Cardioprotective Effects of Transforming Growth Factor-β1 During Early Reoxygenation or Reperfusion Are Mediated by p42/p44 MAPK

Abstract

Apoptosis contributes to myocardial cell death during ischemia and reperfusion, especially during reperfusion. Growth factor “survival” signaling attenuates apoptosis. We therefore examined the effects of transforming growth factor-β1 (TGF-β1) on reperfusion injury and assessed the role of p42/p44 MAPK signaling in TGF-β1–induced protection. Rat ventricular myocytes were subjected to hypoxia and reoxygenation. TGF-β1 (0.2 ng/ml) was applied to cells during reoxygenation and the extent of apoptosis was determined by TUNEL and annexin V binding assays. Further studies were conducted in intact rat hearts subjected to regional ischemia and reperfusion. TGF-β1 (0.2 ng/ml) was perfused during early reperfusion. In cells, incubation with TGF-β1 (0.2 ng/ml) during reoxygenation attenuated the extent of cell membrane damage (trypan blue uptake) and also reduced the numbers of TUNEL-and annexin V–positive cells. Reduction of apoptosis was abrogated by PD98059 (5 μM), an inhibitor of p42/p44 MAPK activation. TGF-β1 activated p42/p44 MAPK transiently in normoxic myocytes. When intact hearts received TGF-β1 (0.2 ng/ml) during early reperfusion, infarct size was reduced from 39.4 ± 3.1% to 17.3 ± 3.1% (p < 0.01). This protective action of TGF-β1 was abrogated by PD98059. These studies are the first to show that TGF-β attenuates cardiac myocyte apoptosis during early reperfusion and limits infarct size through p42/p44 MAPK activation.

Prompt restoration of blood flow to ischemic myocardium is necessary to limit infarct size, but some cells that survive ischemia may be lethally injured by the process of reperfusion ^(1–3). Cardiac myocytes at risk for lethal reperfusion injury represent a target for therapeutic intervention because their preservation could contribute to tissue salvage. The pathologic basis of irreversible cell injury during reperfusion is not clear, but there is evidence that apoptosis occurs in myocardium during ischemia-reperfusion in animal models and in human myocardial infarction ^(4–11). Although apoptosis may be initiated during ischemia ⁽⁵⁾, the process may be particularly relevant and extensive during reperfusion ^(4,12).

Many peptide growth factors modulate apoptosis and several, including transforming growth factor-β1 (TGF-β1), may exert either pro-apoptotic or anti-apoptotic actions according to the target-cell type and the pathophysiologic milieu ^(13–18). There is persuasive experimental evidence that TGF-β1 protects against myocardial ischemia-reperfusion injury when administered exogenously. Previous studies showed that TGF-β1 limited ischemia-reperfusion injury in the rat heart ex vivo and in vivo ⁽¹⁹⁾ and in feline myocardium in vivo ⁽²⁰⁾. These studies suggested that limitation of infarction by TGF-β1 was a consequence of endothelial preservation and neutrophil inhibition. However, we have observed that TGF-β1 protects against infarction when administered before the onset of coronary occlusion in a buffer-perfused rat heart preparation, an action that is neutrophil-independent (unpublished observations). Furthermore, activation of p42/p44 mitogen activated protein kinase (p42/p44 MAPK, also known as extracellular signal-related kinase, ERK) has been identified as a major anti-apoptotic “survival” signaling pathway for several peptide growth factors, including TGF-β1 ^(18,21). We hypothesized that exogenous TGF-β1 exerts a direct cytoprotective action mediated by activation of the p42/p44 MAPK signal transduction pathway in myocytes. We examined this hypothesis in isolated ventricular myocytes subjected to simulated ischemia and reoxygenation and in intact Langendorff-perfused heart subjected to coronary artery occlusion and reperfusion.

METHODS

Work was conducted in accordance with the Guidelines on the Operation of the Animals (Scientific Procedures) Act 1986 published by The Stationery Office (London, U.K.). Human recombinant TGF-β1 was from Calbiochem (Nottingham, U.K.). PD98059 was from Alexis Corp. (Nottingham, U.K.).

Simulated ischemia-reoxygenation in isolated myocytes

Ventricular myocytes were isolated from neonatal (2-day-old) Sprague–Dawley rats ⁽²²⁾. Myocytes were transferred to 24-well gelatin-coated plates. After 24 h, the cell medium was replenished with Dulbecco's modified Eagle's medium supplemented with 1% fetal calf serum for an additional 24 h. Confluent monolayers of spontaneously beating myocytes developed within 2 days. At this stage, the cell preparation had approximately 95% myocyte purity as assessed by anti-desmin immunostaining. Fewer than 5% of cells took up trypan blue. Thus the cell preparation was considered to be pure and viable. The growth medium was replaced with modified Esumi's ischemic buffer containing (in m M): NaCl, 137; KCl, 12; MgCl ₂ , 0.5; CaCl ₂ , 0.9; HEPES, 4; deoxyglucose, 10; and sodium lactate, 20 (pH 6.2). Cells were incubated in a hypoxic chamber at 37°C with humidified 5% CO ₂ and 95% Ar for either 2 h or 6 h. They were then re-oxygenated for 2 h in humidified 21% O ₂ and 5% CO ₂ at 37°C with 1 ml minimal medium, with or without TGF-β1, 0.2 ng/ml, and PD98059, 5 μM. In appropriate experimental groups, TGF-β1 or PD98059 or both were present during the 2 h re-oxygenation period.

Myocyte viability and apoptosis assays

The cells were washed with phosphate-buffered saline (PBS), trypsinized for 2 min in 0.25 mg/ml trypsin in Versene (GIBCO-BRL), and then neutralized with fetal calf serum. Cells were centrifuged (1,000 g for 10 min), the supernatant was aspirated, and the myocytes were re-suspended in 0.1 ml PBS. An equal volume of 0.4% trypan blue in PBS was added and the cells were counted in a hemocytometer under light microscopy. Cells that were without membrane disruption did not take up trypan blue and appeared white under the microscope. Not less than 250 cells were counted, including blue and non-blue cells. The number of blue cells was expressed as a percentage of the total cells counted. This determination was made in triplicate for each well of plated cells.

Apoptosis was assessed by a modified TdT-mediated dUTP-biotin nick end labeling (TUNEL) method and by annexin V labeling ⁽²²⁾. For TUNEL, cardiac myocytes were fixed in 4% paraformaldehyde and then washed with PBS. Terminal deoxynucleotidyl transferase solution, containing 2 m M fluorescein-conjugated dUTP and 10 U terminal deoxynucleotidyl transferase (Boehringer Mannheim, Mannheim, Germany) was added to the cells for 2 h in a 37°C humidified incubator. Washed cells were imaged with fluorescent microscopy and the percentage of TUNEL-positive cells was determined under phase as a percentage of the total cells counted. A minimum of 250 cells was scored in triplicate for each well.

For annexin V labeling, the myocytes were exposed to 2-h simulated ischemia and 2-h reoxygenation as described above. Cells were washed with PBS. To each of the wells was added 0.1 ml annexin V-FITC solution (Alexis, Nottingham, U.K.), containing 10 μg/ml fluorescein isothiocyanate-labeled annexin V (10 μl) diluted in HEPES binding buffer supplemented with 25 m M CaCl ₂ (90 μl). Cells were incubated at room temperature, without light, for 30 min. After incubation with the label, the cells were washed, fixed with 1% paraformaldehyde, and then washed again. Cells were imaged with fluorescent microscopy and the percentage of cells positive for annexin V was expressed as a percentage of total counted cells. A minimum of 250 cells was scored in triplicate for each well.

p42/p44 MAPK activation

Cardiac myocytes were cultured in six-well culture dishes in serum-free medium for 24 h in a humidified normoxic environment. They were incubated with TGF-β1, 0.2 ng/ml, or TGF-β1, 0.2 ng/ml, plus PD98059, 5 μM, for 1–60 min. Control cells had no treatment. Cells were harvested in a lysis buffer and proteins were extracted and separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis as described previously ⁽²²⁾. After transfer to nitrocellulose membranes, proteins were probed with a dual phosphospecific mouse monoclonal antibody to p42/p44 MAPK (1:1000 dilution) that reacts with the phosphorylated tyrosine-204 residue of p42/p44 MAPK (SC 7383; Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A). Nonphosphospecific antibody was used to determine equal protein content.

Infarct size study

Acute myocardial infarction was induced in isolated buffer-perfused rat hearts as described previously ⁽²³⁾. Hearts were excised under pentobarbital anesthesia from male Sprague–Dawley rats (Harlan Olac, Bicester, U.K.) weighing 300–350 g and were Langendorff-perfused at 1,000 mm H ₂ O pressure with modified Krebs–Henseleit buffer. Intracardiac temperature was maintained at 36.5–37.5°C throughout the perfusion protocol. A left ventricular balloon was inflated to a preload of 8–10 mm Hg. Regional ischemia was induced for 35 min by occlusion of the left main coronary artery followed by reperfusion for 120 min. Negative staining with zinc-cadmium sulfide fluorescent particles, 1–10 μm (Duke Scientific, Palo Alto, CA, U.S.A.), was used to delineate the ischemic risk zone. Triphenyltetrazolium chloride staining of heart slices was used to identify the zone of infarction. Infarction within the risk zone was drawn and the percentage calculated using computerized planimetry (Summa Sketch II; Summagraphics, Seymour, CT, U.S.A.). Hearts were assigned to the following treatment protocols. Group 1: After stabilization, control hearts underwent 35 min regional ischemia followed by 120 min reperfusion. Group 2: Buffer containing TGF-β1, 0.2 ng/ml, was perfused from 30 min ischemia until 15 min reperfusion. Thereafter, the perfusate was normal Krebs–Henseleit buffer. Group 3: The selective MEK inhibitor PD98059, 5 μM, was perfused from 30 min ischemia until 15 min reperfusion. Group 4: TGF-β1, 0.2 ng/ml, and PD98059, 5 μM, were co-perfused from 30 min ischemia until 15 min reperfusion.

Statistical analysis

All values are expressed as mean ± SEM. Differences in continuously distributed variables between predetermined experimental groups were analyzed using one-way ANOVA followed by Fisher's protected test of least significant difference. P values of 0.05 were considered to be at the limit of statistical significance.

RESULTS

TGF-β1 at reoxygenation attenuates cardiac myocyte injury and apoptosis

Trypan blue uptake is an index of plasma membrane disruption, a primary feature of oncosis. Figure 1 summarizes trypan blue uptake. Under conditions of normoxic incubation over 8 h, trypan blue uptake was 14.2 ± 2.7%. After 6 h of hypoxia and 2 h reoxygenation, trypan blue uptake in control cells was 57.3 ± 2.3%. Incubation with TGF-β1, 0.2 ng/ml, at reoxygenation significantly reduced this value, with a 24% relative reduction of trypan blue uptake (p < 0.01). Incubation of cells with PD98059, 5 μM, during reoxygenation resulted in a nonsignificant increase in trypan blue uptake but coincubation of TGF-β1 and PD98059 was associated with a statistically significant 20% relative reduction in trypan blue uptake (52.3 ± 3.8% vs. 65.4 ± 4.4% in PD98059 alone group, p < 0.05).

FIG. 1.: Trypan blue uptake in isolated cardiac myocytes subjected to 6-h simulated ischemia and 2-h reoxygenation. Transforming growth factor-β1 (TGF-β1) included in the reoxygenation medium reduced the uptake of trypan blue. Inclusion of PD98059 (PD) in the medium caused a modest increase in trypan blue uptake but co-incubation with TGF-β1 resulted in significant limitation of trypan blue uptake compared with that in the PD98059 group. Data are means ± SEM of 11–30 separate determinations. *p < 0.05; †0.05 > p < 0.1 (one-way ANOVA).

In the second series of isolated cell experiments, apoptosis was assessed by TUNEL (Fig. 2). During normoxic incubation for 8 h, the percentage of cells with TUNEL-positive nuclei was 7.6 ± 2.6%. After 6-h hypoxia and 2-h reoxygenation, the number of TUNEL-positive nuclei under control conditions was 55.0 ± 2.5%. This value was significantly reduced to 35.5 ± 3.2% in cells incubated with TGF-β1 during reoxygenation. PD98059 during reoxygenation did not alter the percentage of TUNEL-positive nuclei but abolished the reduction associated with TGF-β1 treatment (47.2 ± 3.9% in TGF-β1 plus PD98059 group vs. 35.5 ± 3.2% in TGF-β1 group, p < 0.05). This observation suggests that TGF-β1 at reoxygenation reduced the extent of apoptosis via p42/p44 MAPK activation.

FIG. 2.: TUNEL positivity in isolated cardiac myocytes subjected to 6-h simulated ischemia and 2-h reoxygenation. Transforming growth factor-β1 (TGF-β1) at reoxygenation reduced the percentage of cells with TUNEL-positive nuclei. Co-incubation with PD98059 (PD) abrogated the reduction in TUNEL positivity. Mean ± SEM of 6–27 separate determinations. *p < 0.01 (one-way ANOVA).

In further experiments, annexin V labeling of cells was used as an additional marker of apoptosis (Fig. 3). During apoptosis, phosphatidylserine moieties are externalized and annexin V can bind to these groups. In these experiments, we used a shorter period of hypoxia (2 h), insufficient to cause cell disruption because, in the presence of membrane damage, annexin V can enter cells and bind on the inner surface of the cell membranes ⁽³⁵⁾. After 2-h simulated ischemia and 2-h reoxygenation, annexin V binding is a sensitive marker of an early event in apoptosis. TGF-β1 during reoxygenation significantly reduced the extent of annexin V binding from 56.1 ± 2.9% in control cells to 36.0 ± 2.0% in TGF-β1–treated cells (p < 0.01). PD98059 during reoxygenation had no effect on the extent of annexin V binding per se (54.1 ± 7.1%,). However, PD98059 almost completely reversed the reduction in annexin V binding conferred by TGF-β1 during reoxygenation (50.1 ± 7.3% vs. 36.0 ± 2.0% in TGF-β1–treated cells, p < 0.05). This observation provides further support that TGF-β1 at reoxygenation reduced the extent of apoptosis via a p42/p44 MAPK-dependent pathway.

FIG. 3.: Annexin V binding in isolated cardiac myocytes subjected to 2-h simulated ischemia and 2-h reoxygenation. Transforming growth factor-β1 (TGF-β1) included in the reoxygenation medium reduced the percentage of cells with annexin V binding. Co-incubation with PD98059 (PD) abrogated the reduction in annexin V positivity observed with TGF-β1. Mean ± SEM of 7–25 separate determinations. *p < 0.01 (one-way ANOVA).

TGF-β1 activates p42/p44 MAPK in cardiac myocytes

Incubation with TGF-β1, 0.2 ng/ml, caused a transient increase in the phosphorylated form of p42 MAPK above basal levels, which was maximal after 5 min (Fig. 4). Thereafter, there was a gradual decline in the activity of both kinases and after 60-min incubation with TGF-β1 there was no detectable phosphorylated p42/p44 MAPK. Application of PD98059, 5 μM, resulted in loss of phosphorylation under basal conditions and completely prevented the pattern of phosphorylation in response to TGF-β1. These findings provide confirmation that TGF-β1, at the concentration shown to be cytoprotective in isolated cells and intact cardiac tissue, caused rapid and transient activation of p42/p44 MAPK and that PD98059 prevented this activation.

FIG. 4.: Representative immunoblots showing activation of p42/p44 MAPK by transforming growth factor-β1 (TGF-β1) in normoxic myocytes. A. Immunoblots probed with dual phosphospecific antibody to phosphorylated p42/p44 MAPK. On the left, incubation of myocytes with TGF-β1 caused rapid phosphorylation of p42/p44 MAPK, maximal at 5 min with subsequent decline in activity below basal level at 30 and 60 min. On the right, incubation of myocytes with PD98059 decreased basal p42/p44 MAPK activity (time 0 min) and prevented activation when co-incubated with TGF-β1. B. Immunoblots probed with nonphosphospecific antibody to p42/p44 MAPK.

TGF-β1 at reperfusion limits infarct size via a PD98059-sensitive mechanism

To extend the studies undertaken in isolated myocytes, we examined the ability of TGF-β1 to modify infarct size in intact hearts when treatment was commenced 30 min after coronary occlusion (5 min before reperfusion) and continued until 15 min after reperfusion. Neither TGF-β1 nor PD98059 exerted any effects on cardiac contractility or coronary flow. Infarct size data are shown in Figure 5. In control hearts, 35-min coronary artery occlusion and 120-min reperfusion resulted in an infarct-to-risk ratio of 39.4 ± 2.0%. Perfusion with TGF-β1 during early reperfusion caused significant limitation of infarction (infarct-to-risk ratio: 17.3 ± 3.1%, p < 0.01 vs. control). Coadministration of PD98059 with TGF-β1 resulted in abolition of the infarct-limiting effect of TGF-β1 (infarct-to-risk ratio: 34.3 ± 5.5%, p < 0.05 vs. TGF-β1 group). Treatment with PD98059 alone during the same period had no effect on infarct size (infarct-to-risk ratio: 34.8 ± 5.7%).

FIG. 5.: Infarct size in isolated rat hearts subjected to regional ischemia and reperfusion. Transforming growth factor-β1 (TGF-β1) at reperfusion significantly limited infarct size. PD98059 (PD), 5 μM, abolished the protective effects of TGF-β1. Mean ± SEM. *p < 0.01 (one-way ANOVA).

DISCUSSION

There are three principal findings of these studies. First, a protective action of TGF-β1 during reoxygenation was seen in cardiac myocytes using trypan blue dye exclusion and two molecular markers of apoptosis. Second, the anti-apoptotic effect in isolated cardiac myocytes was attenuated by coadministration of PD9059. And third, in isolated hearts, TGF-β1 limited tetrazolium-assessed infarct size when administered during the early reperfusion period and this effect was abolished by co-administration of PD98059, suggesting a key role of p42/p44 MAPK activation. Together, these findings indicate that lethal reperfusion/reoxygenation injury occurs in cardiac myocytes and in intact myocardium and that TGF-β1 exerts a primary cytoprotective action in cardiac tissue that is related to p42/p44 MAPK activation.

Considerable evidence has accumulated that apoptosis contributes to myocardial cell loss during ischemia and reperfusion ^(4–12). Apoptosis occurs more rapidly in tissue subjected to reperfusion than in tissue subjected to ischemia alone ⁽⁴⁾, but the relation between necrosis and apoptosis in intact tissue is often unclear ^(24,25). Because of technical difficulties in assessing infarct size and apoptosis in the same intact heart preparation, we used a surrogate cell model in which apoptosis could be more precisely studied. These studies in isolated cardiac myocytes point toward two mechanisms of protection by TGF-β1. The first mechanism is related to protection against trypan blue uptake, a marker of cell membrane disruption. Under conditions of ischemia-reperfusion, plasma membrane disruption may represent the onset of irreversible cell damage and is a morphologic feature of cells undergoing necrosis. Cells with plasma membrane disruption take up trypan blue very readily. TGF-β1 at reoxygenation conferred modest but significant reduction in trypan blue uptake and this was attenuated by PD98059. If we note that PD98059 alone caused an increase in trypan blue uptake, one interpretation of these data is that the protective action of TGF-β1 against cell membrane disruption was still present under pharmacologic inhibition of p42/p44 MAPK (Fig. 1). This suggests that the modest anti-oncotic effect of TGF-β1 at reoxygenation was independent of p42/p44 MAPK activation. However, compared with the control group, the protective effect of TGF-β1 was attenuated by PD98059. Thus TGF-β1 might protect against oncosis by a mechanism that is MAPK-independent or only partially MAPK-dependent, although this requires further investigation.

The second mechanism through which TGF-β1 may protect at reperfusion is inhibition of apoptosis. In reoxygenated myocytes, TGF-β1 effected reductions in two distinct markers of apoptosis. TUNEL detects internucleosomal DNA fragmentation, a late event in apoptosis. Annexin V binds to phosphatidylserine expressed on the outer leaflet of the plasma membrane, an early event after the onset of the apoptotic program ⁽²⁴⁾. The anti-apoptotic action of TGF-β1 at reoxygenation, assessed by either technique, was abrogated by cotreatment with PD98059, implicating activation of p42/p44 MAPK in the anti-apoptotic mechanism. We confirmed that TGF-β1 causes rapid phosphorylation of p42/p44 MAPK and that PD98059 abolishes this activation. However, p38 MAPK and JNK were not phosphorylated by TGF-β1, 0.2 or 2.0 ng/ml, under the conditions examined here (data not presented). Although kinases other than p/42/p44 MAPK may play roles in determining cellular responses to ischemia-reperfusion, we found no evidence that p38 MAPK or JNK were regulated by TGF-β1 under normoxic conditions.

We are cautious in extrapolating findings in isolated cells to intact myocardium or acute infarction in vivo. The cell preparation studied here is a surrogate model of ischemia-reperfusion. Both the extent of cell membrane disruption and apoptosis and the time courses of these events are probably highly artificial and do not permit ready extrapolation to the situation in vivo. There is presently no consensus about the contribution of apoptotic cell death to acute myocardial infarction in vivo, but the occurrence of TUNEL or annexin V positivity is unlikely to approach the extent observed in isolated cells. The time course of onset and execution of apoptosis in the reperfused myocardium in vivo is unclear, because the techniques most widely used at present provide only a snapshot of the extent of apoptosis-associated events. It is noteworthy that in the intact-heart infarct model described here, the infarct-limiting effect of TGF-β1 at reperfusion was reversed by PD98059. It is likely that tetrazolium staining in whole tissue does not distinguish between the two forms of cell death. At present, we can not comment on the fate of tissue in which early salvage with TGF-β1 was seen. Although previous studies have validated the technique of early macrochemical assessment of infarct size against late histologic assessment of infarct size in experimental infarction, further studies are warranted to confirm that the limitation of infarction seen with TGF-β1 in the present studies is sustained. It is possible that TGF-β1 merely delays or suspends the processes of necrosis and apoptosis for a period. It is also possible that secondary necrosis may occur many hours after reperfusion in tissue in which apoptosis has been prevented. In vivo studies are required to explore these issues further.

There are numerous reports of dual regulation of apoptosis by TGF-β1 in various cultured cell types. These include both pro-apoptotic effects of TGF-β1 ^(13–17) and anti-apoptotic effects ^(14,18). The present studies are the first to describe an anti-apoptotic effect of TGF-β1 in cardiac myocytes. Other growth factors exert anti-apoptotic effects in cardiac myocytes, including insulin-like growth factor-1 (IGF-1) ⁽²⁶⁾, insulin ⁽²⁷⁾, cardiotrophin-1 ^(28,29), and neuregulin-1 ⁽³⁰⁾, and several have been shown to limit acute ischemia-reperfusion injury in intact hearts when administered before coronary occlusion ^(31–33). In previous studies of the cardioprotective effects of TGF-β1, the beneficial effects of the peptide when administered before coronary occlusion or after the onset of coronary occlusion were considered to be due to its ability to promote the release of nitric oxide, thereby inhibiting neutrophil adhesion ^(19,20). The present studies show that cardioprotection by TGF-β1 occurs independently of any actions on blood-borne cells. Furthermore, the timing of TGF-β1 administration is more precisely related to reperfusion than in earlier work. In a previous study ⁽²⁰⁾ TGF-β1 was administered to cats 30 min before reperfusion. Because the feline heart has a significant native coronary collateral circulation, it is conceivable that TGF-β1 could modify the development of tissue injury during ischemia. The present studies demonstrate an action specifically during reperfusion/reoxygenation.

Although there are variations in different experimental systems, the anti-apoptotic actions of peptide growth factors appear to be related to activation of p42/p44 MAPK and phosphatidyl-inositol 3-kinase, which leads to activation of Akt (also known as protein kinase B). There is limited evidence to suggest that the phosphatidyl-inositol 3-kinase/Akt pathway influences p42/p44 MAPK activation, but the extent of “crosstalk” between these two pathways in cardiac myocytes is unclear. The distal molecular mechanisms by which p42/p44 MAPK and phosphatidyl-inositol 3-kinase/Akt inhibit apoptosis are as yet unknown. There is accumulating evidence that p42/p44 MAPK, alone or in conjunction with Akt, regulates various aspects of the apoptotic cascade. For example, in some (noncardiovascular) cell types, inhibition of caspase-3 activation follows p42/p44 MAPK activation ^(34,35). Phosphorylation and inhibition of the Bcl-2 family member, Bad, also occurs after p42/p44 MAPK stimulation ^(36,37) and may be an important pro-survival mechanism. However, the precise distal effector mechanisms relating to cardioprotection by TGF-β1 and other growth factors remain to be determined.

In conclusion, these data point to a primary cytoprotective action of TGF-β1 against the deleterious effects of reperfusion and highlight the critical role of an anti-apoptotic MAPK-dependent pathway. We conclude that manipulation of growth factor “survival” signaling mechanisms may provide a promising route to attenuate lethal reperfusion injury.

Acknowledgment:

This work was supported by British Heart Foundation Program grant (RG/99002) and by a British Heart Foundation personal fellowship (FS/97001) awarded to GFB. MMM was supported by a Royal Society/NATO fellowship. The authors thank the Hatter Foundation for continued support.