OUP user menu

A potential cardioprotective role of hepatocyte growth factor in myocardial infarction in rats

Hideki Ueda, Teruya Nakamura, Kunio Matsumoto, Yoshiki Sawa, Hikaru Matsuda, Toshikazu Nakamura
DOI: http://dx.doi.org/10.1016/S0008-6363(01)00272-3 41-50 First published online: 1 July 2001


Objective: Cardiotrophic growth factors with anti-cell death actions on cardiac myocytes have gained attention for treatment of patients with myocardial infarction. Hepatocyte growth factor (HGF) plays a role in tissue repair and protection from injuries, however, the physiological role of HGF in the myocardium has not been well defined. We asked if HGF would afford to the infarcted myocardium. Methods and results: Mature cardiac myocytes prepared from adult rats expressed barely detectable levels of the c-Met/HGF receptor, however, c-Met receptor expression increased during cultivation, which meant that cardiac myocytes are potential targets of HGF. Addition of hydrogen peroxide remarkably decreased the number of viable mature cardiac myocytes in primary culture, whereas treatment with HGF enhanced survival of the cells subjected to the oxidant stress. Although very low levels of c-Met/HGF receptor and HGF mRNA expression were seen in normal rat hearts, both c-Met/HGF receptor and HGF mRNA levels rapidly increased to much higher levels than normal, when the rats were subjected to myocardial infarction. Immunohistochemical analysis of the c-Met receptor indicated that this receptor was expressed in cardiomyocytes localized in the border regions of the viable myocardium and in non-infarcted regions following myocardial infarction. Conclusion: The c-Met/HGF receptor is induced in cardiomyocytes following myocardial infarction and HGF exhibits protective effect on cardiomyocytes against oxidative stress. Our working hypothesis is that HGF may afford myocardial protection from myocardial infarction.

  • Coronary disease
  • Infarction
  • Growth factors
  • Myocytes
  • Signal transduction

Time for primary review 25 days.

This article is referred to in the Editorial by I.M.C. Dixon (pages 4–6) in this issue.

1. Introduction

Acute myocardial infarction caused by coronary artery occlusion may lead to unfavorable remodeling of the myocardium including left ventricular dilatation and aneurysmal formation, and often severe heart failure [1]. Current therapeutic techniques for treatment of acute myocardial infarction mainly focus on restoration of blood flow through the obstructed coronary artery by means of thrombolytic treatment, mechanical vascularization and bypass surgery. These techniques can decrease mortality rates for patients with acute myocardial infarction, however, these modalities do not always prevent damage to the ischemic myocardium.

Approaches to analyze pathophysiological events following acute myocardial infarction revealed that hypertrophic responses occur in cardiac myocytes in the surviving portion of the ventricle, followed by ventricular dilatation due to an architectural rearrangement of myocytes [2]. The underlying mechanism responsible for cardiac dilatation has been linked to myocyte cell death in infarcted regions [3]. Several studies focused on involvement and application of cardiotrophic growth factors when analyzing pathogenesis and therapeutics following acute myocardial infarction. Particularly, insulin-like growth factor-I (IGF-I), IGF-II, and basic fibroblast growth factor (bFGF) were shown to stimulate hypertrophy of myocytes, to protect against death of myocytes and to attenuate ventricular dilation after acute myocardial infarction [4–6].

Hepatocyte growth factor (HGF), originally identified and cloned as a potent mitogen for mature hepatocytes [7–9], has mitogenic, motogenic, morphogenic, and anti-apoptotic activities in a wide variety of cells, preferentially in most epithelial and endothelial cells [10–12]. One particular importance is the finding that HGF enhances regeneration of organs such as the liver, kidney and lung [11]. Likewise, HGF has angiogenic activity [13–17]. The potent activity of HGF to prevent cell death in distinct types of cells has been documented [18–20].

Because of the preferential, but not exclusive, target cell specificity of HGF on epithelial and endothelial cells and the relatively low expression of both HGF and c-Met/HGF receptor in the normal heart, less attention has been directed to pathophysiological roles of HGF in cardiac tissue, except for a restricted number of recent studies [21–25]. In the present study, we found that c-Met is expressed in cardiac myocytes, HGF exhibits cytoprotective actions on mature cardiac myocytes, and both c-Met receptor and HGF mRNA expression was strongly induced following myocardial infarction. The cardiotrophic action of HGF, as mediated by the c-Met receptor deserves our ongoing attention.

2. Methods

2.1. Materials

Human and rat recombinant HGF was purified from the culture medium of Chinese hamster ovary cells transfected with expression plasmid containing human or rat HGF cDNA [7–9]. Purity of HGF exceeded 98% as determined by SDS–PAGE and subsequent protein staining. Polyclonal antibody against rat HGF was prepared from the serum of a rabbit immunized with rat recombinant HGF and IgG was purified using protein A-Sepharose (Pharmacia Biotech, Uppsala). Anti-rat HGF IgG (1 μg/ml) completely neutralized the biological activities of 1 ng/ml rat HGF. Collagenase-D and dispase were respectively obtained from Molecular Biochemicals (Mannheim) and Godo Shusei (Tokyo). Human recombinant IGF-I and Wortmannin were respectively obtained from Life Technologies (Rockville, MD) and Sigma (St Louis, MO).

2.2. Animal treatment and histopathological analysis

Eight-week-old male Wistar rats were given humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Laboratory Animals prepared by the Institute of Laboratory Animal Resource and published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996).

Myocardial infarction was made as described elsewhere [2]. Briefly, Male Wistar rats were anesthetized with sodium pentobarbital (50 mg/kg body weight), and positive-pressure respiration was applied through an endotracheal tube. The left thorax was opened via fourth intercostal space, and the left coronary artery was ligated 3 mm distal to its origin by 7-0 polypropylene ligature. For histopathological analysis, tissues were fixed in 10% (w/v) formalin, embedded in paraffin and stained with Hematoxylin and Eosin.

2.3. Isolation and cultivation of cells

The heart was placed on a stainless cannula of Langendorff perfusion system for retrograde coronary perfusion and perfused through the aorta with modified Krebs–Henseleit solution gassed with 95% O2/5% CO2 at 100 cm H2O perfusion pressure at 37°C for 10 min. The heart was subsequently perfused with modified Krebs–Henseleit solution gassed with 95% O2/5% CO2 containing 0.12% collagenase-D, 0.01% trypsin, 125 PU/ml dispase and 0.11% bovine serum albumin (BSA), and 0.125 mM CaCl2 for 20–30 min. Ventricles without valves were removed from the cannula, minced, dissociated by pipetting in HEPES Earl's salt (HES). Intact cardiac myocytes were settled by centrifugation at 35×g for 3 min. After removal of the supernatant containing cardiac non-myocytes, precipitated cells were resuspended in HES and the centrifugation was repeated three times. The cells were then resuspended in HES containing 4% BSA and centrifuged at 25×g for 30 sec. The resultant preparation of cardiac myocytes contained 70–80% rod-shaped myocytes and the contamination of non-myocytes was consistently less than 1%. Cardiac myocytes were suspended in 199 medium containing 4% fetal bovine serum (FBS) and seeded on a dish coated with 10 μg/ml laminin. After a 1 h culture, the medium was changed to serum-free medium and the cells were cultured for 24 h. Cardiac endothelial cells were prepared from the supernatant of the first centrifugation of cardiac cells at 35×g (see above). The supernatant was centrifuged at 120×g for 10 min and the precipitated cells enriched with endothelial cells were resuspended in Dulbecco's modified eagle (DME) medium containing 20% FBS and 50 μg/ml endothelial cell growth supplement. The cells were plated on a culture dish coated with 0.03% type I collagen. Cardiac endothelial cells at the fifth passage were used. Contamination of non-endothelial cells in the endothelial culture was less than 3% as determined by immunohistochemistry for von Willebrand factor.

Neonatal rat cardiomyocytes were cultured using methods originally described by elsewhere [26], but with minor modification. After the cells had been cultured for 2 days in DME medium containing 10% FBS, the cells were cultured in serum-free DME medium for 24 h and used for experiments. Purity of the cultured cardiomyocytes exceeds 90%, as determined by immunocytochemical detection of α-sarcomeric actin.

2.4. Cell survival assay

Cardiac myocytes were cultured for 24 h in serum-free condition, then cultured for 12 h in the absence or presence of HGF. H2O2 (200 μM) was added to cultures of cardiac myocytes and the cells were further cultured for 2 h. To assess viability of the myocytes, rod-shaped and striated cells were defined as viable, while round or irregular shaped cells with loss of striations were considered nonviable [27]. In each sample, at least five fields were observed and 100 cells were randomly counted and evaluated at each field. In preliminary experiments, the viability of cardiac myocytes determined by the colorimetric assay with 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide and Trypan Blue dye exclusion assay indicated a good correlation with that determined based on their rod shape, as described above (data not shown).

2.5. Western blot

After serum starvation for 24 h, cardiomyocytes were treated with 10 ng/ml HGF or IGF-I for 10 min and lysed with SDS sample buffer. Cell lysate was subjected to SDS–PAGE on a 12% polyacrylamide gel at 40 μg protein/lane and proteins were electroblotted on polyvinylidene difluoride (PVDF) membrane (Bio-Rad Laboratories, Hercules, CA). After blocking, the membrane was sequentially incubated with anti-phospho-p44/p42 MAP kinase (ERK 1/2) monoclonal antibody (E10; New England BioLabs, Beverly, MA) for 1 h, biotinylated anti-mouse IgG (Vector, Burlingame, CA) for 1 h, horseradish peroxidase-conjugated streptavidin (Amersham, Buckinghamshire) for 1 h and an enhanced chemiluminescence reagent (Amersham). To detect ERK protein, anti-ERK 1 antibody (K-23; Santa Cruz Biotechnology, Santa Cruz, CA) was used. Likewise, to detect total Akt protein and phosphorylated Akt, anti-Akt (# 9272; New England Biolabs) and anti-phosphorylated Akt (4E2; New England Biolabs) were respectively used for the immunoblotting.

2.6. RT–PCR

Total RNA was extracted from cardiac tissues and cells by the guanidinium thiocyanate method [28]. One microgram of total RNA was reverse transcribed into first strand cDNA with a random hexaprimer, using Superscript II reverse transcriptase. One microlitre of first strand cDNA solution was amplified with primers set as follows: 5′-TGT GCG TTC CCC ATC AAA TAT GTC-3′ (forward), 5′-GTC CCA GGC ACA TAT GGT CA-3′ (reverse) for mouse c-Met receptor sequence (Gene Bank accession No. Y00671); 5′-GGT AAG CTT CGG CCT CGT CTC ATA GAC AA-3′ (forward), 5′-TTG AGA TCT GGA GAG AGA ACG AGA GTC AT-3′ (reverse) for rat GAPDH (GeneBank accession No. X02231), using GeneAmp PCR System 9600 (Perkin Elmer) with AmpliTaq DNA polymerase (Perkin Elmer). PCR conditions were as follows: 30 cycles, denaturation at 94°C for 30 s, followed by annealing at 55°C for 60 s and extension at 72°C for 90 s. PCR products were subjected to 2% agarose gel electrophoresis and visualized by ethidium bromide staining. To normalize signals for c-Met, the density value was divided by that for GAPDH. The data were presented as arbitrary unit after densitometric analysis with Fluorchem™(Alpha Innotech Corporation, CA).

Real-time quantitative PCR was performed using TaqMan™ fluorogenic probes for 5′ exonuclease assay as described [29,30]. Amplification reactions contained a first strand cDNA sample, Mastermix®, the following primer sets: 5′-AAG AGT GGC ATC AAG TGC CAG-3′ (forward), 5′-CTG GAT TGC TTG TGA AAC ACC-3′ (reverse) for rat HGF (GeneBank accession No. D90102); 5′-GTA CGG TGT CTC CAG CAT TTT T-3′ (forward), 5′-AGA GCA CCA CCT GCA TGA AG-3′ (reverse) for mouse c-Met receptor; 5′-CCA TCA CTG CCA CTC AGA AGA C-3′ (forward), 5′-TCA TAC TTG GCA GGT TTC TCC A -3′ (reverse) for rat GAPDH, and the following fluorogenic probe: 5′(FAM)-TGA TCC CCC ATG AAC ACA GCT TTT TG -(TAMRA)3′ for rat HGF, 5′(FAM)-ACC ACG AGC ACT GTT TCA ATA GGA CCC-(TAMRA)3′ for rat c-Met receptor, 5′(FAM)-CGT GTT CCT ACC CCC AAT GTA TCC GT-(TAMRA)3′ for rat GAPDH. Reactions were run in the Model 7700 Sequence Detector (Perkin Elmer). PCR conditions were as follows: 40 cycles, denaturation at 95°C for 15 s, followed by annealing and extension at 62°C for 60 s. Samples were matched to a standard curve generated by amplifying the serially diluted product in the same PCR. Standard curves were obtained in both logarithmic regression in each product from 102 to 106 copy number (r>0.995). To correct for variability in RNA recovery and efficiency of reverse transcription, GAPDH cDNA was amplified and quantitated from each cDNA preparation. All samples were done in duplicate.

2.7. Immunocytochemical and immunohistochemical staining

Isolated cells were plated on Nunclon® 4-well Lab-Tek chamber slides (Nunc Inc., Naperville, IL) and cultured for 24 h. The cells were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 10 min, washed twice with PBS, treated with PBS containing 0.1% saponin and then with PBS containing 3% normal goat serum and 2% BSA at room temperature. The cells were successively incubated with the rabbit anti-mouse c-Met/HGF receptor IgG (1 μg/m) for 1 h, 3 μg/ml biotinylated anti-rabbit IgG for 1 h and avidin–biotin complex conjugated alkaline phosphatase for 30 min at room temperature. Enzyme reactions were obtained using substrate solution composed of New Fuchsin with 1 M levamisole.

For immunohistochemical staining, the cardiac tissue was fixed in 10% formalin and embedded in paraffin. Tissue sections were deparaffinized and autoclaved at 120°C for 20 min. After washing twice with PBS, tissue sections were incubated with 0.3% H2O2 for 5 min at room temperature followed by PBS containing 3% normal goat serum and 2% BSA for 30 min at room temperature. The sections were subsequently incubated with rabbit anti-mouse Met IgG (1 μg/m) for 1 h at room temperature, biotinylated anti-rabbit IgG antibody for 1 h and avidin–biotin complex conjugated horseradish peroxidase for 30 min at room temperature.

2.8. Statistical analysis

All values are expressed as mean±S.E.M. The difference in the data was determined with unpaired Student's t-test. A P-value of less than 0.05 was considered statistically significant.

3. Results

3.1. c-Met expression in cultured cardiac myocytes

We first analyzed c-Met/HGF receptor expression in mature cardiac myocytes in primary culture. These myocytes, highly purified from the normal rat hearts, were cultured for 24 h (Fig. 1A) and subjected to c-Met receptor analysis using RT–PCR. When cardiac myocytes and non-myocytes were freshly isolated from the normal hearts, the expression of c-Met receptor mRNA in these types of cells was weak. However, the expression of c-Met receptor mRNA in cardiac myocytes was significantly increased 24 h after the culture (Fig. 1B). The size of RT–PCR product coincided with a predicted size of 675 bp. Cultured cardiac endothelial cells (5th-passaged) also significantly increased the expression of c-Met receptor mRNA. To confirm the expression of c-Met receptor in cultured cardiac myocytes, the cells were used for immunocytochemical detection of c-Met receptor. Although normal rabbit IgG gave no specific staining in these cells, specific staining was seen with the anti-c-Met IgG in cardiac myocytes cultured for 24 h (Fig. 1C, D). Thus, c-Met receptor expression in cardiac myocytes is up-regulated during cultivation of these cells.

Fig. 1

Induction of c-Met receptor expression in mature cardiac myocytes in primary culture. (A) Appearance of mature cardiac myocytes cultured for 24 h. (B) Expression of c-Met mRNA detected by RT–PCR in cardiac myocytes derived from normal adult rat hearts. RNA was prepared from freshly isolated cardiomyocytes (lane 1) and non-myocytes (lane 2), and cardiomyocytes cultured for 24 h (lane 3) and coronary endothelial cells (lane 4). The experiments were independently performed three times and similar results were obtained in each experiment. Density of the each band was divided by that of GAPDH and standardized. ##P<0.01 vs. lane 1, ###P<0.005 vs. lane 2. (C and D) Immunocytochemical detection of c-Met receptor expression in cardiac myocytes, using anti-mouse c-Met receptor antibody (C) or normal rabbit IgG (D). Scale bar=100 μm.

3.2. Cytoprotection and ERK 1/2 activation by HGF in cardiac myocytes

The above results suggested that stress caused by isolation and cultivation of the cells might be involved in induction of c-Met receptor and that c-Met receptor induced in cardiac myocytes might have biological functions. To investigate possible role of HGF in myocardium, we analyzed biological effects of HGF on mature cardiac myocytes in primary culture. Myocytes isolated from the normal rat heart were cultured for 24 h, then cultured for 12 h in the absence or presence of HGF. H2O2 (200 μM) was added to cultures of cardiac myocytes followed by a further 2 h of cultivation. Treatment with H2O2 markedly decreased the number of viable rod-shaped cardiac myocytes in control cultures, whereas the number of rod-shaped cells remained after H2O2-treatment was higher in HGF-treated cardiac myocytes than in control culture (Fig. 2A). The viability of cardiac myocytes decreased from 80±1 to 18±7% as a result of the oxidant stress caused by H2O2-treatment, however, HGF dose-dependently increased the viability of the cells (n=3 in each group) (Fig. 2B). The cytoprotective effect of HGF was seen with 0.3 ng/ml HGF and a two-fold increase in the viability was seen with 3 ng/ml HGF. Taken together, c-Met receptor expression in cardiomyocytes can be induced and HGF has cytoprotective effects on mature cardiac myocytes.

Fig. 2

Enhancement of viability of cardiac myocytes by HGF against oxidant stress induced by H2O2-treatment. (A) Appearance of cardiomyocytes non-treated (control) and treated with H2O2 or H2O2+3 ng/ml human HGF. (B) Viability of cardiomyocytes non-treated treated with H2O2 in the absence or presence of human HGF (*P<0.05 vs. none, **P<0.01 vs. none). Rod-shaped and striated cells were defined as viable, while round or irregular-shaped cells with loss of striations were considered nonviable [27]. Each value represents the mean±S.E.M. of triplicate measurements. The experiments were independently performed three times and similar results were obtained in each experiment.

Since activation of ERK 1/2 and Akt is associated with cytoprotective action of cardiotrophic factors such as IGF-I in cardiac myocytes [31,32], we asked if ERK 1/2 and Akt would be activated by HGF in cardiac myocytes. To prepare cell lysates quantitatively available for analysis in Western immunoblotting, cardiac myocytes obtained from neonatal rats were expanded during cultivation and used for these experiments. HGF had similar cytoprotective effect on H2O2-treated neonatal cardiac myocytes in culture (data not shown). Following serum starvation for 24 h, cells were stimulated with 10 ng/ml HGF or IGF-I for 10 min and cell lysates were respectively subjected to Western blotting. The basal level of weakly phosphorylated ERK 1/2 was evident in control serum-starved cells, whereas treatment of cardiac myocytes with HGF as well as IGF-I stimulated the phosphorylation of ERK 1/2 (Fig. 3A). A specific antibody against HGF inhibited the ERK 1/2 phosphorylation induced by HGF but not IGF-I. The same amount of cell lysates were simultaneously subjected to Western blot by anti-ERK1 antibody to indicate that the level of total ERK protein were comparable. On the other hand, treatment with HGF, as well as HGF plus Wortmannin, a phosphatidylinositol (PI)-3-kinase inhibitor, had no significant effect on the phosphorylation of Akt in cardiac myocytes (Fig. 3B).

Fig. 3

Effect of HGF and IGF-I on phosphorylation of ERK 1/2 (A) and Akt (B) in cardiac myocytes. (A) ERK 1/2 phosphorylation. Cardiac myocytes derived from neonatal rats were serum-starved for 24 h and 10 ng/ml rat HGF or 10 ng/ml IGF-I was added, in the absence or presence of 10 μg/ml anti-rat HGF IgG. Western immunoblotting was done using anti-phospho-p44/p42 MAP kinase monoclonal antibody (upper panel) and anti-ERK1 antibody (lower panel). (B) Akt phosphorylation. Cardiomyocytes were treated as above, except that 10 ng/ml rat HGF or 10 ng/ml IGF-I was added, in the absence or presence of 200 nM Wortmannin. Western immunoblotting was done using anti-phosphorylated Akt antibody (upper panel) and anti-Akt antibody (lower panel). The similar results were obtained in three independent experiments.

3.3. Induction of HGF and c-Met in the infarcted myocardium

To determine the potential involvement of HGF in cardiac injury, we analyzed changes in HGF and c-Met receptor mRNA expression in the heart following myocardial infarction induced in rats. RNA was respectively prepared from the left ventricular free wall, including the infarcted area and interventricular septum, and analyzed using the real-time quantitative RT–PCR. Although expression of c-Met receptor mRNA in the intact heart was low, this expression increased after myocardial infarction (n=3 at each time-point) (Fig. 4A). c-Met receptor mRNA levels in the left ventricular free wall increased from 6 h after the treatment and continued to increase for up to 7 days. The maximal increase was 40-fold at 7 days after the treatment and c-Met receptor mRNA levels decreased thereafter. The c-Met receptor mRNA levels in the interventricular septum also increased from 6 h after the treatment and a maximally increased expression continued after 24 h until 2 weeks after treatment. On the other hand, HGF mRNA levels also increased in a distinct profile to that of c-Met receptor mRNA levels (Fig. 4B). HGF mRNA levels in the left ventricle free wall increased from 6 h after the treatment and increased to 10-fold higher levels at 72 h. The HGF mRNA levels in the left ventricular free wall thereafter decreased, yet was higher than that in normal heart 2 weeks after the treatment. HGF mRNA levels in the interventricular septum changed relatively in a similar manner to that seen in the left ventricular free wall: increase was from 12 h after myocardial infarction and reached a peak 72 h after the treatment. These results indicate that both HGF and c-Met/HGF receptor mRNA levels increased in the infarcted rat heart.

Fig. 4

Changes in c-Met receptor and HGF mRNA levels in cardiac tissues and cardiomyocytes following myocardial infarction. (A and B) Changes in expression of c-Met receptor (A) and HGF (B) mRNA levels in left ventricular free wall (LVFW) and interventricular septum (IVS) after the left coronary artery ligation. Changes in c-Met and HGF mRNA levels were determined by means of real-time quantitative RT–PCR. In all groups, three rats were used to quantify mRNA level in each time-point. Each value represents the mean±S.E.M. of triplicate experiments. HGF and c-Met cDNA copy numbers were divided by GAPDH cDNA copy numbers. *P<0.05, **P<0.01, ***P<0.005 vs. Sham/LVFW. (C) Increased expression of c-Met mRNA in cardiac myocytes and non-myocytes. Cardiac myocytes and non-myocytes were isolated from the left ventricle of normal rats or rats subjected to left coronary artery ligation for 24 h. c-Met receptor mRNA levels were analyzed by RT–PCR and subsequent staining of amplified DNA with ethidium bromide. Lane 1, cardiomyocytes from normal hearts; lane 2, non-myocytes from normal hearts; lane 3, cardiomyocytes from the infarcted hearts; lane 4, coronary endothelial cells from the infarcted hearts.

To determine if c-Met receptor mRNA would increase in cardiac myocytes following myocardial infarction, we analyzed expression of c-Met receptor mRNA in highly purified cardiac myocytes and cardiac cells except for myocytes (cardiac non-myocytes), using RT–PCR. Cardiac myocytes and cardiac non-myocytes were isolated from normal and infarcted hearts at 24 h after myocardial infarction. Although expression of the c-Met receptor was barely detectable in cardiac myocytes isolated from the normal heart, the expression of c-Met receptor mRNA was obvious in cardiac myocytes isolated from the heart 24 h after cardiac infarction (Fig. 4C). Likewise, c-Met receptor expression in cardiac non-myocytes also increased after the cardiac infarction. Therefore, cardiac myocytes become target cells of HGF, in response to myocardial infarction.

3.4. Localization of c-Met receptor in myocardium

To confirm the induction of c-Met receptor expression in cardiomyocytes, localization of c-Met receptor protein in the rat heart was analyzed immunohistochemically (Fig. 5). Consistent with the low level expression of c-Met receptor mRNA in normal heart, c-Met receptor expression in the normal heart was faintly detectable, using immunohistochemical analysis (not shown). On the other hand, c-Met receptor-positive cells were evident in the heart 24 h after the induced myocardial infarction. Both cardiac myocytes and vascular endothelial cells in border regions and also in non-infarcted distant regions were positive for c-Met receptor expression 24 h after myocardial infarction (Fig. 5A, D), whereas most of these cells in the infarcted region were negative for immunohistochemical detection of the c-Met receptor. c-Met receptor expression in viable region, as well as the lack of c-Met receptor expression in infarcted region, were more clearly seen in immunohistochemical analysis in border regions. Fig. 5A and Fig. 5C, respectively, show c-Met receptor expression and Hematoxylin and Eosin-stained tissue in border regions between the infarcted and non-infarcted regions, in serial sections. Hematoxylin and eosin staining showed cells in the upper half were characterized by disappearance of the striated pattern in myocytes, nuclear aggregation and acidic staining of cytoplasm, indicating that these cells were not viable after myocardial infarction (Fig. 5C). Cells located in the lower half were c-Met-positive, whereas those in the upper half were c-Met negative (Fig. 5A). Cells in infarcted regions might not have been viable, hence incapable of protein synthesis. Immunohistochemical staining, using normal IgG, showed no staining (Fig. 5B, E). These results suggested that mature cardiac myocytes as well as cardiac vascular endothelial cells might be target cells of HGF, when myocardial was ischemically injured. A previous report [22] showed that cardiac endothelial cells expressed the c-Met receptor while cardiac myocytes did not express the receptor both in normal hearts and hearts subjected to myocardial infarction. As Ono et al. used an anti-human c-Met antibody, such may explain the discrepancy.

Fig. 5

Expression of c-Met receptor in cardiac tissues subjected to myocardial infarction. Cardiac tissues were obtained at 24 h after left coronary artery ligation. Immunohistochemical staining was done in cardiac tissues of border region between non-infarcted and infarcted regions (A, B and C) and of non-infarcted intact regions (D, E and F), respectively using anti-c-Met IgG (A and D) and normal rabbit IgG (B and E). Serial sections used for immunohistochemistry were also stained by Hematoxylin and Eosin (HE) (C and F). Scale bar=100 μm.

4. Discussion

Despite a number of studies on ‘trophic’ actions of HGF for regeneration and/or maintenance of organs, including the liver, kidney, lung, stomach, and cerebrum, less attention has been directed to the physiological role of HGF in the heart. We reported that HGF gene transfection into the myocardium attenuated ischemia-reperfusion injury, although related mechanisms remained to be determined [23]. In the present study, we found that although expression of both HGF and c-Met receptor were low in the normal myocardium, mature cardiac myocytes do express the c-Met receptor in response to myocardial infarction and during cultivation, and that HGF had direct cytoprotective effects on mature cardiac myocytes. Although a previous report showed up-regulation of c-Met receptor expression in cardiac endothelial cells following cardiac ischemia-reperfusion injury [22], this is the first report to show the induction of the c-Met receptor in mature cardiac myocytes following myocardial infarction and cytoprotective actions of HGF in the cells.

Other workers reported that protective effects of cardiotrophic factors such as IGF-I and leukemia inhibitory factor involve ERK activation [31,33]. Inhibition of ERK phosphorylation by a specific inhibitor resulted in increased cell death in cardiomyocytes and ischemia-reperfusion injury in isolated rat hearts [34,35]. In addition to the ERK-related pathway, stimulation of Akt downstream of PI3-kinase is involved in protection of cardiomyocytes by IGF-I [32]. Since HGF shares similar protective effect with IGF-I, we analyzed the phosphorylation state of ERK1/2 and Akt by these factors. Consistent with previous reports, IGF-1 stimulated both ERK1/2 and Akt phosphorylation in our experiments. However, HGF did induce ERK1/2 phosphorylation but did not induced Akt phosphorylation. Thus, the ERK pathway but not PI3-kinase-Akt pathway seems to be involved in HGF-dependent cytoprotection in cardiac myocytes. Therefore, the intracellular signaling pathways leading to cardiomyocyte protection by HGF and IGF-I differ, at least in part, although both signals triggered by HGF and IGF-I are mediated receptor tyrosine kinases. In this context, it was noted that Bag-1, a partner of Bcl-2/Bcl-xL [36], is specifically associated with the c-Met/HGF receptor and expression of Bag-1 promotes the cell survival activity of HGF [18]. Possible involvement of Bag-1 and the Bcl-2/Bcl-xL-related pathway in HGF-dependent myocardial protection remains to be defined.

Previous studies focused on expression of c-Met receptor in cardiac endothelial cells after acute myocardial infarction and biological effects of HGF on these cells. Coronary endothelial cells express c-Met receptor and c-Met mRNA expression in the heart increases following cardiac ischemia-reperfusion injury [22], and HGF exhibits mitogenic effect on rat coronary endothelial cells [14]. A recent study demonstrated potential therapeutic angiogenesis induced by transfection and expression of HGF gene delivered to the infarcted myocardium [17]. These results revealed that HGF has a physiological role as well as therapeutic value in cases of myocardial infarction, based on angiogenic and angioprotective actions of HGF. Together with our present finding, biological actions of HGF for both cardiac myocytes and coronary endothelial cells means that HGF may afford myocardial protection.

Myocardial infarction often causes ventricular dilatation characterized by a diminished cardiac performance, poor recovery of function, and increased mortality. The underlying mechanism responsible for cardiac dilatation has been linked to myocyte cell death in the surviving regions [3]. Thus strategies for the treatment of myocardial infarction have focused on protection against myocyte death and the restoration of blood flow. Polypeptide growth factors with cardiotrophic and/or angiogenic activities have been tested for their therapeutic effects in case of myocardial infarction. Administration of IGF-I or IGF-II to laboratory animals improved myocardial function after induced myocardial infarction [5,37] and transgenic expression of IGF-I protected from myocyte death and attenuated ventricular dilatation [4]. We propose that HGF may play the role of a physiological cardiotrophic factor, as well as an angiogenic factor, in maintaining of cardiac function. Our latest report showed that HGF reduced infarct area by suppressing cell death of cardiomyocytes in rats with ischemia-reperfusion injury, thereby suggesting possible therapeutic advantages for treating subjects with myocardial infarction [38].


We thank M. Ohara for helpful comments and Dr U. Koshimizu, H. Ohmichi and S. Kitagawa-Sakakida (Osaka University Medical School) for technical support. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Sports, Culture, Science and Technology of Japan. This work was also supported by the Japan Heart Foundation and IBM Japan Research Grant, and Pfizer Pharmaceuticals Grant for Research on Coronary Artery Disease.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
View Abstract