INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

GROWTH HORMONE HAS GAINED increasing interest in the treatment of intractable heart failure. Clinical trials of growth hormone therapy have been shown to improve hemodynamic function in end-stage heart failure (8, 40). Another growth factor, potentially efficacious in protecting myocardium and available for clinical use, is insulin-like growth factor-I (IGF-I). IGF-I is a local effector of growth hormone that has a structural similarity to proinsulin and a functional similarity to insulin (10, 39). IGF-I provides cytoprotection against various noxious stimuli in a number of cell types, including fibroblasts overexpressing c-Myc (12), in PC12 cells after serum withdrawal (36), neuroblastoma cells under hyperosmotic stress (24), cardiomyocytes treated with cytotoxic chemical doxorubicin (44), and neuronal cells and myocytes subjected to ischemia and reperfusion (20, 43).

Although the underlying mechanisms of cytoprotection exerted by IGF-I have not been clarified because of the complicated nature of IGF-I receptor-linked signal transduction pathways (9), recent studies (26, 36, 45) indicate that death-regulating Bcl-2 family proteins may be effector molecules involved in IGF-I-induced cytoprotection. Bax plays a crucial role in an induction phase of apoptosis, whereas Bcl-2 and Bcl-xL are known to counteract the proapoptotic action of Bax (28, 29, 33, 47). Accumulating evidence suggests that these proteins participate in determining the life and death of cells subjected to ischemia and reperfusion. A positive correlation between overexpression of antiapoptotic Bcl-2 family proteins and their protective effects against ischemia-reperfusion injury has been suggested in the heart. Acute myocardial infarction causes various degrees of expression of Bcl-2 in viable myocytes in border areas of infarcted myocardium, whereas nonviable myocytes up-regulate Bax expression (2, 27). Nevertheless, direct evidence that IGF-I-induced myocardial protection is accompanied by alteration of Bcl-2 family protein expression has not been provided. Recent studies have revealed that mitochondrial localization of Bcl-2 family proteins is pivotal in regulating cell death signaling cascade (38, 50). Cytoprotective action of Bcl-2 and Bcl-xL, but contradictory function of Bax and Bad, led us to investigate the effects of IGF-I on expression of these Bcl-2 family proteins in the mitochondria in relation to myocardial protection against ischemia and reperfusion in the rat heart.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal preparation. Male Sprague-Dawley rats weighing 250-300 g and fed a standard diet were used for experiments. These animals received humane care according to the animal welfare regulations of the Kansai Medical University. Rats of the IGF-I-treated group received intraperitoneal injection of 1 mg of human recombinant IGF-I (generous gift from Fujisawa Pharmaceutical; Osaka, Japan) dissolved in 1 ml PBS. This concentration of IGF-I was chosen on the basis of the rat study of IGF-I on blood glucose levels (Asada T, Sato M, Nakano K, Sakata Y, Horiai H, Mori Y, and Mukousaka M, Report DIR-940051 issued by Fujisawa Pharmaceutical). The study demonstrated that IGF-I injected subcutaneously at doses above 1 mg/kg caused a significant decrease in the blood glucose level and 3.2 mg/kg IGF-I, which is an equivalent dose to that employed in our study, produced a submaximal response without inducing lethal hypoglycemia. Rats of the control group received intraperitoneal injection of 1 ml PBS only, as a vehicle.

Isolation of mitochondria. After washout of blood from the coronary circulation, the hearts were removed from the perfusion apparatus and immersed in ice-cold isolation buffer (buffer A) with the following composition: 150 mM KCl, 5 mM MOPS, 2 mM EGTA and 0.1% bovine serum albumin, pH 7.25 at 4°C. After atria and major blood vessels were discarded, ventricles were finely minced and gently homogenized with a Teflon homogenizer in the isolation buffer. The homogenate was centrifuged at 800 g for 5 min at 4°C. The supernatant was retrieved and centrifuged at 7,000 g for 10 min. The resulting pellet designated as the mitochondrial fraction was resuspended in the buffer to a protein concentration of 30-50 mg/ml and kept on ice. Protein concentration was determined by the Bradford method using BSA as a standard. The concentrations of Ca²⁺ in incubation buffer were calculated according to the formula described by Fabiato and Fabiato (7).

Measurements of cytochrome oxidase activity and oxygen consumption. Cytochrome oxidase activity was measured according to the method described by Cortese et al. (3). Mitochondria were resuspended in the medium containing (in mM) 220 mannitol, 70 sucrose, 2.5 K₂HPO₄, 2.5 MgCl₂, and 0.5 EDTA. Carbonyl cyanide 3-chlorophenylhydrazone, rotenone, and antimycin A were then added to block mitochondrial respiration through complex III. Reaction was started by adding ascorbate/N,N,N',N'-tetramethyl-p-phenylenediamine as an electron donor. Oxygen consumption was measured with a biological oxygen meter (YSI Japan; Tokyo, Japan).

Western blot analysis for Bcl-2 family proteins. The mitochondrial fraction was mixed with 1 ml of radioimmunoprecipitation assay buffer [10 mM sodium phosphate, 150 mM NaCl (pH 7.4), 1 mM EDTA, 1 mM EGTA, 1% (wt/vol) sodium deoxycholate, 1% Triton X-100, 0.1% (wt/vol) SDS, 100 µg/ml phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM sodium orthovanadate]. Equal amounts of the mixture protein (60 µg) were loaded on reducing SDS/10-20% gradient polyacrylamide gels [buffer composition: 4% (wt/vol) SDS, 20% (wt/vol) glycerol, 4% (wt/vol) -mercaptoethanol, 0.2 M Tris · HCl (pH 6.8), and 0.02% (wt/vol) bromophenol blue]. The proteins were transferred on to nylon membranes and blocked for 1 h. The membrane was then incubated with antibodies against Bcl-2 family proteins (Santa Cruz Biotechnology; Santa Cruz, CA) rabbit polyclonal anti-Bcl-2 antibodies (1:1,500 dilution), rabbit polyclonal anti-Bcl-xS/L antibodies (1:1,500 dilution), rabbit polyclonal anti-Bax antibodies (1:2,000 dilution), and rabbit polyclonal anti-Bad antibodies (1:2,000 dilution) for 1 h. These antibodies possess cross reactivity with the rat proteins according to the manufacturer. After membrane washing, horseradish peroxidase-conjugated secondary antibodies were used (1:2,000 dilution, Dako; Kyoto, Japan). Bound antibodies were detected using enhanced chemiluminescence with a kit from Amersham (Arlington Heights, IL).

Immunofluorescent confocal microscopy for Bcl-2 family proteins. The heart on the perfusion apparatus was fixed by perfusion with 10 ml of 10% formaldehyde in 0.1 M phosphate buffer, pH 7.2. The formaldehyde-fixed left ventricle (LV) was embedded in paraffin, cut into transverse sections (4 µm thick), and deparaffinized with a graded series of xylene and ethanol solutions. The sections were placed in a glass jar filled with 10 mM citrate buffer, pH 6.0, and exposed to a 500 W microwave for 5 min twice. The sections were incubated in absolute methanol containing 0.3% hydrogen peroxide for 30 min, and nonspecific binding was blocked with normal goat serum. The sections were then incubated overnight with the primary antibodies as employed in Western blot analysis with 1:20 dilution at 4°C. Afterward, the slides were washed with PBS and incubated with the secondary antibody, fluorescein isothiocyanate-conjugated goat anti-rabbit F(ab')₂ IgG (Zymed Laboratories; San Francisco, CA), at a 1:100 dilution for 1 h at room temperature. The fluorescence staining was visualized with confocal laser microscopy (Fluoview, Olympus; Tokyo, Japan). Adjacent sections were immunostained against ventricular myosin using monoclonal mouse anticardiac myosin antibodies (Biogenesis; Poole, UK) as a primary antibody and fluorescein isothiocyanate-conjugated rabbit anti-mouse IgG (DAKO JAPAN; Tokyo, Japan) as a secondary antibody to identify myocytes.

Measurement of large amplitude swelling in mitochondria. Large amplitude swelling of mitochondria was measured as an index of permeability transition (PT) (22). Mitochondria were resuspended in buffer A and supplemented with 5 mM KH₂PO₄ and various concentrations of Ca²⁺. State 4 respiration was induced by the addition of 5 mM succinate during monitoring of optical density at 540 nm (OD₅₄₀) for 15 min at 23°C. The decline rate of OD₅₄₀, which represents mitochondrial swelling, was expressed as OD₅₄₀/mg of mitochondrial protein.

Measurement of cytochrome c release from mitochondria. Isolated mitochondria were resuspended in buffer A and supplemented with 5 mM KH₂PO₄ and various concentrations of Ca²⁺. State 4 respiration was induced by addition with 5 mM succinate at 23°C. State 2 respiration was conducted at a Ca²⁺ concentration of 1 µM. After 15 min incubation, the mitochondrial suspension was quickly placed on ice and centrifuged at 3,000 g for 15 min at 4°C. Each 10 µl of the supernatant was carefully obtained and blotted with rabbit polyclonal anti-cytochrome c antibodies (generous gift from Dr. Kanji Tomioka, Process Technology and Research Laboratories, Research Institute, Kaneka, Hyogo, Japan). Densitometry assays were performed as described above.

Measurement of ATP synthesis by mitochondria. Mitochondria were resuspended in buffer A and supplemented with 5 mM KH₂PO₄, 5 mM succinate, and 5 mM ADP, and then incubated for 15 min at 37°C. The reaction was stopped by adding an equal volume of 12% trichloroacetic acid, and the reaction mixture was centrifuged at 1,200 g for 15 min at 4°C. The supernatant was neutralized with 6 N KOH and measured for ATP by an enzymatic method using a kit obtained from Sigma Chemical (Tokyo, Japan).

Measurement of LV function. The rats treated for 24 h with 1 mg/kg IGF-I or the vehicle were anesthetized with intraperitoneal injection with pentobarbital sodium (100 mg/kg). The hearts were rapidly excised, placed in the Langendorff apparatus, and perfused as described earlier. Isovolumic LV pressure was measured with a latex balloon inserted into the LV via a left atrial incision. The balloon was filled with bubble-free saline solution and attached to a pressure transducer connected to a multichannel recorder. The balloon volume was set to produce a LV end-diastolic pressure of between 5 and 10 mmHg at the baseline. At the end of the experiment, the heart was removed from the perfusion apparatus, the wet weight of the heart was measured, and then the heart was heated to dryness at 80°C to measure the wet-to-dry ratio.

Measurement of creatine kinase release. Coronary effluent was collected during reperfusion to determine creatine kinase (CK) release to the effluent. CK activity was measured by an enzymatic assay method (32).

TUNEL assay. After 2 h of reperfusion, the heart was perfusion-fixed and the LV was paraffin-embedded as described earlier. The deparaffinized sections were assayed for in situ terminal transferase labeling (TUNEL) using Apop Tag Plus (Oncor; Gaithersburg, MD). Negative control slides were processed with the terminal deoxynucleotidyl transferase enzyme excluded. The occurrence of apoptosis in myocytes and nonmyocytes was demonstrated in the sections first stained with TUNEL (fluorescein isothiocyanate staining). The sections were then incubated with monoclonal mouse anticardiac myosin antibodies (Biogenesis) as a primary antibody followed by incubation with tetrarhodamine isothiocyanate-conjugated rabbit anti-mouse IgG as a secondary antibody and viewed with double immunofluorescence confocal laser microscopy. The number of TUNEL-positive myocytes and nonmyocytes was counted on 60 high power fields (magnification ×600) from the endocardium through the epicardium of the LV free wall in three sections from each heart.

Statistical analysis. All numerical data are presented as means ± SE. Statistical comparisons were done by an analysis of variance and Scheffé's multiple comparison test. P values < 0.05 were considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of IGF-I on cytochrome oxidase activity. Activity of cytochrome oxidase, a marker enzyme of mitochondria, was measured in the mitochondrial fraction obtained from the heart treated with IGF-I or the vehicle for 24 h. There was no significant difference in cytochrome oxidase activity in the mitochondrial fraction between the IGF-I-treated and the vehicle-treated hearts (Fig. 1). This observation suggests that the purity of mitochondria in both groups of mitochondrial suspension was identical.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Cytochrome oxidase activity of mitochondria. Mitochondria were isolated from the heart treated with the vehicle or insulin-like growth factor-I (IGF-I) for 24 h and measured for cytochrome oxidase activity as described in MATERIALS AND METHODS. Each bar represents the mean ± SE of 5 experiments.

Effects of IGF-I on expression of Bcl-2, Bcl-xL, Bax, and Bad. Bcl-2 was not detected by Western blot analysis with a loading condition performed in the present study in the mitochondrial fraction obtained from the heart treated with the vehicle or IGF-I, whereas HL-60 provided as a positive control contained abundant Bcl-2 (Fig. 2). Bcl-2 was also not detected in the cytosol and the particulate fractions (not shown). The antibodies against Bcl-2 employed in the present study have been confirmed to be reactive with the rat lymphoid cells by immunohistochemical analysis (not shown). In contrast, the mitochondrial fraction contained a detectable amount of Bcl-xL under the baseline condition. Bcl-xL appeared as a doublet just under 30 kDa. The nature of two immunoreactive bands is currently unknown, but may represent phosphorylated (Fig. 2, top band) and nonphosphorylated (bottom band) forms of Bcl-xL as demonstrated in some malignant cells by Poruchynsky et al. (37). IGF-I, but not the vehicle treatment, increased Bcl-xL in the mitochondria. Bcl-xL expression was increased appreciably 12 h after the treatment with IGF-I, reached a maximum level after 24 h, and returned to the baseline after 36 h. Quantitative analysis demonstrated that the IGF-I treatment increased Bcl-xL content in mitochondria 2.0- and 2.5-fold after 12 and 24 h, respectively, compared with the time-matched, vehicle-treated heart (Fig. 3A). The mitochondrial fraction also contained abundant Bax (Fig. 2). In contrast to Bcl-xL, Bax expression was attenuated in the mitochondria obtained from the hearts pretreated with IGF-I. The maximum attenuation of Bax was observed between 12 and 24 h after the IGF-I treatment and the expression was returned to the baseline after 36 h. Quantitative analysis demonstrated that Bax contents in the mitochondrial fractions 12 and 24 after the IGF-I treatment were 2.2- and 2.0-fold lower, respectively, than the time-matched, vehicle-treated heart (Fig. 3B). Whereas Bcl-xL was not detected in the cytosol and the particulate fractions, a small amount of Bax was present in the cytosol, although there was no measurable change in the amount of Bax in this fraction after the IGF-I treatment (not shown). An appreciable amount of Bad expression was also observed in the mitochondria (Fig. 2). Quantitative analysis showed that the amount of Bad in the mitochondria after the IGF-I treatment was not significantly different from that in the time-matched control (Fig. 3C). In addition, although Bad was present in the cytosol fraction in an amount similar to the mitochondrial fraction, the relative distribution of Bad in the cytosol and the mitochondrial fractions was not altered by pretreatment with IGF-I (not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Western blot analysis for Bcl-2, Bcl-xL, Bax, and Bad in the mitochondrial fraction obtained from the heart treated with IGF-I or the vehicle for the indicated time. PC, a positive control derived from the cell extracts of HL-60.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Quantitative analysis of Bcl-xL (A), Bax (B), and Bad (C) in the mitochondrial fraction obtained from the heart treated with IGF-I (hatched bar) or the vehicle (open bar) for 24 h. Each bar represents the mean ± SE of 5 experiments. *P < 0.05, **P < 0.01 compared with the vehicle-treated mitochondria.

Myocardial localization of Bcl-2 family proteins. Immunofluorescence confocal laser microscopy for Bcl-2 family proteins showed that Bcl-2 expression was not observed in the vehicle-treated heart (Fig. 4A) nor was it enhanced by treatment with IGF-I for 24 h (Fig. 4E). Bcl-xL expression was visible in some of myocytes in the vehicle-treated heart (Fig. 4B) and was greatly enhanced in the majority of cardiac cells, including myocytes and nonmyocytes by 24-h treatment with IGF-I (Fig. 4F). Identification of myocytes and nonmyocytes were performed on serial sections immunostained with cardiac myosin. Contrary to Bcl-xL, Bax immunostaining was potentiated in myocytes and nonmyocytes of the vehicle-treated heart (Fig. 4C) but was attenuated in these cells of the IGF-I-treated heart (Fig. 4G). The punctate pattern of immunostaining for Bcl-xL and Bax suggests, but does not prove, mitochondrial localization of these Bcl-2 family proteins. A more diffuse and less punctate pattern of Bad immunostaining was observed in both the vehicle-treated and the IGF-I-treated hearts, but the immunostaining pattern was not different between these two groups of hearts (Fig. 4, D and H)

View larger version (72K): [in this window] [in a new window]   Fig. 4.   Immunofluorescence confocal laser microscopy for Bcl-2 family proteins in the heart treated for 24 h with the vehicle (A-D) or IGF-I (E-H). A and E, Bcl-2; B and F, Bcl-xL; C and G, Bax;, D and H, Bad. Arrow, myocyte; arrow head, nonmyocyte. Magnification ×600.

Effect of IGF-I on large amplitude swelling of mitochondria. Mitochondria isolated from the vehicle-treated heart showed Ca²⁺ concentration-dependent decrease of OD₅₄₀, an index of large amplitude swelling, under state 4 respiration (Fig. 5). The occurrence of large amplitude swelling became prominent 15 min after incubation. In contrast, mitochondria isolated from the heart pretreated with IGF-I for 24 h almost completely inhibited large amplitude swelling induced under a high Ca²⁺ concentration (1 µM).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Large amplitude swelling (LAS) of isolated mitochondria. LAS is expressed as decline in optical density (OD)540. Filled symbols (, no Ca2+; , 10 nM Ca2+, , 100 nM Ca2+; , 1,000 nM Ca2+) indicate the mitochondria obtained from the heart treated with the vehicle for 24 h. Open symbols (, no Ca2+; , 10 nM Ca2+; , 100 nM Ca2+; , 1,000 nM Ca2+) indicate the mitochondria obtained from the heart treated with IGF-I for 24 h. Each symbol represents the mean ± SE of 5 experiments. *P < 0.05 compared with the vehicle-treated mitochondria at respective Ca2+ concentrations.

Effect of IGF-I on cytochrome c release from mitochondria. Cytochrome c release was not observed in nonrespiring (state 2) mitochondria even under a high Ca²⁺ concentration (Fig. 6). There was only a small amount of cytochrome c release from the mitochondria incubated under state 4 respiration without added Ca²⁺. However, cytochrome c release from the respiring mitochondria was increased in a Ca²⁺ concentration-dependent manner. Cytochrome c release was significantly inhibited in the mitochondria obtained from the heart pretreated with IGF-I for 24 h at any given Ca²⁺ concentrations.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Cytochrome c (cy. c) release from isolated mitochondria. Top: Western blot analysis for cy. c in the supernatant of mitochondrial suspension after 15-min incubation at state 2 or state 4 respiration under various Ca2+ concentrations. Mitochondria were isolated from the heart treated with IGF-I or the vehicle for 24 h. Bottom: quantitative analysis for cy. c release from the mitochondria at state 2 or state 4 respiration under various Ca2+ concentrations. Solid bars, IGF-I-treated mitochondria; open bars, vehicle-treated mitochondria. Each bar represents the mean ± SE of 5 experiments. *P < 0.05 compared with the vehicle-treated mitochondria.

Effect of IGF-I on ATP synthesis by mitochondria. There was no significant difference in basal ATP contents between the vehicle-treated and IGF-I-treated mitochondria (not shown). However, ATP-generating activity was significantly increased in the mitochondria obtained from the heart treated with IGF-I for 24 h (Fig. 7).

View larger version (9K): [in this window] [in a new window]   Fig. 7.   ATP-generating activity of isolated mitochondria. Mitochondria were isolated from the heart treated with the vehicle or IGF-I for 24 h and were incubated in the ATP-generating buffer as described in MATERIALS AND METHODS. Each bar represents the mean ± SE of 5 experiments. *P < 0.05 compared with vehicle.

Effect of IGF-I on LV function. Since Western blot analysis and immunohistochemistry demonstrated maximum expression of Bcl-xL and attenuation of Bax 24 h after the treatment with IGF-I, the effect of pretreatment for 24 h with IGF-I on the recovery of myocardial function after 25 min of normothermic global ischemia was evaluated in the isolated and perfused rat heart (Table 1). None of the baseline measurements of LV function, i.e., heart rate, LV developed pressure (LVDP), LV end-diastolic pressure (LVEDP), and coronary flow were significantly different between the vehicle-treated and the IGF-I-treated heart. LVEDP markedly increased 25 min after ischemia without a significant intergroup difference, suggesting an equivalent degree of ischemic contracture development by pretreatment with IGF-I. However, IGF-I-pretreated hearts exhibited significantly greater recovery of LV systolic function as evidenced by an increase in LVDP and heart rate and LV diastolic function as evidenced by a decline of LVEDP during reperfusion compared with the vehicle-treated heart. Maximum recovery of LV function was observed between 30 and 60 min after reperfusion, and LV systolic function declined gradually thereafter in both groups of hearts, although LVDP and heart rate remained significantly better in the IGF-I-treated heart 2 h after reperfusion. Coronary flow was decreased after reperfusion in both groups of hearts without a significant intergroup difference throughout a reperfusion period.

Effects of IGF-I on CK release. Using the Langendorff preparation, we also examined the effect of pretreatment for 24 h with IGF-I on CK release during reperfusion after 25 min of normothermic global ischemia (Fig. 8). CK release peaked 10 min after reperfusion and declined gradually thereafter in the hearts after 25 min of ischemia. However, CK release during reperfusion was significantly inhibited in the hearts pretreated with IGF-I compared with the hearts pretreated with the vehicle only.

View larger version (12K): [in this window] [in a new window]   Fig. 8.   Creatine kinase (CK) release in the isolated and perfused rat heart. The rat heart treated with IGF-I or the vehicle for 24 h was isolated and buffer perfused. The heart was subjected to 25 min of normothermic global ischemia followed by 120 min of reperfusion or was perfused continuously as a time-matched control. , Vehicle (n = 10 rat hearts); , IGF-I (n = 8 rat hearts); , time-matched control (n = 5 rat hearts). Each symbol represents the mean ± SE. *P < 0.05 compared with vehicle.

Effects of IGF-I on apoptosis. We then evaluated the effect of pretreatment for 24 h with IGF-I on the occurrence of apoptosis 2 h after reperfusion. Double immunofluorescent microscopy using TUNEL and antibodies against cardiac myosin discriminate TUNEL-positive myocytes from TUNEL-positive nonmyocytes. The time-matched control heart showed few TUNEL-positive myocytes and nonmyocytes (<0.1% of total myocyte and nonmyocyte populations) (Fig. 9). However, both TUNEL-positive myocytes and nonmyocytes increased 2 h after reperfusion in the hearts pretreated with the vehicle only. Fewer TUNEL-positive myocytes and nonmyocytes were found after reperfusion in the hearts pretreated with IGF-I. Quantitative analysis showed that the number of TUNEL-positive myocytes and nonmyocytes was significantly reduced by IGF-I-pretreatment compared with the vehicle pretreatment.

View larger version (61K): [in this window] [in a new window]   Fig. 9.   Terminal transferase labeling (TUNEL) staining in the isolated and perfused rat heart. The rat heart treated with IGF-I or the vehicle for 24 h was isolated and buffer-perfused. The heart was subjected to 25 min of normothermic global ischemia followed by 120 min of reperfusion or was perfused continuously as a time-matched control. The heart was perfusion-fixed for TUNEL assay as described in MATERIALS AND METHODS. TUNEL-positive nuclei are stained with green and myocytes are stained with red. Arrow, myocyte; arrowhead, nonmyocyte. Magnification ×600. The number of TUNEL-positive myocytes and nonmyocytes was counted on 60 high-power fields (magnification ×600) from the endocardium through the epicardium of the left ventricular free wall in three sections from each heart (right). Open bar, myocytes; solid bar, nonmyocytes. Each bar represents the mean ± SE of 5 experiments (control), 10 experiments (vehicle), and 8 experiments (IGF-I). *P < 0.05, **P < 0.01 compared with vehicle.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The major finding of the present study was that pretreatment of the rat with a single intraperitoneal injection of IGF-I at a dose of 1 mg induces enhanced Bcl-xL expression in the heart mitochondria. Although IGF-I increases the expression of Bcl-xL in several different cell types (5, 21, 36, 49), to our knowledge, Bcl-xL increased by growth factors and cytokines has not been reported in cardiomyocytes. Bax expression, on the contrary, was attenuated in the mitochondria after the IGF-I treatment. Bax expression is known to be attenuated by IGF-I treatment in cardiomyocytes (44, 45). Such differential regulation of Bcl-xL and Bax levels in the mitochondria reached maximum 24 h after the IGF-I treatment. Bcl-xL was undetectable in the cytosol and particulate fractions before and after treatment with IGF-I. Although Bax was present in the cytosol, there was no measurable change in the amount of Bax after the IGF-I treatment. Confocal laser microscopy showed that enhanced immunostaining of Bcl-xL and attenuated immunostaining of Bax occurred in both myocytes and nonmyocytes. Unlike Bcl-xL and Bax, Bcl-2 was not detected in any subcellular fractions by Western blot analysis or by immunofluorescent confocal microscopy before and after the IGF-I treatment in the rat heart. Our inability to detect Bcl-2 was not due to the antibody employed in the present study as mentioned earlier. On the other hand, an appreciable amount of Bad was present under the basal condition in both the cytosol and the mitochondrial fractions. The amount and the relative distribution of Bad in these fractions were not altered by pretreatment with IGF-I. Thus the findings of the present study indicate that the IGF-I treatment differentially regulates the amount of Bcl-xL and Bax in the mitochondria, thereby increasing the quantitative ratio of antiapoptotic Bcl-xL to proapoptotic Bax proteins in this organelle, which is considered to be an important determinant for cell survival (17, 21, 28, 38, 49, 50).

The changes in Bcl-xL and Bax expression accompanied by inhibition of Ca²⁺ induced large amplitude swelling and cytochrome c release and increase in ATP synthesis in the isolated mitochondria. These beneficial effects of IGF-I on mitochondrial function in vitro was associated with improved recovery of myocardial function, reduced CK release and apoptosis during reperfusion in the isolated, and perfused rat heart preparation. The putative roles of Bcl-2 family proteins in protecting mitochondrial function and cytochrome c release suggest that the observed myocardial protection against ischemia-reperfusion injury conferred by pretreatment with IGF-I may be causally related to the differential regulation of Bcl-2 family proteins. However, because mitochondrial oxidative phosphorylation is shut down during ischemia due to the lack of oxygen, it is difficult to perceive that the favorable change of Bcl-2 family proteins by IGF-I can provide a significant benefit during ischemia by improving mitochondrial function. Thus the timing of beneficial action by IGF-I pretreatment should be most often after reperfusion and not during ischemia. It should also be noted that because isolated mitochondria are likely to be originated from both myocyte and nonmyocyte populations in our study, relative contribution of these cell populations in IGF-I-induced modulation of mitochondrial parameters remains unclear. However, it is likely that inhibition of apoptosis in nonmyocyte populations may also contribute to the alleviation of ischemia-reperfusion injury.

The critical issue concerning the form of cardiomyocyte cell death during reperfusion is whether cardiomyocytes die solely from apoptosis. It has been suggested that apoptosis is an independent contributor of cardiomyocyte cell death during reperfusion (15). Consistent with this hypothesis is the fact that treatment with the caspase inhibitor Z-Val-Ala-Asp(OMe)-fluoromethylketone led to an inhibition of apoptosis associated with a reduction of infarct size and an improvement of hemodynamic functions (48). However, contradictory results have also been provided by Ohno et al. (31) who have proposed that apoptotic cardiomyocytes observed in the infarcted myocardium after ischemia and reperfusion may be necrotic myocytes with DNA fragmentation. Our recent study (35) using a double-labeling technique with annexin V and propidium iodide, has supported the later hypothesis that the majority of apoptotic myocytes also underwent necrosis. Although such a controversy remains to be settled, it is clear that myocardial ischemia and reperfusion could provoke both forms of cell death.

Bcl-2 family proteins are known to play a crucial role in regulating the life and death of the cell at the mitochondrial level. A large number of Bcl-2 family proteins resides in the outer membrane of mitochondria and these proteins are thought to regulate mitochondrial PT pore opening (23, 38, 50). The PT pore is a large conductance, cyclosporin-inhibitable channel in the inner membrane of mitochondria. PT pore opening causes dissipation of H⁺ gradient across the membrane. The loss of electrochemical gradient results in uncoupling of the respiratory chain and subsequent abrogation of ATP synthesis via F₀-F₁ ATPase. Depriving the cell of ATP is a primary step for necrosis. PT pore opening, on the other hand, is associated with the release of cytochrome c and the apoptosis inducing factor from mitochondria (18, 23, 25). Cytochrome c is a component of the mitochondrial electron transfer chain and is localized in intermembrane space in mitochondria. Cytochrome c released into cytosol can activate a cysteine protease (caspase) cascade leading to apoptosis in the presence of Apaf-1 and dATP (13, 51). Although a causal relationship between PT pore opening and cytochrome c release has not been demonstrated unequivocally (17), it appears that antiapoptotic Bcl-2 family proteins, such as Bcl-2 and Bcl-xL, interfere with mitochondrial PT pore opening and the release of cytochrome c and apoptosis-inducing factor (16, 18, 38). The proapoptotic Bcl-2 family protein Bax on the contrary, accelerates cytochrome c release by interacting with Bcl-2 and Bcl-xL (41,46) and by a mechanism independent of interaction with these proteins (14). Bad, on the other hand, binds strongly to Bcl-xL and displaces Bax from Bcl-xL, resulting in free Bax and thereby introducing cell death (47). Thus mitochondria appear to play a key role in both necrosis and apoptosis, and this organelle may be a point of convergence of the pathways that mediate these morphologically distinct forms of cell death.

Our present study confers significant therapeutic implications in IGF-I treatment against heart failure and myocardial ischemia-reperfusion injury. Mitochondria isolated from the failing heart exhibits a deteriorated function as documented by a decreased respiratory control index and ATP synthetic activity (5, 30). Recently, cytochrome c release from mitochondria has been demonstrated in the human heart with end-stage cardiomyopathy associated with the increase in the number of apoptotic cardiomyocytes (4). Moreover, recent studies (6, 11) have demonstrated that PT pore opening occurs during postischemic reperfusion concomitant with intracellular Ca²⁺ overload and oxidative stress, two major causes of myocardial reperfusion injury (34). It is, therefore, suggested that IGF-I treatment could alleviate heart failure and myocardial ischemia-reperfusion injury by inhibiting two distinct forms of cell death, i.e., necrosis and apoptosis, through the preservation of functional integrity of mitochondria and the inhibition of death signaling generation by this organelles.