Contribution Of Extracellular HSP60to The Inflammation In Ischemic Myocardium

Heat shock protein60(HSP60), a member of the heat shock proteins family, is normallylocated in the mitochondria and cytoplasm. We previously found that HSP60appeared in theserum and the cardiac plasma membrane fraction after permanent coronary artery occlusion inrats, indicating the translocation of HSP60into extracellular space under myocardial ischemia.Exogenous HSP60(exHSP60) has been reported to induce inflammatory responses throughmechanisms both dependent and independent of toll-like receptor (TLR)2and TLR4inimmune cells. Both TLR2and TLR4are type I transmembrane receptors expressed by cardiacmyocytes and other cell types, in addition to immune cells. Whether endogenous HSP60(enHSP60) that appears in extracellular space under myocardial ischemia inducesinflammation in cardiac myocytes remains unknown. The present work tested the hypothesisthat HSP60in extracellular space may contribute to the inflammation in ischemic cardiacmyocytes through activating TLR2and/or TLR4. We presume that extracellular HSP60mayrepresent an important trigger for ischemic myocardial inflammation in the absence ofmicrobial pathogens. This study examined the pro-inflammatory effects of exHSP60and theunderlying signaling pathway mediated by TLR2/4in cultured H9c2rat ventricular cell line,and examined the contribution of enHSP60in extracellular space to cardiac inflammation inH9c2cells exposed to ischemia, as well as the intact rats subjected to coronary artery ligation.MethodsExperiments in H9c2rat ventricular cell line: H9c2cell were cultured with high-glucoseDMEM containing10%fetal bovine serum. When the cells were80%confluent, they wereserum-deprived for12h for synchronization, and then exposed to HSP60or ischemia.Ischemia was induced by exposing cells to1%O2–94%N2–5%CO2in serum-freelow-glucose DMEM for12h. HSP60levels in cell-culture supernatant were measured byELISA method to examine whether HSP60appears in the extracellular space in response toischemia. To determine the inflammatory responses, the mRNA and protein levels of TNF-αand IL-6were determined by real-time RT-PCR and ELISA, respectively. The interactionbetween HSP60, TLR2/4, and their downstream adaptor proteins MyD88/Trif was examined by co-immunoprecipitation (co-IP) experiments. The activities of p38, JNK and NF-κB, theeffector molecules downstream of the TLR2/4was examined by Western blot andimmunostaining. The effects of inhibiting TLR2/4, MyD88, p38, JNK, NF-κB on theinflammation responses induced by exHSP60and ischemia were observed by using specificblockers in H9c2cardiomyocytes.Experiments in intact rat model: Myocardial ischemia was induced by ligation of theleft-anterior descending coronary artery (LAD) for4h in anaesthetized rats, Serum HSP60(extracellular HSP60) was detected by ELISA. To examine the inflammation responseinduced by myocardial ischemia, TNF-α and IL-6mRNA levels in myocardial ischemic areawere determined by RT-PCR, serum content of TNF-α and IL-6were detected by ELISA.Co-IP experiments were performed to investigate the association between enHSP60, TLR2/4,and MyD88/Trif. Serum HSP60(extracellular HSP60) was neutralized by using anti-HSP60antibodies (20μg/kg, intravenously injected15min before LAD ligation) to observe theeffects on ischemia-induced inflammation.Results1. ExHSP60induced inflammatory cytokine production in H9c2cardiomyocytes viaTLR4-MyD88-p38/NF-kB pathwayIn H9c2cells, exHSP60significantly increased the expression and release of TNF-α andIL-6. Time-course studies showed significantly increase at1h, and the greatest increase at6h.A dose response was observed for exHSP60in H9c2cells. Five mg/mL exHSP60increasedthe mRNA levels and the release of TNF-α and IL-6more than1mg/mL exHSP60afterexHSP60incubated for6h.A series of experiments was done to investigate the signaling pathway(s) activated byexHSP60. First, we examined the interaction of exHSP60with TLR2/4by co-IP experiments.When the lysates of H9c2cells were precipitated with antibodies against HSP60, TLR4butnot TLR2was co-precipitated. In control, untreated cells, only a trace amount of TLR4wasco-precipitated with HSP60. In contrast, in cells treated with exHSP601mg/mL for6h, a significant amount of TLR4was co-precipitated. Secondly, we investigated the adaptorproteins, MyD88and Trif, which TLR4signals through. We precipitated cell lysates withanti-TLR4antibodies, and observed that MyD88, but not Trif was co-precipitated, with agreater amount in exHSP60-treated cells. Thirdly, we examined the activities of p38, JNK andNF-κB, the effector molecules downstream of the TLR4-MyD88pathway. These studiesshowed that p38and JNK were activated by exHSP60, as shown by the increasedphosphorylation of p38and JNK in immunoblotting. NF-κB was also activated by exHSP60,as shown by p65translocation from the cytoplasm to the nucleus. Fourthly, we examined theeffects of inhibiting upstream signaling molecules on exHSP60-induced p38, JNK and NF-κBactivation, as well as cytokine production. It was showed that TLR4siRNA (100nM),anti-TLR4antibody (5μg/mL) and MyD88homodimerization inhibitory peptide (InhiMyD88,100μM) prevented the activation of p38, JNK and NF-κB, and significantlyinhibited TNFα and IL-6mRNA expression in exHSP60-treated myocytes. The p38inhibitorSB203580(20μM) and NF-kB inhibitor PDTC (100μM) also inhibited TNFα and IL-6expression. However, the negative control siRNA (NC siRNA,100nM), the control antibody(IgG,5μg/mL), anti-TLR2antibody (5μg/mL), Cntl MyD88(100μM), and the JNK inhibitorSP600125(20μM) did not affect the pro-inflammatory effects of exHSP60.The above results showed that exHSP60stimulated cytokine production viaTLR4-MyD88-p38/NF-kB pathway, but not TLR2and Trif pathway.2. EnHSP60in the extracellular space contributes to cytokine production induced byischemia in cultured cardiomyocytes via the TLR4-MyD88-p38/NF-κB pathwayHSP60was barely detectable by ELISA in the media of normal H9c2myocytes.However,3.7±0.7ng/mL of HSP60was detected in the media after12h of ischemia,indicating that ischemia caused the translocation of HSP60into extracellular space. Toexamine whether the extracellular enHSP60contributed to cytokine production induced byischemia, anti-HSP60antibodies (10μg/mL) were used to neutralize enHSP60. Thissignificantly inhibited cytokine induction by ischemia.The association between enHSP60and TLR2/4, and between TLR4and MyD88/Trif was investigated further with co-IP experiments. In accordance with the results for exHSP60, alarge amount of TLR4co-precipitated with enHSP60from ischemic myocytes. However, nosignificant binding was seen between TLR2and enHSP60. Additionally, TLR4co-precipitated with MyD88, but not Trif. The amount of MyD88bound with TLR4wasmuch greater in ischemic myocytes than control cells. In accordance with the ability ofexHSP60to activate p38and JNK, enHSP60contributed to the activation of p38and JNK inischemic myocytes. The activity of p38and JNK was increased by13.2and3.7foldrespectively under ischemia, which was significantly reduced to5.3and1.6fold of control byanti-HSP60antibodies (10μg/mL). Anti-TLR4antibodies (5μg/mL) and Inhi MyD88(100μM) also inhibited p38and JNK activation.Signal inhibitors were also used to investigate the role of key proteins in cytokineproduction. Anti-HSP60(10μg/mL), anti-TLR4(5μg/mL), TLR4siRNA (100nM), InhiMyD88(100μM), SB203580(20μM), and PDTC (100μM) all significantly inhibited boththe increase in TNFα and IL-6production and release, while SP600125(20μM) had noeffect.The above results showed the presence of extracellular HSP60under ischemia, and theinduction of cytokine production by ischemia-induced in H9c2cells, which was dependent onenHSP60-TLR4-MyD88-p38/NF-κB.3. EnHSP60in circulation contributed to cytokine production induced by myocardialischemia in intact ratsIn anaesthetized rats, HSP60was detectable by ELISA in the serum after4h of LADligation (8.3±1.5ng/mL), but undetectable after sham surgery. Co-IP experiments detectedsignificant interaction between HSP60and TLR4, but not HSP60and TLR2, in ischemicmyocardium. Significant binding between TLR4and MyD88was detected in both sham andischemic myocardium, with more binding in the latter. However, no binding was observedbetween TLR4and Trif in either sham or ischemic myocardium.In ischemic myocardium, the activity of p38was increased by53.3fold, and the mRNAlevel of TNFα and IL-6was increased by3.7and3.3fold, respectively. The serum content of TNFα and IL-6was increased from16.9to81.8pg/mL, and from34.0to164.7pg/mL,respectively. By using anti-HSP60antibodies (20μg/kg, intravenously injected15min beforeLAD ligation) to neutralize serum HSP60that appeared after LAD ligation, we observedreduced activation of p38and decreased generation of TNFα and IL-6, confirming thatenHSP60(extracellular) in circulation contributed to p38activation and cytokine productionin the setting of myocardial ischemia.ConclusionThe present study provides direct evidence that extracellular HSP60activates TLR4andMyD88, but not TLR2or Trif, in cardiomyocytes. During myocardial ischemia, HSP60released from cardiomyocytes stimulates inflammatory cytokine production viaTLR4-MyD88-p38/NF-κB pathway. We provide novel insights into the underlyingmechanisms by which myocardial ischemia causes inflammation in the absence of recognizedpathogens, which may facilitate the intervention of cardiac inflammation caused by ischemia.