The protein kinase C pathway mediates cardioprotection induced by cardiac-specific overexpression of fibroblast growth factor-2

Stacey L. House, Susan J. Melhorn, Gilbert Newman, Thomas Doetschman, Jo El J. Schultz

Abstract

Elucidation of protective mechanisms against ischemia-reperfusion injury is vital to the advancement of therapeutics for ischemic heart disease. Our laboratory has previously shown that cardiac-specific overexpression of fibroblast growth factor-2 (FGF2) results in increased recovery of contractile function and decreased infarct size following ischemia-reperfusion injury and has established a role for the mitogen-activated protein kinase (MAPK) signaling cascade in the cardioprotective effect of FGF2. We now show an additional role for the protein kinase C (PKC) signaling cascade in the mediation of FGF2-induced cardioprotection. Overexpression of FGF2 (FGF2 Tg) in the heart resulted in decreased translocation of PKC-δ but had no effect on PKC-α, -ε, or -ζ. In addition, multiple alterations in PKC isoform translocation occur during ischemia-reperfusion injury in FGF2 Tg hearts as assessed by Western blot analysis and confocal immunofluorescent microscopy. Treatment of FGF2 Tg and nontransgenic (NTg) hearts with the PKC inhibitor bisindolylmaleimide (1 μmol/l) revealed the necessity of PKC signaling for FGF2-induced reduction of contractile dysfunction and myocardial infarct size following ischemia-reperfusion injury. Western blot analysis of FGF2 Tg and NTg hearts subjected to ischemia-reperfusion injury in the presence of a PKC pathway inhibitor (bisindolylmaleimide, 1 μmol/l), an mitogen/extracellular signal-regulated kinase/extracellular signal-regulated kinase (MEK/ERK) pathway inhibitor (U-0126, 2.5 μmol/l), or a p38 pathway inhibitor (SB-203580, 2 μmol/l) revealed a complicated signaling network between the PKC and MAPK signaling cascades that may participate in FGF2-induced cardioprotection. Together, these data suggest that FGF2-induced cardioprotection is mediated via a PKC-dependent pathway and that the PKC and MAPK signaling cascades are integrally connected downstream of FGF2.

  • ischemia-reperfusion injury
  • cardiac dysfunction
  • myocardial infarction
  • kinase signaling cross talk

the elucidation of cardioprotective mechanisms may lead to a reduction of the substantial morbidity and mortality experienced as a result of cardiac ischemia-reperfusion injury in the United States. Many different cardioprotective strategies have been discovered in animal models, including growth factors, opioids, ion channel inhibitors, and signaling modulators. Our laboratory has previously shown in a mouse model of global low-flow ischemia-reperfusion injury that overexpression of fibroblast growth factor-2 (FGF2) specifically in the heart results in reduced postischemic contractile dysfunction and myocardial infarction (24). Understanding the signaling mechanisms that mediate FGF2's cardioprotective effect is necessary for the development of FGF2 as a potential therapeutic agent. Recently, our laboratory demonstrated that the MAPK signaling cascade was essential to FGF2-induced cardioprotection (25), but other signaling pathways may also play a role in the mediation of FGF2's cardioprotective effect.

The protein kinase C (PKC) signaling cascade has been implicated to play a role in ischemia-reperfusion injury by several different studies. Ischemic preconditioning is abrogated by multiple PKC inhibitors, including staurosporine, polymixin B, and chelerythrine (33, 40, 70). Activation of PKC signaling with phorbol esters or diacylglycerol-like agents also provides cardioprotection in some studies (64, 70) but aggravates myocardial damage in others (6, 27). These contradictory reports may be a result of using pharmacological agents with nonselective actions or the different biological roles of specific PKC isoforms.

Information concerning the role of specific PKC isoforms in ischemia-reperfusion injury has provided some insight into this controversy. Ischemic preconditioning has been shown to induce translocation of PKC-δ, -ε, and -ν (47, 50, 55). PKC-ε transgenic mice have a cardioprotective phenotype (12), and PKC-ε knockout mice are incapable of being preconditioned by brief periods of ischemia (60). PKC-δ has also been shown to mediate cardioprotection provided by ischemia (31, 71) or opioids (19) and to provide protection to cardiomyocytes in vitro (71). However, administration of a PKC-δ inhibitor during reperfusion prevents reperfusion injury in isolated hearts (28), illustrating the controversy that exists concerning the role of specific PKC isoforms in cardioprotection. Interestingly, the use of a PKC-δ inhibitor along with a PKC-ε translocation agonist results in additive cardioprotective effects, suggesting multiple PKC isoforms may work in coordination to provide protection from ischemia-reperfusion injury (28).

This study set forth to determine if the PKC signaling pathway contributes to the cardioprotective effect of cardiac-specific overexpression of FGF2. Our results demonstrate PKC pathway activation is critical to FGF2-mediated cardioprotection from postischemic contractile dysfunction and myocardial infarction. Overexpression of FGF2 (FGF2 Tg) in the heart results in inhibition of PKC-δ, which may contribute to this cardioprotection. In addition, alterations in PKC isoform activation and subcellular localization occur in FGF2 Tg hearts throughout exposure to ischemia and reperfusion injury. This study also demonstrates a complicated signaling network between the PKC and mitogen-activated protein kinase (MAPK) signaling cascades that mediates FGF2-induced cardioprotection. These results further identify cardioprotective signaling mechanisms downstream of FGF2 that provide essential information for the development of FGF2 as a therapeutic target for coronary artery disease and ischemia-reperfusion injury.

METHODS

Animals.

Generation of mice with a cardiac-specific overexpression of FGF2 has been previously described (24). Mice were housed in a pathogen-free facility and handled in accordance with standard use protocols, animal welfare regulations, and the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All protocols were approved by the University of Cincinnati Institutional Animal Care and Use Committee. Nontransgenic (NTg) mice and mice with a cardiac-specific overexpression of all four isoforms of human FGF2 (FGF2 Tg) were randomly assigned to the present study [60 min ischemia, 120 min reperfusion studies (21 NTg and 32 FGF2 Tg), PKC levels in nonischemic hearts (8 NTg and 16 FGF2 Tg), time course studies (36 NTg and 64 FGF2 Tg), immunofluorescence (16 NTg and 32 FGF2 Tg), and PKC-MAPK cross talk studies (64 NTg and 74 FGF2 Tg)]. Exclusion criteria were based on signs of aortic or pulmonary leaks in the work-performing heart preparation. Two independently derived FGF2 Tg lines were used in all experiments to ensure that results were not because of random transgene insertion affecting other genetic loci. Because similar results were obtained in both FGF2 Tg lines, all figures depict combined data from both transgenic lines.

Isolated work-performing heart model of global low-flow ischemia.

Age-(10–12 wk) and sex-matched NTg mice and two FGF2 Tg lines with a cardiac-specific overexpression of FGF2 were anesthetized with pentobarbital sodium (80 mg/kg ip), and the heart was removed from the thoracic cavity. The heart was then perfused in an isolated work-performing heart preparation and subjected to global ischemia-reperfusion injury as previously described (24). Two ischemia protocols were used following 30 min of baseline equilibration time: either 30 min low-flow ischemia followed by 30 min of reperfusion or 60 min low-flow ischemia followed by 120 min reperfusion (Fig. 1).

Fig. 1.

Schematic of low-flow ischemia protocols. A: time course protocol for protein kinase C (PKC) activation during ischemia-reperfusion injury: 30 min equilibration followed by 30 min global low-flow ischemia and 30 min reperfusion. Arrows indicate time when hearts were arrested and homogenized for Western immunoblotting. B: drug infusion protocol for PKC inhibition studies: 30 min equilibration followed by 60 min global low-flow ischemia and 120 min reperfusion. Pharmacological agents were administered for 15 min before and after the beginning of ischemia and 15 min before and after the beginning of reperfusion. C: protocol for mitogen-activated protein kinase (MAPK) and PKC cross talk analysis. Pharmacological agents were administered as described above, but hearts were arrested following 30 min equilibration and 5 min ischemia or 30 min equilibration, 60 min global low-flow ischemia, and 5 min reperfusion.

Inhibition of the PKC pathway.

Bisindolylmaleimide (1 μmol/l; Calbiochem) was used to provide inhibition of the PKC pathway in a model of irreversible ischemia-reperfusion injury (60 min global low-flow ischemia and 120 min reperfusion). Studies using varying concentrations (1–10 μmol/l) of bisindolylmaleimide in the ischemia-reperfusion model were performed, and a concentration of 1 μmol/l was chosen because this concentration inhibited PKC activation without any adverse effects (i.e., cardiac arrhythmias). Bisindolylmaleimide was administered for the last 15 min of equilibration and the first 15 min of ischemia and again for the last 15 min of ischemia and the first 15 min of reperfusion (Fig. 1B), consistent with previously reported pharmacological studies in this ischemia-reperfusion model (25).

Measurement of infarct size.

Bisindolylmaleimide-treated and vehicle-treated NTg and FGF2 Tg hearts were perfused with warmed (37°C) 1% 2,3,5-triphenyltetrazolium chloride (Sigma Aldrich; see Ref. 34) following the 60 min ischemia, 120 min reperfusion study to distinguish viable vs. necrotic tissue as previously described (24).

Western blot analysis of PKC isoform translocation.

Nonischemic NTg and FGF2 Tg hearts were analyzed to determine the translocation state of PKC isoforms in response to cardiac-specific overexpression of FGF2. In addition, NTg and FGF2 Tg hearts subjected to various time points of ischemia-reperfusion injury were analyzed to determine the timing of alterations in PKC isoform translocation during ischemia-reperfusion injury. Time points studied included Sham, 5 min ischemia, 15 min ischemia, 30 min ischemia, and 30 min of ischemia followed by 5, 15, or 30 min of reperfusion (Fig. 1A). To determine the level of translocation of PKC isoforms, NTg and FGF2 Tg hearts that were nonischemic or subjected to various time points of ischemia-reperfusion injury were snap-frozen in liquid nitrogen. Hearts were then homogenized in buffer containing 25 mmol/l Tris, 4 mmol/l EGTA, 2 mmol/l EDTA, 5 mmol/l dithiothreitol, phenylmethylsulfonyl fluoride (PMSF), and protease inhibitor cocktail (Roche). Samples were then centrifuged at 100,000 g at 4°C for 30 min. The supernatants contained the cytosolic fraction, and the pellets were then homogenized in the above buffer with 1% Triton added. This homogenate was then centrifuged at 100,000 g at 4°C for 30 min, and the supernatant was considered the total membrane fraction. Each fraction (50 μg) was subjected to SDS-PAGE, and Western blot analysis using antibodies (1:1,000 dilution; Santa Cruz Biotechnologies) specific to PKC-α, -δ, -ε, and -ζ was performed. The level of translocation is expressed as the ratio of the amount of a specific PKC isoform found in the membrane fraction to that found in the cytosolic fraction.

Confocal immunofluorescent microscopy.

FGF2 Tg and NTg hearts that were subjected to sham treatment, 5 min ischemia, or 30 min ischemia and 5 min reperfusion were embedded and frozen in Tissue-Tek optimum cutting temperature compound. Hearts were then sectioned transversely (5–10 μm) and stored at −80°C. Sections were fixed in 3.7% paraformaldehyde and permeabilized in 0.2% Triton X. Nonspecific protein binding was then blocked by incubation of the sections in 10% normal goat serum (Vector Laboratories) and 10% normal donkey serum (Vector Laboratories), and sections were probed with rabbit polyclonal antibodies specific for PKC-α (1:100 dilution), PKC-δ (1:50 dilution), PKC-ε (1:100 dilution), or PKC-ζ (1:50 dilution; Santa Cruz Biotechnologies). Sections counterstained for sarcoplasmic reticulum were also incubated in mouse monoclonal anti-ryanodine receptor antibody (1:100 dilution; Affinity Bioreagents) at this time. PKC staining was visualized using appropriate Cy5-conjugated secondary antibodies (1:500 dilution; Molecular Probes). Mitochondria, nuclei, and the actin cytoskeleton were then labeled in these heart sections using commercially available organellar stains from Molecular Probes [MitoTracker Green FM (MTG) 1:10,000 dilution, popo-3-iodide 1:1,000 dilution, and Alexxa 433 phalloidin 1:40 dilution, respectively]. To label mitochondria, heart sections were incubated with MTG, which passively diffuses across the plasma membrane, accumulating in mitochondria by covalently binding to free thiol groups on cysteine residues (22, 56, 58, 62). MTG dye can selectively stain mitochondria both in live cells and in cells that have been fixed (22). Sections were mounted in Vectashield mounting medium (Vector Laboratories). Fluorescence was visualized using a confocal microscope, and digital images were captured with Zeiss LSM Image software.

PKC-MAPK cross talk analysis.

To identify cross talk within the MAPK cascade or between the MAPK and PKC pathways, U-0126 (2.5 μmol/l), SB-203580 (2 μmol/l), or bisindolylmaleimide (1 μmol/l) were administered during ischemia-reperfusion injury, and drug- or vehicle-treated NTg or FGF2 Tg hearts were snap-frozen in liquid nitrogen after being subjected to 30 min equilibration followed by either 5 min ischemia or 60 min ischemia and 5 min reperfusion (Fig. 1C). One-half of the heart was then homogenized and subjected to SDS-PAGE and Western blot analysis to determine PKC isoform translocation as described in Western blot analysis of PKC isoform translocation.

The remaining one-half of each heart was then homogenized and subjected to SDS-PAGE and Western blot analysis for determination of extracellular signal-regulated kinase (ERK) and p38 phosphorylation as follows. Hearts were homogenized in buffer containing 25 mmol/l HEPES, 150 mmol/l NaCl, 1% Triton X-100, 5 mmol/l EDTA, 1% glycerol, 1 mmol/l sodium orthovanadate, 25 mmol/l β-glycerolphosphate, 50 mmol/l sodium fluoride, 0.5 μmol/l okadaic acid, 100 μmol/l calpain inhibitor, diisopropylfluorophosphate, Pefabloc Stock 1 and 2 (Roche), Sigma phosphatase inhibitor, Roche complete mini EDTA-free protease inhibitor cocktail, and PMSF. Homogenates were then centrifuged at 3,000 g to remove cell debris. Each supernatant (100 μg) was then subjected to SDS-PAGE, and Western blot analysis was performed using phosphospecific rabbit polyclonal ERK and p38 antibodies (1:1,000 dilution; Cell Signaling) or with antibodies to total ERK (1:1,000 dilution, mouse monoclonal; BD Transduction Laboratories) and total p38 (1:1,000 dilution, rabbit polyclonal; Santa Cruz Biotechnologies). Equal protein loading was assured by Commassie staining. Densitometry of protein bands was performed using a Fluorchem 8800 gel imager (Alpha Innotech).

Statistical analysis.

All values are expressed as means ± SE. Differences at various time points of percent functional recovery were compared using a two-way ANOVA for time and treatment with repeated measures followed by a Student's t-test. Myocardial infarct size was compared using a one-way ANOVA followed by a Student's t-test. Western immunoblotting data were compared using a Student's t-test. Statistical significance was determined by a P < 0.05.

RESULTS

Our laboratory has previously shown the cardioprotective effect of FGF2 overexpression to be mediated by the MAPK cascade (25), but FGF2 signals through other cellular pathways, making it possible that other factors might modulate FGF2-induced cardioprotection. The PKC pathway was the first signal transduction cascade identified for FGF2 (20). This pathway has also been shown to mediate cardioprotection in response to many different stimuli, including ischemic preconditioning (11), suggesting that it could potentially play a role in the cardioprotection elicited by FGF2. Multiple isoforms exist within the PKC family with several, including PKC-α, -β, -γ, -δ, -ε, -ζ, and -ν, localized to adult heart tissue (59). This study focused on the role of PKC-α, -δ, -ε, and -ζ in FGF2-induced cardioprotection, since these isoforms have been shown to be activated by FGF2 in other tissues or to have a significant role in cardioprotective signaling.

Effects of cardiac-specific overexpression of FGF2 on PKC isoform activation.

To elucidate whether cardiac-specific overexpression of FGF2 primes the PKC cascade, the expression level and activation state of PKC isoforms were examined in nonischemic FGF2 Tg and NTg hearts. Expression of PKC-ε was significantly decreased in nonischemic FGF2 Tg hearts compared with the NTg, whereas PKC-α, PKC-δ, and PKC-ζ remained unchanged in nonischemic FGF2 Tg hearts (data not shown). Activation of PKC isoforms involves both autophosphorylation and translocation to various membrane compartments where PKC isoforms are able to encounter and phosphorylate target substrates. Whenever possible, both translocation and phosphorylation of PKC isoforms have been studied to assess multiple aspects of PKC activation. There were no significant differences in the level of translocation of PKC-α, -ε, and -ζ in nonischemic FGF2 Tg hearts compared with NTg hearts. PKC-δ, however, showed significantly decreased translocation in FGF2 Tg hearts (P < 0.05, Fig. 2, A and B). Similarly, PKC-δ was significantly less phosphorylated in FGF2 Tg hearts, whereas there are no differences in the phosphorylation state of the other PKC isoforms in FGF2 Tg hearts (data not shown). These data suggest that cardiac-specific overexpression of FGF2 primes the heart through reduction in the activation state of PKC-δ.

Fig. 2.

PKC isoform translocation in nonischemic (A and B) and ischemic-reperfused (C–G) nontransgenic (NTg) mouse hearts and hearts of mice with a cardiac-specific overexpression of all four isoforms of human fibroblast growth factor-2 (FGF2 Tg). A and B: significantly decreased translocation of PKC-δ is seen in nonischemic FGF2 Tg hearts, but PKC-α, -ε, and -ζ translocation is unchanged. C–G: time course of PKC isoform translocation in FGF2 Tg and NTg hearts in response to ischemia-reperfusion injury. C and D: PKC-α shows significantly decreased translocation in FGF2 Tg hearts throughout ischemia and the beginning of reperfusion. C, E, and G: FGF2 Tg hearts have decreased translocation of PKC-δ and PKC-ζ during ischemia. C and F: PKC-ε shows increased translocation in FGF2 Tg hearts in response to reperfusion injury. Translocation is expressed as a ratio of the amount of a specific PKC isoform found in a membrane fraction vs. that found in the cytosolic fraction. I, ischemia; R, reperfusion; n = 4–10 hearts/group. *P < 0.05 vs. NTg at same time point.

Because PKC isoforms have been shown to be activated during ischemia-reperfusion injury (57, 66), time course experiments were performed in which the translocation and phosphorylation of PKC isoforms were assessed in FGF2 Tg and NTg hearts subjected to different time points of ischemia-reperfusion injury. FGF2 Tg hearts showed significantly decreased PKC-α translocation compared with NTg hearts during ischemia and early reperfusion injury (P < 0.05, Fig. 2, C and D). Similarly, the level of translocation of PKC-δ and -ζ was also significantly reduced in FGF2 Tg hearts during early ischemia (P < 0.05, Fig. 2, C, E, and G). PKC-ε showed significantly enhanced translocation in FGF2 Tg hearts during early reperfusion injury (P < 0.05, Fig. 2, C and F). Western blot analysis of the phosphorylation state of these PKC isoforms in FGF2 Tg and NTg hearts subjected to ischemia-reperfusion injury revealed similar results (data not shown). These data indicate modulation of PKC signaling is altered in response to cardiac-specific overexpression of FGF2 during both ischemia and reperfusion injury.

Activation of the PKC pathway mediates FGF2-induced cardioprotection from global low-flow ischemia-reperfusion injury.

To ascertain the necessity of PKC signaling for FGF2-induced cardioprotection, a pharmacological inhibition of the PKC pathway was employed. Bisindolylmaleimide, a nonselective PKC inhibitor that blocks the ATP-binding site of PKC-α, -β, -δ, -ε, and -ζ (45, 67), was administered to NTg and FGF2 Tg hearts subjected to irreversible ischemia-reperfusion injury. Inhibition of the PKC pathway with bisindolylmaleimide reduced postischemic contractile recovery in FGF2 Tg hearts but has no significant effect on the recovery of function in NTg hearts (P > 0.05, Fig. 3A), suggesting that PKC signaling is necessary for FGF2-induced cardioprotection from postischemic contractile dysfunction. Similarly, infarct size was significantly increased following irreversible ischemia-reperfusion injury in bisindolylmaleimide-treated FGF2 Tg hearts compared with vehicle-treated FGF2 Tg hearts (P < 0.05, Fig. 3B). Infarct size was unaffected by PKC pathway inhibition in NTg hearts. These data suggest that activation of PKC isoform(s) is necessary for both the preservation of cardiac contractile function and limitation of myocardial cell death provided by overexpression of FGF2 in the heart.

Fig. 3.

Inhibition of the PKC pathway with bisindolylmaleimide significantly abrogates FGF2-mediated cardioprotection. A: cardiac-specific overexpression of FGF2 significantly increases postischemic recovery of +dP/dt, a measure of systolic function. Inhibition of the PKC pathway abrogates this effect. %Recovery of +dP/dt is the dP/dt after 60 min ischemia/120 min reperfusion divided by baseline +dP/dt; n = 7–13 for each group. B: overexpression of FGF2 in the heart significantly decreases myocardial infarct size following irreversible ischemia-reperfusion injury. Inhibition of the PKC pathway blocks this protection. %Infarct size is the percent of myocardial cell death in the entire heart; n = 4–5 for each group. *P < 0.05 vs. NTg of the same treatment group. †P < 0.05 vs. vehicle-treated cohort.

Subcellular distribution of PKC isoforms in NTg and FGF2 Tg hearts subjected to ischemia-reperfusion injury.

Localization of PKC isoforms is thought to reflect substrate specificity and biological functions of the individual PKC isoforms. Confocal immunofluorescence studies were undertaken to identify the localization of PKC-α, -δ, -ε, and -ζ in NTg and FGF2 Tg hearts subjected to sham treatment, 5 min ischemia, and 30 min ischemia with 5 min of reperfusion. PKC-α shows primarily cytosolic and sarcolemmal localization, with perinuclear localization occurring in response to ischemia and reperfusion in FGF2 Tg hearts but not NTg hearts (Fig. 4A). Mitochondrial localization during ischemia and sarcoplasmic reticulum localization during reperfusion is also observed for PKC-α in FGF2 Tg hearts (Fig. 4, B and C). PKC-δ shows cytoplasmic and sarcolemmal localization as well in sham-treated hearts, with perinuclear staining appearing in early ischemia and early reperfusion injury for both FGF2 Tg and NTg hearts (data not shown). Sarcoplasmic reticulum localization of PKC-δ is also apparent for both FGF2 Tg and NTg hearts (data not shown). No mitochondrial localization of PKC-δ is observed in any of the groups or time points of ischemia-reperfusion injury (data not shown). PKC-ε is located in the cytosol, sarcolemmal, nuclear, and sarcoplasmic reticulum cellular compartments for both FGF2 Tg and NTg hearts (data not shown). Intercalated disks and mitochondria also show PKC-ε localization during reperfusion injury in FGF2 Tg hearts (Fig. 4, D and E). PKC-ζ is located in the cytosol, nuclei, mitochondria, and sarcoplasmic reticulum of both NTg and FGF2 Tg hearts during sham treatment and ischemia (data not shown); however, nuclear localization of PKC-ζ is absent in FGF2 Tg hearts during reperfusion (Fig. 4F).

Fig. 4.

Representative confocal immunofluorescence images of PKC-α, -ε, and -ζ localization in NTg and FGF2 Tg hearts subjected to sham treatment, 5 min ischemia, or 30 min ischemia, 5 min reperfusion. A: PKC-α is primarily localized in the cytosol and sarcolemmal of sham-treated hearts. During early ischemia, FGF2 Tg hearts also show PKC-α translocation to the perinuclear region. During early reperfusion injury, both FGF2 Tg and NTg hearts show PKC-α localization at the sarcolemmal and perinuclear regions of cardiomyocytes. B: NTg hearts subjected to ischemia show cytosolic and sarcolemmal staining for PKC-α, but FGF2 Tg hearts show translocation to mitochondria at this time point. During early reperfusion injury, mitochondrial localization of PKC-α is not observed, and localization is cytosolic and sarcolemmal in both NTg and FGF2 Tg hearts. C: PKC-α is localized at the sarcoplasmic reticulum during early reperfusion injury in FGF2 Tg hearts. D: PKC-ε is primarily localized in the cytosol of hearts subjected to sham treatment or 5 min ischemia with some sarcolemmal and nuclear staining. During early reperfusion injury, PKC-ε localization is also seen at intercalated disks in FGF2 Tg hearts. E: mitochondrial localization of PKC-ε is observed during reperfusion injury in FGF2 Tg hearts. F: PKC-ζ is primarily localized in the cytosol and nucleus of hearts subjected to sham treatment or 5 min ischemia. During early reperfusion injury, PKC-ζ is still localized in the cytosol and nucleus of NTg hearts, but nuclear localization is absent in FGF2 Tg hearts subjected to reperfusion injury. White arrows show examples of site-specific cellular localization of PKC isoforms in NTg or FGF2 Tg hearts; n = 4–8/group.

Signaling cross talk between the MAPK and PKC signaling cascades during ischemia-reperfusion injury.

Multiple studies have noted signaling cross talk between the MAPK and PKC cascades. Our laboratory has previously shown the MAPK cascade to mediate FGF2-induced cardioprotection (25), and we now demonstrate a role for PKC signaling in the mediation of FGF2's cardioprotective effect. Studies were performed to determine if signaling cross talk occurs between these two cascades during ischemia-reperfusion injury, which might mediate FGF2-induced cardioprotection. The phosphorylation of MAPK family members, ERK and p38, was determined in NTg and FGF2 Tg hearts in the presence of PKC pathway inhibition to elucidate if PKC activation affects either ERK or p38 phosphorylation. During early ischemic injury, PKC pathway inhibition with bisindolylmaleimide increased ERK phosphorylation in both NTg and FGF2 Tg hearts, whereas ERK phosphorylation was significantly decreased in response to PKC pathway inhibition during early reperfusion injury (P < 0.05, Fig. 5A). Similarly, differential regulation of p38 by the PKC pathway occurred during either early ischemia or early reperfusion injury. PKC pathway inhibition resulted in significantly decreased phosphorylation of p38 during early ischemia in both FGF2 Tg and NTg hearts (P < 0.05, Fig. 5B). FGF2 Tg hearts, however, showed significantly increased levels of phospho-p38 during early reperfusion injury in response to bisindolylmaleimide treatment (P < 0.05, Fig. 5B). These data suggest PKC-mediated alterations in MAPK phosphorylation in FGF2 Tg hearts that are regulated by the presence of ischemia and/or reperfusion injury.

Fig. 5.

Western blot analysis of cross talk between the PKC and MAPK signaling cascades as evaluated by PKC inhibition. A: inhibition of the PKC pathway with bisindolylmaleimide results in increased phosphorylation of extracellular signal-regulated kinase (ERK) during early ischemic injury but decreased ERK phosphorylation during early reperfusion injury. B: PKC pathway inhibition decreased phosphorylation of p38 during early ischemic injury but increased p38 phosphorylation during early reperfusion injury; n = 4–15 for each group. *P < 0.05 vs. NTg of the same treatment group. †P < 0.05 vs. vehicle-treated cohort.

To determine if MAPK family member activation affects the activation of PKC isoforms during ischemia-reperfusion injury, NTg and FGF2 Tg hearts were subjected to ischemia-reperfusion injury in the absence or presence of either ERK or p38 pathway inhibition, and the level of translocation of PKC-α, -δ, -ε, and -ζ was assessed in these hearts. As was seen previously (Fig. 2B), after 5 min of ischemia, PKC-α translocation was significantly reduced in vehicle-treated FGF2 Tg hearts compared with NTg hearts (P < 0.05, Fig. 6A). p38 pathway inhibition resulted in increased PKC-α translocation in FGF2 Tg hearts subjected to early ischemic injury without altering PKC-α translocation in NTg hearts. This suggests that the p38 pathway inhibits PKC-α translocation during ischemia in FGF2 Tg hearts. Similarly, FGF2 Tg hearts showed decreased translocation of PKC-α during early reperfusion injury that is increased significantly upon inhibition of p38 signaling (P < 0.05, Fig. 6A). PKC-α translocation, however, was not significantly altered in FGF2 Tg hearts subjected to reperfusion injury in the presence of ERK pathway inhibition. PKC-δ translocation was significantly reduced in FGF2 Tg hearts compared with NTg hearts at 5 min ischemia (P < 0.05, Fig. 6B), as previously observed (Fig. 2C). Neither ERK nor p38 pathway inhibition significantly altered the level of PKC-δ translocation in either FGF2 Tg or NTg hearts during early ischemia or early reperfusion. For PKC-ε, no significant alterations in the level of translocation occurred in response to ERK or p38 pathway inhibition following 5 min of ischemia (Fig. 6C). During early reperfusion injury, PKC-ε showed significantly enhanced translocation in FGF2 Tg hearts compared with NTg hearts (P < 0.05, Fig. 6C). Inhibition of p38 did not affect this relationship, but mitogen/extracellular signal-regulated kinase (MEK)-ERK pathway inhibition significantly increased PKC-ε translocation in NTg hearts during early reperfusion injury. PKC-ζ translocation was significantly decreased in FGF2 Tg hearts subjected to 5 min of ischemia (P < 0.05, Fig. 6D), as was seen previously (Fig. 2E). ERK and p38 pathway inhibition increased PKC-ζ translocation in FGF2 Tg hearts without affecting NTg hearts during early ischemic injury (P < 0.05, Fig. 6D). During early reperfusion, PKC-ζ translocation was significantly increased in FGF2 Tg hearts. ERK and p38 inhibition significantly increased PKC-ζ translocation in NTg hearts subjected to early reperfusion injury to the level of FGF2 Tg hearts (P < 0.05, Fig. 6D). Together these data reveal a complicated signaling network between the MAPK and PKC cascades that is regulated both by overexpression of FGF2 in the heart and by the time point of ischemia-reperfusion injury studied. This complicated interplay between the MAPK and PKC pathways may serve to coordinate the downstream effects of FGF2 in eliciting cardioprotection from ischemia-reperfusion injury.

Fig. 6.

Western blot analysis of cross talk between the MAPK and PKC signaling cascades as assessed by MAPK inhibition. A: PKC-α translocation is significantly reduced in FGF2 Tg hearts during both early ischemia and early reperfusion injury. p38 pathway inhibition results in increased PKC-α translocation in FGF2 Tg hearts during both early ischemia and reperfusion. ERK pathway inhibition, however, does not significantly affect PKC-α translocation. B: PKC-δ translocation is significantly reduced in vehicle-treated FGF2 Tg hearts during early ischemia, which is not significantly altered by either ERK or p38 pathway inhibition. There is no influence of MAPK inhibition in either NTg or FGF2 Tg hearts during early reperfusion injury. C: no significant alterations in PKC-ε translocation occur in response to either ERK or p38 pathway inhibition in NTg or FGF2 Tg hearts during early ischemia. During early reperfusion, PKC-ε shows significantly enhanced translocation in FGF2 Tg hearts. ERK pathway inhibition significantly increases PKC-ε translocation in NTg hearts during early reperfusion injury. D: PKC-ζ translocation is significantly decreased in FGF2 Tg hearts during early ischemic injury. Both ERK and p38 pathway inhibition increase PKC-ζ translocation in FGF2 Tg mice during early ischemia. During early reperfusion, however, PKC-ζ shows significantly enhanced translocation in FGF2 Tg hearts. ERK and p38 pathway inhibition significantly increases PKC-ζ translocation in NTg hearts during early reperfusion injury; n = 4–15 for each group. *P < 0.05 vs. NTg of the same treatment group. †P < 0.05 vs. vehicle-treated cohort.

DISCUSSION

This study has evaluated the role of the PKC pathway in FGF2-induced cardioprotection from ischemia-reperfusion injury. Four PKC isoforms (PKC-α, -δ, -ε, and -ζ) were studied because of their known relationship with FGF2 signaling or protection from ischemia-reperfusion injury. PKC isoform activation is a complicated process involving multiple steps, including phosphorylation by upstream kinases (phospho-inositide-dependent kinase 1, PDK-1), autophosphorylation, and translocation to cellular and organellar membranes (13, 18, 44). The phosphorylation state (4, 10, 17, 18, 38) and the level of translocation from a soluble (cytosolic) fraction to a membrane fraction (35, 36) have both been observed to be synonymous with activation.

Nonischemic FGF2 Tg hearts show decreased PKC-δ translocation compared with NTg hearts (Fig. 2, A and B), suggesting inhibition of PKC-δ activation by FGF2 may prime FGF2 Tg hearts in preparation for ischemia-reperfusion injury. PKC-δ translocation has been shown to be deleterious to the cardiac muscle and to mediate ischemic damage (8, 9). PKC-δ is capable of promoting changes in the mitochondrial membrane potential, causing the release of cytochrome c from the mitochondria and inducing apoptosis (39, 43). Some studies, however, have shown a protective effect of PKC-δ (19, 51). In vitro, expression of constitutively active PKC-δ protects cardiomyocytes from ischemic damage (71). Evidence from chronic overexpression of the low-molecular-weight protein isoform of FGF2 showed alterations in PKC-α translocation (63), supporting our hypothesis that FGF2 overexpression modulates cardioprotective signaling in preparation for ischemia-reperfusion injury. The differences in translocation or phosphorylation of a particular PKC isoform may be dependent on which FGF2 protein isoform is predominantly expressed.

PKC isoform translocation was also studied in FGF2 Tg and NTg hearts subjected to ischemia-reperfusion injury because activation of PKC isoforms has been shown to occur in response to ischemia-reperfusion injury (57, 66). PKC-α, -δ, and -ζ show decreased translocation during ischemia in FGF2 Tg hearts, whereas, during early reperfusion injury, FGF2 Tg hearts have decreased PKC-α translocation and increased PKC-ε translocation compared with NTg control hearts (Fig. 2, C–G). These data suggest inactivation or activation of particular PKC isoforms in FGF2 Tg hearts that is dependent on the presence of either ischemia or reperfusion injury. Other groups evaluated PKC translocation in which human recombinant FGF2 was administered before ischemia-reperfusion or during reperfusion and observed translocation of PKC multiple isoforms (-α, -ε, and -ζ) following ischemia-reperfusion injury (29, 53). These alterations in PKC isoform activation may mediate the cardioprotective effects of FGF2, since differences in PKC isoform activation occurring during ischemia-reperfusion injury have been implicated in other forms of cardioprotection. For example, ischemic preconditioning seems to be mediated through the translocation of PKC isoforms from the cytosol to cellular membranes and organelles because colchicine, which inhibits microtubule-mediated translocation, is able to abrogate the reduction in infarct size provided by ischemic preconditioning (1). Multiple models have shown that activation of PKC isoforms with either phorbol myristate or diacylglycerol analogs reproduces the infarct limitation seen with ischemic preconditioning (64, 70). Other studies, however, have reported PKC activation to be deleterious in that it aggravates hypoxic myocardial injury (27) and promotes arrhythmogenesis (6). These discrepancies may be related to the relative nonspecific activities of many PKC pathway inhibitors. In addition, individual PKC isoforms may have opposing effects in terms of protection from or exacerbation of ischemia-reperfusion injury.

To determine if activation of PKC isoforms during ischemia-reperfusion injury contributes to FGF2-induced cardioprotection, PKC inhibition with bisindolylmaleimide was performed in FGF2 Tg and NTg hearts subjected to irreversible ischemia-reperfusion injury. Bisindolylmaleimide is a nonselective inhibitor of multiple PKC isoforms at their catalytic ATP-binding sites (5). PKC inhibition abrogates FGF2-induced cardioprotection from both contractile dysfunction (Fig. 3A) and myocardial infarction (Fig. 3B). Our data are consistent with that of Jiang et al. (29) and Padua et al. (53) who demonstrated a role of PKC signaling and FGF2 in regulation of cardiac function following ischemia-reperfusion injury. Our observation that the PKC pathway is also necessary for FGF2-induced cardioprotection against myocardial infarction is a novel finding. PKC inhibition has also shown the necessity of PKC pathway activation for other forms of cardioprotection, including ischemic preconditioning (33, 40, 70) and pharmacologic-induced cardioprotection (9, 48, 50, 69).

The mechanism by which FGF2 exerts its cardioprotection through activation of the PKC pathway, however, remains to be determined. Multiple substrates have been identified for PKC isoforms, which may play a role in reduction of ischemia-reperfusion injury by FGF2, including proteins involved in metabolic pathways, transcription factors, and the signal transduction pathway (26). Some of these substrates, which may have relevance for myocardial ischemia, include ATP-dependent K+ channels (41), phospholipase A2 (52), Na+/H+ exchanger (68), nitric oxide synthases (7), Na+-K+-ATPase (42), gap junctions (15) and MAPK (37). Cellular localization of PKC isoforms is an important determinant of the substrate specificity and biological functions of individual PKC isoforms. Because immunofluorescence studies have shown that individual PKC isoforms localize to unique subcellular sites in cardiomyocytes upon stimulation (14), this study has used confocal immunofluorescence microscopy to elucidate PKC isoform localization in FGF2 Tg and NTg hearts subjected to ischemia-reperfusion injury, providing insights into the mechanisms by which PKC isoforms may mediate FGF2-induced cardioprotection. Sarcolemmal localization of PKC-α, -δ, and -ε may implicate these isoforms in alterations of cell surface receptors and ion channels that may affect cardiomyocyte signaling and contraction. Nuclear and perinuclear localization (Fig. 4, A and F) of PKC-α, -δ, -ε, and -ζ reveal actions of these isoforms in the regulation of gene transcription. Localization of PKC-α, -δ, -ε, and -ζ suggests roles for these isoforms in the regulation of cardiac contractility through modification of calcium-handling proteins in the sarcoplasmic reticulum (Fig. 4C). Mitochondrial localization (Fig. 4, B and E) of PKC-α, -ε, and -ζ indicates that these isoforms may regulate cellular energetics, metabolism, and apoptosis. For example, others have shown translocation of PKC-ε to mitochondria to result in the inhibition of apoptosis (2). Translocation of PKC-ε to intercalated disks during early reperfusion injury (Fig. 4D) suggests PKC-ε may participate in the regulation of cardiac contractility and intercellular communications. Other studies have shown PKC-ε localized to cross-striated structures and intercalated disks (21, 30) and to enhance cardiac function (30, 31). Similarly, investigators have demonstrated that PKC-ε interacts with cardiac gap junctions to influence cell-cell communications and that this interaction was induced by FGF2 administration (15, 16, 65).

Our laboratory has now demonstrated the necessity of both the MAPK (25) and PKC signaling cascades to the mediation of FGF2-induced cardioprotection from postischemic contractile dysfunction and myocardial infarction. The present study also determined whether the protective actions of these protein kinase pathways may be a result of interplay between PKC and MAPK signaling. PKC pathway inhibition with bisindolylmaleimide leads to increased ERK phosphorylation during early ischemia and decreased ERK phosphorylation during early reperfusion (Fig. 5A). PKC pathway inhibition also results in decreased phosphorylation of p38 during early ischemia and increased phosphorylation of p38 during early reperfusion injury in FGF2 Tg hearts (Fig. 5B), suggesting ischemia- or reperfusion-dependent regulation of both ERK and p38 by the PKC pathway. These data are consistent with other reports of PKC pathway-dependent activation or inactivation of MAPK family members. For example, several studies have shown activation of ERK by PKC-ε (3, 23, 54). In addition, PKC-α has also been shown to mediate activation of ERK and p38 (46, 49), and MEK-1, an upstream kinase for ERK, has been shown to be activated by conventional novel and atypical PKC isoforms (61). Our laboratory has previously reported inhibition of ERK by p38 to occur during early ischemic injury (25), which could potentially play a role in the effect of PKC inhibition on ERK activation. PKC inhibition with bisindolylmaleimide decreases p38 phosphorylation during early ischemia, which would then result in disinhibition of ERK, leading to increased ERK phosphorylation. Further characterization of this signaling cross talk would be necessary to determine if the effect of PKC inhibition during early ischemia is mediated by p38 but is beyond the scope of the present study.

This study has also shown regulation of PKC isoforms by MAPK family members. MEK1/2 inhibition with U-0126 results in increased PKC-ζ translocation during both early ischemia and early reperfusion injury (Fig. 6D) and increased PKC-ε translocation during early reperfusion injury (Fig. 6C), suggesting the ERK signaling cascade regulates the activation state of these two PKC isoforms during ischemia-reperfusion injury. p38 pathway inhibition with SB-203580 results in increased PKC-α and -ζ translocation during both early ischemia and early reperfusion injury (Fig. 6, A and D). Other studies have also suggested regulation of PKC isoform activation by MAPK family members. In chondrocytes, p38-dependent activation of PKC-ζ but not PKC-α has been observed (32).

In summary, this study demonstrated that PKC is an important signaling component for FGF2-induced cardioprotection and that the multiple interconnections between the MAPK and PKC pathway during both ischemia and reperfusion injury may play a role in the mediation of cardioprotection elicited by FGF2.

GRANTS

This work was supported by Grant SDG 23004N from the American Heart Association, a Research Starter Grant from the Pharmaceutical Research and Manufacturers of America, and National Heart, Lung, and Blood Institute Grant HL-075633 to J. J. Schultz.

Acknowledgments

We acknowledge M. Bender and A. Whitaker for excellent animal husbandry and N. Vatamaniuc for assistance with cardiac function data analysis.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract