NDRG2: A Newly Identified Mediator Of Insulin Cardioprotection Against Myocardial Ischemia-reperfusion Injury

BackgroundThe N-myc downstream-regulated gene2(NDRG2), shares roughly60%residueidentity with other members of the NDRG2family of proteins. NDRG2is involved incell growth, differentiation, stress and hormonal responses. Previous studies have shownthat NDRG2expression is regulated by many anabolic and catabolic hormones, such asdexamethasone, androgens and aldosterone. Although NDRG2is highly expressed in theheart, its function in the cardiovascular system is largely unknown. In our previous study,we discovered that the NDRG2expression in the heart is modulated by myocardial I/R, implicating its involvement in cardiac tissue injury following I/R.Cardiovascular events such as acute myocardial infarction are leading causes ofmorbidity and mortality. After detrimental occlusion of coronary artery, restoringcoronary blood flow (reperfusion) is currently the main clinical strategy for salvagingviable myocardium. However, restoring blood flow to the ischemic tissue itself causesmyocardial reperfusion injury, paradoxically reducing the benefit of reperfusion andcausing cell death and functional tissue damage. Thus, clarifying the mechanism ofreperfusion injury and finding new targets to alleviate the resulting tissue damage is aresearch priority.The cardioprotective property of insulin against I/R injury is well-documented.Insulin has also been shown to protect the heart from I/R-related damage by activating aprogram of intracellular protein kinases termed reperfusion injury salvage kinases(RISK), involving the parallel activation of PI3K-Akt, JAK-STAT and the extracellularregulated kinase (ERK) system. However, the mechanism behind insulin-inducedcardioprotection is not completely understood. NDRG2is a substrate of protein kinaseAkt and is involved in insulin signaling transduction mediated by Akt. In our previouswork, NDRG2was shown to play an important role in Akt-mediated protection ofpancreatic β cells against free fatty acid (FFA)-triggered damage. However, the potentialrole of NDRG2in insulin cardioprotection against I/R damage has not been reported.ObjectiveIn the present study, we investigated the expression and phosphorylation of NDRG2in cardiomyocytes and myocardial tissue subjected to myocardial I/R injury as well astheir regulation by insulin administration and molecular mechanism involved.Furthermore, we explored whether NDRG2was involved in Akt-mediatedcardioprotection stimulated by insulin during myocardial I/R injury.Materials and Methods1. To produce ischemia/reperfusion, male Sprague-Dawley (SD) rats weighing140–180g were anesthetized with pentobarbital sodium, and ventilated60times/min with avolume-cycled respirator. During the surgery, blood pressure was monitored byinserting a PE-50catheter filled with heparinized saline and connected to acomputerized data acquisition system into the left femoral artery. Heart rate wasmonitored by ECG. Through a left thoracotomy incision, a ligature was placed aroundthe left anterior descending coronary artery (LAD). After30min of ischemia, theslipknot around the LAD was released. Sham-operated animals were subjected to thesame surgical procedures, except for coronary artery occlusion.2. In the first part of the experiment, in order to observe the change in phosphorylation ofNDRG2in the myocardium, rats were randomly divided into the following groups:sham operation, ischemia only (30min),30min of ischemia followed by1,2,3or4h of reperfusion (n=5). In the second part of the experiment, in order to study theeffect of insulin on the Akt/NDRG2signaling pathway, rats were randomized toreceive one of the following treatments:(1) sham operation;(2) vehicle (0.9%NaCl);(3) GIK (glucose:200g/L, insulin:60U/L, potassium:60mEq/L, intravenousinfusion at4mL/kg per h for4h, beginning5min before reperfusion);(4) GK(glucose potassium)(n=16). Rat heart function was evaluated by echocardiographyimmediately prior to myocardial I/R in order to establish a baseline.3. Neonatal rat cardiomyocyte culture and in vitro simulated ischemia/reperfusion model;Culture of H9C2cardiomyocytes and rat INS-1cells4. Treatment of cells with siRNAs and lentiviral-vectored shRNA targeting NDRG2: TheRNAi sequences targeting rat NDRG2are5-CCAAACGUCCAGCGAUAUUCATT-3. The siRNAs were transfected into cells with Lipofectamine2000(Invitrogen, Carlsbad, CA, USA). A pilot study with the GFP-expressing controllentivirus demonstrated that70–80%of the cells were transduced, as detected byfluorescence microscopy.5. Virus vector inoculation in vivo：Male SD rats were anesthetized with pentobarbitalsodium, intubated and ventilated as described above. A thoracotomy incision wasmade in the fourth intercostal space to expose the heart. Inoculation of rat hearts with NDRG2-silencing lentivirus was conducted by injecting it into the free anterior wallmyocardium of the left ventricle (LV) at three different sites.6. Myocardial infarct size was evaluated by Evans Blue/TTC dual staining；TUNELstaining，Western blot analysis and immunostaining of caspase3active form assayswere used to determine cardiomyocyte apoptosis; Flow cytometric analysis ofcellular apoptosis was performed; Echocardiography was conducted to measure Leftventricular ejection fraction (LVEF) and left ventricular fraction shortening (LVFS).Reverse-Transcription PCR was performed to detect the mRNA level of NDRG2;Immunoblot assay was used to examine the total level and phosphorylated level ofprotein.Results1. The level of phosphorylated NDRG2increased in the early reperfusion period, whileno change of total NDRG2level was observed. However, phosphorylation ofNDRG2gradually decreased and reached the basal level at4h after reperfusion. Asimilar trend was observed in cultured primary neonatal cardiomyocytes afterexposure to simulated I/R, with NDRG2phosphorylation reaching the lowest levelat6h of re-oxygenation.2. Infusion of GK during reperfusion significantly increased the blood glucoseconcentration, but no difference was observed between GIK and vehicle treatment inthe myocardial I/R rats at the end of the4-h reperfusion. Compared with the vehiclegroup, GIK treatments increased the concentration of plasma insulin at the end of the4-h reperfusion.3. Administration of insulin (GIK) at the onset of reperfusion provided a significantanti-apoptotic effect, as evidenced by a significantly lower expression of the keyapoptosis-related protein, active caspase3, in both Western blot andimmunohistochemical analyses, compared with either the I/R+V or I/R+GK group.In addition, a similar trend could also be observed by TUNEL staining. The number of apoptotic cells within the myocardium was significantly less than that of the othertwo groups.4. Administration of insulin significantly decreased the infarct size after24h ofreperfusion compared with the I/R+V group, and no significant difference in AAR(AAR) was found between groups. In addition, left ventricular function wasevaluated by echocardiography after30min of ischemia and72h of reperfusion.Insulin administration significantly enhanced LVEF as well as LVFS, compared withthe I/R+V group and I/R+GK group.5. Insulin treatment significantly reversed the decrease of NDRG2phosphorylation atThr348and Akt phosphorylation at Ser473induced by30min of ischemia and4h ofreperfusion in vivo. In cultured neonatal cardiomyocytes subjected to simulated I/R,insulin treatment significantly increased the level of Akt phosphorylation at Ser473,which was blocked by the PI3K inhibitor wortmannin. Insulin also increased thephosphorylation of NDRG2at Thr348, which was abolished by both wortmanninand the specific Akt inhibitor.6. The anti-apoptotic effect of insulin on cardiomyocytes was significantly abolishedby wortmannin or Akt inhibitor co-treatment with insulin, as evidenced byup-regulation of active caspase3and increased TUNEL (green) staining of cells.Moreover, H9C2cells were likewise treated and analyzed by flow cytometry, whichshowed that wortmannin or Akt inhibitor facilitated cellular apoptosis and impairedthe cardioprotective effect of insulin.7. After transfection with siRNA targeting NDRG2for48h, NDRG2expression wasexamined in rat INS-1cells. Western blotting showed that NDRG2siRNA couldsuppress endogenous NDRG2expression in INS-1cells. Based on this siRNAsequence, a lentivirus vector carrying the NDRG2shRNA was constructed for moreefficiently delivery into rat cardiomyocytes. Both mRNA and protein levels ofNDRG2were significantly down-regulated after transduction of culturedcardiomyocytes with the NDRG2shRNA lentivirus. 8. Although insulin induced the activation of Akt in NDRG2-silenced cardiomyocytes,they showed lower phosphorylation levels of NDRG2at Threonine348and sufferedmore severe apoptosis, as indicated by up-regulation of active caspase3expression,compared with cardiomyocytes transduced with the control shRNA (scramblelentivirus). Meanwhile, insulin produced a similar protective effect in thecardiomyocyte cell line H9C2transfected with the scramble lentivirus. However, thecardioprotective effect of insulin in NDRG2knockdown cardiomyocytes wassignificantly reduced, as indicated by a higher apoptotic rate in NDRG2-silencedcardiomyocytes.9. Twenty-four hours after lentivirus injection, moderate inflammatory reaction wasobserved in myocardium, as demonstrated by CD11b immunohistological staining.Ninety-six hours after injection with the NDRG2-silencing lentivirus, myocardialtissues around the injection sites and area remote to the injection site were harvestedfor evaluation of NDRG2expression. PCR and Western blot analyses showed drasticreductions of both NDRG2mRNA and protein in heart tissue transduced withNDRG2shRNA lentivirus, while that transduced with scramble virus showed nodifferences compared with the control. Immunohistochemical staining demonstratedthe down-regulation of NDRG2in the ischemic area, such as different regions of LVfree wall, while there is no NDRG2reduction observed in remote area such as theventricular septum and right ventricular wall. Meanwhile, lentivirus injections didnot induce alteration of total NDRG2and phosphorylated NDRG2expression in theremote myocardium.10. In GIK treated animals, compared with scramble lentivirus-treated hearts,NDRG2-silencing transduced hearts showed a trend of myocardial I/R-aggravatedapoptosis, as demonstrated by enhanced expression of active caspase3and increasedTUNEL (green) staining in cardiomyocytes. In addition, in agreement with the invitro experiments above, NDRG2silencing partly attenuated the beneficial effects ofinsulin, as evidenced by increased infarct size and compromised left ventricular function.11. Immunohistological detection showed no significant down-regulation of NDRG2protein in the border area compared with remote area. TUNEL staining showed nosignificant difference in the apoptotic index of border zone cardiomyocytes betweenGIK+scramble lentivirus group and GIK+NDRG2shRNA lentivirus group.ConclusionThe investigation into the role of NDRG2phosphorylation in the cardioprotectiveeffect of insulin against I/R injury in this study led to several major findings. First,myocardial NDRG2phosphorylation was found to be decreased following I/R, possiblyrelated to the change in Akt phosphorylation. Second, treatment with insulin duringreperfusion significantly increased myocardial NDRG2phosphorylation via aPI3K/Akt-dependent mechanism. Third, siRNA-mediated NDRG2knockdown facilitatedapoptosis of I/R-induced cardiomyocytes. Most importantly, NDRG2knockdownsignificantly blunted the protection of the myocardium afforded by insulin treatment.These results demonstrate that NDRG2is a novel player in the insulin-inducedcardioprotection against myocardial I/R injury.