Myocardial protection from ischemia/reperfusion injury by endogenous and exogenous HGF

Abstract

Using a rat model of ischemia/reperfusion injury, we demonstrate here that HGF is cardioprotective due to its antiapoptotic effect on cardiomyocytes. Following transient myocardial ischemia and reperfusion, c-Met/HGF receptor expression rapidly increased in the ischemic myocardium, an event accompanied by a dramatic increase in plasma HGF levels in the infarcted rats. When endogenous HGF was neutralized with a specific antibody, the number of myocyte cell deaths increased markedly, the infarct area expanded, and the mortality increased to 50%, as compared with a control group in which there was no mortality. Plasma from the myocardial infarcted rats had cardioprotective effects on primary cultured cardiomyocytes, but these effects were significantly diminished by neutralizing HGF. In contrast, recombinant HGF administration reduced the size of infarct area and improved cardiac function by suppressing apoptosis in cardiomyocytes. HGF rapidly augmented Bcl-xL expression in injured cardiomyocytes both in vitro and in vivo. As apoptosis of cardiomyocytes is one of the major contributors to the pathogenesis in subjects with ischemia/reperfusion injury, prevention of apoptosis may prove to be a reasonable therapeutic strategy. Supplements of HGF, an endogenous cardioprotective factor, may be found clinically suitable in treating subjects with myocardial infarction.

Figure 1

Changes in pathophysiology and HGF/c-Met expression in rats with ischemia/reperfusion injury. (a–d) Histopathology of the heart. At 6 hours after treatment, infiltration with inflammatory cells was seen in the LVFW (a, b). At 24 hours after ischemia/reperfusion, a more diffuse infiltration of inflammatory cells was evident (c), and the infarct lesion had expanded widely at 48 hours (d) Bars: a, c, and d, 200 μm; b, 50 μm. (e) Change in CPK activity in sera. I/R, ischemia/reperfusion injury. ^AP < 0.001, ^BP < 0.01, ^CP < 0.05. (f) Plasma HGF levels determined by ELISA. ^AP < 0.001, ^BP < 0.01, ^CP < 0.05. (g) Change in c-Met mRNA expression in myocardium, as determined by quantitative RT-PCR. ^AP < 0.001, ^BP < 0.01, ^CP < 0.05 compared with IVS; ^DP < 0.001, ^EP < 0.01 compared with sham LV. (h–m) Immunohistochemical findings for c-Met. (h–k) Double immunohistochemistry for α-sarcomeric actin and c-Met in the heart resected 48 hours after reperfusion. Photographs of the LVFW (h, i) and IVS (j, k) of a section are shown. Red (h, j) and green (i, k) fluorescence, respectively, indicate α-sarcomeric actin and c-Met labeling. (l and m) Immunostaining for c-Met in the sham-operated myocardium (l) and the border region between infarcted and noninfarcted area (m). To detect background staining, anti–c-Met antibody was preabsorbed with antigenic synthetic peptide in a serial section (inset in m). Bars: 100 μm.

Figure 2

c-Met expression and cardioprotection by HGF in vitro. (a) c-Met protein expression determined by Western blot. Cardiomyocytes from neonatal rats were cultured for 3 days (lane 1), and further cultured in serum-free condition for 24 hours without (lane 2) or with (lane 3) hydrogen peroxide (H₂O₂). (b) Survival of cardiomyocytes in culture. The cells were pretreated with HGF for 1 hour, then treated with 50 μM H₂O₂ for 1 hour and further cultured for 6 hours. ^AP < 0.05 vs. 0 ng/ml of HGF. (c) Activation of ERK-1/2 (p44/p42 mitogen-activated protein kinases) in cultured cardiomyocytes by HGF, as determined by immunoblotting with anti–phospho-ERK-1/2 antibody (top). Immunoblotting with anti–ERK-1 antibody indicates that the total amount of ERK-1/2 protein is similar in two lanes (bottom). (d) Quantification of ERK-1/2 activity. Intensity of the bands was analyzed by densitometry. Each value represents the mean ± SEM of quadruplicate experiments.

Figure 3

Adverse effects of neutralization of endogenous HGF on the ischemia/reperfusion injury model. (a) Specificity of the neutralizing antibody to HGF. Plasma from a rat with ischemia/reperfusion injury was immunoprecipitated with normal IgG (lane 1) or anti–rat HGF IgG (lane 2), and immunoreactive proteins were detected by Western blot, using biotinylated anti–rat HGF IgG. (b) Immunohistochemical staining of infarcted hearts with α-sarcomeric actin to depict the infarct area and its quantification. Anti–rat HGF IgG (n = 10) or normal IgG (n = 10) was injected 20 minutes before coronary occlusion, and every 12 hours after reperfusion. Forty-eight hours after operation, rats were killed and histological and biochemical analyses were made. Arrowheads indicate the α-sarcomeric actin–negative infarct area (original magnification, ×40). ^AP < 0.05. (c) Change in TUNEL-positive cardiomyocytes by neutralization of endogenous HGF 48 hours after reperfusion. Bars: 100 μm. ^BP < 0.01. (d) Survival of rats injected with anti-HGF IgG (n = 10) or normal IgG (n = 10) after reperfusion. There was a significant difference in survival between the two groups (P < 0.01).

Figure 4

Survival of cultured cardiomyocytes in the presence of plasma from a normal rat (Normal plasma) or plasma from a rat with ischemia/reperfusion injury (I/R plasma). Plasma was nontreated or treated with anti–rat HGF antibody for 2 hours, and the cells were cultured in the absence or presence of the plasma for 2 hours. Cells were treated with H₂O₂ for 1 hour and further cultured for 6 hours. ^AP < 0.01 vs. normal plasma; ^BP < 0.05 vs. I/R plasma.

Figure 5

Amelioration of ischemia/reperfusion injury by HGF. Recombinant human HGF (n = 8) or saline (n = 8) was injected immediately after and every 12 hours after reperfusion. After 48 hours, rats were killed. (a) α-Sarcomeric actin staining done to depict the infarct area (original magnification, ×40). Arrowheads indicate the α-actin–negative infarct area. (b and c) Changes in infarct area (b) and serum CPK activity (c). ^AP < 0.01, ^BP < 0.05. (d) Change in cardiac functions after ischemia/reperfusion injury. ^AP < 0.01, ^BP < 0.05. LVSP, left ventricular systolic pressure; LVEDP, left ventricular end-diastolic pressure; max dP/dt, maximal rate of left ventricular pressure rise.

Figure 6

Antiapoptotic effect of HGF on the ischemia/reperfusion injury model. Hearts were excised 3 hours and 48 hours after reperfusion. (a) Distribution of TUNEL-positive cardiomyocytes 3 hours after reperfusion (bars: 100 μm) and changes in TUNEL-positive cells by HGF-treatment. ^AP < 0.05. (b) Immunohistochemical staining for Bcl-xL in the rat heart and semiquantitative analysis (bars: 100 μm). Bar graphs represent percent of Bcl-xL–positive myocytes observed at least in ten fields per section. ^AP < 0.05. (c) Change in Bcl-xL expression in myocardial extracts detected by Western blot. Lane 1, sham-operated LV; lane 2, ischemia/reperfused LV with saline treatment; lane 3, ischemia/reperfused LV with HGF-treatment; lane 4, normal rat spleen for positive control. (d) Bcl-xL expression in cardiomyocytes in culture. Cardiomyocytes were cultured in serum-free medium in the presence or absence of HGF for the indicated period. Expression of Bcl-xL was detected by Western blot.