecipi - metabolic parameters are shown in Table  1. In LETO, tated proteins were analyzed by Western blotting initially body weight (531 ± 7 g), fasting blood glucose level and with rabbit anti-Bcl-2 antibody and then with mouse serum total cholesterol level were not changed by treat- monoclonal anti-Beclin-1 antibody after stripping. ment with vildagliptin or exenatide. OLETF had larger body weight (627  ±  14  g) and higher fasting blood glu- mRNA quantification cose level than those of LETO (Table  1), as we previ- Total RNA was isolated from myocardial tissues by using ously reported [4, 34, 37]. Treatment with vildagliptin or an RNeasy Fibrous Tissue Mini Kit (Qiagen, Valencia, exenatide at the dose used in the present study did not Table 1 Metabolic parameters after treatment LETO OLETF Vehicle Vildagliptin Exenatide Vehicle Vildagliptin Exenatide Vildagliptin +  Chloroquine Blood glucose (mg/dl) 108 ± 5 106 ± 6 117 ± 12 205 ± 25* 201 ± 13* 251 ± 17* 197 ± 22* † † Serum insulin (ng/ml) 2.5 ± 0.5 1.3 ± 0.2 3.1 ± 0.9 5.8. ± 0.9* 2.5 ± 0.5 1.3 ± 0.8 4.1 ± 0.9 Serum TC (mg/dl) 82 ± 5 85 ± 5 95 ± 4 117 ± 3 120 ± 6 109 ± 12 96 ± 7 Values are mean ± SEM. N = 4–11. TC total cholesterol. * P < 0.05 vs. LETO vehicle; P < 0.05 vs. OLETF vehicle. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 5 of 16 change body weight and blood glucose level in OLETF. LV function at 12 h after coronary ligation also, but clear Serum insulin level was significantly higher in OLETF cardiac images for quantitative assessment could not than in LETO, and both vildagliptin and exenatide be obtained at that time point because of intra-thoracic decreased insulin level in OLETF without reduction in air, fluid and chest wall damage related to open-chest blood glucose level (Table  1). Treatment of OLETF with surgery. chloroquine in addition to vildagliptin affected neither insulin level nor blood glucose level compared to the lev- Vildagliptin improved survival in OLETF in a els in those treated with vildagliptin alone. Despite lack chloroquine‑sensitive manner of effects on blood glucose level, vildagliptin increased The survival rate during a period of 48  h after MI in serum level of the active form of GLP-1; GLP-1 level was LETO was 82%, which was comparable with the rate in below the detection range (<2.0 pmol/L) in 4 of a total of our previous study [4]. Pretreatment with vildagliptin or five samples from the vehicle-treated OLETF, but it was exenatide for 2  weeks before MI did not affect the sur - 5.0 ± 0.9 pmol/L in vildagliptin-treated OLETF (Fig. 2a). vival rate in LETO (86 and 80%, respectively) (Fig.  3a). In OGTTs, levels of blood glucose and serum insu- As in our previous study [4], the survival rate was sig- lin before and after glucose administration were higher nificantly lower in OLETF (32%, Fig.  3b) than that in in vehicle-treated OLETF than in LETO (Fig.  2b, c). LETO. Pretreatment of OLETF with vildagliptin signifi - Treatment with vildagliptin did not change glucose lev- cantly increased the survival rate (80%) to a level similar els before and after glucose administration in OLETF to that in LETO (Fig. 3b). Exenatide treatment tended to (Fig.  2b). On the other hand, insulin level was signifi - improve the survival rate in OLETF, but the difference did cantly lower in the vildagliptin-treated OLETF than in not reach statistical significance. In chloroquine-treated the vehicle-treated OLETF (Fig. 2c). LETO, the survival rate (62%, Fig. 3c) tended to be lower than that in vehicle-treated LETO (Fig. 3a). Survival rates Cardiac function was not modified by vildagliptin or of chloroquine-treated OLETF were similar, regardless exenatide at baseline of vildagliptin treatment or no vildagliptin treatment (33 Before induction of MI, heart rate was lower and left and 39%, respectively, Fig. 3c), to the survival rate in vehi- ventricular (LV) dimension was larger in OLETF than cle-treated OLETF (Fig. 3b). Autopsies of rats died within in LETO, though there were no differences in LV ejec - 48 h after MI revealed no case of cardiac rupture. tion fraction and fractional shortening between the two Infarct sizes 48  h after the permanent coronary occlu- groups (Table  2). Treatment with vildagliptin, exenatide, sion were 35–40% of the left ventricle and were compa- or chloroquine did not alter these echocardiographic rable among the treatment groups (Fig.  3d–f ). In LETO, parameters in either LETO or OLETF. We tried to assess hemodynamic parameters (heart rate and blood pressures) Fig. 2 Serum active GLP-1 level and oral glucose tolerance test in LETO and OLETF. a Serum level of the active form glucagon-like peptide (GLP-1) in OLETF treated with the vehicle or vildagliptin. Values in 4 of 5 samples from the vehicle-treated group were under the detection limit (<2.0 pmol/L) of the assay. Blood glucose (b) and serum insulin (c) during an oral glucose tolerance test (2 g glucose per kg body weight) in LETO (triangle) and OLETF treated with the vehicle (closed circle) or vildagliptin (open circle). *p < 0.05 vs. LETO and p < 0.05 vs OLETF vehicle at each time point. N = 5 in each group. Vilda vildagliptin. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 6 of 16 Table 2 Echocardiographic data at baseline LETO OLETF Vehicle Vildagliptin Exenatide Chloroquine Vehicle Vildagliptin Exenatide Vildagliptin +  Chloroquine HR (bpm) 384 ± 6 386 ± 7 385 ± 4 375 ± 6 312 ± 7* 316 ± 4* 315 ± 6* 324 ± 9* LVEF (%) 69.4 ± 1.5 70.8 ± 1.3 68.4 ± 1.2 73.0 ± 1.3 65.4 ± 1.5 67.7 ± 0.7 68.5 ± 1.3 69.9 ± 1.3 FS (%) 35.0 ± 1.3 35.9 ± 1.0 34.0 ± 0.8 37.6 ± 1.5 32.1 ± 1.1 33.4 ± 0.5 34.4 ± 1.0 35.2 ± 1.1 IVST (mm) 1.72 ± 0.04 1.69 ± 0.02 1.76 ± 0.03 1.75 ± 0.04 1.81 ± 0.04 1.74 ± 0.03 1.73 ± 0.03 1.85 ± 0.02 PWT (mm) 1.66 ± 0.06 1.73 ± 0.05 1.64 ± 0.02 1.76 ± 0.06 1.85 ± 0.05 1.79 ± 0.04 1.83 ± 0.06 1.84 ± 0.03 LVEDD (mm) 7.04 ± 0.12 6.71 ± 0.19 7.11 ± 0.12 6.68 ± 0.12 8.05 ± 0.18* 7.69 ± 0.10* 7.88 ± 0.12* 7.65 ± 0.15* LVESD (mm) 4.60 ± 0.15 4.34 ± 0.13 4.69 ± 0.09 4.16 ± 0.14 5.49 ± 0.17* 5.08 ± 0.10* 5.17 ± 0.12* 4.94 ± 0.07* LVEDV (ml) 0.80 ± 0.04 0.75 ± 0.04 0.83 ± 0.04 0.69 ± 0.03 1.17 ± 0.07* 1.02 ± 0.04* 1.10 ± 0.04* 1.00 ± 0.06* LVESV (ml) 0.26 ± 0.02 0.22 ± 0.01 0.26 ± 0.02 0.19 ± 0.02 0.41 ± 0.03* 0.37 ± 0.04* 0.35 ± 0.02* 0.30 ± 0.01* Values are mean ± SEM. N = 12–33. HR heart rate, bpm beats per minute, LVEF left ventricular ejection fraction, FS fractional shortening, IVST interventricular septal thickness, PWT posterior wall thickness, LVEDD left ventricular end-diastolic dimension, LVESD left ventricular end-systolic dimension, LVEDV left ventricular end-diastolic volume, LVESV left ventricular end- systolic volume. * P < 0.05 vs. LETO Vehicle. vinculin ratio was significantly increased after MI in LETO, at 12  h after MI in immunoblot experiments were com- but such a response was not observed in OLETF (Fig.  4a). parable between the treatment groups (Table  3). Heart Changes in LC3-II/LC3-I ratio were similar to those in rates before and after MI and blood pressure at 12 h after LC3-II/vinculin, though the difference did not reach statis MI were lower in OLETF than in LETO, but treatment - with either vildagliptin or exenatide did not significantly tical significance. Vildagliptin significantly increased LC3- change these hemodynamic parameters in OLETF. Taken II level and tended to increase LC3-II/LC3-I after MI in together, these results indicate that difference in infarct OLETF (Fig. 4b). Although p62 protein is often used as an size or hemodynamic response does not underlie the dif index of autophagic flux [38], its level was not changed by MI in either LETO or OLETF (Fig. 4c). Neither vildagliptin ference in survival rate among the treatment groups. nor exenatide changed p62 protein level after MI in OLETF To examine whether the loading condition in the non- (Fig. 4d). mRNA levels of p62 in the myocardium were also infarcted region after MI was modulated by vildagliptin, we measured BNP mRNA level in the remote myocar similar in LETO and OLETF regardless of MI and treat - - dium 12  h after MI (Fig.  3g). BNP mRNA level was sig- ment with vildagliptin or exenatide (data not shown). nificantly increased after MI in both LETO and OLEFT, Autophagic activity in the non-infarcted region of but such an increase in BNP mRNA after MI was not the myocardium was assessed also by immunostaining of autophagosomes with anti-LC3 antibody (Fig.  5). In observed in OLETF treated with vildagliptin (Fig.  3g). Since augmented adrenergic activity is one of the features LETO, the number of LC3-positive dots was significantly of heart failure, we determined phosphorylation of vas increased after MI by 66%, but such an increase was not observed in OLETF. Not only vildagliptin but also exena odilator-stimulated phosphoprotein (VASP) at Ser157, - a protein kianse A phosphorylation site, in the myocar- tide significantly increased the number of LC3 dot after dium. The levels of phospho-VASP after MI were simi - MI in OLETF (Fig.  5), indicating that vildagliptin and lar in LETO and OLETF, but treatment of OLETF with exenatide restored autophagic induction in the non- infarcted myocardium after MI in OLETF. vildagliptin significantly decreased phospho-VASP levels (Fig.  3h). These findings suggest that vildagliptin attenu - ated both ventricular overloading and augmented adren- Activation of AMPK in response to MI was impaired ergic drive after MI in OLETF. in OLETF Since AMPK is known to activate autophagy by phos- Autophagic response in the non‑infarcted region of the phorylating ULK1 at Ser317 [25], we assessed AMPKα myocardium after MI was impaired in OLETF phosphorylation at Thr172 and phosphorylation of Marker molecules of autophagic activities in the non- its downstream target proteins, acetyl-CoA carboxy- infarcted region of the heart after MI are shown in Fig.  4. lase (ACC) and ULK1, in the non-infarcted region after LC3-II levels in the heart without infarction (i.e., sham MI (Fig.  6a–c). Levels of phospho-Thr172-AMPKα, operation) were similar in LETO and OLETF. LC3-II/ phospho-Ser79-ACC and phospho-Ser317-ULK1 were Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 7 of 16 Fig. 3 Eec ff ts of vildagliptin, exenatide, and chloroquine on survival after MI. Kaplan–Meier survival analysis of LETO (a), OLETF (b), and rats treated with chloroquine (c) after left coronary artery occlusion. *p < 0.05 vs. Vehicle-treated group. Infarct size measured at 48 h after MI in LETO (d), OLETF (e), and rats treated with chloroquine (f). g Quantification of BNP mRNA levels normalized to β-actin in the non-infarcted myocardium sampled 12 h after MI. N = 3–6 in each group. h Summary data of immunoblotting for phospho-Ser157 VASP in samples from LETO, vehicle- or vildagliptin- treated OLETF after MI. N = 9–12 in each group. Vilda vildagliptin, Exe exenatide, CQ chloroquine. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 8 of 16 Fig. 4 Vildagliptin restored autophagic induction in the non-infarcted area after MI in OLETF. a Representative images of Western blotting for LC3 protein (left) and summary data of LC3-II level and LC3-II/LC3-I ratio (right) in samples from sham-operated hearts (Sham) or the non-infarcted myocardium after MI in LETO and OLETF. b Representative images of Western blotting for LC3 protein (left) and summary data of LC3-II level and LC3-II/LC3-I ratio (right) in samples from the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). c Representative blots (left) and summary data (right) of Western blotting for p62 protein in samples from sham-operated hearts (Sham) or the non-infarcted myocardium after MI in LETO and OLETF. d Representative blots (left and middle) and summary data (right) of Western blotting for p62 protein in OLETF treated with the vehicle, Vilda, or Exe after MI. N = 8–10 in each group. *p < 0.05. a.u. arbitrary units, NS not significant. Akt/mTORC1 activity after MI was attenuated in the OLETF significantly increased at 12 h after MI in LETO. In con - hearts trast, such responses of AMPKα, ACC and ULK1 were not detected in OLETF. Although vildagliptin and exena Alterations in Akt/mTORC1 signaling, a negative regu- latory mechanism of autophagy [25], in OLETF were tide restored the increase in LC3-positive autophago- examined by immunoblotting. The level of Ser473-Akt somes after MI in OLETF (Fig. 5), neither agent restored phosphorylation of Thr172-AMPKα, Ser79-ACC or phosphorylation was significantly elevated after MI in Ser317-ULK1 after MI in OLETF (Fig. 6d–f ). the non-infarcted myocardium in LETO (Fig.  7a). The Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 9 of 16 Table 3 Hemodynamic data 12 h after myocardial infarction LETO OLETF Sham MI Sham MI Vehicle Vehicle Vildagliptin Exenatide Vehicle Vehicle Vildagliptin Exenatide SBP (mmHg) 119 ± 5 105 ± 6 115 ± 4 114 ± 5 104 ± 6 90 ± 3* 96 ± 4* 93 ± 5* DBP (mmHg) 88 ± 4 81 ± 6 85 ± 4 82 ± 6 73 ± 7 64 ± 3* 65 ± 4* 64 ± 4* Heart rate (bpm) 423 ± 7 401 ± 8 410 ± 11 400 ± 10 332 ± 13* 321 ± 6* 331 ± 7* 326 ± 9* Values are measn ± SE. N = 8–15. SBP systolic blood pressure, DBP diastolic blood pressure, bpm beats per minute. * P < 0.05 vs. LETO Sham. phosphorylation was associated with increases in lev- els of phospho-mTOR at Ser2448 and phospho-S6 at Ser235/236 (Fig.  7b, c). However, such an activation of Akt after MI was not observed in OLETF; phospho-Akt levels were similar in sham-operated and MI-induced OLETF (Fig.  7a). Phospho-mTOR and phospho-S6 lev- els were increased after MI in OLETF, but their levels remained significantly lower than those in LETO (Fig.  8b, c). Neither vildagliptin nor exenatide increased phospho- rylation of Ser473-Akt, Ser2448-mTOR, and Ser235/236- S6 after MI in OLETF (Fig. 7d–f ). Beclin‑1–Bcl‑2 interaction was enhanced in OLETF Since vildagliptin improved the response of autophagic activity to MI in OLETF (Figs.  4, 5) without normaliza- tion of AMPK phosphorylation and Akt/mTORC1 sign- aling (Figs.  6, 7), we examined whether Beclin-1–Bcl-2 interaction is modified in OLETF or by vildagliptin treat - ment. Beclin-1 is an essential component for activation of autophagy, and its autophagy-promoting activity is inhibited by binding to an anti-apoptotic protein, Bcl-2 [39]. There were no significant differences in Beclin-1 and Bcl-2 protein levels between LETO and OLETF regard- less of MI (Fig. 8a), and neither vildagliptin nor exenatide changed levels of these proteins in MI-induced OLETF (Fig. 8b). However, Beclin-1–Bcl-2 interaction was signif- icantly augmented in OLETF with MI compared to that in OLETF with a sham operation, and their interaction was attenuated by vildagliptin (Fig. 8c–e). Discussion In the present study, treatment with vildagliptin at a Fig. 5 Immunofluorescent analysis of LC3 protein in the non- dose that did not lower plasma glucose level significantly infarcted myocardium after MI. a Representative immunofluores- improved survival of OLETF after acute MI (Figs.  2, 3). cence images of LC3 protein in LV sections from the non-infarcted By using telemetric monitoring of heart rate and blood myocardial area after MI. The image without the primary antibody did not show any green dots. b Quantification of LC3 dots per field pressure, we previously demonstrated that increased (568 µm × 426 µm). A total of 40 fields from five hearts were analyzed mortality during the acute phase after MI in OLETF is in each group. *p < 0.05. Vilda vildagliptin, Exe exenatide, NS not due to progressive heart failure but not lethal arrhyth- significant. mia [4]. Although vildagliptin did not modify ventricular Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 10 of 16 Fig. 6 Analysis of the AMPK/ULK1 pathway. Representative images (left) and summary data (right) of Western blotting for phospho-Thr172 and total AMPKα (a), phospho-Ser79 and total acetyl-CoA carboxylase (ACC) (b), and phospho-Ser317-ULK1 (c) in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. Representative blots (left) and summary data (right) of Western blotting for phospho-Thr172 and total AMPKα (d), phospho-Ser79 and total acetyl-CoA carboxylase (ACC) (e), and phospho-Ser317-ULK1 (f) in the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). N = 9–10 in each group. *p < 0.05. a.u. arbitrary units, NS not signifi- cant. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 11 of 16 Fig. 7 Analysis of Akt/mTORC1 activity. Representative blots (left) and summary data (right) of Western blotting for phospho-Ser473 and total Akt (a), phospho-Ser2448 and total mTOR (b), and phospho-Ser235/236 and total S6 (c) in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. Representative blots (left) and summary data (right) of Western blotting for phospho-Ser473 and total Akt (d), phospho-Ser2448 and total mTOR (e), and phospho-Ser235/236 and total S6 (f) in the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). N = 9–10 in each group. *p < 0.05. a.u. arbitrary units. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 12 of 16 Fig. 8 Eec ff t of vildagliptin on Beclin-1/Bcl-2 interaction after MI in OLETF. a Representative images (left) and summary data (right) of Western blotting for Beclin-1 and Bcl-2 in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. b Eec ff ts of vildagliptin (Vilda) and exenatide (Exe) on Beclin-1 and Bcl-2 levels in the non-infarcted myocardium after MI in OLETF. Representative immunoblot- ting images (left) and summary data (right) are shown. c Myocardial lysates were immunoprecipitated (IP) with anti-Beclin-1 antibody or rabbit IgG followed by immunoblotting with anti-Bcl-2 and Beclin-1 antibodies. d, e Beclin-1/Bcl-2 interaction was increased after MI in OLETF, which was attenuated by vildagliptin (Vilda). N = 5 in each group. *p < 0.05. a.u. arbitrary units. function under baseline conditions (Table  2), it sup- but a  recent study has shown that a GLP-1 receptor pressed MI-induced upregulation of BNP expression and agonist provides cardioprotection by a mechanism inde- cardiac adrenergic activity in OLETF (Fig.  3g, h). Thus, pendent of the GLP-1 receptor in the cardiomyocyte [31]. suppression of heart failure progression after MI is the On the other hand, DPP-4 is involved in degradation of most likely explanation for reduction in acute mortality multiple peptides such as substance P and stromal cell- after MI in OLETF by vildagliptin. derived factor-1 [27, 41], and these properties of DPP-4 Serum active GLP-1 level was elevated by vildagliptin inhibitors might underlie the differences in survival rate (Fig.  2a), suggesting that GLP-1 mediated the improved (Fig.  3b) and changes in LC3-II protein level (Fig.  4b) survival by vildagliptin. On the other hand, treatment after MI between vildagliptin-treated and exenatide- with a GLP-1 analog, exenatide, tended to improve treated OLETF. post-MI survival in OLETF, but the effect was statisti - In the literature, a study by French et  al. [42] is the cally insignificant. We do not have a clear explanation only study in which the changes in autophagy after MI for the different outcomes in the vildaglitpin-treated in diabetic mice and non-diabetic mice were compared. and exenatide-treated groups. However, a difference in In that study, autophagic activity was not increased after the mechanism of cardioprotection is a possibility. The MI not only in the diabetic heart but also in the non-dia- GLP-1 receptor is localized in the cardiomyocyte [40], betic heart. The negative results are in contrast to results Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 13 of 16 of several studies showing that autophagic activity was The results of the present study supported the notion increased after MI in the healthy control myocardium [5, that activation of autophagy by vildagliptin during the 8–12]. The reason why French et  al. could not detect an acute phase after MI contributed to improved survival alteration of autophagy after MI even in the non-diabetic in OLETF. However, roles of autophagy in the heart may myocardium is unclear, but use of entire risk zone tissue be different depending on the phase and type of cardiac consisting of infarcted and non-infarcted cells might have stress. Matsui et  al. [5] showed that autophagy is pro- obscured changes in autophagy in the viable myocardium tective for cardiomyocyte survival during ischemia but after MI. is rather detrimental during reperfusion. Zhu et  al. [47] We focused on the non-ischemic region of the showed that sustained activation of autophagy during infarcted heart since that region plays a crucial role pressure overload is detrimental to cell survival. They in compensation for the lost function of the infarcted showed that cardiac function after thoracic aortic band- region and undergoes adaptive and maladaptive post-MI ing was preserved in beclin-1 heterozygous knockout remodeling [43, 44]. There was no significant difference mice, whereas cardiomyocyte-specific overexpression in LC3-II or p62 levels or LC3-positive autophagosomes of beclin-1 worsened cardiac function. In contrast, sus- between LETO and OLETF under baseline conditions tained elevation of autophagy may be protective for (i.e., sham-operated groups in Figs.  4, 5), suggesting that post-infarcted ventricular function and remodeling. Sup- autophagic activities were similar in the diabetic myocar- pression of autophagy by bafilomycin A1 or chloroquine dium and non-diabetic myocardium under non-stressed has been shown to exacerbate cardiac function after MI, conditions. However, there was a significant difference but activation of autophagy by an mTORC1 inhibitor, between LETO and OLETF in autophagic response rapamycin, or by caloric restriction was protective [8, 12]. after MI. In LETO, an increase in autophagic activity Maejima et  al. [11] also reported that cardiac function was observed in the non-infarct region at 12  h after MI at 6  weeks after MI was worsened in beclin-1 heterozy- (Figs. 4, 5). A protective role of the increase in autophagic gous knockout mice. Since the survival rate of OLETF at activity after MI has been demonstrated by findings that 48 h after MI was only 32% (Fig.  3b), we did not include inhibition of autophagy by bafilomycin A or genetic dele - assessment of autophagic activity at the later phase after tion of beclin-1 aggravated remodeling and dysfunction MI in OLETF in the present study. of the ventricle after MI [8, 11]. Importantly, the mor- Streptozotocin-induced diabetes and high-fat diet tality rate after MI in OLETF was not further increased have been shown to reduce LC3-II in the myocardium, by inhibiting autophagy with chloroquine (Fig.  3). These which was associated with suppressed phosphorylation results support the notion that impaired autophagic of AMPK, a positive regulator of autophagy [16, 18]. In response in the non-infarcted region of the infarcted contrast, LC3-II, p62 or AMPK phosphorylation in the heart contributes to increase in acute mortality after MI heart without MI was not different between OLETF and in OLETF. LETO in this study. However, Lee et  al. [48] reported There are two possible mechanisms for suppression 50% reduction in AMPK phosphorylation in the myo- of heart failure by autophagy: reduction of reactive cardium of OLETF at the age of 28  weeks. A possible oxygen species (ROS) production and improvement of explanation for the discrepant results is more advanced myocardial energy status. Damaged organelles partici- stage of T2DM in OLETF in the study by Lee et al. [48]. pating in ROS generation, including mitochondria, are Despite similar ages, OLETF in their study had slightly sequestrated and removed by the autophagic process, larger body weight compared with that in this study and and autophagy plays a role in suppression of ROS gen- showed significantly increased interstitial fibrosis in the eration [17, 45]. In fact, ROS generation from damaged heart [48], though such an increased collagen deposition mitochondria is involved in exacerbation of ventricu- in the myocardium was not detected by histochemistry lar dysfunction [46]. An impact of autophagy on myo- or determination of mRNA levels of collagens I and III cardial energy status has been shown by findings that in OLETF used in our studies [4, 33]. Some difference in myocardial ATP content after MI was increased by aug- rearing conditions (possibly the amount or calories per mentation of autophagy [9]. Diabetes impairs mecha- volume of the chow provided) might underlie the differ - nisms regulating ATP supply, and our recent study [33] ence in the phenotype of OLETF at similar ages. Never- showed that reduced myocardial reserve of ATP sup- theless, it is possible that suppression of baseline AMPK ply, leading to diastolic dysfunction, was disclosed by phosphorylation and autophagy occurs in OLETF at an increased afterload in OLETF. How impaired response advanced stage of T2DM. of autophagy relates to dysregulation of ATP supply Increased AMPK phosphorylation with or without mechanisms in diabetic hearts may warrant further suppressed mTOR phosphorylation was associated with investigation. promotion of autophagy after MI in non-diabetic mice Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 14 of 16 [9, 16]. The responses of autophagy, AMPK and mTOR Conclusions to MI were confirmed in LETO (Figs.  4, 5, 6, 7). However, Treatment with vildagliptin at a dose that elevated serum in OLETF, phosphorylation of AMPK was not increased GLP-1 without normalization of plasma glucose level and activation of the mTOR/S6 pathway was 60–70% of reduced acute mortality after MI in a rat model of T2DM that in LETO (Figs. 6, 7). In addition, we found that inter- to the level in non-diabetic controls. The beneficial effect action of Beclin-1 and Bcl-2, which reportedly inhibits of vildagliptin was sensitive to chloroquine and closely Beclin-1-dependent autophagy [11, 15, 39], was signifi - associated with restoration of the autophagic response cantly increased in the myocardium of OLETF (Fig.  8). in the non-infarcted myocardium to MI, suggesting an Restoration of the adaptive responses of both LC3-II and involvement of impaired autophagy in T2DM-induced autophagosomes after MI in OLETF by vildagliptin was increase in post-MI mortality. associated with suppression of Beclin-1–Bcl-2 interac- Additional file tion but not with improved phosphorylation of AMPK, mTOR or S6. These findings suggest that increased Additional file 1: Table S1. Antibodies. Description of data: a list of Beclin-1–Bcl-2 interaction was responsible for T2DM- antibodies used in this study. induced loss of adaptive autophagy in the non-ischemic myocardium after MI. How vildagliptin suppressed Becin-1–Bcl-2 interac- Authors’ contributions Participated in research design: AK, TaM, MT and TeM. Conducted experiments: tion in the myocardium of OLETF remains unclear. HM, AK, TaM, HK, SI, TT, MO and KN. Performed data analysis: HM, AK, TaM, MT, Among molecules that regulate Beclin-1–Bcl-2 interac- TY, HK and TeM. Performed statistical analyses: HM and AK. Wrote or contrib- tion, AMPK-JNK activation has been reported to induce uted to the writing of the manuscript: AK, TaM and TeM. All authors read and approved the final manuscript. disruption of Beclin-1–Bcl-2 interaction through phos- phorylation of Bcl-2 at Ser70 [15], whereas activation Author details of mammalian sterile 20-like kinase 1 (Mst1) promoted Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, South-1, West-16, Chuo-ku, Sap- Beclin-1/Bcl-2 interaction by phosphorylation of Bec- poro 060-8543, Japan. Department of Pharmacology, Sapporo Medical lin-1 at Thr108 [11]. In this study, vildagliptin did not University School of Medicine, Sapporo 060-8543, Japan. restore phosphorylation of AMPK (Fig.  6) or JNK (data Acknowledgements not shown). Hence, there is the possibility that vildaglip- This study was supported by Grants for Education and Research 2012–2014 tin suppressed Mst1 expression or activity, preventing from Sapporo Medical University and a grant from Novartis Pharma AG. interaction of Beclin-1 and Bcl-2 in the myocardium of Novartis Pharma AG did not play any role in the collection, analysis and inter- pretation of data or the writing of the manuscript. OLETF. Unfortunately, we could not examine this pos- sibility since phospho-Mst1 (Thr183) protein in the Compliance with ethical guidelines myocardium of OLETF could not be detected by use of Competing interests commercially available antibodies. This study was supported in part by a grant from Novartis Pharma AG. Contrary to our expectations, the dose of vildagliptin and Grants for Education and Research 2012–2014 from Sapporo Medical we used in the present study was not sufficient for reduc - University. ing glucose levels in OLETF (Table 1; Fig. 2b), though the Received: 27 April 2015 Accepted: 24 July 2015 dose of vildagliptin increased serum active GLP-1 level in OLETF (Fig.  2a). Therefore, the effects of vildagliptin on cardiac autophagy and mortality cannot be explained by its effect on glycemic control. Although DPP-4 inhibi - References tors are known to enhance glucose-stimulated insulin 1. Marso SP, Miller T, Rutherford BD, Gibbons RJ, Qureshi M, Kalynych A release [49], reduction in plasma insulin level by DPP-4 et al (2007) Comparison of myocardial reperfusion in patients undergo- inhibitors together with improved metabolic parameters ing percutaneous coronary intervention in ST-segment elevation acute myocardial infarction with versus without diabetes mellitus (from the has also been reported [50, 51]. It is notable that DPP-4 EMERALD Trial). Am J Cardiol 100:206–210 inhibitors (saxagliptin and alogliptin) failed to reduce 2. De Luca G, Dirksen MT, Spaulding C, Kelbæk H, Schalij M, Thuesen L cardiovascular mortality in diabetic patients at high car- et al (2013) Impact of diabetes on long-term outcome after primary angioplasty: insights from the DESERT cooperation. Diabetes Care diovascular risk in two large clinical trials [52, 53]. The 36:1020–1025 effect of vildagliptin on acute mortality after MI in this 3. Miki T, Itoh T, Sunaga D, Miura T (2012) Eec ff ts of diabetes on myocardial study (Fig. 3) is difficult to easily reconcile with the nega - infarct size and cardioprotection by preconditioning and postcondition- ing. Cardiovasc Diabetol 11:67 tive results in the clinical trials [52, 53]. However, there 4. Takada A, Miki T, Kuno A, Kouzu H, Sunaga D, Itoh T et al (2012) Role of ER is the possibility that the optimal dose of DPP-4 for stress in ventricular contractile dysfunction in type 2 diabetes. PLoS One cardiprotection is lower than that for glycemic control. 7:e39893 Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 15 of 16 5. Matsui Y, Takagi H, Qu X, Abdellatif M, Sakoda H, Asano T et al (2007) 26. Jing Yin J, Bo Li Y, Ming Cao M, Wang Y (2013) Liraglutide improves the sur- Distinct roles of autophagy in the heart during ischemia and reperfu- vival of INS-1 cells by promoting macroautophagy. Int J Endocrinol Metab sion: roles of AMP-activated protein kinase and Beclin 1 in mediating 11:184–190 autophagy. Circ Res 100:914–922 27. Shigeta T, Aoyama M, Bando YK, Monji A, Mitsui T, Takatsu M et al (2012) 6. Nakai A, Yamaguchi O, Takeda T, Higuchi Y, Hikoso S, Taniike M et al Dipeptidyl peptidase-4 modulates left ventricular dysfunction in chronic (2007) The role of autophagy in cardiomyocytes in the basal state and in heart failure via angiogenesis-dependent and -independent actions. response to hemodynamic stress. Nat Med 13:619–624 Circulation 126:1838–1851 7. Zhai P, Sciarretta S, Galeotti J, Volpe M, Sadoshima J (2011) Differential 28. Ravassa S, Barba J, Coma-Canella I, Huerta A, López B, González A et al roles of GSK-3β during myocardial ischemia and ischemia/reperfusion. (2013) The activity of circulating dipeptidyl peptidase-4 is associated with Circ Res 109:502–511 subclinical left ventricular dysfunction in patients with type 2 diabetes 8. Kanamori H, Takemura G, Goto K, Maruyama R, Tsujimoto A, Ogino A et al mellitus. Cardiovasc Diabetol 12:143 (2011) The role of autophagy emerging in postinfarction cardiac remod- 29. Noyan-Ashraf MH, Momen MA, Ban K, Sadi AM, Zhou YQ, Riazi AM et al elling. Cardiovasc Res 91:330–339 (2009) GLP-1R agonist liraglutide activates cytoprotective pathways and 9. Kanamori H, Takemura G, Goto K, Tsujimoto A, Ogino A, Takeyama T et al improves outcomes after experimental myocardial infarction in mice. (2013) Resveratrol reverses remodeling in hearts with large, old myo- Diabetes 58:975–983 cardial infarctions through enhanced autophagy-activating AMP kinase 30. Sauvé M, Ban K, Momen MA, Zhou YQ, Henkelman RM, Husain M et al pathway. Am J Pathol 182:701–713 (2010) Genetic deletion or pharmacological inhibition of dipeptidyl 10. Kubli DA, Zhang X, Lee Y, Hanna RA, Quinsay MN, Nguyen CK et al (2013) peptidase-4 improves cardiovascular outcomes after myocardial infarc- Parkin protein deficiency exacerbates cardiac injury and reduces survival tion in mice. Diabetes 59:1063–1073 following myocardial infarction. J Biol Chem 288:915–926 31. Ussher JR, Baggio LL, Campbell JE, Mulvihill EE, Kim M, Kabir MG et al 11. Maejima Y, Kyoi S, Zhai P, Liu T, Li H, Ivessa A et al (2013) Mst1 inhibits (2014) Inactivation of the cardiomyocyte glucagon-like peptide-1 recep- autophagy by promoting the interaction between Beclin1 and Bcl-2. Nat tor (GLP-1R) unmasks cardiomyocyte-independent GLP-1R-mediated Med 19:1478–1488 cardioprotection. Mol Metab 3:507–517 12. Watanabe T, Takemura G, Kanamori H, Goto K, Tsujimoto A, Okada H et al 32. Hausenloy DJ, Whittington HJ, Wynne AM, Begum SS, Theodorou L, (2014) Restriction of food intake prevents postinfarction heart failure Riksen N et al (2013) Dipeptidyl peptidase-4 inhibitors and GLP-1 reduce by enhancing autophagy in the surviving cardiomyocytes. Am J Pathol myocardial infarct size in a glucose-dependent manner. Cardiovasc 184:1384–1394 Diabetol 12:154 13. Delbridge LM, Mellor KM, Taylor DJ, Gottlieb RA (2015) Myocardial 33. Kouzu H, Miki T, Tanno M, Kuno A, Yano T, Itoh T et al (2015) Excessive autophagic energy stress responses—macroautophagy, mitophagy, and degradation of adenine nucleotides by up-regulated AMP deaminase glycophagy. Am J Physiol Heart Circ Physiol 308:H1194–H1204 underlies afterload-induced diastolic dysfunction in the type 2 diabetic 14. Yamahara K, Yasuda M, Kume S, Koya D, Maegawa H, Uzu T (2013) The role heart. J Mol Cell Cardiol 80:136–145 of autophagy in the pathogenesis of diabetic nephropathy. J Diabetes 34. Miki T, Miura T, Hotta H, Tanno M, Yano T, Sato T et al (2009) Endoplasmic Res 2013:193757 reticulum stress in diabetic hearts abolishes erythropoietin-induced 15. He C, Zhu H, Li H, Zou MH, Xie Z (2013) Dissociation of Bcl-2-Beclin1 myocardial protection by impairment of phospho-glycogen synthase complex by activated AMPK enhances cardiac autophagy and protects kinase-3beta-mediated suppression of mitochondrial permeability transi- against cardiomyocyte apoptosis in diabetes. Diabetes 62:1270–1281 tion. Diabetes 58:2863–2872 16. Sciarretta S, Zhai P, Shao D, Maejima Y, Robbins J, Volpe M et al (2012) 35. Shinmura K, Tamaki K, Sano M, Murata M, Yamakawa H, Ishida H et al Rheb is a critical regulator of autophagy during myocardial ischemia: (2011) Impact of long-term caloric restriction on cardiac senescence: pathophysiological implications in obesity and metabolic syndrome. caloric restriction ameliorates cardiac diastolic dysfunction associated Circulation 125:1134–1146 with aging. J Mol Cell Cardiol 50:117–127 17. Wang B, Yang Q, Sun YY, Xing YF, Wang YB, Lu XT et al (2014) Resveratrol- 36. Jungraithmayr W, De Meester I, Matheeussen V, Baerts L, Arni S, Weder enhanced autophagic flux ameliorates myocardial oxidative stress injury W (2012) CD26/DPP-4 inhibition recruits regenerative stem cells via in diabetic mice. J Cell Mol Med 18:1599–1611 stromal cell-derived factor-1 and beneficially influences ischaemia/rep - 18. Xie Z, Lau K, Eby B, Lozano P, He C, Pennington B et al (2011) Improve- erfusion injury in mouse lung transplantation. Eur J Cardiothorac Surg ment of cardiac functions by chronic metformin treatment is associated 41:1166–1173 with enhanced cardiac autophagy in diabetic OVE26 mice. Diabetes 37. Hotta H, Miura T, Miki T, Togashi N, Maeda T, Kim SJ et al (2010) Angio- 60:1770–1778 tensin II type 1 receptor-mediated upregulation of calcineurin activity 19. Xu X, Kobayashi S, Chen K, Timm D, Volden P, Huang Y et al (2013) underlies impairment of cardioprotective signaling in diabetic hearts. Circ Diminished autophagy limits cardiac injury in mouse models of type 1 Res 106:129–132 diabetes. J Biol Chem 288:18077–18092 38. Zheng Q, Su H, Ranek MJ, Wang X (2011) Autophagy and p62 in cardiac 20. Eguchi M, Kim YH, Kang KW, Shim CY, Jang Y, Dorval T et al (2012) proteinopathy. Circ Res 109:296–308 Ischemia-reperfusion injury leads to distinct temporal cardiac remodeling 39. Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N et al (2005) in normal versus diabetic mice. PLoS One 7:e30450 Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell 21. Hou X, Hu Z, Xu H, Xu J, Zhang S, Zhong Y et al (2014) Advanced glyca- 122:927–939 tion endproducts trigger autophagy in cadiomyocyte via RAGE/PI3K/ 40. Kim M, Platt MJ, Shibasaki T, Quaggin SE, Backx PH, Seino S et al (2013) AKT/mTOR pathway. Cardiovasc Diabetol 13:78 GLP-1 receptor activation and Epac2 link atrial natriuretic peptide secre- 22. Noyan-Ashraf MH, Shikatani EA, Schuiki I, Mukovozov I, Wu J, Li RK et al tion to control of blood pressure. Nat Med 19:567–575 (2013) A glucagon-like peptide-1 analog reverses the molecular pathol- 41. Ussher JR, Drucker DJ (2014) Cardiovascular actions of incretin-based ogy and cardiac dysfunction of a mouse model of obesity. Circulation therapies. Circ Res 114:1788–1803 127:74–85 42. French CJ, Zaman AT, McElroy-Yaggy KL, Neimane DK, Sobel BE 23. Huisamen B, Genade S, Lochner A (2008) Signalling pathways activated (2011) Absence of altered autophagy after myocardial ischemia by glucagon-like peptide-1 (7-36) amide in the rat heart and their role in in diabetic compared with nondiabetic mice. Coron Artery Dis protection against ischaemia. Cardiovasc J Afr 19:77–83 22:479–483 24. Zeng Y, Li C, Guan M, Zheng Z, Li J, Xu W et al (2014) The DPP-4 inhibitor 43. French BA, Kramer CM (2007) Mechanisms of post-infarct left ventricular sitagliptin attenuates the progress of atherosclerosis in apolipoprotein-E- remodeling. Drug Discov Today Dis Mech 4:185–196 knockout mice via AMPK- and MAPK-dependent mechanisms. Cardiovasc 44. Pfeffer MA, Braunwald E (1990) Ventricular remodeling after myocardial Diabetol 13:32 infarction. Experimental observations and clinical implications. Circula- 25. Kim J, Kundu M, Viollet B, Guan KL (2011) AMPK and mTOR regulate tion 81:1161–1172 autophagy through direct phosphorylation of Ulk1. Nat Cell Biol 45. Lee J, Giordano S, Zhang J (2012) Autophagy, mitochondria and oxidative 13:132–141 stress: cross-talk and redox signaling. Biochem J 441:523–540 Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 16 of 16 46. Ide T, Tsutsui H, Kinugawa S, Suematsu N, Hayashidani S, Ichikawa K et al 51. Apaijai N, Pintana H, Chattipakorn SC, Chattipakorn N (2013) Eec ff ts of (2000) Direct evidence for increased hydroxyl radicals originating from vildagliptin versus sitagliptin, on cardiac function, heart rate variability superoxide in the failing myocardium. Circ Res 86:152–157 and mitochondrial function in obese insulin-resistant rats. Br J Pharmacol 47. Zhu H, Tannous P, Johnstone JL, Kong Y, Shelton JM, Richardson JA et al 169:1048–1057 (2007) Cardiac autophagy is a maladaptive response to hemodynamic 52. White WB, Cannon CP, Heller SR, Nissen SE, Bergenstal RM, Bakris GL et al stress. J Clin Invest 117:1782–1793 (2013) Alogliptin after acute coronary syndrome in patients with type 2 48. Lee JE, Yi CO, Jeon BT, Shin HJ, Kim SK, Jung TS et al (2012) & #x03B1;- diabetes. N Engl J Med 369:1327–1335 Lipoic acid attenuates cardiac fibrosis in Otsuka Long-Evans Tokushima 53. Scirica BM, Bhatt DL, Braunwald E, Steg PG, Davidson J, Hirshberg B, fatty rats. Cardiovasc Diabetol 11:111 SAVOR-TIMI 53 Steering Committee and Investigators et al (2013) 49. Burkey BF, Li X, Bolognese L, Balkan B, Mone M, Russell M et al (2005) Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes Acute and chronic effects of the incretin enhancer vildagliptin in insulin- mellitus. N Engl J Med 369:1317–1326 resistant rats. J Pharmacol Exp Ther 315:688–695 50. Dobrian AD, Ma Q, Lindsay JW, Leone KA, Ma K, Coben J et al (2011) Dipeptidyl peptidase IV inhibitor sitagliptin reduces local inflammation in adipose tissue and in pancreatic islets of obese mice. Am J Physiol Endocrinol Metab 300:E410–E421 Submit your next manuscript to BioMed Central and take full advantage of:  Convenient online submission  Thorough peer review  No space constraints or color figure charges  Immediate publication on acceptance  Inclusion in PubMed, CAS, Scopus and Google Scholar  Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cardiovascular Diabetology Springer Journals

Inhibition of DPP-4 reduces acute mortality after myocardial infarction with restoration of autophagic response in type 2 diabetic rats

Download PDF
16 pages

Loading next page...
 
/lp/springer-journals/inhibition-of-dpp-4-reduces-acute-mortality-after-myocardial-P3bB81cpni

References (56)

Publisher
Springer Journals
Copyright
2015 Murase et al.
eISSN
1475-2840
DOI
10.1186/s12933-015-0264-6
Publisher site
See Article on Publisher Site

Abstract

Background: Type 2 diabetes mellitus ( T2DM) worsens the outcome after myocardial infarction (MI). Here, we hypothesized that inhibition of dipeptidyl peptidase-4 (DPP-4) improves survival after MI in T2DM by modifying autophagy in the non-infarcted region of the heart. Methods and results: Under baseline conditions, there was no significant difference between levels of myocardial autophagy marker proteins in OLETF, a rat model of T2DM, and in LETO, a non-diabetic control. However, in contrast to the response in LETO, LC3-II protein and LC3-positive autophagosomes in the non-infarcted region of the myo- cardium were not increased after MI in OLETF. The altered autophagic response in OLETF was associated with lack of AMPK/ULK-1 activation, attenuated response of Akt/mTOR/S6 signaling and increased Beclin-1–Bcl-2 interaction after MI. Treatment with vildagliptin (10 mg/kg/day s.c.), a DPP-4 inhibitor, suppressed Beclin-1–Bcl-2 interaction and increased both LC3-II protein level and autophagosomes in the non-infarcted region in OLETF, though it did not normalize AMPK/ULK-1 or mTOR/S6 signaling. Plasma insulin level, but not glucose level, was significantly reduced by vildagliptin at the dose used in this study. Survival rate at 48 h after MI was significantly lower in OLETF than in LETO (32 vs. 82%), despite similar infarct sizes. Vildagliptin improved the survival rate in OLETF to 80%, the benefit of which was abrogated by chloroquine, an autophagy inhibitor. Conclusions: The results indicate that vildagliptin reduces T2DM-induced increase in post-MI acute mortality pos- sibly by restoring the autophagic response through attenuation of Bcl-2-Beclin-1 interaction. Keywords: Type 2 diabetes, Autophagy, DPP-4 inhibitor, Myocardial infarction, Mortality of the myocardium, often called “diabetic cardiomyo- Background pathy” [3]. Augmentation of contractile function in the Diabetes mellitus (DM) is associated with poor outcome non-infarcted region is crucial in acute compensation for after acute myocardial infarction (MI) even in the era of the lost contraction in the infarct region. u Th s, pre-exist - reperfusion therapy [1, 2]. The poor outcome after MI in patients with DM has been explained by extensive ather ing contractile dysfunction in patients with DM, if any, potentially compromises such a compensatory response osclerotic lesions, increased myocardial susceptibility to after MI and increases mortality. Recently, we confirmed ischemia/reperfusion injury, and contractile dysfunction DM-induced increase in acute mortality after MI in an animal model of obese type 2 DM (T2DM), Otsuka- *Correspondence: [email protected] 1 Long-Evans-Tokushima Fatty rats (OLETF). The mortal - Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, South-1, West-16, Chuo-ku, ity rate at 48 h after MI was significantly higher in OLETF Sapporo 060-8543, Japan due to lethal heart failure than in non-diabetic control Full list of author information is available at the end of the article © 2015 Murase et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 2 of 16 rats, Long-Evans-Tokushima-Otsuka rats (LETO), while To examine the hypotheses, we used OLETF as a model infarct sizes were similar in OLETF and LETO [4]. Inter- of T2DM in the present study as in previous studies estingly, OLETF had preserved ventricular contractility [4, 33, 34]. Results of the experiments showed that a with mildly impaired relaxation under baseline condi- DPP-4 inhibitor, vildagliptin, significantly improved tions [4]. The findings indicate that the altered response survival after MI in OLETF and that the protective of the non-infarcted myocardium, rather than base- effect of vildagliptin was closely associated with resto - line ventricular contractility, contributes to the T2DM- ration of the autophagic response in the non-infarcted induced increase in acute mortality after MI. myocardium. Autophagy is a cellular process of lysosome-medi- ated degradation of cytoplasmic components or dam- Methods aged organelles in response to cellular stress [5–13]. It Animals and experimental protocol has been suggested that autophagy has roles in critical The present study was conducted in strict accordance adaptive mechanisms in the heart under hemodynamic with the Guideline for the Care and Use of Labora- stress conditions such as pressure overload [6] and loss tory Animals published by the US National Institute of of a contractile region by MI [5, 7–11]. In the case of MI, Health (NIH publication No. 85-23, revised 1996) and autophagic activity was augmented in the non-infarcted was approved by the Animal Use Committee of Sap- remote area and border area but not in the infarcted poro Medical University. Protocols of the experiments myocardium after MI, and the autophagic activity pro- are summarized in Fig.  1. Male LETO and OLETF at gressively increased in the remote area during a 3-week ages of 25–30  weeks were used in all experiments. period after MI [8, 9]. The increase in autophagic activity LETO were pretreated with saline, vildagliptin (10  mg/ has protective effects against remodeling and dysfunc - kg/day), a DPP-4 inhibitor, or exenatide (10  μg/kg/day), tion of the ventricle after MI [8, 9]. However, autophagy a GLP-1 receptor agonist, for 2  weeks, or with chloro- has been shown to be impaired by DM in non-cardiac quine (10 mg/kg/day), an autophagy inhibitor [9, 12, 35], [14] and cardiac tissues [15–20], though advanced gly- for 1 week. Vildagliptin was kindly provided by Novartis cogen endproducts reportedly activate autophagy in Pharma AG (Basel, Switzerland). The pharmacological cardiomyocytes [21]. Whether the increase in myocar- agents were administered via osmotic minipumps (Alzet, dial autophagic activity after MI is impaired by DM and Cupertino, CA, USA), not via drinking water contain- whether such an impairment, if any, is treatable by phar- ing the agents, because the amount of water rats drink macological agents have not been clarified. per day is not consistent. The dose of vildagliptin was We hypothesized that autophagic response of the selected on the basis of a result in a previous report [36], myocardium to MI is impaired by T2DM and that inhi- and we confirmed that this dose of vildagliptin signifi - bition of dipeptidyl peptidase-4 (DPP-4) would attenu- cantly increased the serum level of GLP-1 (see “Results”). ate the T2DM-induced increase in post-MI mortality OLETF also received saline, vildagliptin, exenatide or by restoring the autophagic response. The rationale chloroquine as did LETO, and an additional group of for the hypotheses is four-fold. First, a significant asso - OLETF received both vildagliptin and chloroquine. In ciation of preserved ventricular function and activa- separate groups of rats, blood samples were collected via tion of autophagy has been demonstrated for different the carotid artery under anesthesia 12  h after fasting to types of cardiac stress in non-diabetic animals [5–11]. examine the effects of these pharmacological agents on Second, DM impairs intracellular signaling mecha- metabolic parameters. nisms relevant to autophagy, including PI3K-Akt sign- aling, in the myocardium [3]. Third, activation of the Oral glucose tolerance test glucagon-like peptide-1 (GLP-1) receptor or treatment An oral glucose tolerance test (OGTT) was performed in with a DDP-4 inhibitor triggers AMP-activated protein LETO treated with a vehicle and OLETF treated with a kinase (AMPK) signaling [22, 23, 24], which facilitates vehicle or vildagliptin (10  mg/kg/day) for 2  weeks. After autophagy [5, 9, 15, 18, 25]. In fact, a GLP-1 analog, fasting for 12  h, rats were administered glucose (2  g/kg liraglutide, has been shown to promote autophagy in body weight) by gavage, and blood glucose and insulin non-cardiac cells [26]. Fourth, it has been reported levels before and after glucose administration were meas- that the activity of circulating DPP-4 is associated with ured by using a Glutest-mint (Sanwa Kagaku Kenkyusho, left ventricular dysfunction in patients [27, 28]. Con- Nagoya, Japan) and a rat insulin RIA kit (Linco Research versely, inhibition of DPP-4 and use of a GLP-1 analog Inc, St. Charles, MO, USA), respectively. Blood for GLP-1 prevented cardiomyopathy, improved cardiac function assay using a GLP-1 (Active) ELISA kit (Millipore) was and post-MI survival rate, and attenuated ventricular collected before glucose administration in sampling tubes remodeling in DM and non-DM animals [27, 29–32]. containing a DPP-4 inhibitor. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 3 of 16 Fig. 1 Experimental protocols. Experimental protocols for survival study (a) and cardiac tissue sampling (b). Vilda vildagliptin, Exe exenatide, CQ chloroquine, LV left ventricle. Echocardiography Cardiac tissue sampling after MI Echocardiography was performed before induction of MI Since the mortality rate at 24–48  h after MI was high as previously reported [4]. in OLETF at ages of 25–30  weeks [4], myocardial tis- sue sampling for biochemical analyses and immunohis- Induction of MI and mortality monitoring tochemistry was performed at 12  h after MI. Rats were Rats were prepared for induction of MI as in our previous anesthetized and ventilated, and blood pressure and study [4]. In brief, rats were anesthetized with sodium heart rate were monitored by a catheter placed in the pentobarbital (40  mg/kg, i.p.), and the level of anesthe- carotid artery. The chest was re-opened and the hearts sia was continuously monitored during the experiment were excised and immediately immersed in ice-cold and an additional dose of pentobarbital was administered saline. The myocardium in the non-infarcted region was when necessary. Rats were then intubated and ventilated quickly excised in the saline, frozen in liquid nitrogen, with a rodent respirator (model 683, Harvard Apparatus, and stored at −80°C until use for biochemical and histo- South Natick, MA, USA). After left thoracotomy, a mar- logical analyses. ginal branch of the left coronary artery was permanently ligated by using a 5–0 silk thread to induce MI. We used Immunohistochemistry a permanent occlusion model of MI to avoid the possi- Frozen heart tissues were embedded in OCT compound bility that pharmacological pretreatments modify infarct (Tissue-Tek) and snap-frozen in liquid nitrogen. After the size and induce an inter-group difference in mechanical tissues had been sectioned at 8  μm in thickness with a stress on the non-infarcted region. The surgical wounds cryostat at −20°C, the sections were incubated with rab- were repaired and the rats were returned to their cages. bit polyclonal anti-LC3 antibody (MBL, PM036, 1:250) in All rats were allowed ad-lib access to water but restricted PBS containing 1% BSA and 0.3% triton X-100 overnight from food for 12 h. Survival rate of rats was determined at 4°C. The samples were then incubated with an Alexa at 24 and 48  h after MI. Rats that had survived at 48  h Fluor 488 anti-rabbit IgG antibody (Invitrogen) for 1  h after MI were euthanized by a pentobarbital overdose at room temperature. After nuclei had been stained with and heart tissue was excised and fixed in 10% formalde - Hoechst33342 (Dojindo, Kumamoto, Japan), samples hyde for infarct size analysis. were mounted on slides for image analysis. Fluorescence Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 4 of 16 images were obtained using a FLoid Cell Imaging Sta- CA, USA). First-strand cDNA was synthesized using a tion (Life Technologies). The number of LC3 dots was SuperScript VILO cDNA Synthesis Kit (Life Technolo- counted and analyzed in 40 randomly selected fields from gies). DNA amplification was performed in StepOne five hearts in each group. (Life Technologies). Analyses of B-type natriuretic pep- tide (BNP) and β-actin mRNA levels were performed by Immunoblotting using Taqman gene expression assays (Rn00676450_g1 Frozen tissue samples were homogenized in ice-cold Nppb and Rn00667869_m1 Actb, respectively). For buffer (CelLytic MT Cell Lysis Reagent) including pro- p62 and 18S, we used Power SYBR PCR Master Mix tease and phosphatase inhibitor cocktails (Nacalai Tesque, (Applied Biosystems, Inc) and the following oligonucle- Inc., Kyoto, Japan). The homogenate was centrifuged at otide primers: for rat p62, 5′-ATCAGCCTCTGGTGG 15,000g for 15 min at 4°C to obtain the supernatant. Equal GAGAT-3′ and 5′-CCCATCCACAGGTGAACTCC-3′; amounts of protein were analyzed by immunoblot assays for rat 18S, 5′-CGGACAGGATTGACAGATTG-3′ using specific antibodies (see Additional file  1: Table S1). and 5′-CAAATCGCTCCACCAACTAA-3′. All assays Intensities of individual bands were quantified by using were performed in duplicate and by the standard curve Image J software (National Institutes of Health). method using serial cDNA dilution. Beclin‑1–Bcl‑2 interaction Statistical analyses Frozen myocardial tissue samples were homogenized in Data are presented as mean ± SEM. Differences between ice-cold buffer containing 20 mM Tris (pH 7.4), 137 mM treatment groups were assessed by one-way analysis of NaCl, 10% glycerol, 0.3% CHAPS, and 2 mM EDTA sup- variance (ANOVA) followed by the Student–Newman– plemented with protease and phosphatase inhibitor cock- Keuls post hoc test for multiple comparisons. Differences tails. The homogenates were then centrifuged at 15,000g in time course between two groups in OGTT were ana- for 15  min to obtain supernatants. After quantification lyzed by 2-way ANOVA for repeated measures followed of protein concentration, the lysates were incubated for by the Student–Newman–Keuls post hoc test for multi- 30  min with 40  μl of protein A magnetic beads (New ple comparisons. Survival rates after MI were compared England Biolads, Ipswich, MA, USA) to remove endog- by Kaplan–Meier curves and log-rank statistics. For all enous IgG. Equal amounts (2,000  μg) of lysates were tests, p < 0.05 was considered statistically significant. incubated with either 6  μg of rabbit anti-Beclin-1 anti- body or normal rabbit IgG overnight at 4°C. The mixture Results was then incubated with 50 μl of fresh beads for 1 h. The Metabolic profiles beads were washed three times with PBS containing pro- Data for metabolic profiles of LETO and OLETF and tease inhibitor cocktail and re-suspended in SDS sample the effects of vildagliptin, exenatide and chloroquine on loading buffer followed by denaturation. Immunoprecipi - metabolic parameters are shown in Table  1. In LETO, tated proteins were analyzed by Western blotting initially body weight (531 ± 7 g), fasting blood glucose level and with rabbit anti-Bcl-2 antibody and then with mouse serum total cholesterol level were not changed by treat- monoclonal anti-Beclin-1 antibody after stripping. ment with vildagliptin or exenatide. OLETF had larger body weight (627  ±  14  g) and higher fasting blood glu- mRNA quantification cose level than those of LETO (Table  1), as we previ- Total RNA was isolated from myocardial tissues by using ously reported [4, 34, 37]. Treatment with vildagliptin or an RNeasy Fibrous Tissue Mini Kit (Qiagen, Valencia, exenatide at the dose used in the present study did not Table 1 Metabolic parameters after treatment LETO OLETF Vehicle Vildagliptin Exenatide Vehicle Vildagliptin Exenatide Vildagliptin +  Chloroquine Blood glucose (mg/dl) 108 ± 5 106 ± 6 117 ± 12 205 ± 25* 201 ± 13* 251 ± 17* 197 ± 22* † † Serum insulin (ng/ml) 2.5 ± 0.5 1.3 ± 0.2 3.1 ± 0.9 5.8. ± 0.9* 2.5 ± 0.5 1.3 ± 0.8 4.1 ± 0.9 Serum TC (mg/dl) 82 ± 5 85 ± 5 95 ± 4 117 ± 3 120 ± 6 109 ± 12 96 ± 7 Values are mean ± SEM. N = 4–11. TC total cholesterol. * P < 0.05 vs. LETO vehicle; P < 0.05 vs. OLETF vehicle. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 5 of 16 change body weight and blood glucose level in OLETF. LV function at 12 h after coronary ligation also, but clear Serum insulin level was significantly higher in OLETF cardiac images for quantitative assessment could not than in LETO, and both vildagliptin and exenatide be obtained at that time point because of intra-thoracic decreased insulin level in OLETF without reduction in air, fluid and chest wall damage related to open-chest blood glucose level (Table  1). Treatment of OLETF with surgery. chloroquine in addition to vildagliptin affected neither insulin level nor blood glucose level compared to the lev- Vildagliptin improved survival in OLETF in a els in those treated with vildagliptin alone. Despite lack chloroquine‑sensitive manner of effects on blood glucose level, vildagliptin increased The survival rate during a period of 48  h after MI in serum level of the active form of GLP-1; GLP-1 level was LETO was 82%, which was comparable with the rate in below the detection range (<2.0 pmol/L) in 4 of a total of our previous study [4]. Pretreatment with vildagliptin or five samples from the vehicle-treated OLETF, but it was exenatide for 2  weeks before MI did not affect the sur - 5.0 ± 0.9 pmol/L in vildagliptin-treated OLETF (Fig. 2a). vival rate in LETO (86 and 80%, respectively) (Fig.  3a). In OGTTs, levels of blood glucose and serum insu- As in our previous study [4], the survival rate was sig- lin before and after glucose administration were higher nificantly lower in OLETF (32%, Fig.  3b) than that in in vehicle-treated OLETF than in LETO (Fig.  2b, c). LETO. Pretreatment of OLETF with vildagliptin signifi - Treatment with vildagliptin did not change glucose lev- cantly increased the survival rate (80%) to a level similar els before and after glucose administration in OLETF to that in LETO (Fig. 3b). Exenatide treatment tended to (Fig.  2b). On the other hand, insulin level was signifi - improve the survival rate in OLETF, but the difference did cantly lower in the vildagliptin-treated OLETF than in not reach statistical significance. In chloroquine-treated the vehicle-treated OLETF (Fig. 2c). LETO, the survival rate (62%, Fig. 3c) tended to be lower than that in vehicle-treated LETO (Fig. 3a). Survival rates Cardiac function was not modified by vildagliptin or of chloroquine-treated OLETF were similar, regardless exenatide at baseline of vildagliptin treatment or no vildagliptin treatment (33 Before induction of MI, heart rate was lower and left and 39%, respectively, Fig. 3c), to the survival rate in vehi- ventricular (LV) dimension was larger in OLETF than cle-treated OLETF (Fig. 3b). Autopsies of rats died within in LETO, though there were no differences in LV ejec - 48 h after MI revealed no case of cardiac rupture. tion fraction and fractional shortening between the two Infarct sizes 48  h after the permanent coronary occlu- groups (Table  2). Treatment with vildagliptin, exenatide, sion were 35–40% of the left ventricle and were compa- or chloroquine did not alter these echocardiographic rable among the treatment groups (Fig.  3d–f ). In LETO, parameters in either LETO or OLETF. We tried to assess hemodynamic parameters (heart rate and blood pressures) Fig. 2 Serum active GLP-1 level and oral glucose tolerance test in LETO and OLETF. a Serum level of the active form glucagon-like peptide (GLP-1) in OLETF treated with the vehicle or vildagliptin. Values in 4 of 5 samples from the vehicle-treated group were under the detection limit (<2.0 pmol/L) of the assay. Blood glucose (b) and serum insulin (c) during an oral glucose tolerance test (2 g glucose per kg body weight) in LETO (triangle) and OLETF treated with the vehicle (closed circle) or vildagliptin (open circle). *p < 0.05 vs. LETO and p < 0.05 vs OLETF vehicle at each time point. N = 5 in each group. Vilda vildagliptin. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 6 of 16 Table 2 Echocardiographic data at baseline LETO OLETF Vehicle Vildagliptin Exenatide Chloroquine Vehicle Vildagliptin Exenatide Vildagliptin +  Chloroquine HR (bpm) 384 ± 6 386 ± 7 385 ± 4 375 ± 6 312 ± 7* 316 ± 4* 315 ± 6* 324 ± 9* LVEF (%) 69.4 ± 1.5 70.8 ± 1.3 68.4 ± 1.2 73.0 ± 1.3 65.4 ± 1.5 67.7 ± 0.7 68.5 ± 1.3 69.9 ± 1.3 FS (%) 35.0 ± 1.3 35.9 ± 1.0 34.0 ± 0.8 37.6 ± 1.5 32.1 ± 1.1 33.4 ± 0.5 34.4 ± 1.0 35.2 ± 1.1 IVST (mm) 1.72 ± 0.04 1.69 ± 0.02 1.76 ± 0.03 1.75 ± 0.04 1.81 ± 0.04 1.74 ± 0.03 1.73 ± 0.03 1.85 ± 0.02 PWT (mm) 1.66 ± 0.06 1.73 ± 0.05 1.64 ± 0.02 1.76 ± 0.06 1.85 ± 0.05 1.79 ± 0.04 1.83 ± 0.06 1.84 ± 0.03 LVEDD (mm) 7.04 ± 0.12 6.71 ± 0.19 7.11 ± 0.12 6.68 ± 0.12 8.05 ± 0.18* 7.69 ± 0.10* 7.88 ± 0.12* 7.65 ± 0.15* LVESD (mm) 4.60 ± 0.15 4.34 ± 0.13 4.69 ± 0.09 4.16 ± 0.14 5.49 ± 0.17* 5.08 ± 0.10* 5.17 ± 0.12* 4.94 ± 0.07* LVEDV (ml) 0.80 ± 0.04 0.75 ± 0.04 0.83 ± 0.04 0.69 ± 0.03 1.17 ± 0.07* 1.02 ± 0.04* 1.10 ± 0.04* 1.00 ± 0.06* LVESV (ml) 0.26 ± 0.02 0.22 ± 0.01 0.26 ± 0.02 0.19 ± 0.02 0.41 ± 0.03* 0.37 ± 0.04* 0.35 ± 0.02* 0.30 ± 0.01* Values are mean ± SEM. N = 12–33. HR heart rate, bpm beats per minute, LVEF left ventricular ejection fraction, FS fractional shortening, IVST interventricular septal thickness, PWT posterior wall thickness, LVEDD left ventricular end-diastolic dimension, LVESD left ventricular end-systolic dimension, LVEDV left ventricular end-diastolic volume, LVESV left ventricular end- systolic volume. * P < 0.05 vs. LETO Vehicle. vinculin ratio was significantly increased after MI in LETO, at 12  h after MI in immunoblot experiments were com- but such a response was not observed in OLETF (Fig.  4a). parable between the treatment groups (Table  3). Heart Changes in LC3-II/LC3-I ratio were similar to those in rates before and after MI and blood pressure at 12 h after LC3-II/vinculin, though the difference did not reach statis MI were lower in OLETF than in LETO, but treatment - with either vildagliptin or exenatide did not significantly tical significance. Vildagliptin significantly increased LC3- change these hemodynamic parameters in OLETF. Taken II level and tended to increase LC3-II/LC3-I after MI in together, these results indicate that difference in infarct OLETF (Fig. 4b). Although p62 protein is often used as an size or hemodynamic response does not underlie the dif index of autophagic flux [38], its level was not changed by MI in either LETO or OLETF (Fig. 4c). Neither vildagliptin ference in survival rate among the treatment groups. nor exenatide changed p62 protein level after MI in OLETF To examine whether the loading condition in the non- (Fig. 4d). mRNA levels of p62 in the myocardium were also infarcted region after MI was modulated by vildagliptin, we measured BNP mRNA level in the remote myocar similar in LETO and OLETF regardless of MI and treat - - dium 12  h after MI (Fig.  3g). BNP mRNA level was sig- ment with vildagliptin or exenatide (data not shown). nificantly increased after MI in both LETO and OLEFT, Autophagic activity in the non-infarcted region of but such an increase in BNP mRNA after MI was not the myocardium was assessed also by immunostaining of autophagosomes with anti-LC3 antibody (Fig.  5). In observed in OLETF treated with vildagliptin (Fig.  3g). Since augmented adrenergic activity is one of the features LETO, the number of LC3-positive dots was significantly of heart failure, we determined phosphorylation of vas increased after MI by 66%, but such an increase was not observed in OLETF. Not only vildagliptin but also exena odilator-stimulated phosphoprotein (VASP) at Ser157, - a protein kianse A phosphorylation site, in the myocar- tide significantly increased the number of LC3 dot after dium. The levels of phospho-VASP after MI were simi - MI in OLETF (Fig.  5), indicating that vildagliptin and lar in LETO and OLETF, but treatment of OLETF with exenatide restored autophagic induction in the non- infarcted myocardium after MI in OLETF. vildagliptin significantly decreased phospho-VASP levels (Fig.  3h). These findings suggest that vildagliptin attenu - ated both ventricular overloading and augmented adren- Activation of AMPK in response to MI was impaired ergic drive after MI in OLETF. in OLETF Since AMPK is known to activate autophagy by phos- Autophagic response in the non‑infarcted region of the phorylating ULK1 at Ser317 [25], we assessed AMPKα myocardium after MI was impaired in OLETF phosphorylation at Thr172 and phosphorylation of Marker molecules of autophagic activities in the non- its downstream target proteins, acetyl-CoA carboxy- infarcted region of the heart after MI are shown in Fig.  4. lase (ACC) and ULK1, in the non-infarcted region after LC3-II levels in the heart without infarction (i.e., sham MI (Fig.  6a–c). Levels of phospho-Thr172-AMPKα, operation) were similar in LETO and OLETF. LC3-II/ phospho-Ser79-ACC and phospho-Ser317-ULK1 were Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 7 of 16 Fig. 3 Eec ff ts of vildagliptin, exenatide, and chloroquine on survival after MI. Kaplan–Meier survival analysis of LETO (a), OLETF (b), and rats treated with chloroquine (c) after left coronary artery occlusion. *p < 0.05 vs. Vehicle-treated group. Infarct size measured at 48 h after MI in LETO (d), OLETF (e), and rats treated with chloroquine (f). g Quantification of BNP mRNA levels normalized to β-actin in the non-infarcted myocardium sampled 12 h after MI. N = 3–6 in each group. h Summary data of immunoblotting for phospho-Ser157 VASP in samples from LETO, vehicle- or vildagliptin- treated OLETF after MI. N = 9–12 in each group. Vilda vildagliptin, Exe exenatide, CQ chloroquine. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 8 of 16 Fig. 4 Vildagliptin restored autophagic induction in the non-infarcted area after MI in OLETF. a Representative images of Western blotting for LC3 protein (left) and summary data of LC3-II level and LC3-II/LC3-I ratio (right) in samples from sham-operated hearts (Sham) or the non-infarcted myocardium after MI in LETO and OLETF. b Representative images of Western blotting for LC3 protein (left) and summary data of LC3-II level and LC3-II/LC3-I ratio (right) in samples from the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). c Representative blots (left) and summary data (right) of Western blotting for p62 protein in samples from sham-operated hearts (Sham) or the non-infarcted myocardium after MI in LETO and OLETF. d Representative blots (left and middle) and summary data (right) of Western blotting for p62 protein in OLETF treated with the vehicle, Vilda, or Exe after MI. N = 8–10 in each group. *p < 0.05. a.u. arbitrary units, NS not significant. Akt/mTORC1 activity after MI was attenuated in the OLETF significantly increased at 12 h after MI in LETO. In con - hearts trast, such responses of AMPKα, ACC and ULK1 were not detected in OLETF. Although vildagliptin and exena Alterations in Akt/mTORC1 signaling, a negative regu- latory mechanism of autophagy [25], in OLETF were tide restored the increase in LC3-positive autophago- examined by immunoblotting. The level of Ser473-Akt somes after MI in OLETF (Fig. 5), neither agent restored phosphorylation of Thr172-AMPKα, Ser79-ACC or phosphorylation was significantly elevated after MI in Ser317-ULK1 after MI in OLETF (Fig. 6d–f ). the non-infarcted myocardium in LETO (Fig.  7a). The Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 9 of 16 Table 3 Hemodynamic data 12 h after myocardial infarction LETO OLETF Sham MI Sham MI Vehicle Vehicle Vildagliptin Exenatide Vehicle Vehicle Vildagliptin Exenatide SBP (mmHg) 119 ± 5 105 ± 6 115 ± 4 114 ± 5 104 ± 6 90 ± 3* 96 ± 4* 93 ± 5* DBP (mmHg) 88 ± 4 81 ± 6 85 ± 4 82 ± 6 73 ± 7 64 ± 3* 65 ± 4* 64 ± 4* Heart rate (bpm) 423 ± 7 401 ± 8 410 ± 11 400 ± 10 332 ± 13* 321 ± 6* 331 ± 7* 326 ± 9* Values are measn ± SE. N = 8–15. SBP systolic blood pressure, DBP diastolic blood pressure, bpm beats per minute. * P < 0.05 vs. LETO Sham. phosphorylation was associated with increases in lev- els of phospho-mTOR at Ser2448 and phospho-S6 at Ser235/236 (Fig.  7b, c). However, such an activation of Akt after MI was not observed in OLETF; phospho-Akt levels were similar in sham-operated and MI-induced OLETF (Fig.  7a). Phospho-mTOR and phospho-S6 lev- els were increased after MI in OLETF, but their levels remained significantly lower than those in LETO (Fig.  8b, c). Neither vildagliptin nor exenatide increased phospho- rylation of Ser473-Akt, Ser2448-mTOR, and Ser235/236- S6 after MI in OLETF (Fig. 7d–f ). Beclin‑1–Bcl‑2 interaction was enhanced in OLETF Since vildagliptin improved the response of autophagic activity to MI in OLETF (Figs.  4, 5) without normaliza- tion of AMPK phosphorylation and Akt/mTORC1 sign- aling (Figs.  6, 7), we examined whether Beclin-1–Bcl-2 interaction is modified in OLETF or by vildagliptin treat - ment. Beclin-1 is an essential component for activation of autophagy, and its autophagy-promoting activity is inhibited by binding to an anti-apoptotic protein, Bcl-2 [39]. There were no significant differences in Beclin-1 and Bcl-2 protein levels between LETO and OLETF regard- less of MI (Fig. 8a), and neither vildagliptin nor exenatide changed levels of these proteins in MI-induced OLETF (Fig. 8b). However, Beclin-1–Bcl-2 interaction was signif- icantly augmented in OLETF with MI compared to that in OLETF with a sham operation, and their interaction was attenuated by vildagliptin (Fig. 8c–e). Discussion In the present study, treatment with vildagliptin at a Fig. 5 Immunofluorescent analysis of LC3 protein in the non- dose that did not lower plasma glucose level significantly infarcted myocardium after MI. a Representative immunofluores- improved survival of OLETF after acute MI (Figs.  2, 3). cence images of LC3 protein in LV sections from the non-infarcted By using telemetric monitoring of heart rate and blood myocardial area after MI. The image without the primary antibody did not show any green dots. b Quantification of LC3 dots per field pressure, we previously demonstrated that increased (568 µm × 426 µm). A total of 40 fields from five hearts were analyzed mortality during the acute phase after MI in OLETF is in each group. *p < 0.05. Vilda vildagliptin, Exe exenatide, NS not due to progressive heart failure but not lethal arrhyth- significant. mia [4]. Although vildagliptin did not modify ventricular Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 10 of 16 Fig. 6 Analysis of the AMPK/ULK1 pathway. Representative images (left) and summary data (right) of Western blotting for phospho-Thr172 and total AMPKα (a), phospho-Ser79 and total acetyl-CoA carboxylase (ACC) (b), and phospho-Ser317-ULK1 (c) in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. Representative blots (left) and summary data (right) of Western blotting for phospho-Thr172 and total AMPKα (d), phospho-Ser79 and total acetyl-CoA carboxylase (ACC) (e), and phospho-Ser317-ULK1 (f) in the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). N = 9–10 in each group. *p < 0.05. a.u. arbitrary units, NS not signifi- cant. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 11 of 16 Fig. 7 Analysis of Akt/mTORC1 activity. Representative blots (left) and summary data (right) of Western blotting for phospho-Ser473 and total Akt (a), phospho-Ser2448 and total mTOR (b), and phospho-Ser235/236 and total S6 (c) in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. Representative blots (left) and summary data (right) of Western blotting for phospho-Ser473 and total Akt (d), phospho-Ser2448 and total mTOR (e), and phospho-Ser235/236 and total S6 (f) in the non-infarcted myocardium after MI in OLETF treated with the vehicle, vildagliptin (Vilda), or exenatide (Exe). N = 9–10 in each group. *p < 0.05. a.u. arbitrary units. Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 12 of 16 Fig. 8 Eec ff t of vildagliptin on Beclin-1/Bcl-2 interaction after MI in OLETF. a Representative images (left) and summary data (right) of Western blotting for Beclin-1 and Bcl-2 in samples from sham-operated heats or the non-infarcted myocardium after MI in LETO and OLETF. b Eec ff ts of vildagliptin (Vilda) and exenatide (Exe) on Beclin-1 and Bcl-2 levels in the non-infarcted myocardium after MI in OLETF. Representative immunoblot- ting images (left) and summary data (right) are shown. c Myocardial lysates were immunoprecipitated (IP) with anti-Beclin-1 antibody or rabbit IgG followed by immunoblotting with anti-Bcl-2 and Beclin-1 antibodies. d, e Beclin-1/Bcl-2 interaction was increased after MI in OLETF, which was attenuated by vildagliptin (Vilda). N = 5 in each group. *p < 0.05. a.u. arbitrary units. function under baseline conditions (Table  2), it sup- but a  recent study has shown that a GLP-1 receptor pressed MI-induced upregulation of BNP expression and agonist provides cardioprotection by a mechanism inde- cardiac adrenergic activity in OLETF (Fig.  3g, h). Thus, pendent of the GLP-1 receptor in the cardiomyocyte [31]. suppression of heart failure progression after MI is the On the other hand, DPP-4 is involved in degradation of most likely explanation for reduction in acute mortality multiple peptides such as substance P and stromal cell- after MI in OLETF by vildagliptin. derived factor-1 [27, 41], and these properties of DPP-4 Serum active GLP-1 level was elevated by vildagliptin inhibitors might underlie the differences in survival rate (Fig.  2a), suggesting that GLP-1 mediated the improved (Fig.  3b) and changes in LC3-II protein level (Fig.  4b) survival by vildagliptin. On the other hand, treatment after MI between vildagliptin-treated and exenatide- with a GLP-1 analog, exenatide, tended to improve treated OLETF. post-MI survival in OLETF, but the effect was statisti - In the literature, a study by French et  al. [42] is the cally insignificant. We do not have a clear explanation only study in which the changes in autophagy after MI for the different outcomes in the vildaglitpin-treated in diabetic mice and non-diabetic mice were compared. and exenatide-treated groups. However, a difference in In that study, autophagic activity was not increased after the mechanism of cardioprotection is a possibility. The MI not only in the diabetic heart but also in the non-dia- GLP-1 receptor is localized in the cardiomyocyte [40], betic heart. The negative results are in contrast to results Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 13 of 16 of several studies showing that autophagic activity was The results of the present study supported the notion increased after MI in the healthy control myocardium [5, that activation of autophagy by vildagliptin during the 8–12]. The reason why French et  al. could not detect an acute phase after MI contributed to improved survival alteration of autophagy after MI even in the non-diabetic in OLETF. However, roles of autophagy in the heart may myocardium is unclear, but use of entire risk zone tissue be different depending on the phase and type of cardiac consisting of infarcted and non-infarcted cells might have stress. Matsui et  al. [5] showed that autophagy is pro- obscured changes in autophagy in the viable myocardium tective for cardiomyocyte survival during ischemia but after MI. is rather detrimental during reperfusion. Zhu et  al. [47] We focused on the non-ischemic region of the showed that sustained activation of autophagy during infarcted heart since that region plays a crucial role pressure overload is detrimental to cell survival. They in compensation for the lost function of the infarcted showed that cardiac function after thoracic aortic band- region and undergoes adaptive and maladaptive post-MI ing was preserved in beclin-1 heterozygous knockout remodeling [43, 44]. There was no significant difference mice, whereas cardiomyocyte-specific overexpression in LC3-II or p62 levels or LC3-positive autophagosomes of beclin-1 worsened cardiac function. In contrast, sus- between LETO and OLETF under baseline conditions tained elevation of autophagy may be protective for (i.e., sham-operated groups in Figs.  4, 5), suggesting that post-infarcted ventricular function and remodeling. Sup- autophagic activities were similar in the diabetic myocar- pression of autophagy by bafilomycin A1 or chloroquine dium and non-diabetic myocardium under non-stressed has been shown to exacerbate cardiac function after MI, conditions. However, there was a significant difference but activation of autophagy by an mTORC1 inhibitor, between LETO and OLETF in autophagic response rapamycin, or by caloric restriction was protective [8, 12]. after MI. In LETO, an increase in autophagic activity Maejima et  al. [11] also reported that cardiac function was observed in the non-infarct region at 12  h after MI at 6  weeks after MI was worsened in beclin-1 heterozy- (Figs. 4, 5). A protective role of the increase in autophagic gous knockout mice. Since the survival rate of OLETF at activity after MI has been demonstrated by findings that 48 h after MI was only 32% (Fig.  3b), we did not include inhibition of autophagy by bafilomycin A or genetic dele - assessment of autophagic activity at the later phase after tion of beclin-1 aggravated remodeling and dysfunction MI in OLETF in the present study. of the ventricle after MI [8, 11]. Importantly, the mor- Streptozotocin-induced diabetes and high-fat diet tality rate after MI in OLETF was not further increased have been shown to reduce LC3-II in the myocardium, by inhibiting autophagy with chloroquine (Fig.  3). These which was associated with suppressed phosphorylation results support the notion that impaired autophagic of AMPK, a positive regulator of autophagy [16, 18]. In response in the non-infarcted region of the infarcted contrast, LC3-II, p62 or AMPK phosphorylation in the heart contributes to increase in acute mortality after MI heart without MI was not different between OLETF and in OLETF. LETO in this study. However, Lee et  al. [48] reported There are two possible mechanisms for suppression 50% reduction in AMPK phosphorylation in the myo- of heart failure by autophagy: reduction of reactive cardium of OLETF at the age of 28  weeks. A possible oxygen species (ROS) production and improvement of explanation for the discrepant results is more advanced myocardial energy status. Damaged organelles partici- stage of T2DM in OLETF in the study by Lee et al. [48]. pating in ROS generation, including mitochondria, are Despite similar ages, OLETF in their study had slightly sequestrated and removed by the autophagic process, larger body weight compared with that in this study and and autophagy plays a role in suppression of ROS gen- showed significantly increased interstitial fibrosis in the eration [17, 45]. In fact, ROS generation from damaged heart [48], though such an increased collagen deposition mitochondria is involved in exacerbation of ventricu- in the myocardium was not detected by histochemistry lar dysfunction [46]. An impact of autophagy on myo- or determination of mRNA levels of collagens I and III cardial energy status has been shown by findings that in OLETF used in our studies [4, 33]. Some difference in myocardial ATP content after MI was increased by aug- rearing conditions (possibly the amount or calories per mentation of autophagy [9]. Diabetes impairs mecha- volume of the chow provided) might underlie the differ - nisms regulating ATP supply, and our recent study [33] ence in the phenotype of OLETF at similar ages. Never- showed that reduced myocardial reserve of ATP sup- theless, it is possible that suppression of baseline AMPK ply, leading to diastolic dysfunction, was disclosed by phosphorylation and autophagy occurs in OLETF at an increased afterload in OLETF. How impaired response advanced stage of T2DM. of autophagy relates to dysregulation of ATP supply Increased AMPK phosphorylation with or without mechanisms in diabetic hearts may warrant further suppressed mTOR phosphorylation was associated with investigation. promotion of autophagy after MI in non-diabetic mice Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 14 of 16 [9, 16]. The responses of autophagy, AMPK and mTOR Conclusions to MI were confirmed in LETO (Figs.  4, 5, 6, 7). However, Treatment with vildagliptin at a dose that elevated serum in OLETF, phosphorylation of AMPK was not increased GLP-1 without normalization of plasma glucose level and activation of the mTOR/S6 pathway was 60–70% of reduced acute mortality after MI in a rat model of T2DM that in LETO (Figs. 6, 7). In addition, we found that inter- to the level in non-diabetic controls. The beneficial effect action of Beclin-1 and Bcl-2, which reportedly inhibits of vildagliptin was sensitive to chloroquine and closely Beclin-1-dependent autophagy [11, 15, 39], was signifi - associated with restoration of the autophagic response cantly increased in the myocardium of OLETF (Fig.  8). in the non-infarcted myocardium to MI, suggesting an Restoration of the adaptive responses of both LC3-II and involvement of impaired autophagy in T2DM-induced autophagosomes after MI in OLETF by vildagliptin was increase in post-MI mortality. associated with suppression of Beclin-1–Bcl-2 interac- Additional file tion but not with improved phosphorylation of AMPK, mTOR or S6. These findings suggest that increased Additional file 1: Table S1. Antibodies. Description of data: a list of Beclin-1–Bcl-2 interaction was responsible for T2DM- antibodies used in this study. induced loss of adaptive autophagy in the non-ischemic myocardium after MI. How vildagliptin suppressed Becin-1–Bcl-2 interac- Authors’ contributions Participated in research design: AK, TaM, MT and TeM. Conducted experiments: tion in the myocardium of OLETF remains unclear. HM, AK, TaM, HK, SI, TT, MO and KN. Performed data analysis: HM, AK, TaM, MT, Among molecules that regulate Beclin-1–Bcl-2 interac- TY, HK and TeM. Performed statistical analyses: HM and AK. Wrote or contrib- tion, AMPK-JNK activation has been reported to induce uted to the writing of the manuscript: AK, TaM and TeM. All authors read and approved the final manuscript. disruption of Beclin-1–Bcl-2 interaction through phos- phorylation of Bcl-2 at Ser70 [15], whereas activation Author details of mammalian sterile 20-like kinase 1 (Mst1) promoted Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, South-1, West-16, Chuo-ku, Sap- Beclin-1/Bcl-2 interaction by phosphorylation of Bec- poro 060-8543, Japan. Department of Pharmacology, Sapporo Medical lin-1 at Thr108 [11]. In this study, vildagliptin did not University School of Medicine, Sapporo 060-8543, Japan. restore phosphorylation of AMPK (Fig.  6) or JNK (data Acknowledgements not shown). Hence, there is the possibility that vildaglip- This study was supported by Grants for Education and Research 2012–2014 tin suppressed Mst1 expression or activity, preventing from Sapporo Medical University and a grant from Novartis Pharma AG. interaction of Beclin-1 and Bcl-2 in the myocardium of Novartis Pharma AG did not play any role in the collection, analysis and inter- pretation of data or the writing of the manuscript. OLETF. Unfortunately, we could not examine this pos- sibility since phospho-Mst1 (Thr183) protein in the Compliance with ethical guidelines myocardium of OLETF could not be detected by use of Competing interests commercially available antibodies. This study was supported in part by a grant from Novartis Pharma AG. Contrary to our expectations, the dose of vildagliptin and Grants for Education and Research 2012–2014 from Sapporo Medical we used in the present study was not sufficient for reduc - University. ing glucose levels in OLETF (Table 1; Fig. 2b), though the Received: 27 April 2015 Accepted: 24 July 2015 dose of vildagliptin increased serum active GLP-1 level in OLETF (Fig.  2a). Therefore, the effects of vildagliptin on cardiac autophagy and mortality cannot be explained by its effect on glycemic control. Although DPP-4 inhibi - References tors are known to enhance glucose-stimulated insulin 1. Marso SP, Miller T, Rutherford BD, Gibbons RJ, Qureshi M, Kalynych A release [49], reduction in plasma insulin level by DPP-4 et al (2007) Comparison of myocardial reperfusion in patients undergo- inhibitors together with improved metabolic parameters ing percutaneous coronary intervention in ST-segment elevation acute myocardial infarction with versus without diabetes mellitus (from the has also been reported [50, 51]. It is notable that DPP-4 EMERALD Trial). Am J Cardiol 100:206–210 inhibitors (saxagliptin and alogliptin) failed to reduce 2. De Luca G, Dirksen MT, Spaulding C, Kelbæk H, Schalij M, Thuesen L cardiovascular mortality in diabetic patients at high car- et al (2013) Impact of diabetes on long-term outcome after primary angioplasty: insights from the DESERT cooperation. Diabetes Care diovascular risk in two large clinical trials [52, 53]. The 36:1020–1025 effect of vildagliptin on acute mortality after MI in this 3. Miki T, Itoh T, Sunaga D, Miura T (2012) Eec ff ts of diabetes on myocardial study (Fig. 3) is difficult to easily reconcile with the nega - infarct size and cardioprotection by preconditioning and postcondition- ing. Cardiovasc Diabetol 11:67 tive results in the clinical trials [52, 53]. However, there 4. Takada A, Miki T, Kuno A, Kouzu H, Sunaga D, Itoh T et al (2012) Role of ER is the possibility that the optimal dose of DPP-4 for stress in ventricular contractile dysfunction in type 2 diabetes. PLoS One cardiprotection is lower than that for glycemic control. 7:e39893 Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 15 of 16 5. Matsui Y, Takagi H, Qu X, Abdellatif M, Sakoda H, Asano T et al (2007) 26. Jing Yin J, Bo Li Y, Ming Cao M, Wang Y (2013) Liraglutide improves the sur- Distinct roles of autophagy in the heart during ischemia and reperfu- vival of INS-1 cells by promoting macroautophagy. Int J Endocrinol Metab sion: roles of AMP-activated protein kinase and Beclin 1 in mediating 11:184–190 autophagy. Circ Res 100:914–922 27. Shigeta T, Aoyama M, Bando YK, Monji A, Mitsui T, Takatsu M et al (2012) 6. Nakai A, Yamaguchi O, Takeda T, Higuchi Y, Hikoso S, Taniike M et al Dipeptidyl peptidase-4 modulates left ventricular dysfunction in chronic (2007) The role of autophagy in cardiomyocytes in the basal state and in heart failure via angiogenesis-dependent and -independent actions. response to hemodynamic stress. Nat Med 13:619–624 Circulation 126:1838–1851 7. Zhai P, Sciarretta S, Galeotti J, Volpe M, Sadoshima J (2011) Differential 28. Ravassa S, Barba J, Coma-Canella I, Huerta A, López B, González A et al roles of GSK-3β during myocardial ischemia and ischemia/reperfusion. (2013) The activity of circulating dipeptidyl peptidase-4 is associated with Circ Res 109:502–511 subclinical left ventricular dysfunction in patients with type 2 diabetes 8. Kanamori H, Takemura G, Goto K, Maruyama R, Tsujimoto A, Ogino A et al mellitus. Cardiovasc Diabetol 12:143 (2011) The role of autophagy emerging in postinfarction cardiac remod- 29. Noyan-Ashraf MH, Momen MA, Ban K, Sadi AM, Zhou YQ, Riazi AM et al elling. Cardiovasc Res 91:330–339 (2009) GLP-1R agonist liraglutide activates cytoprotective pathways and 9. Kanamori H, Takemura G, Goto K, Tsujimoto A, Ogino A, Takeyama T et al improves outcomes after experimental myocardial infarction in mice. (2013) Resveratrol reverses remodeling in hearts with large, old myo- Diabetes 58:975–983 cardial infarctions through enhanced autophagy-activating AMP kinase 30. Sauvé M, Ban K, Momen MA, Zhou YQ, Henkelman RM, Husain M et al pathway. Am J Pathol 182:701–713 (2010) Genetic deletion or pharmacological inhibition of dipeptidyl 10. Kubli DA, Zhang X, Lee Y, Hanna RA, Quinsay MN, Nguyen CK et al (2013) peptidase-4 improves cardiovascular outcomes after myocardial infarc- Parkin protein deficiency exacerbates cardiac injury and reduces survival tion in mice. Diabetes 59:1063–1073 following myocardial infarction. J Biol Chem 288:915–926 31. Ussher JR, Baggio LL, Campbell JE, Mulvihill EE, Kim M, Kabir MG et al 11. Maejima Y, Kyoi S, Zhai P, Liu T, Li H, Ivessa A et al (2013) Mst1 inhibits (2014) Inactivation of the cardiomyocyte glucagon-like peptide-1 recep- autophagy by promoting the interaction between Beclin1 and Bcl-2. Nat tor (GLP-1R) unmasks cardiomyocyte-independent GLP-1R-mediated Med 19:1478–1488 cardioprotection. Mol Metab 3:507–517 12. Watanabe T, Takemura G, Kanamori H, Goto K, Tsujimoto A, Okada H et al 32. Hausenloy DJ, Whittington HJ, Wynne AM, Begum SS, Theodorou L, (2014) Restriction of food intake prevents postinfarction heart failure Riksen N et al (2013) Dipeptidyl peptidase-4 inhibitors and GLP-1 reduce by enhancing autophagy in the surviving cardiomyocytes. Am J Pathol myocardial infarct size in a glucose-dependent manner. Cardiovasc 184:1384–1394 Diabetol 12:154 13. Delbridge LM, Mellor KM, Taylor DJ, Gottlieb RA (2015) Myocardial 33. Kouzu H, Miki T, Tanno M, Kuno A, Yano T, Itoh T et al (2015) Excessive autophagic energy stress responses—macroautophagy, mitophagy, and degradation of adenine nucleotides by up-regulated AMP deaminase glycophagy. Am J Physiol Heart Circ Physiol 308:H1194–H1204 underlies afterload-induced diastolic dysfunction in the type 2 diabetic 14. Yamahara K, Yasuda M, Kume S, Koya D, Maegawa H, Uzu T (2013) The role heart. J Mol Cell Cardiol 80:136–145 of autophagy in the pathogenesis of diabetic nephropathy. J Diabetes 34. Miki T, Miura T, Hotta H, Tanno M, Yano T, Sato T et al (2009) Endoplasmic Res 2013:193757 reticulum stress in diabetic hearts abolishes erythropoietin-induced 15. He C, Zhu H, Li H, Zou MH, Xie Z (2013) Dissociation of Bcl-2-Beclin1 myocardial protection by impairment of phospho-glycogen synthase complex by activated AMPK enhances cardiac autophagy and protects kinase-3beta-mediated suppression of mitochondrial permeability transi- against cardiomyocyte apoptosis in diabetes. Diabetes 62:1270–1281 tion. Diabetes 58:2863–2872 16. Sciarretta S, Zhai P, Shao D, Maejima Y, Robbins J, Volpe M et al (2012) 35. Shinmura K, Tamaki K, Sano M, Murata M, Yamakawa H, Ishida H et al Rheb is a critical regulator of autophagy during myocardial ischemia: (2011) Impact of long-term caloric restriction on cardiac senescence: pathophysiological implications in obesity and metabolic syndrome. caloric restriction ameliorates cardiac diastolic dysfunction associated Circulation 125:1134–1146 with aging. J Mol Cell Cardiol 50:117–127 17. Wang B, Yang Q, Sun YY, Xing YF, Wang YB, Lu XT et al (2014) Resveratrol- 36. Jungraithmayr W, De Meester I, Matheeussen V, Baerts L, Arni S, Weder enhanced autophagic flux ameliorates myocardial oxidative stress injury W (2012) CD26/DPP-4 inhibition recruits regenerative stem cells via in diabetic mice. J Cell Mol Med 18:1599–1611 stromal cell-derived factor-1 and beneficially influences ischaemia/rep - 18. Xie Z, Lau K, Eby B, Lozano P, He C, Pennington B et al (2011) Improve- erfusion injury in mouse lung transplantation. Eur J Cardiothorac Surg ment of cardiac functions by chronic metformin treatment is associated 41:1166–1173 with enhanced cardiac autophagy in diabetic OVE26 mice. Diabetes 37. Hotta H, Miura T, Miki T, Togashi N, Maeda T, Kim SJ et al (2010) Angio- 60:1770–1778 tensin II type 1 receptor-mediated upregulation of calcineurin activity 19. Xu X, Kobayashi S, Chen K, Timm D, Volden P, Huang Y et al (2013) underlies impairment of cardioprotective signaling in diabetic hearts. Circ Diminished autophagy limits cardiac injury in mouse models of type 1 Res 106:129–132 diabetes. J Biol Chem 288:18077–18092 38. Zheng Q, Su H, Ranek MJ, Wang X (2011) Autophagy and p62 in cardiac 20. Eguchi M, Kim YH, Kang KW, Shim CY, Jang Y, Dorval T et al (2012) proteinopathy. Circ Res 109:296–308 Ischemia-reperfusion injury leads to distinct temporal cardiac remodeling 39. Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N et al (2005) in normal versus diabetic mice. PLoS One 7:e30450 Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell 21. Hou X, Hu Z, Xu H, Xu J, Zhang S, Zhong Y et al (2014) Advanced glyca- 122:927–939 tion endproducts trigger autophagy in cadiomyocyte via RAGE/PI3K/ 40. Kim M, Platt MJ, Shibasaki T, Quaggin SE, Backx PH, Seino S et al (2013) AKT/mTOR pathway. Cardiovasc Diabetol 13:78 GLP-1 receptor activation and Epac2 link atrial natriuretic peptide secre- 22. Noyan-Ashraf MH, Shikatani EA, Schuiki I, Mukovozov I, Wu J, Li RK et al tion to control of blood pressure. Nat Med 19:567–575 (2013) A glucagon-like peptide-1 analog reverses the molecular pathol- 41. Ussher JR, Drucker DJ (2014) Cardiovascular actions of incretin-based ogy and cardiac dysfunction of a mouse model of obesity. Circulation therapies. Circ Res 114:1788–1803 127:74–85 42. French CJ, Zaman AT, McElroy-Yaggy KL, Neimane DK, Sobel BE 23. Huisamen B, Genade S, Lochner A (2008) Signalling pathways activated (2011) Absence of altered autophagy after myocardial ischemia by glucagon-like peptide-1 (7-36) amide in the rat heart and their role in in diabetic compared with nondiabetic mice. Coron Artery Dis protection against ischaemia. Cardiovasc J Afr 19:77–83 22:479–483 24. Zeng Y, Li C, Guan M, Zheng Z, Li J, Xu W et al (2014) The DPP-4 inhibitor 43. French BA, Kramer CM (2007) Mechanisms of post-infarct left ventricular sitagliptin attenuates the progress of atherosclerosis in apolipoprotein-E- remodeling. Drug Discov Today Dis Mech 4:185–196 knockout mice via AMPK- and MAPK-dependent mechanisms. Cardiovasc 44. Pfeffer MA, Braunwald E (1990) Ventricular remodeling after myocardial Diabetol 13:32 infarction. Experimental observations and clinical implications. Circula- 25. Kim J, Kundu M, Viollet B, Guan KL (2011) AMPK and mTOR regulate tion 81:1161–1172 autophagy through direct phosphorylation of Ulk1. Nat Cell Biol 45. Lee J, Giordano S, Zhang J (2012) Autophagy, mitochondria and oxidative 13:132–141 stress: cross-talk and redox signaling. Biochem J 441:523–540 Murase et al. Cardiovasc Diabetol (2015) 14:103 Page 16 of 16 46. Ide T, Tsutsui H, Kinugawa S, Suematsu N, Hayashidani S, Ichikawa K et al 51. Apaijai N, Pintana H, Chattipakorn SC, Chattipakorn N (2013) Eec ff ts of (2000) Direct evidence for increased hydroxyl radicals originating from vildagliptin versus sitagliptin, on cardiac function, heart rate variability superoxide in the failing myocardium. Circ Res 86:152–157 and mitochondrial function in obese insulin-resistant rats. Br J Pharmacol 47. Zhu H, Tannous P, Johnstone JL, Kong Y, Shelton JM, Richardson JA et al 169:1048–1057 (2007) Cardiac autophagy is a maladaptive response to hemodynamic 52. White WB, Cannon CP, Heller SR, Nissen SE, Bergenstal RM, Bakris GL et al stress. J Clin Invest 117:1782–1793 (2013) Alogliptin after acute coronary syndrome in patients with type 2 48. Lee JE, Yi CO, Jeon BT, Shin HJ, Kim SK, Jung TS et al (2012) & #x03B1;- diabetes. N Engl J Med 369:1327–1335 Lipoic acid attenuates cardiac fibrosis in Otsuka Long-Evans Tokushima 53. Scirica BM, Bhatt DL, Braunwald E, Steg PG, Davidson J, Hirshberg B, fatty rats. Cardiovasc Diabetol 11:111 SAVOR-TIMI 53 Steering Committee and Investigators et al (2013) 49. Burkey BF, Li X, Bolognese L, Balkan B, Mone M, Russell M et al (2005) Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes Acute and chronic effects of the incretin enhancer vildagliptin in insulin- mellitus. N Engl J Med 369:1317–1326 resistant rats. J Pharmacol Exp Ther 315:688–695 50. Dobrian AD, Ma Q, Lindsay JW, Leone KA, Ma K, Coben J et al (2011) Dipeptidyl peptidase IV inhibitor sitagliptin reduces local inflammation in adipose tissue and in pancreatic islets of obese mice. Am J Physiol Endocrinol Metab 300:E410–E421 Submit your next manuscript to BioMed Central and take full advantage of:  Convenient online submission  Thorough peer review  No space constraints or color figure charges  Immediate publication on acceptance  Inclusion in PubMed, CAS, Scopus and Google Scholar  Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

Cardiovascular Diabetology – Springer Journals

Published: Dec 1, 2015

Keywords: diabetes; angiology; cardiology

There are no references for this article.