Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Platelet activation plays an important role in the pathogenesis of myocardial ischemia (Aiken et al., 1981; Conti and Mehta, 1987). Although platelets participate in thrombus formation and contribute to coronary occlusion, the deposition of platelets in the ischemic-reperfused myocardium is secondary to neutrophil deposition, and platelets are not deleterious to the myocardium (Bednar et al., 1985; Mullane and McGiff, 1985). Recent studies from our laboratory have demonstrated that washed platelets protect isolated hearts against reperfusion injury. The mechanisms of cardioprotection involve direct positive inotropic effect of platelets, platelet-elicited nitric oxide release from the myocardial microvasculature, antioxidant effect of platelets, platelet-derived adenosine, and platelet-released transforming growth factor-₁ (TGF-₁) (Yang et al., 1993, 1994; Yang and Mehta, 1994; Mehta et al., 1999).

Apoptosis is an important mechanism of cell death necessary for normal development, and is also seen in many pathological states. Its hallmark is cleavage of genomic DNA into nucleosomal fragments of 160 to 200 base pairs. The outcome of myocardial infarction depends on myocardial injury, including apoptosis (Buja and Entman, 1998). Reperfusion after transient myocardial ischemia activates apoptosis in cardiomyocytes (Gottlieb et al., 1994). The precise trigger of apoptosis during ischemia-reperfusion is unknown, but the cytokines, tumor necrosis factor-, interleukin-1, and interleukin-6, have been implicated (Polunovsky et al., 1994).

The modulation of growth factor-related signals represents a novel strategy for the treatment of cardiac and vascular diseases. TGF-₁, a member of the growth factor family and mainly produced from platelets, has a cardioprotective activity against myocardial ischemia-reperfusion injury (Lefer et al., 1990; Roberts and Sporn, 1993). Although TGF-₁ induces/promotes apoptosis in some carcinoma/tumor cell lines (Chuang et al., 1994; Khosla et al., 1994; Mathieu et al., 1995; Perry et al., 1995; MacDonald et al., 1996; Yamamoto et al., 1996), Henrich-Noack et al. (1996) showed that TGF-₁ protects neuronal tissue against degeneration caused by transient global ischemia, and suggested that this protective effect is associated with the antioxidant and antiapoptotic effects of TGF-₁. Kawakami et al. (1996) reported that TGF-₁ inhibits Fas antigen-mediated apoptosis of rheumatoid synovial cells in vitro. Another recent study showed that TGF-₁ prevents cytokine-mediated induction of metalloelastase in macrophages (Feinberg et al., 1998). We postulated that platelet-mediated cardioprotection during ischemia may relate to the antiapoptotic effect of TGF-₁. Accordingly, we investigated the relationship among platelet-mediated cardioprotective effect, TGF-₁, and apoptosis in cultured adult rat myocytes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Myocyte Isolation

Myocytes were isolated from adult Sprague-Dawley rats (200-300 g) by a slight modification of the procedure of Claycomb and Palazzo (1980). All procedures were carried out under aseptic conditions in a laminar flow hood. Rats were anesthetized with sodium pentobarbital (60 mg/kg i.p.) and heparinized with 1000 U of sodium heparin/kg. The thorax was then opened and the heart was removed and placed into 25 ml of ice-cold Ca²⁺-free Krebs-Henseleit (K-H) buffer (perfusion medium, composition: NaCl 118 mM, KCl 4.7 mM, KH₂PO₄ 1.2 mM, MgSO₄ 1.2 mM, NaHCO₃ 25 mM, and glucose 11 mM, pH 7.4). Within 1 min, the heart was transferred to a perfusion apparatus and perfused via aorta with oxygen-saturated (95% O₂/5% CO₂) Ca²⁺-free K-H buffer at 37°C, at a rate of 5 to 6 ml/min for 5 min, to wash out residual blood. The heart was then perfused with 100 ml of oxygen-saturated (95% O₂/5% CO₂) Ca²⁺-free K-H buffer containing 1 mg/ml collagenase type XI (Sigma), kept at 37°C at a rate of 5 to 6 ml/min to rinse out the intravascular space. The perfusion medium was recirculated and the perfusion was continued until the heart became soft.

Atria and large vessels were dissected off and the ventricles were transferred to a plastic container and minced into small pieces. These tissue fragments were mixed in 7.5 ml of perfusion medium containing 1 mg/ml collagenase, incubated for 15 min with shaking in a water bath at 37°C, then 7.5 ml of enzyme-free perfusion medium was added. The large tissue fragments were allowed to settle and the supernatant containing the liberated cells was decanted into a graduated 15-ml tube and centrifuged at approximately 10 g for 5 min to sediment the cells. The supernatant was aspirated, 10 ml of fresh perfusion medium was added, and the cells were gently resuspended and recentrifuged for 5 min. This procedure was repeated until very few tissue fragments remained. All the cells were then combined and further washed and freed of nonmuscle cells by allowing them to settle in a centrifuge tube for approximately 10 to 15 min. The result of these manipulations is a homogeneous preparation of adult ventricular cardiac muscle cells.

Cell Culture

To each 0.1 ml of sedimented cells, 1.9 ml of perfusion medium was added, and the cell number was counted. Approximately 2.5 × 10⁵ cells were placed into 25-cm² plastic culture flasks containing 5 ml of culture medium (Dulbecco's modified Eagle's medium containing Eagle's salts, amino acids, 10% fetal bovine serum, 100 U/ml penicillin, and 0.1 mg/ml streptomycin; Sigma), which was pre-equilibrated for 1 to 2 h in a CO₂ incubator. The flasks were gently shaken to evenly distribute the cells and then maintained in a 37°C incubator in a humidified atmosphere containing 95% air and 5% CO₂. At 2 h and then again at 20 to 24 h after placing the cells in culture, the contents of the flasks were decanted into new flasks. This procedure virtually eliminates fibroblast contamination because fibroblasts attach to the surface of the culture flask very rapidly. The medium was changed in 5 to 7 days (live cells all attached to the surface of the culture flask), and then every other day thereafter, as described by Claycomb and Palazzo (1980).

Preparation of Aggregated Platelet Supernatant

Rat blood withdrawn from the common carotid artery was collected in 3.8% sodium citrate (9:1, v/v) and centrifuged at 150g for 10 min at room temperature to obtain platelet-rich plasma, which was centrifuged again at 1000g for 20 min at 4°C. Platelets were washed with Tris-sodium-glucose buffer (composition: Tris-HCl 15 mM, NaCl 134 mM, glucose 5 mM, and EDTA 1 mM, pH 7.4) and suspended in K-H buffer (10⁹ cells/ml). Platelet suspension was then transferred to a glass tube to induce platelet aggregation. Aggregated platelet medium was then centrifuged at 1000g for 20 min at 4°C to obtain the supernatant of aggregated platelets (Mehta et al., 1999).

Exposure of Myocytes to Hypoxia-Reoxygenation

In a pilot experiment, culture medium (under normoxia, i.e., 95% air/5% CO₂, PO₂ ~ 150 mm Hg) was replaced with 95% N₂/5% CO₂ pre-equilibrated culture medium. The culture flasks were placed in a 37°C incubator in a humidified atmosphere containing 95% N₂/5% CO₂ for 0, 48, or 72 h, PO₂ ~ 30 mm Hg (hypoxia). Parallel sets of cells were exposed to 48 or 72 h of hypoxia and then placed in a 37°C incubator in a humidified atmosphere containing 95% air and 5% CO₂ for 3 h (reoxygenation). Then, the culture medium was collected for lactate dehydrogenase (LDH) measurement. The myocytes were detached by incubation with trypsin (0.25 mg/ml), settled on positive electron-charged glass slides by cytospin, and examined for apoptosis by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) stain. These experiments showed that the optimal hypoxia time for induction of apoptosis was 48 h, and the degree of apoptosis increased substantially after reoxygenation (see Results).

In subsequent experiments, cultured adult rat myocytes were exposed to hypoxia for 48 h followed by 3 h of reoxygenation. Myocytes not exposed to hypoxia served as control. Various groups of myocytes were incubated with 0.5 to 5 ng/ml human recombinant TGF-₁, or aggregated rat platelet supernatant (from 2-3 × 10⁷ platelets/ml, containing ~0.5 ng/ml TGF-₁) 10 min before hypoxia. At the end of the hypoxia-reoxygenation period, myocytes were examined for cell necrosis (by trypan blue stain and LDH release), apoptosis (phase contrast microscopy, TUNEL stain, and DNA laddering), and Bcl-2 and Fas expression (Western blotting).

Determination of LDH in Culture Medium

LDH activity in culture medium was determined by using Cyto Tox 96 NonRadioactive Cytotoxicity Assay kit (Promega, Madison, WI).

Determination of Apoptosis

Quantification of Apoptosis by In Situ TUNEL Stain. This assay was performed using ApopTag Plus In Situ Apoptosis Detection Kit (Oncor, Gaithersburg, MD). At least 500 cells from randomly selected fields were counted manually to determine the percentage of apoptotic cells (Li et al., 1998).

Extraction and Electrophoresis of DNA Fragmentation (DNA Laddering). Cultured rat adult myocytes (1 × 10⁶) were removed from culture flasks, and washed twice in PBS. DNA was extracted using Apoptosis Lysate Kit (Chemicon, Temecula, CA). Twenty micrograms of protein was electrophoresed on a 1.8% agarose gel, and the gels were photographed (Li et al., 1998).

Quantitation of Fas Protein Expression in Myocardium by Western Analysis

Myocytes were lysed in boiling lysis buffer (0.1% SDS, 0.1% Triton X-100, 10 mM Tris-HCl, pH 7.4) and centrifuged at 10,000 rpm for 30 min at 4°C. The lysate protein from myocytes (10-15 µg/lane) was separated by 8% SDS-polyacrylamide gel electrophoresis using a Bio-Rad Mini-Protean cell (Bio-Rad Laboratories, Hercules, CA), transferred to nitrocellulose filters (Amersham Life Science, Arlington Heights, IL), and then immunoblotted with a rabbit polyclonal antibody against Fas at 1:200 dilution (1 µg/ml). Anti-rabbit IgG conjugated with AP was used as a second antibody at 1:2500 dilution. The blots were detected with AP substrate as described earlier (Li et al., 1998).

Quantitation of Bcl-2 Protein Expression in Myocardium by Immunoprecipitation

Myocytes were lysed in boiling lysis buffer (1% SDS, 0.1% Triton X-100, 10 mM Tris-HCl, pH 7.4) and centrifuged at 10,000 rpm for 30 min at 4°C. 400 µg of protein sample in 1 ml of lysis buffer was placed in 1.5 ml microcentrifuge tube, mixed with 10 µl (2 µg) of rabbit polyclonal antibody to Bcl-2, and incubated at 4°C for 1 h, then mixed with 20 µl of protein G plus-agarose (Santa Cruz Biotechnologies, Inc., Santa Cruz, CA) and incubated at 4°C on a rotating device overnight. The immunoprecipitates were collected by centrifugation at 3000 rpm at 4°C for 5 min, washed three times with lysis buffer, resuspended in 20 µl of electrophoresis buffer, boiled for 3 min, and subjected to Western blotting by using rabbit polyclonal antibody against Bcl-2. The rest of the procedure was the same as described above.

Statistical Analysis

All data were obtained from at least five separate experiments, and expressed as mean ± S.D. Data were compared with ANOVA followed by Scheffè F test or Newman Kuel's-test for paired and unpaired observations. A P value <.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In preliminary experiments, hypoxia alone was found to cause a time-dependent increase in the number of apoptotic myocytes. The number of apoptotic cells was 19 ± 3% after 48 h of hypoxia and 39 ± 5% after 72 h of hypoxia compared with 5 ± 1% in control cells, based on a count of 500 cells. Three hours of reoxygenation after 48 h of hypoxia further increased the number of apoptotic cells (34 ± 8 versus 19 ± 3% in hypoxia for 48 h, P < .05), whereas reoxygenation after 72 h of hypoxia did not further increase the number of apoptotic cells. These data were confirmed by in situ TUNEL staining and DNA laddering, as shown in Figs. 1 to 3.

View larger version (111K): [in this window] [in a new window]   Fig. 1.   Identification of apoptosis in myocytes by in situ TUNEL staining. C, control cells; H, cell exposed to hypoxia for 48 h; HR, cells exposed to hypoxia for 48 h followed by reoxygenation for 3 h; (T+HR) cells exposed to TGF-1 1 ng/ml and hypoxia for 48 h and reoxygenation for 3 h; (P+HR) cell exposed to aggregated platelet supernatant (from ~2 × 107 platelets/ml) and hypoxia for 48 h and reoxygenation for 3 h; and N, negative control. Cells with dark purple to blue stain with fragmented nuclei are apoptotic cells. (original magnification, 200×)

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Summary of the data on hypoxia (48 h)-reoxygenation (3 h)-induced apoptosis in cultured adult rat myocytes and the effect of platelets and TGF-1. Forty-eight hours of hypoxia significantly increased the number of apoptotic myocytes compared with the control group. Three hours of reoxygenation after 48 h of (H-R) hypoxia further increased the number of apoptotic cells. Addition of TGF-1 (0.5 ng/ml) or aggregated platelet supernatant (from ~2 to 3 × 107 platelets/ml) before hypoxia significantly reduced the H-R-induced myocyte apoptosis. n = 5 to 10 in each group. Data is mean ± S.D.

View larger version (79K): [in this window] [in a new window]   Fig. 3.   DNA fragmentation in myocyte lysate in one experiment. Hypoxia for 48 h resulted in minimal DNA laddering. Three hours of reoxygenation after 48 h of hypoxia markedly increased DNA laddering. Preincubation of myocytes with TGF-1 0.5 ng/ml, or aggregated platelet supernatant reduced DNA fragmentation despite hypoxia and reoxygenation.

Incubation of myocytes with human recombinant TGF-₁ markedly reduced the number of apoptotic cells despite hypoxia-reoxygenation. The effect of 0.5 ng/ml recombinant TGF-₁ was same as that of 5 ng/ml of recombinant TGF-₁. Presence of aggregated rat platelet supernatants in cultured myocytes during the period of hypoxia-reoxygenation reduced the number of apoptotic myocytes in a similar fashion as did recombinant TGF-₁, as determined by TUNEL staining (Fig. 2). Both TGF-₁ and platelet supernatants decreased DNA laddering on gel electrophoresis, but the results were less marked as compared with TUNEL staining data (Fig. 3).

Hypoxia-reoxygenation caused an increase in myocyte Fas expression, and a decrease in myocyte Bcl-2 expression (Fig. 4). Presence of recombinant TGF-₁ or aggregated platelet supernatant to the culture flasks before hypoxia blocked the decrease in Bcl-2 expression in myocytes after hypoxia-reoxygenation (Fig. 4). On the other hand, the effects on Fas expression were less marked (Fig. 4).

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Myocyte Fas and Bcl-2 expression determined by Western blotting. Control myocytes showed modest Fas and Bcl-2 signals; hypoxia (48 h)-reoxygenation (3 h) increased myocyte Fas signal and decreased Bcl-2 signal; addition of TGF-1 0.5 ng/ml or aggregated platelet supernatant (from ~2 to 3 × 107 platelets/ml) to the culture medium before hypoxia restored the myocyte Fas and Bcl-2 signals. These results are representative of five separate experiments.

Hypoxia-Reoxygenation-Induced Cell Injury and Protective Effect of Platelets and TGF-₁

Exposure of myocytes to 48 h of hypoxia caused modest membrane injury, and reoxygenation of the hypoxic myocytes further increased injury, as determined by LDH released into the culture medium as well as by trypan blue exclusion (Fig. 5). Presence of recombinant TGF-₁ in the culture flasks before hypoxia markedly reduced hypoxia-reoxygenation-induced myocyte LDH release, and these results were confirmed by trypan blue exclusion (Fig. 5). However, platelet supernatant did not affect LDH release or trypan blue staining (Fig. 5). The injury data (trypan blue stained cells and LDH release) in response to TGF-₁ paralleled the data on apoptosis (TUNEL staining). The effects of platelet supernatant on prevention of cell injury were less evident as compared with those of TGF-₁.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Hypoxia (48 h)-reoxygenation (3 h) induced cell necrosis in cultured adult rat myocytes, determined by LDH release and trypan blue staining. Hypoxia-reoxygenation caused increase in LDH levels in the culture medium and the number of trypan blue positive cells. Addition of TGF-1, but not aggregated platelet supernatant, to the culture medium before hypoxia modestly attenuated the hypoxia-reoxygenation-induced number of trypan blue stained myocytes and LDH level in the culture medium. (n = 5-10 in each group).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Hypoxia-Reoxygenation and Cell Injury in Cardiomyocytes

It has long been known that reperfusion of previously ischemic tissues leads to an additional injury beyond that caused by ischemia alone (Hearst, 1977; Thompson and Hess, 1986). In the last decade, distinct modes of cell injury in response to ischemia-reperfusion have been described; among these is the process of apoptosis (programmed cell death) (Buja and Entman, 1998). Recent studies from our laboratory have suggested that platelets, via release of TGF-₁, have the potential to limit myocardial injury that follows ischemia-reperfusion (Mehta et al., 1999). To study the regulation of myocyte injury by TGF-₁, the current study was undertaken in cultured rat cardiomyocytes. A major emphasis of ischemia-reperfusion injury was to study the process of apoptosis.

This study demonstrated that hypoxia alone causes time-dependent apoptosis, and reoxygenation increases the number of apoptotic cells beyond that in response to hypoxia alone. Notably, reoxygenation did not increase apoptosis when the duration of hypoxia was increased to 72 h, probably because of extensive cell necrosis by prolonged hypoxia. In addition to an increase in apoptosis by reoxygenation, there was evidence of cell death as determined by LDH release and trypan blue staining.

Apoptosis in Ischemic-Reperfused Myocardial Tissues

The hallmark of apoptosis is cleavage of genomic DNA into nucleosomal fragments of 160 to 200 bases pairs, which can be recognized by DNA laddering on gel electrophoresis as well as laddering of nuclear fragments by TUNEL stain. Because these methodologies may not entirely distinguish between apoptosis and cell necrosis, it has been proposed that several methods must be used conjunctively to determine apoptosis (Buja and Entman, 1998). Indeed, use of gel electrophoresis and in situ TUNEL staining confirmed that a process akin to apoptosis occurs in cardiac myocytes exposed to ischemia-reperfusion. These data support the observations of Gottlieb et al. (1994), who demonstrated association of reperfusion injury with apoptosis in hearts of rabbits exposed to ischemia-reperfusion. It is noteworthy that the required duration of hypoxia as well as reoxygenation for induction of apoptosis is much greater in myocyte cultures than in the in vivo setting.

Although mechanisms controlling apoptosis in cardiac cells have not been defined, it is known that alterations in certain genes are associated with apoptosis. For example, bcl-2 protein expression is decreased (Garcia et al., 1992; Kane et al., 1993; Haendeler et al., 1996) and that of Fas protein increased in apoptosis (Li et al., 1998). In the present study, we demonstrate that hypoxia-reoxygenation is associated with a reduction in bcl-2 and an increase in Fas protein.

Growth Factors and Myocardial Protection

An important role of several growth factors in cardioprotection is becoming increasingly evident in experimental and clinical studies (Roberts et al., 1992; Padua et al., 1995; Fazio et al., 1996). TGF-₁, one of the growth factors, helps to maintain the rate of myocyte contraction in cultured cells (Padua et al., 1995), inhibits circulating neutrophils from adhering to the endothelium (Freese et al., 1992), and reduces the amount of superoxide anions (Roberts and Sporn, 1993). Lefer et al. (1990) reported that TGF-₁ given i.v. protected isolated feline cardiac tissues against ischemia-reperfusion injury. They demonstrated that TGF-₁ preserved ischemia-reperfusion-induced reduction in endothelium-dependent relaxation of coronary artery, prevented superoxide anion generation without superoxide anion scavenger action, and decreased tumor necrosis factor- release in the coronary effluent during reperfusion after coronary artery occlusion. They attributed the cardioprotective effect of TGF-₁ to its inhibitory effect on free radical generation and preservation of vasomotor tone. Our earlier studies in isolated rat hearts showing markedly beneficial effect of TGF-₁ complement their observations (Mehta et al., 1999). Another growth factor, basic fibroblast growth factor, also improves survival and prevents degeneration of cells (Freese et al., 1992). Padua et al. (1995) recently reported that basic fibroblast growth factor exerts a cardioprotective effect against ischemia-reperfusion injury, which could be mediated by augmentation of function of viable myocardium or to reduced myocyte injury after ischemia-reperfusion. Fazio et al. (1996) showed beneficial effect of human growth hormone in patients with dilated cardiomyopathy. It is possible that these growth factors, including TGF-₁, exert cardioprotective effect via inhibitory effects on the release of free radicals (Lefer et al., 1990) and inflammatory cytokines (Ranges et al., 1987), preservation of nitric oxide activity (Lefer et al., 1990), and reduction myocardial lipid peroxidation (Padua et al., 1995) after ischemia-reperfusion. In the present study, TGF-₁ also protected against hypoxia-reoxygenation-induced myocyte necrosis (LDH release and trypan blue staining), and these effects paralleled the inhibition of apoptosis.

TGF-₁ also plays an important role in the modulation of apoptosis. Most reports on TGF-₁ and apoptosis relate to carcinoma or other tumor cell lines. In most carcinoma/tumor cell lines (Chuang et al., 1994; Khosla et al., 1994; Mathieu et al., 1995; Perry et al., 1995; MacDonald et al., 1996; Yamamoto et al., 1996), TGF-₁ induces or promotes apoptosis. TGF-₁ has also been reported to cause apoptosis in cultured hepatocytes (Benedetti et al., 1995), human umbilical vein endothelial cells (Tsukada et al., 1995), endometrial stroma (Moulton, 1994), and murine osteoclasts (Hughes et al., 1996). An important study by Henrich-Noack et al. (1996) showed that TGF-₁, in a surprisingly low dose range (0.5, 4, 50 ng i.c.v., or 4 ng intrahippocampally), protected hippocampal neurons against degeneration caused by transient global ischemia, and suggested that this protective effect could well be associated with the antioxidant and antiapoptotic effects of TGF-₁ demonstrated in vitro. Kawakami et al. (1996) reported that TGF-₁ inhibits Fas antigen-mediated apoptosis of rheumatoid synovial cells in vitro. The observations suggest that although TGF-₁ is capable of inducing apoptosis in carcinoma cells, hepatocytes, endothelial cells, and osteoclasts, it can also exert antiapoptotic effect in cardiac myocytes subjected to ischemia and inflammation, most probably by regulating Fas and bcl-2 expression (Fig. 4).

Antiapoptotic Effect of Platelets

Almost two decades ago, it became evident from work in several laboratories, including ours, that platelet hyperactivity plays a critical role in the induction of thrombosis in the narrowed coronary arteries (Conti and Mehta, 1987). Although platelet accumulation and activation in the coronary artery have been proposed to initiate ischemic events and exert a detrimental effect on cardiac performance (Aiken et al., 1981), the role of platelets in reperfusion injury has not been determined until recently. Recent studies from our laboratory have demonstrated that washed platelets can protect isolated hearts against reperfusion injury, and the antioxidant effect of platelets and platelet-released TGF-₁ may be the most important mechanisms of platelet-mediated cardioprotection (Yang et al., 1993, 1994; Yang and Mehta, 1994; Mehta et al., 1999). In the present study, we show that aggregated platelet supernatant inhibits hypoxia-reoxygenation-induced myocyte apoptosis. The effects of platelet supernatant on myocyte necrosis (LDH release and trypan blue staining) were less marked than those of TGF-₁. The reduction in apoptosis was evident from in situ TUNEL staining (Fig. 2) and to a lesser extent on gel electrophoresis (Fig. 3). The protective effect of platelet supernatants against myocyte apoptosis was qualitatively similar to that of recombinant TGF-₁. Incubation of myocytes with recombinant TGF-₁ caused bcl-2 expression to increase and Fas expression to decrease. On the other hand, platelet supernatants had a pronounced effect on bcl-2 expression only. Earlier studies have indicated that forced up-regulation of bcl-2 with the use of gene transfer techniques can inhibit apoptosis in rat smooth muscle cells (Tsukada et al., 1995) and murine endothelial cells (Kondo et al., 1994). Our earlier studies (Mehta et al., 1999) indicated that 2 to 3 × 10⁷ platelets release ~0.5 ng TGF-₁, the amount that causes significant protection of isolated heart or cultured rat myocytes from hypoxia-reoxygenation-mediated injury. The qualitative similarity of data with platelet supernatant or recombinant TGF-₁ suggests that TGF-₁ is an important stable mediator released from platelets that protects tissues during hypoxia-reoxygenation. There may well be mediators other than TGF-₁ in aggregating platelet supernatants, which regulate myocyte survival and function and result in differences from the effect of TGF-₁ alone. In future studies, TGF-₁ antibodies may be used to remove TGF-₁ from the platelet supernatants and delineate the role of this growth factor in preventing myocyte injury.

In summary, hypoxia-reoxygenation results in both cellular necrosis and apoptosis in cultured adult rat myocytes along with a decrease in cellular bcl-2 expression and an increase in Fas level. Aggregated platelet supernatant and TGF-₁ demonstrate strong protective effect against hypoxia-reoxygenation-induced apoptosis in myocytes, and TGF-₁ also significantly protected myocytes from hypoxia-reoxygenation-induced necrosis in myocytes. If these observations are reproduced in other laboratories, therapy with TGF-₁ and other growth factors may be considered in the treatment of reperfusion injury.