Dipeptidyl Peptidase-4 Inhibition Attenuates Arrhythmias via a Protein Kinase A-Dependent Pathway in Infarcted Hearts

Abstract

Background: The effect of dipeptidyl peptidase-4 (DPP-4) inhibitors on arrhythmias remains unknown. The aim of this study was to investigate whether sitagliptin attenuates arrhythmias through inhibiting nerve growth factor (NGF) expression, focusing on cyclic adenosine monophosphate (cAMP) downstream signaling such as protein kinase A (PKA) and exchange protein directly activated by cAMP (Epac).

Methods and Results: Male Wistar rats were randomized to either vehicle or sitagliptin for 4 weeks starting 24 h after ligating the coronary artery. Post-infarction was associated with increased oxidative stress. Measurement of myocardial norepinephrine levels revealed a significant elevation in vehicle-treated rats compared with sham. Compared with the vehicle, infarcted rats treated with sitagliptin had significantly increased cAMP levels, decreased DPP-4 activity, oxidative stress, NGF levels and immunofluorescence-stained sympathetic hyperinnervation. Arrhythmic scores were significantly lower in the sitagliptin-treated infarcted rats than in vehicle. Ex vivo studies showed that sitagliptin increased the phosphorylated cAMP response element-binding protein (CREB), which can be reversed by H-89 (a PKA inhibitor), not brefeldin A (an Epac inhibitor). Heme oxygenase-1 (HO-1) expression was increased by a PKA agonist but not by an Epac agonist. HO-1 expression was attenuated in KG-501 (a CREB inhibitor)-treated infarcted rats in the presence of a PKA agonist.

Conclusions: Sitagliptin protects ventricular arrhythmias by attenuating NGF-induced sympathetic innervation via upregulation of HO-1 expression in a cAMP/PKA/CREB-dependent antioxidant pathway in non-diabetic infarcted rats. (Circ J 2015; 79: 2461–2470)

Dipeptidyl peptidase-4 (DPP-4) inhibitors are a novel class of drugs for the treatment of hyperglycemia. DPP-4 inhibitors act by inhibiting the degradation of glucagon-like peptide-1 (GLP-1) and glucose insulinotropic peptide by the plasma DPP-4 enzyme.¹ The increased availability of these hormones stimulates glucose-dependent insulin release.¹ DPP-4 inhibitors have been shown to provide cardioprotection in an insulin-independent manner via ancillary pathways.¹ This cardioprotective effect of DPP-4 inhibitors was also observed in normoglycemic rats.²

Cardiac arrhythmias are associated with increased risks of cardiovascular complications and sudden death.³ Reactive oxygen species (ROS) have a strong impact on ventricular remodeling after myocardial infarction (MI).^4,5 Very recently, we demonstrated that oxidative stress is increased in the remote myocardium and plays a critical role in ventricular remodeling after MI.^4,5 During the chronic stage of MI, regional increase of sympathetic nerves was commonly observed at the remote zone.⁶ A nerve sprouting, due to increased sympathetic reinnervation, has been shown to be an important contributing factor in the occurrence of ventricular arrhythmias and sudden cardiac death in the healing or healed stage of MI in both animals^7,8 and humans.⁹ Nerve growth factor (NGF) is a prototypic member of the neurotrophin family, members of which are critical for the differentiation, survival, and synaptic activity of the peripheral sympathetic and sensory nervous systems.¹⁰ The NGF promoter contains activator protein-1,¹¹ which is subjected to redox regulation through its conserved cysteine residue.¹² Previous studies have shown that peroxynitrite, the byproduct of •NO and superoxide (O₂˙^–), is an important trigger of NGF formation.¹³ A brief exposure to peroxynitrite induces NGF expression and secretion in astrocytes.¹³

DPP-4 inhibitors may increase cyclic adenosine monophosphate (cAMP) levels by activating GLP-1-induced adenylate cyclase.¹⁴ cAMP is a key intracellular signal that regulates neuronal differentiation, survival, neurite length, and neurite guidance in a variety of neurons.¹⁵ Downstream effectors of cAMP include protein kinase A (PKA) and the exchange protein activated by cAMP (Epac). While most studies of cAMP signaling have focused on PKA, cAMP has been shown to regulate gene transcription, cellular proliferation, and cytokine signaling through the Epac pathway.¹⁶ Epac belongs to a family of cAMP-regulated guanine nucleotide exchange factors that mediate PKA-independent signal transduction properties of the second messenger cAMP.¹⁷ Epac has been shown to either antagonize¹⁸ or synergize with PKA.¹⁹ Previous studies have shown that cAMP mediates axonal growth cone attraction or repulsion by distinctly activating PKA or Epac, depending on intracellular levels of cAMP.²⁰ High levels mediate attraction in an Epac-dependent manner, whereas lower levels result in repulsion in a PKA-dependent manner.²⁰ However, the specific involvement of cAMP effectors, PKA or Epac, on neurite outgrowth is still a subject of debate. Activation of these pathways in the cytoplasm ultimately affects nuclear transcription factors. One critical stimulus-induced transcription factor is the cAMP response element-binding protein (CREB).²¹ Transcriptional activity of CREB is positively regulated by phosphorylation of a critical serine residue, Ser133. Once phosphorylated, CREB recruits the CREB binding protein to the promoter regions of cAMP-responsive targeting genes such as heme oxygenase-1 (HO-1). HO-1, a stress-inducible protein, is the rate-limiting enzyme of heme degradation, and it is associated with protection against cellular injury and oxidative stress.²² HO-1 thus provides a relevant and sensitive index by which to assess alterations in a cellular redox state.²²

Administration of DPP-4 inhibitors has been shown to attenuate increased ROS in diabetic db/db mouse islets.²³ Very recently, the GLP-1 receptor agonist, exendin-4, has been shown to attenuate ROS production.¹⁴ The GLP-1 receptor couples with Gs and activates adenylate cyclase to upregulate cAMP. There were contradictory observations regarding the GLP-1-induced cAMP downstream molecules such as PKA and Epac. GLP-1 prevents ROS-induced endothelial cell senescence only through the activation of PKA.²⁴ In contrast, the GLP-1 receptor agonist, exendin-4, attenuated ROS formation in an Epac-dependent manner in diabetic Goto-Kakizaki rat islets.¹⁴ Thus, it appears that in different cell types and different targets, different pathways are involved in GLP-1-induced antioxidation mechanisms. However, it remains unknown whether PKA and Epac collaborate in modulating cAMP-induced NGF expression upon DPP-4 inhibition in the infarcted heart. Thus, we assessed: (1) whether sitagliptin, a DPP-4 inhibitor, after infarction modulates NGF expression; (2) whether sitagliptin-induced superoxide changes are PKA or Epac dependent in a rat MI model by the use of inhibitors of PKA and Epac as well as the measurement of cyclic nucleotides; and (3) whether PKA activation increases the phosphorylation of CREB (Ser-133).

Methods

Ethics

The animal experiment was approved and conducted in accordance with local institutional guidelines for the care and use of laboratory animals at the China Medical University and conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

Animals

Experiment 1 (In Vivo) Healthy non-diabetic Male Wistar rats (300–350 g) were subjected to ligation of the left anterior descending artery, resulting in infarction of the left ventricular (LV) free wall. Twenty-four hours after inducing MI, survived rats were randomly assigned into either the vehicle (saline) or sitagliptin (5 mg/kg per day; Merck, NJ, USA) groups that were administered orally by gastric gavage once a day. The study duration was designed to be 4 weeks. For further details on the methods, please refer to Supplementary Methods.

Experiment 2 (Ex Vivo) To test the relative importance of PKA and Epac in sitagliptin-related NGF levels, we used inhibitors of PKA and Epac in an ex vivo model. Four weeks after induction of MI by coronary ligation, infarcted rat hearts were isolated and subjected to no treatment (vehicle), sitagliptin (5 μmol/L), sitagliptin+H-89 (0.1 μmol/L, a highly specific inhibitor of PKA), and sitagliptin+brefeldin A (100 μmol/L, an Epac-signaling inhibitor). The doses of sitagliptin, H-89, and brefeldin A have been shown to be effective in modulating biological activities.^25,26 H-89 and brefeldin A were all from Sigma (St Louis, MO, USA). The heart was perfused with a non-circulating modified Tyrode’s solution containing (in mmol/L): glucose 5.5, NaCl 117.0, NaHCO₃ 23.0, KCl 4.6, NaH₂PO₄ 0.8, MgCl₂ 1.0, and CaCl₂ 2.0, equilibrated at 37℃ and oxygenated with a 95% O₂ to 5% CO₂ gas mixture.²⁷ The drugs were infused for 60 min. At the end of the study, all hearts (n=5 each group) were used for performing Western analysis for p-CREB (Ser-133) and NGF protein at the remote zone (>2 mm outside the infarct).

Experiment 3 (Ex Vivo) To confirm the relative importance of PKA and Epac in HO-1 expression, we used activators of PKA and Epac in an ex vivo model. Four weeks after induction of MI by coronary ligation, infarcted rat hearts were isolated and subjected to no treatment (vehicle), N6-benzoyladenosine 3’,5-cyclic monophosphate (1 mmol/L, N6Bz, a PKA-specific activator; BIOLOG Life Science Institute), 8-(4-chlrorophenylthio)-2’-O-methyladenosine-3’,5’-cyclic monophosphate) (1 mmol/L, 8-CPT, an Epac-specific activator; Calbiochem), N6Bz+2-naphthol-AS-E-phosphate (10 μmol/L, KG-501, a p-CREB inhibitor; Sigma-Aldrich), or 8-CPT+KG-501. Each heart was perfused with the same protocol as Experiment 2. The doses of N6Bz, 8-CPT and KG-501 were used, as previously described.^28,29 It has been reported that KG-501 is a small molecule that binds to the transcription coactivator, CBP, and blocks the interaction of CBP with the active form of CREB, p-CREB.²⁹ At the end of the study, hearts (n=5 per group) were used for real-time quantitative reverse transcription-polymerase chain reaction (RT-PCR) of HO-1 at the remote zone.

Experiment 4 (Ex Vivo) To evaluate the importance of peroxynitrite in sitagliptin-related NGF levels, we performed an ex vivo experiment. Four weeks after induction of MI by coronary ligation, infarcted rat hearts were isolated and subjected to no treatment (vehicle), sitagliptin (5 μmol/L), or a combination of sitagliptin and 3-morpholinosydnonimine (37 μmol/L, SIN-1, a peroxynitrite generator). Each heart was perfused with the same protocol as Experiment 2. The dose of SIN-1 was used as previously described.³⁰ In order to preclude non-specific actions to SIN-1, the relatively low concentration of SIN-1 was used. At the end of the study, hearts (n=5 per group) were used for Western blot of NGF at the remote zone.

Hemodynamics and Infarct Size Measurements

Hemodynamic parameters and infarct size were measured in anesthetized rats at the end of the study, as described in detail in Supplementary Methods.

In Vivo Electrophysiological Studies

Following arterial pressure measurement, the rats were intubated. Electrophysiological studies were performed with electrodes sewn to the epicardial surface of the right ventricular outflow tract. To determine the ventricular effective refractory period (VERP), single extrastimuli were introduced at progressively shorter intervals. The VERP was the longest S1S2 interval that did not evoke a premature ventricular depolarization. For detailed methods, please refer to Supplementary Methods.

Real-Time RT-PCR of HO-1 and NGF

mRNAs were quantified by using real-time RT-PCR, with cyclophilin as a loading control. For detailed methods, please refer to Supplementary Methods.

Western Blot Analysis of p-CREB (Ser-133) and NGF

Samples were obtained from the remote zone at week 4 after infarction. Rabbit polyclonal antibodies to NGF (Chemicon, CA, USA), p-CREB (Ser-133, Cell Signaling) and CREB (Cell Signaling) were used. Western blotting procedures were described previously.⁴ Experiments were replicated three times and results were expressed as the mean value.

Immunofluorescent Studies of Tyrosine Hydroxylase, Growth Associated Factor 43 and Neurofilament

In order to investigate the spatial distribution and quantification of sympathetic nerve fibers, analysis of immunofluorescent staining was performed on the LV muscle from the remote zone. The analysis of the immunofluorescent staining is described in detail in Supplementary Methods.

In Situ Detection of Superoxide

Myocardial intracellular superoxide production was determined by using in situ dihydroethidium (DHE; Invitrogen Molecular Probes, Eugene, OR, USA) fluorescence, as previously described.⁵ For a detailed method, please refer to Supplementary Methods.

Laboratory Measurements

We measured the DPP-4 activity and active GLP-1 levels in plasma at the end of the study to confirm that continuous administration of sitagliptin was indeed associated with suppression of plasma DPP-4 activity, and an increase in the active GLP-1 levels. EDTA plasma was used to measure active GLP-1 (Millipore Corporation, Billerica, MA, USA) and DPP-4 activity (Quantizume AssaySystem, BIOMOL International, Plymouth Meeting, PA, USA). Insulin was measured by an ultrasensitive rat enzyme immunoassay (Mercodia, Uppsala, Sweden).

Although cardiac innervation was detected by immunofluorescent staining of tyrosine hydroxylase, growth-associated factor 43, and neurofilament, it did not imply that the nerves are functional. Thus, to examine the sympathetic nerve function after administering sitagliptin, we measured LV norepinephrine levels from the remote zone using a commercial ELISA kit (Noradrenalin ELISA, IBL Immuno-Biological Laboratories Co, Hamburg, Germany).

Myocardial cAMP was measured by using an enzyme-linked immunoassay kit (R&D Systems, Abingdon, UK), and the protein content was determined by the BCA protein assay kit and cAMP levels were expressed as pmol/g LV tissue.

Superoxide production by myocardium from the remote zone was measured using lucigenin (5 μmol/L bis-N-methylacridinium nitrate; Sigma, St. Louis, MO, USA) enhanced chemiluminescence, as previously described.⁵ The specific chemiluminescence signal was calculated after subtraction of background activity, and expressed as counts per minute per milligram weight (cpm/mg).

To estimate myocardial peroxynitrite formation, we measured free nitrotyrosine (as a marker for peroxynitrite formation) by using an ELISA (Cayman Chemical, Ann Arbor, MI, USA) in myocardial homogenates.

Statistical Analysis

Results are presented as mean±SD. Statistical analysis was performed using the SPSS statistical package (SPSS, version 12.0, Chicago, IL, USA). Differences among the groups of rats were tested by an ANOVA. In case of a significant effect, the measurements between the groups were compared with Bonferroni’s correction. Electrophysiological data (scoring of programmed electrical stimulation-induced arrhythmias) were compared by a Kruskal-Wallis test followed by a Mann-Whitney test. The significant level was assumed at a value of P<0.05.

Results

Part 1. In Vivo Study (Experiment 1)

Differences in mortality between the 2 infarcted groups were not found throughout the study. Sitagliptin had little effect on cardiac gross morphology in the sham-operated rats. Four weeks after infarction, the infarcted area of the LV was very thin, and was totally replaced by fully differentiated scar tissue. The weight of the LV inclusive of the septum remained essentially constant for 4 weeks between the infarcted groups (Table). Compared with vehicle-treated infarcted rats in sitagliptin-treated infarcted rats, the maximal rate of LV +dP/dt and –dP/dt was significantly increased, and the lung weight/ body weight ratio was significantly lower, which is consistent with favorable LV remodeling. LV end-systolic pressure and infarct size did not differ between the infarcted groups. No changes in plasma glucose and insulin levels were observed between the 2 infarcted groups.

Table. Cardiac Morphology, Hemodynamics and Glucose, Insulin, and Tissue NE Levels at the End of Study

Parameters	Sham		Infarction treated with
	Saline	Sitagliptin	Saline	Sitagliptin
No. of rats	10	10	11	12
Body weight, g	405±10	401±9	395±11	404±12
Heart rate, bpm	397±9	379±10	401±11	404±12
LVESP, mmHg	102±5	100±4	96±6	99±5
LVEDP, mmHg	4±2	4±2	18±5*	17±6*
+dp/dt , mmHg/s	7,384±265	7,564±243	2,374±255*	3,084±223*,**
−dp/dt, mmHg/s	6,382±217	6,492±225	2,217±235*	2,635±214*,**
Infarct size, %	–	–	40±2	41±2
LVW/BW, mg/g	2.52±0.21	2.49±0.22	3.15±0.32*	3.22±0.35*
RVW/BW, mg/g	0.59±0.12	0.62±0.10	1.03±0.11*	0.98±0.12*
LungW/BW, mg/g	4.02±0.48	3.97±0.35	5.98±0.94*	4.65±0.67**
Glucose, mg/dl	85±7	84±6	90±5	87±7
Insulin, μu/ml	18±9	21±10	52±11*	58±12*
NE, μg/g protein	1.38±0.38	1.27±0.32	2.78±0.22*	1.26±0.31

Values are presented as mean±SD. BW, body weight; LungW, lung weight; LVEDP, left ventricular end-diastolic pressure; LVESP, left ventricular end-systolic pressure; LVW, left ventricular weight; NE, norepinephrine levels from remote myocardium; RVW, right ventricular weight. *P<0.025 compared with respective sham; **P<0.025 compared with the saline-treated infarcted group.

Plasma GLP-1 and DPP-4 Activity and Myocardial cAMP, Superoxide, Nitrotyrosine, and Norepinephrine Levels

Plasma DPP-4 activity and GLP-1 levels were determined to confirm the successful oral delivery of sitagliptin. This was associated with a significant increase of the active GLP-1 level in the sitagliptin-treated group (Figure 1A). Plasma DPP-4 activity was significantly reduced by 63% in the sitagliptin-treated group compared to the vehicle group (Figure 1B).

Figure 1.

Plasma (A) glucagon-like peptide-1 (GLP-1) and (B) dipeptidylpeptidase-4 (DPP-4) activity and myocardial (C) cyclic adenosine monophosphate (cAMP), (D) superoxide and (E) nitrotyrosine levels from the remote zone. *P<0.01 compared with sham- and vehicle-treated infarcted rats; ^†P<0.01, compared with sham and sitagliptin.

Compared with sham, ventricular remodeling after MI was associated with a significant increase in cAMP content (239±52 in sham vs. 378±58 pmol/g LV, P<0.001, Figure 1C). Treatment with sitagliptin had a significant increase of cAMP content compared with vehicle.

Myocardial superoxide production, as assessed by lucigenin-enhanced chemiluminescence, was markedly increased in remote LV myocardium after MI as compared with sham (P<0.001, Figure 1D). Superoxide was significantly decreased in sitagliptin-treated rats to the level of sham.

Myocardial nitrotyrosine in vehicle-treated infarcted rats significantly increased as compared to sham (P<0.001, Figure 1E). Myocardial nitrotyrosine in sitagliptin-treated infarcted rats can be significantly reduced compared with vehicle.

To investigate the possible role of cardiac norepinephrine synthesis, we determined the LV norepinephrine levels. Sitagliptin administration did not affect tissue norepinephrine concentrations in sham-operated rats. LV norepinephrine levels were significantly upregulated 2.01-fold in the vehicle-treated infarcted rats in comparison with sham (2.78±0.22 vs. 1.38±0.38 µg/g protein, P<0.001, Table). When compared with vehicle-treated infarcted rats, sitagliptin-treated infarcted rats had a significantly lower LV norepinephrine level.

DHE Staining in Myocardium

DHE reacts with superoxide radicals to form ethidium bromide, which in turn intercalates with DNA to provide nuclear fluorescence as a marker of superoxide radical generation. As shown in Figure 2A, postinfarction remodeling markedly enhanced the intensity of the DHE staining in the remote zone in the vehicle-treated rats compared with sham. However, the intensity of the fluorescent signal in the sitagliptin group was significantly reduced relative to the vehicle group.

Figure 2.

(A) Detection of superoxide in the myocardium by dihydroethidium (DHE) staining (magnification 400×). Compared with sham, the DHE fluorescence intensity in the myocardium of the vehicle-treated infarcted group was significantly increased. DHE staining (%) at the remote zone. (B) Immunofluorescent staining for tyrosine hydroxylase from the remote regions. Tyrosine hydroxylase-positive nerve fibers are located between myofibrils and are oriented in a longitudinal direction as that of the myofibrils. (C) Immunofluorescent staining for growth-associated protein 43 from the remote regions. (A) Sham group; (B) infarction treated with vehicle group; (C) infarction treated with sitagliptin group. Bar=50 μm. Nerve density area fraction (%) at the remote zone. Each column and bar represents mean±SD. The number of animals in each group is indicated in parentheses. *P<0.05, compared with sham and sitagliptin.

Immunofluorescent Analyses

The tyrosine hydroxylase-immunostained nerve fibers appeared to be oriented in the longitudinal axis of adjacent myofibers (Figure 2B). Tyrosine hydroxylase-positive nerve density was significantly increased in the vehicle-treated infarcted rats than that in sham group. Sitagliptin-treated rats showed lower nerve density at the remote regions than vehicle-treated rats (0.23±0.11% vs. 0.09±0.03% in sitagliptin group, both P<0.001). Similar to the tyrosine hydroxylase results, densities of growth-associated protein 43- (Figure 2C) and neurofilament-positive (data not shown) nerves were significantly attenuated in the sitagliptin-treated infarcted rats compared with those in the vehicle-treated infarcted group. These morphometric results mirrored those of norepinephrine contents.

NGF Protein and mRNA Expression

Western blot shows that NGF levels were significantly upregulated 2.1-fold at the remote zone in the vehicle-treated infarcted rats than in sham-operated rats (P<0.001, Figure 3A). When compared with vehicle-treated infarcted rats, sitagliptin-treated infarcted rats had significantly lower NGF levels.

Figure 3.

(A) Western blot analysis of nerve growth factor (NGF) (MW: 13 kDa) in homogenates of the left ventricular (LV) from the remote zone. When compared with vehicle-treated infarcted rats, sitagliptin-treated infarcted rats had significantly lower NGF levels at the remote zone by quantitative analysis. Relative abundance was obtained by normalizing the density of the NGF protein against that of β-actin. Results are presented as mean±SD of 3 independent experiments. (B) Left ventricular NGF mRNA levels. Each mRNA was corrected for a mRNA level of cyclophilin. Each column and bar represents the mean±SD. *P<0.05 compared with sham and sitagliptin. (C) Inducibility quotient of ventricular arrhythmias by programmed electrical stimulation 4 weeks after MI in an in vivo model. *P<0.05 compared with sham and sitagliptin. (D) Infarction significantly reduced the ventricular effective refractory period (VERP). Sitagliptin administration significantly attenuated the shortening of VERP compared with vehicle-treated infarcted rats. ^†P=0.004 compared with sitagliptin.

PCR amplification of the cDNA revealed that the NGF mRNA levels showed a 2.7-fold upregulation at the remote zone in the vehicle-treated infarcted rats compared with sham-operated rats (P<0.001, Figure 3B). In sitagliptin-treated infarcted rats, the NGF mRNA levels were significantly decreased compared with those in the vehicle-treated infarcted rats.

Electrophysiological Stimulation

To further elucidate the physiological effect of attenuated sympathetic hyperinnervation, ventricular pacing was performed. Arrhythmia score in sham-operated rats was very low (0.1±0.3, Figure 3C). In contrast, ventricular tachyarrhythmias consisting of ventricular tachycardia and ventricular fibrillation were inducible by programmed stimulation in vehicle-treated infarcted rats. Sitagliptin treatment significantly decreased the inducibility of ventricular tachyarrhythmias compared with those in the vehicle-treated infarcted group.

Besides, sitagliptin had no statistically significant effect on VERP in the sham-operated groups (data not shown). In the vehicle-treated infarcted group, VERP was significantly reduced by 17±2 ms compared with sham-operated rats (46±3 in vehicle vs. 63±4 ms, P<0.001, Figure 3D), thus creating the dispersion of the VERP in the heart during this ischemic period. Sitagliptin treatment significantly reduced the infarction-induced shortening of VERP by 35% (52±3 vs. 46±3 ms in vehicle, P=0.004). Thus, sitagliptin was noted to have less dispersion of the VERP during ischemia.

Part 2. Ex Vivo Study

Effect of PKA and Epac Signaling on Sitagliptin-Induced CREB and NGF Levels (Experiment 2) To study whether cAMP-regulated PKA and Epac might cooperate to augment sitagliptin-induced CREB and NGF levels, we perfused the infarcted hearts with H-89 and brefeldin A (Figure 4). The level of sitagliptin-induced CREB phosphorylation was increased to 1.52-fold over the vehicle. Moreover, this effect of sitagliptin on CREB phosphorylation was inhibited by coadministration of H-89 (a PKA inhibitor), not brefeldin A (an Epac inhibitor). Compared with sitagliptin alone, the PKA inhibitor significantly increased NGF levels to 2.05-fold. Intriguingly, the Epac antagonist did not change NGF levels induced by sitagliptin.

Figure 4.

Experiment 2. In a rat isolated heart model, the effect of protein kinase A (PKA) and exchange protein directly activated by cAMP (Epac) on phosphorylated cAMP response element-binding protein (p-CREB) (Ser-133) and nerve growth factor (NGF) levels, is shown. Significant increased p-CREB (Ser-133) can be reversed after adding H-89. Brefeldin A (Bre) did not affect the p-CREB (Ser-133) and NGF levels. Relative abundance was obtained by normalizing the density of the NGF protein against that of β-actin. Each point is an average of 3 separate experiments (n=5 per group). *P<0.05 compared with groups treated with vehicle, and sitagliptin (sita)+H-89.

Effect of PKA on HO-1 Expression (Experiment 3) Treatment with a PKA inhibitor confirmed that CREB phosphorylation by sitagliptin occurs through the PKA pathway. We further determined the effect of PKA on the CREB target gene, HO-1. As shown in Figure 5, N6Bz, not 8-CPT, increased HO-1 mRNA expression, which can be reversed by administering KG-501, a specific antagonist that disrupts the CREB:CBP complex and attenuates target gene induction. This result further confirms that PKA can activate CREB, which in turn upregulates the downstream target gene, HO-1.

Figure 5.

Experiment 3. In a rat isolated heart model, the effect of protein kinase A (PKA) and exchange protein directly activated by cAMP (Epac) agonists on heme oxygenase-1 (HO-1) mRNA expression is shown. Significant increased HO-1 expression induced by N6Bz can be abolished by KG-501. 8-CPT did not affect HO-1 expression. Each point is an average of 3 separate experiments (n=5 per group). *P<0.01 compared with groups treated with vehicle, 8-CPT, N6Bz+KG-501, and 8-CPT+KG-501.

Effect of Superoxide on NGF Levels (Experiment 4) To elucidate the role of superoxide in modulating NGF, SIN-1 was assessed in an ex vivo model. Figure 6 shows that SIN-1 significantly increased levels of NGF compared with sitagliptin alone, thus confirming the role of superoxide in mediating NGF levels.

Figure 6.

Experiment 4. In a rat isolated heart model, the effect of superoxide on nerve growth factor (NGF) levels is shown. SIN-1 significantly increased levels of NGF compared with sitagliptin alone. *P<0.01 compared with groups treated with vehicle and sitagliptin (sita)+SIN-1.

Discussion

Our present study shows for the first time that chronic treatment for 4 weeks with sitagliptin leads to attenuated NGF expression, probably through a cAMP/PKA-dependent HO-1 pathway, independently from its glucose-lowering action. The induction of HO-1 along with an increase of CREB phosphorylation by PKA is responsible for attenuated sympathetic innervation and arrhythmias against oxidant stress. These results were concordant for beneficial effects of sitagliptin, as documented structurally by reduction in cardiac nerve sprouting, molecularly by myocardial NGF protein and mRNA levels, biochemically by tissue cAMP, superoxide and norepinephrine levels, pharmacologically by inhibition of PKA and Epac, and electrophysiologically by reduced extent of infarction-induced VERP shortening and improvement of fatal ventricular tachyarrhythmias.

The present study illustrates an additional role of sitagliptin, showing that sitagliptin attenuated NGF-induced sympathetic innervation through a cAMP/PKA-dependent pathway, which upregulated CREB phosphorylation and HO-1 expression. The effect of sitagliptin on attenuated sympathetic innervation was supported by 3 lines of evidence, as shown in Figure 7:

Figure 7.

Schematic representation illustrates the involvements of dipeptidylpeptidase-4 (DPP-4) inhibitors and its related components in nerve growth factor (NGF)-related sympathetic innervation in postinfarcted rats. DPP-4 inhibitors induce cyclic adenosine monophosphate (cAMP) production through glucagon-like peptide-1 (GLP-1)-induced adenylyl cyclase activation. Thereby, cAMP triggers protein kinase A (PKA) activation and subsequent cAMP response element-binding protein (CREB) phosphorylation, which upregulates heme oxygenase-1 (HO-1) expression and attenuates superoxide levels to inhibit NGF expression.

1) Substantial evidence indicates that the balance between oxidants and antioxidants is severely disturbed in post-infarcted myocardial tissues. Treatment with the DPP-4 inhibitor, sitagliptin, raised GLP-1 levels and attenuated ROS production, as assessed by myocardial superoxide and nitrotyrosine, and nuclear oxidative stress, as assessed with the use of DHE staining.

2) Sitagliptin attenuated the NGF levels through a cAMP/PKA-dependent pathway. Sitagliptin administration increased myocardial cAMP levels. Sitagliptin administration significantly attenuated NGF levels, which can be reversed after coadministration of H-89 (a PKA inhibitor). To address the possibility that Epac may influence NGF levels, we treated infarcted rats with an Epac inhibitor (brefeldin A). The selective Epac inhibitor did not increase or reduce NGF levels, suggesting that the effects of cAMP are likely to be mediated by PKA. By what mechanisms might Epac and PKA signaling be differentially regulated? A number of studies have shown that Epac and PKA are activated at different cAMP concentrations. PKA responds to relatively low levels of cAMP, and is activated in the low-to-mid nanomolar range,³¹ levels similar to this study. Our results were consistent with the notion that when cAMP levels were low, Epac may not be activated and cAMP signaling is instead mediated by PKA, inducing growth cone repulsion.¹⁹ Sitagliptin increases intracellular cAMP by activating GLP-1-induced adenylate cyclase. Although DPP-4 inhibition in vivo provides cardioprotection through an increase in GLP-1 signaling, DPP-4 inhibition restores GLP-1 signaling within the physiological, not pharmacological, range. Myocardial benefits of GLP-1 were carried out using either native GLP-1 or recombinant GLP-1 analogues at high concentrations or in a way that induced supraphysiological GLP-1 signaling. The supraphysiological GLP-1 levels by exendin-4 may explain, at least in part, the different mechanisms responsible for attenuated ROS in diabetic Goto-Kakizaki rat islets.²³ Furthermore, the rise in cAMP treated with sitagliptin is within the physiological range and limited to those subcellular regions where adenylate cyclase is localized (eg, cell membrane), by the maximal activity level of adenylate cyclase, and by cAMP degradation by cyclic nucleotide phosphodiesterases. Depending on their relative abundance, distribution, partners, and localization, PKA and Epac may act independently, cooperate, or oppose each other in regulating a specific cellular function.³² Because our results using specific inhibitors reveal that PKA mediates inhibitory effects on NGF, it appears likely that the PKA pathway is of physiological relevance for NGF levels.

3) PKA signaling may regulate HO-1 expression through CREB activation. mRNA expression of the HO-1 gene is induced by the PKA-stimulating agent, N6Bz. Addition of a CREB:CBP antagonist, KG-501, suppressed N6Bz-induced HO-1 expression. The Epac agonist did not affect the expression of HO-1. These results suggest that PKA and CREB activation were involved in cAMP-induced HO-1 expression. The HO-1 gene is a CREB-dependent gene.³³ Increasing evidence has demonstrated that HO-1 catalysis provides the potent antioxidant activity. HO-1 activates the expression of mitochondrial superoxide dismutase in neonatal rat astroglia.³⁴ In addition, the HO-1-deficient mice exhibit serious damage to iron metabolism, resulting in organ oxidative injury.³⁵ The specific role of PKA was further supported by the ex vivo observation that sitagliptin treatment induced CREB phosphorylation, which can be reversed by administering H-89 in infarcted rats. Thus, our findings indicate the close correlation between PKA activation and HO-1 expression in sitagliptin-treated hearts. Indeed, our results were consistent with the findings of Immenschuh et al,³⁶ showing HO-1 gene induction by PKA activation. This extends a previous observation in which in vivo vildagliptin treatment resulted in HO-1 induction of cardiac tissue.²⁴

In this study, we demonstrated attenuated sympathetic reinnervation in sitagliptin-treated hearts. The detailed mechanisms by which sitagliptin affected sympathetic reinnervation remain undefined; however, several factors can be excluded. First, infarct sizes. The degree of sympathetic reinnervation has been related to the infarct sizes.³⁷ Successful fiber reinnervation appears dependent on repopulating sheaths with Schwann cells, which would be injured according to the size of infarction. This possibility was excluded in this study due to similar infarct sizes between the infarcted groups. Second, insulin concentrations. Our results showed a robust phenomenon despite any potentially confounding effect of insulin-mediated sympathetic reinnervation. Initially, it was surmised that DPP-4 inhibitors, acting as a potent incretin, could increase levels of insulin, thereby suggesting a possible mechanism to explain our findings in the in vivo study. However, although a rise in insulin and a decrease in glucose is possible, our results showed that sitagliptin-mediated stimulation of insulin release is not significant because insulinotropic effects of sitagliptin are glucose dependent. Importantly, the attenuated NGF levels observed in our study are reproduced in both the in vivo and ex vivo models; the latter being specifically relevant because in this setting, there is an absence of circulating insulin. Any residual insulin in these hearts at the time of harvesting from the rats would be lost from the tissues during the stabilization period on the Langendorff apparatus, implying that the protective effects that we observe may not be a direct consequence of insulin itself.

Conclusions

These data provide new evidence that sitagliptin protects fatal arrhythmias by attenuating NGF-induced sympathetic innervation via the cAMP/PKA/CREB signaling pathway. The results of the present study support the concept of pleiotropic antioxidant properties of sitagliptin, suggesting that DPP-4 inhibitors might have antiarrhythmic benefits in diabetic patients.

Acknowledgments

This work was supported by grants from China Medical University (CMU102-S-01 and CMU103-S-07) and the Ministry of Science and Technology (MOST 104-2314-B-039-021), Taiwan.

Disclosure

The authors declare that there are no conflicts of interest associated with this manuscript.

Supplementary Files

Supplementary File 1

Methods

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-15-0515

References

• 1.    Matsubara  J,  Sugiyama  S,  Akiyama  E,  Iwashita  S,  Kurokawa  H,  Ohba  K, et al. Dipeptidyl peptidase-4 inhibitor, sitagliptin, improves endothelial dysfunction in association with its anti-inflammatory effects in patients with coronary artery disease and uncontrolled diabetes. Circ J 2013; 77: 1337–1344.
• 2.    Hausenloy  DJ,  Whittington  HJ,  Wynne  AM,  Begum  SS,  Theodorou  L,  Riksen  N, et al. Dipeptidyl peptidase-4 inhibitors and GLP-1 reduce myocardial infarct size in a glucose-dependent manner. Cardiovasc Diabetol 2013; 12: 154.
• 3.    Murakoshi  N,  Aonuma  K. Epidemiology of arrhythmias and sudden cardiac death in Asia. Circ J 2013; 77: 2419–2431.
• 4.    Lee  TM,  Lai  PY,  Chang  NC. Effect of N-acetylcysteine on sympathetic hyperinnervation in post-infarcted rat hearts. Cardiovasc Res 2010; 85: 137–146.
• 5.    Lee  TM,  Lin  SZ,  Chang  NC. Antiarrhythmic effect of lithium in rats after myocardial infarction by activation of Nrf2/HO-1 signaling. Free Radic Biol Med 2014; 77: 71–81.
• 6.    Vracko  R,  Thorning  D,  Frederickson  RG. Fate of nerve fibers in necrotic, healing, and healed rat myocardium. Lab Invest 1990; 63: 490–501.
• 7.    Cao  JM,  Fishbein  MC,  Han  JB,  Lai  WW,  Lai  AC,  Wu  TJ, et al. Relationship between regional cardiac hyperinnervation and ventricular arrhythmias. Circulation 2000; 101: 1960–1969.
• 8.    Cao  JM,  Chen  LS,  KenKnight  BH,  Ohara  T,  Lee  MH,  Tsai  J, et al. Nerve sprouting and sudden cardiac death. Circ Res 2000; 96: 816–821.
• 9.    Ieda  M,  Fukuda  K,  Hisaka  Y,  Kimura  K,  Kawaguchi  H,  Fujita  J, et al. Endothelin-1 regulates cardiac sympathetic innervation in the rodent heart by controlling nerve growth factor expression. J Clin Invest 2004; 113: 876–884.
• 10.    Snider  WD. Functions of the neurotrophins during nervous system development: What the knockouts are teaching us. Cell 1994; 77: 627–638.
• 11.    Colangelo  AM,  Johnson  PF,  Mocchetti  I. Beta-adrenergic receptor-induced activation of nerve growth factor gene transcription in rat cerebral cortex involves CCAAT/enhancer-binding protein delta. Proc Natl Acad Sci USA 1998; 95: 10920–10925.
• 12.    Abate  C,  Patel  L,  Rauscher  FJ 3rd,  Curran  T. Redox regulation of fos and jun DNA-binding activity in vitro. Science 1990; 249: 1157–1161.
• 13.    Vargas  MR,  Pehar  M,  Cassina  P,  Estévez  AG,  Beckman  JS,  Barbeito  L. Stimulation of nerve growth factor expression in astrocytes by peroxynitrite. In Vivo 2004; 18: 269–274.
• 14.    Mukai  E,  Fujimoto  S,  Sato  H,  Oneyama  C,  Kominato  R,  Sato  Y, et al. Exendin-4 suppresses SRC activation and reactive oxygen species production in diabetic Goto-Kakizaki rat islets in an Epac-dependent manner. Diabetes 2011; 60: 218–226.
• 15.    Xu  N,  Engbers  J,  Khaja  S,  Xu  L,  Clark  JJ,  Hansen  MR. Influence of cAMP and protein kinase A on neurite length from spiral ganglion neurons. Hear Res 2012; 283: 33–44.
• 16.    Ding  WQ,  Dong  M,  Ninova  D,  Holicky  EL,  Stegall  MD,  Miller  LJ. Forskolin suppresses insulin gene transcription in islet beta-cells through a protein kinase A-independent pathway. Cell Signal 2003; 15: 27–35.
• 17.    de Rooij  J,  Zwartkruis  FJ,  Verheijen  MH,  Cool  RH,  Nijman  SM,  Wittinghofer  A, et al. Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 1998; 396: 474–477.
• 18.    Huston  E,  Lynch  MJ,  Mohamed  A,  Collins  DM,  Hill  EV,  MacLeod  R, et al. EPAC and PKA allow cAMP dual control over DNA-PK nuclear translocation. Proc Natl Acad Sci USA 2008; 105: 12791–12796.
• 19.    Petersen  RK,  Madsen  L,  Pedersen  LM,  Hallenborg  P,  Hagland  H,  Viste  K, et al. Cyclic AMP (cAMP)-mediated stimulation of adipocyte differentiation requires the synergistic action of Epac- and cAMP-dependent protein kinase dependent processes. Mol Cell Biol 2008; 28: 3804–3816.
• 20.    Murray  AJ,  Tucker  SJ,  Shewan  DA. cAMP-dependent axon guidance is distinctly regulated by Epac and protein kinase A. J Neurosci 2009; 29: 15434–15444.
• 21.    Montminy  MM. Transcriptional regulation by the phosphorylation-dependent factor CREB. Nat Rev Mol Cell Biol 2001; 2: 599–609.
• 22.    Alam  J,  Cook  JL. Transcriptional regulation of the heme oxygenase-1 gene via the stress response element pathway. Curr Pharm Des 2003; 9: 2499–2511.
• 23.    Cheng  Q,  Law  PK,  de Gasparo  M,  Leung  PS. Combination of the dipeptidyl peptidase IV inhibitor LAF237 [(S)-1-[(3-hydroxy-1-adamantyl)ammo] acetyl-2-cyanopyrrolidine] with the angiotensin II type 1 receptor antagonist valsartan [N-(1-oxopentyl)-N-[[2_-(1H-tetrazol-5-yl)-[1,1_-biphenyl]-4-yl]methyl]-L-valine] enhances pancreatic islet morphology and function in a mouse model of type 2 diabetes. J Pharmacol Exp Ther 2008; 327: 683–691.
• 24.    Oeseburg  H,  de Boer  RA,  Buikema  H,  van der Harst  P,  van Gilst  WH,  Silljé  HH. Glucagon-like peptide 1 prevents reactive oxygen species-induced endothelial cell senescence through the activation of protein kinase A. Arterioscler Thromb Vasc Biol 2010; 30: 1407–1414.
• 25.    Sauvé  M,  Ban  K,  Momen  MA,  Zhou  YQ,  Henkelman  RM,  Husain  M, et al. Genetic deletion or pharmacological inhibition of dipeptidyl peptidase-4 improves cardiovascular outcomes after myocardial infarction in mice. Diabetes 2010; 59: 1063–1073.
• 26.    Rocher  A,  Caceres  AI,  Almaraz  L,  Gonzalez  C. Epac signalling pathways are involved in low PO₂ chemoreception in carotid body chemoreceptor cells. J Physiol 2009; 587: 4015–4027.
• 27.    Lee  TM,  Lin  MS,  Chang  NC. Effect of ATP-sensitive potassium channel agonists on ventricular remodeling in healed rat infarcts. J Am Coll Cardiol 2008; 51: 1309–1318.
• 28.    Somekawa  S,  Fukuhara  S,  Nakaoka  Y,  Fujita  H,  Saito  Y,  Mochizuki  N. Enhanced functional gap junction neoformation by protein kinase A-dependent and Epac-dependent signals downstream of cAMP in cardiac myocytes. Circ Res 2005; 97: 655–662.
• 29.    Best  JL,  Amezcua  CA,  Mayr  B,  Flechner  L,  Murawsky  CM,  Emerson  B, et al. Identification of small-molecule antagonists that inhibit an activator: Coactivator interaction. Proc Natl Acad Sci USA 2004; 101: 17622–17627.
• 30.    Dutka  TL,  Mollica  JP,  Lamb  GD. Differential effects of peroxynitrite on contractile protein properties in fast- and slow-twitch skeletal muscle fibers of rat. J Appl Physiol 2011; 110: 705–716.
• 31.    Dostmann  WR,  Taylor  SS. Identifying the molecular switches that determine whether (Rp)-cAMPS functions as an antagonist or an agonist in the activation of cAMP-dependent protein kinase. Biochemistry 1991; 30: 8710–8716.
• 32.    Métrich  M,  Berthouze  M,  Morel  E,  Crozatier  B,  Gomez  AM,  Lezoualc’h  F. Role of the cAMP-binding protein Epac in cardiovascular physiology and pathophysiology. Pflügers Arch 2010; 459: 535–546.
• 33.    Lonze  BE,  Ginty  DD. Function and regulation of CREB family transcription factors in the nervous system. Neuron 2002; 35: 605–623.
• 34.    Frankel  D,  Mehindate  K,  Schipper  HM. Role of heme oxygenase-1 in the regulation of manganese superoxide dismutase gene expression in oxidatively-challenged astroglia. J Cell Physiol 2000; 185: 80–86.
• 35.    Poss  KD,  Tonegawa  S. Reduced stress defense in heme oxygenase 1-deficient cells. Proc Natl Acad Sci USA 1997; 94: 10925–10930.
• 36.    Immenschuh  S,  Kietzmann  T,  Hinke  V,  Wiederhold  M,  Katz  N,  Muller-Eberhard  U. The rat heme oxygenase-1 gene is transcriptionally induced via the protein kinase A signaling pathway in rat hepatocyte cultures. Mol Pharmacol 1998; 53: 483–491.
• 37.    Gulati  AK. Evaluation of acellular, and cellular nerve grafts in repair of rat peripheral nerve. J Neurosurg 1988; 68: 117–123.