Involvement of Endothelin 1 in Remote Preconditioning-Induced Cardioprotection through connexin 43 and Akt/GSK-3β Signaling Pathway

The present study was aimed to explore the role of endothelins in remote preconditioning (RP)-induced myocardial protection in ischemia-reperfusion (IR) injury. RP stimulus was given by subjecting hind limb to four cycles of ischemia and reperfuion (5 minutes each) using blood pressure cuff in male rats. Following RP, hearts were isolated and subjected to 30 minutes of ischemia and 120 minutes of reperfusion on Langendorff apparatus. The extent of myocardial injury was determined by measuring the levels of LDH-1, CK-MB and cardiac troponin T (cTnT) in coronary effluent; caspase-3 activity and Bcl 2 expression in heart (apoptosis); infarct size by triphenyl tetrazolium chloride and contractility parameters including left ventricular developed pressure, dp/dtmax dp/dtmin and heart rate. RP reduced ischemia reperfusion-induced myocardial injury, increased the levels of endothelin 1 (in blood), Akt-P, GSK-3β-P and P-connexin 43 (in hearts). Pretreatment with ETA receptor antagonist, BQ 123 (1 and 2 mg/kg), ETB receptor antagonist, BQ 788 (1 and 3 mg/kg) and dual inhibitor of ETA and ETB receptor, bonsentan (25 and 50 mg/kg) abolished these effects of RP. However, the effects of bonsentan were more pronounced in comparison to BQ 123 and BQ 788. It is concluded that RP stimulus may release endothelin 1 in the blood, which may activate myocardial ETA and ETB receptors to trigger cardioprotection through connexin 43 and Akt/GSK-3β pathway.

, which are present on the different body parts including heart 19 . Endothelin 1 and endothelin 2 activate ET A receptors to a comparable degree; however, endothelin 3 does not activate ET A receptors. On the other hand, all endothelins have same affinity for ET B receptors 20 .
It is interesting to note that the stimuli or agents that produce myocardial injury are reported to precondition the heart. Ischemia 3 , free radicals 21 and calcium 22 are well documented to produce myocardial injury. However, there has been a large of studies showing that these agents also precondition the heart and protect it from subsequent sustained ischemic injury 3,21,22 . Similarly, there have been studies suggesting that endothelin may contribute in inducing myocardial injury during ischemia and reperfusion 23,24 . On the other hand, exogenous administration of endothelin has also been shown to produce preconditioning like effects in heart 25,26 . Accordingly, it is hypothesized that the endothelium of the blood vessels of the remote organ may respond to short episodes to ischemia by releasing endothelins, which may act as cardioprotective humoral factors. However, the role of endothelins in remote ischemic preconditioning-induced cardioprotection is not explored yet. Therefore, the present was performed to explore the role of endothelins in remote preconditioning-induced myocardial protection in ischemia reperfusion injury in rats.

Material and Methods
Animals, Chemicals and Drugs. Wistar albino male rats (250-300 g) were employed in the study and rats were purchased from Jilin University. The animal studies were approved by the Institutional Review Board and Institutional Animal Care and Use Committee of the Jilin University, and all of the experiments were performed in accordance with the approved protocols. BQ-123, a selective ET A receptor antagonist, and BQ788, a selective ET B receptor antagonist, were procured from EMD Bioscience, USA. Bonsentan, a dual blocker of ET A and ET B receptors, was procured from Sigma-Aldrich, USA. The diagnostic kits for quantitative estimation of LDH-1, CK-MB and cardiac troponin T were procured from Jiancheng Reagent Co, Nanjing, China. ELISA kits for quantitative estimation of endothelins, caspase 3, bcl-2, phosphorylated GSK-3β, Akt and connexin 43 were procured from Life BioSpan Biosciences, USA. BQ-123 was solubilized in water, while BQ788 and bonsentan were dissolved in DMSO. The doses of BQ 123 27 , BQ 788 28 and bonsentan 29 were selected the basis of published literature.

Remote Preconditioning and Isolated Heart preparation of Langendorff System. Thiopentone
(35 mg kg −1 , i.p.) was used to anesthetize the animals and left hind limb was tied with a blood pressure cuff at the inguinal level. The cuff was alternatively inflated (up to 150 mm of Hg) and deflated for 5 minutes to induce ischemia and reperfusion, respectively. Four such episodes (each episode consisting of 5 minutes ischemia and 5 minutes reperfusion) constituted remote preconditioning stimuli 30 . After last episode of remote preconditioning, rat was sacrificed and heart was isolated to perfuse with physiological solution (Kreb's Henseleit solution) on the Langendorff system 31 . Global ischemia was produced for 30 minutes by blocking perfusion to the heart and reperfusion was given for 120 minutes by restoring perfusion to the heart. A balloon, connected to catheter and pressure transducer, was inserted in the left ventricle to record contractility parameters (mentioned below). The coronary perfusate was collected at different time intervals to measure the biochemical parameters and assess the extent of myocardial injury.
Determination of Biochemical Parameters of Heart Injury. The injury to heart was assessed by measuring the release of specific biomarkers from the heart into the coronary perfusate. The measurement of LDH-1, CK-MB and cardiac troponins (cTnT) were done using commercial kits. The release of LDH-1 from heart was determined by measuring its levels in the coronary perfusate immediately before global ischemia and 30 minutes after reperfusion. The levels of CK-MB and cTnT were measured in coronary effluent before global ischemia and immediately after reperfusion.

Determination of Myocardial Infarct
Size. The extent of heart injury and its modulation with remote preconditioning was determined macroscopically by determining myocardial infarction with triphenyltetrazolium chloride (TTC) staining. TTC staining is one of the standard methods of assessing cell viability. TTC stains viable cells red, while dead cells remain unstained. In this method, heart was cut in thin slices and stained with TTC. Thereafter, area of infarcted (unstained) and viable portions (stained) were assessed using planimetry 32 .

Measurement of Functional Parameters of Heart.
A balloon, connected to catheter and pressure transducer, was inserted in the left ventricles of rat hearts to determine heart contractility including left ventricular developed pressure (LVDP), dp/dt max (rate of contraction), dp/dt min (rate of relaxation) and heart rate. These contractility parameters were measured throughout the experiment and comparison was made between pre-ischemic and post-ischemic contractility parameters.
Determination of Endothelin levels in the Plasma. After last episode of remote ischemic stimulus, the blood was removed from rats and the plasma was separated by centrifugation. In ischemia reperfusion group, blood was withdrawn immediately before heart isolation. The levels of endothelin 1, endothelin 2 and endothelin 3 in the plasma were measured using commercially available ELISA kits. Determination of Apoptosis Markers, GSK-3β-P, Akt-P and P-connexin 43 in the Heart. After global ischemia-reperfusion on the Langendorff apparatus, the hearts were isolated. The half portion of the heart was homogenized in the lysis solution containing NP-4 and protease inhibitor. The other half portion of the heart was used to measure myocardial infarction using TTC staining. The homogenized mixture was centrifuged and the supernatant was used to measure the levels of p-GSK-3β-S9, Akt-P, P-connexin 43, bcl-2 along with caspase 3 activity using commercially available ELISA kits. Caspase 3 activity was standardized with respect to total protein content 33 . Experimental Groups. Animals were distributed in various groups (n = 6) to meet the objectives of study and these included: i. Ischemia reperfusion injury: Isolated rat hearts were subjected to thirty minutes of ischemia and one twenty minutes of reperfusion on the Langendorff apparatus; ii. Remote preconditioning: Four episodes of ischemia and reperfusion (each of five minutes) were given in rats before prolonged ischemia and reperfusion on the Langendorff apparatus; iii. and iv. BQ-123(1 and 2 mg/kg i.p.) in remote preconditioning: BQ-123 was administered in rats thirty minutes before remote preconditioning; v. and vi BQ788 (1 and 3 mg/kg i.p.) in remote preconditioning: BQ-788 was administered in rats thirty minutes before remote preconditioning; vii. and viii. Bonsentan (25 and 50 mg/kg i.p.) in remote preconditioning: Bonsentan was administered in rats thirty minutes before remote preconditioning; ix. DMSO (0.5 ml/kg) in remote preconditioning: 10% DMSO (solvent of BQ 788 and bonsentan) was administered in rats thirty minutes before remote preconditioning; x. Normal Rats: Neither treatment nor ischemia reperfusion was given to animals. The hearts were isolated to measure the basal levels of GSK3β-P, Akt-P, P-connexin 43, bcl-2 and caspase 3 activity.
Statistical Analysis. The data were presented in the form of mean ± standard deviation (S.D.). Two-way ANOVA was employed to compare the data of LDH-1, CK-MB, cTnT, LVDP, +dp/dt max , −dp/dt min, and heart rate. One-way ANOVA was used to compare the data of infarct size, P-Akt, P-GSK-3β, P-connexin 43, bcl-2 and caspase 3. Unpaired t test was used to analyse the data of endothelin levels. Tukey's test was employed for post hoc analysis. Statistical significance was calculated by fixing p < 0.05.

Results
Ischemia reperfusion produced injury to isolated rat hearts. In isolated rat hearts, prolonged ischemia followed by reperfusion produced marked injury to heart as evident by an increase in LDH-1 levels in the coronary perfusate at 30 minutes of reperfusion in comparison to basal levels (before ischemia) (Fig. 1a). Similarly, there was an increase in the levels of other biochemical markers including CK-MB and cTnT in the coronary perfusate during reperfusion period in comparison to basal levels (before ischemia) (Figs 1b and 2a). Moreover, in ischemia-reperfusion subjected hearts, there was a marked decrease in TTC staining area suggesting the significant rise in dead tissue area or myocardial infarction (Figs 2b and 3a). Ischemia-reperfusion injury triggered apoptosis as depicted by increase in caspase-3 activity and decrease in bcl-2 expression in hearts (Fig. 4a,b). The functional studies also depicted a marked increase in myocardial injury due to prolonged episodes of ischemia and reperfusion. In the present study, there was a marked reduction in heart contractility parameters including LVDP, dp/dt max and dp/dt min during reperfusion phase in comparison to pre-ischemic phase ( Tables 1, 2 and 3). Moreover, prolonged ischemia and reperfusion also reduced heart rate in a significant manner during reperfusion period ( Table 4).

Ischemia reperfusion altered the levels of GSK-3β-P, Akt-P and P-connexin 43 in hearts.
Along with myocardial injury, prolonged ischemia and reperfusion also led to significant alterations in the phosphorylated forms of Akt, GSK-3β and connexin 43 in heart. In comparison to normal hearts (not subjected to ischemia reperfusion episodes), there was a marked reduction in the levels of Akt-P, GSK-3β-P and P-connexin 43 in ischemia and reperfusion subjected isolated rat hearts (Figs 5a,b and 6).

Remote preconditioning decreased myocardial injury in ischemia reperfusion subjected hearts.
Four short cycles of ischemia and reperfusion (5 minutes each) to the hind limb in the form of remote preconditioning protected the isolated rat hearts from prolonged ischemia-reperfusion injury. The release of specific biochemical markers of myocardial injury including LDH-1, CK-MB and cTnT was markedly reduced from remote preconditioning subjected hearts in comparison to ischemia reperfusion group (Figs 1a,b and 2a). Moreover, there was an increase in the extent of TTC staining in remote preconditioning subjected hearts suggesting a decrease in myocardial infarction (Figs 2b and 3a). There was significant decrease in capspase-3 activity and increase in Bcl-2 expression in hearts suggesting decrease in apoptosis (Fig. 4a,b). Furthermore, remote prconditioning also led to improvement in heart contractility parameters including LVDP, dp/dt max , dp/dt min and heart rate suggesting the functional improvement in heart (Tables 1, 2, 3 and 4).
Remote preconditioning modulated the levels of endothelin 1 in plasma and GSK-3β-P, Akt-P and P-connexin 43 in heart. Following four short episodes of hind limb ischemia and reperfusion (remote ischemic preconditioning stimulus), there was an increase in the levels of endothelin 1 in the plasma in comparison to ischemia-reperfusion group. However, there was no change in the levels of endothelin 2 and endothelin 3 (Fig. 3b). Thirty minutes of global ischemia and 120 minutes of reperfusion decreased the levels of phosphorylated forms of Akt, GSK-3β and connexin 43 in the heart in comparison to normal hearts. However, remote preconditioning stimulus led to increase in the levels of Akt-P, GSK-3β-P and P-connexin 43 in the heart following prolonged ischemia and reperfusion (Figs 5a,b and 6).
Endothelin receptor antagonists modulated the cardioprotective effects of remote preconditioning. Pretreatment with BQ 123 (a selective inhibitor of ET A receptors) and BQ 788 (a selective inhibitor of ET B receptors) before remote preconditioning (thirty minutes prior to first episode of remote preconditioning stimulus) mitigated the cardioprotective effects of remote preconditioning. Administration of both BQ 123 (1 and 2 mg/kg) and BQ 788 (1 and 3 mg/kg) led to increase in the levels of LDH-1, CK-MB and cTnT in coronary perfusate in remote preconditioning groups (Figs 1a,b and 2a). There was also an increase in myocardial infarct size (Fig. 3a), increase in caspase-3 activity (Fig. 4a), decrease in Bcl-2 expression (Fig. 4b) and decrease in contractility parameters (Tables 1, 2, 3 and 4). Moreover, these pharmacological agents decreased the levels of Akt-P, GSK-3β-P and P-connexin 43 in the hearts of remote preconditioning group (Figs 5a,b and 6). Pretreatment with bonsentan (a non-selective inhibitor of ET A and ET B receptors) mimicked the effects of BQ 123 and BQ 788 in remote preconditioning groups. However, the effects of bonsentan (25 mg/kg and 50 mg/kg) on remote preconditioning were much pronounced in comparison to BQ 123 and BQ 788. On other words, bonsentan abolished the effects of remote preconditioning to a greater extent as compared to BQ 123 and BQ 788. However, administration of vehicle (10% DMSO) did not influence the parameters of myocardial injury and kinases in heart in ischemia reperfusion subjected rats.

Discussion
In the present study, we observed remote preconditioning reduced ischemia-reperfusion-induced cardiac injury, as indicated by reduction in the levels of LDH-1, CK-MB and cTnT in the coronary perfusate. These are specific biomarkers of cardiac injury and their release takes place during rupturing of cell membrane 34 . Remote preconditioning also attenuated ischemia-reperfusion induced increase in apoptosis as indicated by decrease in caspase-3 activity and increase in bcl-2 expression. It also reduced the extent of myocardial infarction (demonstrated by TTC staining) and improved myocardial contractility including LVDP, dp/dt max dp/dt min and heart rate. Remote preconditioning-induced myocardial protection is widely supported by preclinical as well as clinical studies 35,36 .
Moreover, remote preconditioning led to an increase in the levels of endothelin 1 in the blood, without altering the levels of endothelin 2 or endothelin 3. Endothelin-1 is the most important member of endothelin family and it is primarily released from the endothelium. It may be possible to suggest that during short cycles of ischemia and reperfusion to hind limb, the endothelial cells may respond to ischemia by releasing endothelin 1, which may travel to the heart as a humoral factor. There have been previous studies documenting that exogenous administration of endothelin 1 produces preconditioning like effects on heart and exerts cardioprotection 25,26 . Moreover, the protective effects of endothelin-1 pretreatment on neonatal rat cardiomyocytes against hypoxia-induced injury has also been reported 37 . Endothelin-1 has also been shown to promote the survival of mesenchymal stem cells 38 . The activation of endothelin 1 signaling pathway has been shown to exert cardioprotection against chronic intermittent hypoxia 24,39 . However, it is the first report documenting that endothelin 1 may be the possible humoral factor released during remote preconditioning to protect heart from sustained ischemia reperfusion injury in rats. The cardioprotective role of endothelin 1 in remote preconditioning was affirmed by pretreatment of rats with selective as well as non-selective ET A and ET B antagonists before remote preconditioning. Pretreatment with BQ-123 (a selective ET A receptor antagonist); BQ-788 (a selective ET B receptor antagonist) and bonsentan (nonselective ET A and ET B receptor antagonist) mitigated the protective effects of remote preconditioning on heart. It again emphasizes that remote preconditioning may increase the release of endothelin 1, which may travel to the heart through blood to activate myocardial ET A and ET B receptors to trigger cardioprotection. It is also worth mentioning that the attenuating effects of bonsentan on cardioprotection were significantly higher in comparison to BQ 123 and BQ 788. It suggests the crucial role of both ET A and ET B receptors in triggering cardioprotection during remote preconditioning.
To explore the molecular mechanisms involved in remote preconditioning, the expression levels of phosphorylated forms of pro-survival protein kinases Akt, GSK-3β along with gap junction protein, connexin 43 were measured in hearts. Remote preconditioning significantly restored the levels of Akt-P, GSK-3β-P and P-connexin 43 in ischemia-reperfusion subjected rats. Akt is a principal pro-survival protein kinase and Akt-P is the active form of this enzyme. Akt phosphorylates other downstream regulator enzymes including GSK-3β. In contrast to other enzymes, GSK-3β exists in the active form in the non-phosphorylated state and it is inactivated on phosphorylation 40 . Thus, an increase in Akt-P and GSK-3β-P levels during remote preconditioning suggests that there is an activation of Akt and inhibition of GSK-3β enzyme during remote preconditioning. Studies have documented that activation of pro-survival protein kinase Akt is important in remote preconditioning-induced cardioprotection 41 . There have been studies suggesting that inhibition of GSK-3β protects the heart from ischemia-reperfusion injury 40 . Moreover, an increase in the phosphorylation of GSK-3β and its inactivation has been documented in ischemic preconditioning as well as in remote preconditioning-induced cardioprotection 40,42,43 . Connexin 43 (also termed as Gap junction alpha-1 protein) is a critical component of gap junctions, which allow intercellular communication to take place and electrical impulses can pass between cells through these junctions 44 . There have been studies showing that an increase in gap junction coupling and preservation of phosphorylation of connexin 43 may participate in hypoxic preconditioning and remote preconditioning-induced cardioprotection 45,46 .
In the present study, pretreatment with BQ123, BQ788 and bonsentan attenuated the levels of pro-survival kinases Akt-P, GSK-3β-P and P-connexin 43. The effects of bonsentan were more significant as compared to BQ 123 and BQ 788, suggesting the key role of both ET A and ET B receptors in modulating the levels of downstream protein kinases and connexin 43. Therefore, it may be proposed that endothelin may activate both ET A and ET B receptors located on the heart to activate Akt, inhibit GSK-3β and increase P-connexin 43. There have been studies documenting that endothelin 1 may activate Akt phosphorylation and inhibit GSK-3β to produce its biological actions [47][48][49] . It is shown that the actions of endothelin-1 on astrocytes are dependent on connexin 43 levels 50 . Furthermore, there has been a study showing that endothelin 1 produces phosphorylation of Akt and GSK-3β through connexin 43 in cardiomyocytes 51 . Therefore, it may be proposed that remote preconditioning Figure 3. Consequences of I/R, RP and endothelin receptor antagonists on (a) infarct size measured by TTC staining; 3(b) endothelin levels in coronary effluent, measured immediately before sacrificing rats. *p < 0.05 vs. I/R; **p < 0.05 vs. RP. For infarction, one way ANOVA was followed by Tukey's test and statistic parameters were n = 6, F(7, 40) = 304.7, p < 0.0001. For endothelin levels, unpaired two tail t-test was applied to compare ET levels in normal and RP: For ET1, n = 6, t = 9.204, df = 10, p < 0.0001; For ET2, n = 6, t = 1.746, df = 10, p = 0.1114; For ET3, n = 6, t = 1.464, df = 10, p = 0.1739. may increase the levels of endothelin 1 in the blood, which may subsequently activate myocardial ET A and ET B receptors to trigger cardioprotective signaling involving activation of Akt, inhibition of GSK-3β and preservation of connexin 43 phosphorylation (Fig. 7).

Conclusion
Remote preconditioning stimulus may activate the endothelial lining to release endothelin 1 in the blood, which may activate myocardial ET A and ET B receptors to trigger cardioprotection through connexin 43 and Akt/GSK-3β signaling pathway.