Molecular Mechanisms Of Cardiomyocyte Apoptosis Induced By Ischemia/reperfusion

Acute cardiac infarction caused by complete coronary obstruction proposes one of the biggest threats on human life. To now, promptly and effectively restoring coronary blood flow is believed to be the most successful strategy to reduce the injury. However, such reperfusion could by itself induce cell death and approximately accumulate 50 % of the final infarct, which is termed myocardial reperfusion (IR) injury. This phenomenon may explain the high death rate and incidence of heart failure after acute myocardial infarction, despite optimal myocardial reperfusion was carried out in most cases. Actually, ischemia/ reperfusion damage in the myocardium is a messy affair with complex network of molecules and pathways involved in the competition between attack and self-protection as evidenced by a mass of papers. Hence, identification and characterization of the essential molecule contributing to IR injury or its counterpart would be critical to establish strategies maximizing the benefit of myocardial reperfusion both in basic research and under clinical settings. Here, we choose two molecule, scaffolding protein Homer 1 and RNA binding protein QKI, to explore their possible roles during myocardial ischemia/reperfusion.I.Homer 1 protein Homer proteins are scaffolding proteins expressed in brain, heart and skeletal muscle, kidney, etc. All Homer isoforms bind to proteins containing a proline-rich motif (PPxxFr) through their N-terminal EVH1 domain, which is required for protein–protein interaction. All long Homer isoforms can form homo- or heteromultimers with themselves and other Homer family members, respectively, through the C-terminal CC domain containing leucine zipper motifs. Because of their distinct dimerization properties, long forms of Homer function both as scaffolds of multiprotein complexes and mediators of signaling pathways. Accumulated evidences in nervous system show that activation of ERK is dependent on Homer 1b/c and Homer 1a could interfer the ERK activation. In contrast, short forms of Homer, such as Homer1a, lack the dimerization domain and behave as dominant negatives. Homer isoforms, especially Homer 1, were proved to be abundantly expressed in myocardium, but there function is still not clear. In this study, we ask what the expression patterns of Homer 1 variants are during myocardial ischemia/reperfusion; what does such expression patterns mean to the ERK activation, cell death and the final cardiac function. The main results are as follows:1. Ischemia/reperfusion induced Homer 1a accumulation in vitro and in vivo.It has been reported that Homer 1a and 1b/c transcripts exist in heart, and proteins were detected with respective antibodies. We used commercial antibodies to determine Homer 1 expression in neonatal cardiomyocytes (NCM), H9C2 cells. Homer 1b/c and 1a could be both detected under normal culture conditions, but Homer 1a is only expressed with neural activity. We performed mimicked IR in NCM and H9C2 cells. Under culture medium deprivation, mimicked ischemia or ischemia/reperfusion (IR), Homer 1b/c expression remained unchanged, but Homer 1a could be up-regulated by these stimuli, among which IR induced the tiptop of Homer 1a expression.2. Silencing Homer 1a elevated susceptibility to IR-induced apoptosis of H9C2 cells.To evaluate what it means to the cardiomyocytes that Homer 1a accumulated under IR condition, we designed siRNAs to knockdown Homer 1a in H9C2 cells. Then the cells were subjected to 9-hour ischemia plus 6-hour reperfusion, apoptosis was determined by flow cytometry. Apoptosis rate was 43.6±3.4 % in ncRNA cells, 21.2±3.3 % in siRNA-a (P<0.05 vs ncRNA). These data indicated that Homer 1a up-regulation enhanced IR-induced apoptosis.3. Over-expression of Homer 1a promoted IR-induced apoptosis in primary cardiomyocytesTo confirm whether Homer 1a expression promoted apoptosis, adeno- Homer 1a was constructed to investigate the direct effects of Homer 1a on IR- induced apoptosis in NCM. Apoptosis was determined by flowcytometry assay after the infected cells were subjected to IR stress. IR induced 52.8±2.3 % apoptosis in adeno-Homer 1a determined by flow cytometry, notably higher than that in cells with adeno-RFP 31.7±4.5 (%).4. Silencing Homer 1a enhanced IR-induced ERK1/2 phosphorylation in H9C2 cellsERK1/2 have been identified as kinases that could propose protection to heart, and constitutively activating MEK1/2 (ERK1/2 upstream kinases) could reduce IR-induced infarction. IR-induced ERK1/2 activation was determined after the genes were silenced. Four-hour ischemia plus 20-min reperfusion could substantially activate ERK1/2, which was exaggerated by knockdown of Homer 1a.5. Over-expression of Homer 1a inhibited ERK1/2 activation during IRTo examine whether Homer 1a proposes an inhibition on ERK1/2 activation, phosphorylation of ERK1/2 in NCMs that had been over-expressed with Homer 1a by adenovirus was subjected to 4-hour ischemia and 20-min reperfusion. The IR treatment induced substantial increase of ERK1/2 phosphorylation in cells with control virus, while the induction was greatly suppressed in cells with adeno-Homer 1a.The above observations provide the first evidence that expression of Homer 1a was induced by ischemia/reperfusion, which promoted ischemia/reperfusion in cardiac myocytes, possibly via the inhibition of anti-apoptotic ERK1/2 cascade.II.RNA binding protein QKIQKI belongs to signal transduction and activation of RNA (STAR) family protein. The major QKI isoforms include QKI-5, QKI-6 and QKI-7, which are derived from the qkI primary transcript via extensive alternative splicing of the C-terminal coding exons; each harbors a single hnRNP K-homology (KH) RNA-binding domain. QKI proteins are abundantly expressed in brain and heart; they take effects by regulating metabolism of messenger RNA.The role of QKI in nervous systerm has been better defined, which involves controling mRNA homeostasis during myelinogenesis, and the deficiency of QKI results in misregulation of its RNA targets, which in turn leads to hypomyelination. But the functions of QKI in myocardium are still obscure. Because QKI could work as RNA binding protein by recgnizing target RNA 3'UTR specificially, we analysized a mass of possible targets of QKI, among which was the proapototic FOXO transcriptional factors that have 3 conservative QKI binding sites in the 3'UTR of mRNA. Hence, we observed the roles of QKI in ischemia/ reperfusion-incuced apoptosis and possible mechanism involving pro-apoptotic transcriptional factor FOXO1. The main results are as follows:1. Mimicked ischemia/reperfusion suppressed QKI-5 expressionIn our study a pan-QKI monoclonal antibody was used to determine QKI expression in neonatal cardiomyocytes (NCM), H9C2 cells and adult myocardium, detecting two bands of 40-kDa and 38-kDa with western blotting. Then we investigated the subcellular localization of QKI protein with immunofluorescence, showing that QKI is dominantly present in nucleus and slightly stained in cytoplasm. Since QKI-5 is only present in nucleus, QKI-7 in cytoplasm, and QKI-6 in both, our results suggest the major QKI isoforms are QKI-5 and -6 in cardiomyocytes. Concomitantly, QKI expressions under mimicked ischemia/reperfusion were also investigated by western blotting and immunofluorescence. In NCM and H9C2 cells, ischemia was mimicked by replacement culture medium with Tyrode's buffer and deprivation of oxygen, while reperfusion was simulated by restoration of culture medium and oxygen. Deprivation of culture medium alone could inhibit QKI-5 expression slightly, but have no effects on QKI-6 expression. Complete ischemia (medium and oxygen withdrawal) for 4 hours decreased QKI-5 expression sharply. Restoring nutrients and oxygen not only failed to resume QKI-5 expression, but further suppressed it nearly to zero. The phenomenon was also observed in immunofluorescence assay. In contrast, QKI-6 expression was not influenced by all these treatment. QKI-5 and QKI-6 expressions were all suppressed by ischemia/reperfusion, which might be attributed to the multiple cell constitution.2. Silencing QKI expression elevated the sensitivity to mimicked ischemia/ reperfusion-induced apoptosis in H9C2 cellsTo evaluate what it means to the cardiomyocytes that QKI expression declines under IR condition, we performed RNA interference to silence QKI expression in H9C2 cells. After subjected to 9-hour ischemia plus 6-hour reperfusion, cells with QKI siRNA present an apoptosis rate of 54.9±3.5 % as showed by flow cytometry, significantly higher than that in negative control cells 33.2±2.4%(P<0.01). These results suggest QKI presence is critical for defense of cardiomyocytes.3. Over-expression of QKI-5 and QKI-6 inhibited mimicked ischemia/ reperfusion-induced apoptosis in primary cardiomyocytesAdenoviruses expressing QKI-5 and -6 were constructed to investigate the direct effects of QKI on IR-induced apoptosis. NCMs were infected with the viruses, which harvested adequate infection efficiency and expression efficacy. Apoptosis was determined by flowcytometry and PARP assay after the infected cells were subjected to IR stress. IR induced an apoptosis rate of 14.7±2.2 % in the cells infected with adeno-QKI-6, notably lower than that in cells with adeno-CMV (34.0±2.3 %, P<0.01). PARP assay confirmed the results in flowcytometry. Besides, PARP assay showed adeno-QKI-5 could also inhibit IR-induced apoptosis as indicated by reduced PARP degradation. 4. Silencing QKI exaggerated the induction of FOXO1 expression by ischemia/ reperfusionAlthough QKI knockdown increased apoptotic sensitivity of H9C2 cells, it is still unknown whether FOXO1 was involved in the increase of apoptotic sensitivity. We found silencing QKI did not affect FOXO1 expression under normal conditions. IR could dramatically elevate FOXO1 expression, while the FOXO1 elevation was more prominent in cells with QKI siRNA. Hence, it is reasonable to postulate that there exists a negative regulation of FOXO1 by QKI.5. Over-expression of QKI-5 and -6 decreased FOXO1 expression level and promoted its exclusion from nucleusTo confirm the existence of QKI-mediated negative regulation on FOXO1 expression, we determined the effects of aden-QKI-5 and -6 on FOXO1 expression and subcellular localization. As mentioned above, FOXO1 was expressed at a relative low level under normal conditions, which was not influenced by over-expressing QKI-5 or -6. IR greatly raised expression of FOXO1 in cells infected with control virus; however, IR failed to induce FOXO1 up-regulation in cells infected with QKI-5 or -6. In addition, immunofluorescence was applied to determine the subcellular distribution of FOXO1. FOXO1 was slightly stained both in cytoplasm and nucleus without IR stress. While facing IR, the nucleus FOXO1 expression was greatly elevated in cells with adeno-CMV. In contrast, addition of adeno-QKI5 failed IR in the induction of FOXO1 expression in nucleus.The above observations provide the first evidence that expression of QKI-5 and -6 proposed potent anti-apoptotic effects against ischemia/reperfusion in cardiac myocytes, possibly via the inhibition of pro-apoptotic transcriptional factor. Besides, the current study suggests the decrease of QKI-5 of cardiomyocytes during mimicked IR led to loss of control of pro-apoptotic factors and subsequent apoptosis at least partially.