Protein kinase Cε and the antiadrenergic action of adenosine in rat ventricular myocytes

Koji Miyazaki, Satoshi Komatsu, Mitsuo Ikebe, Richard A. Fenton, James G. Dobson Jr.


Adenosine-induced antiadrenergic effects in the heart are mediated by adenosine A1 receptors (A1R). The role of PKCε in the antiadrenergic action of adenosine was explored with adult rat ventricular myocytes in which PKCε was overexpressed. Myocytes were transfected with a pEGFP-N1 vector in the presence or absence of a PKCε construct and compared with normal myocytes. The extent of myocyte shortening elicited by electrical stimulation of quiescent normal and transfected myocytes was recorded with video imaging. PKCε was found localized primarily in transverse tubules. The A1R agonist chlorocyclopentyladenosine (CCPA) at 1 μM rendered an enhanced localization of PKCε in the t-tubular system. The β-adrenergic agonist isoproterenol (Iso; 0.4 μM) elicited a 29–36% increase in myocyte shortening in all three groups. Although CCPA significantly reduced the Iso-produced increase in shortening in all three groups, the reduction caused by CCPA was greatest with PKCε overexpression. The CCPA reduction of the Iso-elicited shortening was eliminated in the presence of a PKCε inhibitory peptide. These results suggest that the translocation of PKCε to the t-tubular system plays an important role in A1R-mediated antiadrenergic actions in the heart.

  • β-adrenergic
  • cardiomyocyte shortening
  • transfection
  • t tubules
  • PKCε translocation

adenosine is known to exert numerous effects in the heart. This endogenous nucleoside via adenosine A1 receptors (A1R) causes antiadrenergic (12), antiarrthymogenic (7, 19), and preconditioning (15, 17) actions in the myocardium. The antiadrenergic action of adenosine involves reductions in β-adrenergic catecholamine-induced increases in adenylyl cyclase activity (30, 43, 45), cAMP formation (10, 51), PKA activation (12), myocardial protein phosphorylation (18), intracellular Ca2+ transient magnitude (19), and cardiac atrial (11, 42) and ventricular (12, 46, 51) contractility. Both the A1R-mediated antiadrenergic actions (40) and preconditioning (29, 49) appear to involve PKC activity. There are at least 12 isoforms of PKC known to exist in tissues (35), and activation fosters their translocation to anchor proteins in various subcellular locations (37). PKCε along with PKC-α and PKCδ are the dominant isoforms present in adult rat cardiomyocytes (9, 47). Because A1R stimulation has been reported to activate PKCδ in freshly isolated rat ventricular myocytes (22), the possibility that this adenosine receptor activates PKCε was explored.

The present study was undertaken to ascertain whether PKCε plays a role in the antiadrenergic effects caused by A1R stimulation. This possibility was investigated by overexpressing PKCε in isolated primary cultured rat ventricular myocytes and determining whether the enzyme could be activated/translocated by A1R stimulation in a manner consistent with its involvement in the antiadrenergic action of adenosine. The findings indicate that the A1R-elicited activation of PKCε is associated with potentiation of the antiadrenergic actions of adenosine in ventricular myocytes transfected with PKCε.


Isolation of ventricular myocytes.

Myocytes were isolated from rat hearts using collagenase and hyaluronidase according to procedures previously reported (45). The harvested rectangular myocytes displayed no spontaneous contractions, excluded Trypan blue, and were capable of contracting upon electrical stimulation. These myocytes were transfected with PKCε as described below. Male Sprague-Dawley rats (Harlan; Indianapolis, IN, or Charles River; Wilmington, MA) 3–4 mo of age and weighing 250–325 g were used to obtain ventricular myocytes for these studies. The rats were maintained and used in accordance with recommendations in the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Pub. No. 85-23, Revised 1996) and the guidelines of the Institutional Animal Care and Use Committee of the University of Massachusetts Medical School.

Construction of the vector.

Rabbit PKCε cDNA was cloned into the BamH1site of the pEGFP-N1 mammalian expression vector (Clontec). The green fluorescent protein (GFP) moiety was placed at the COOH-terminal end of the PKCε molecule to avoid the potential disruption of the regulatory properties of PKCε and the binding to the cellular elements that is thought to involve the NH2-terminal portion of the molecule. To avoid the effect of the COOH-terminal GFP moiety on the proper folding of the PKCε molecule, a flexible linker sequence consisting of a 15-amino acid residue (LQSTVPRARDPPVAT) was inserted between the two moieties.


Isolated ventricular myocyte transfection was performed using GeneSHUTTLE-20 (Quantum Biotechnologies; Montreal, Quebec, Canada) according to the instructions of the supplier. Briefly, 1 μg of pEGFPN1-PKCε DNA was diluted with 100 μl of minimal essential medium (MEM) and transferred to the diluted liposome solution containing 5 μl of GeneSHUTTLE-20 in 100 μl of MEM. After incubation for 30 min, the DNA/liposome complex solutions were added to the ventricular myocytes in a 35-mm dish. Cells were used in experiments after overnight incubation at 37°C with a 5% CO2 atmosphere.

Cos7 cells were transfected with the GFP-PKCε expression vector or pEGFPN1 by electroporation as described previously by us (36). In brief, the cells were treated with ice-cold 5% trichloroacetic acid followed by sonication. The samples were then dissolved in a 5% SDS-0.5 M NaHCO3 buffer and subjected to SDS-PAGE followed by Western blotting.

Extraction, immunoprecipitation, and assay of PKC.

PKCε-GFP was extracted from transfected Cos7 cells in buffer containing 0.5 M KCl, 10% glycerol, 0.025% Triton X-100, 1 mM ATP, 1 mM PMSF, 0.2 mg/ml N-α-p-tosyl-l-lysine chloromethyl ketone, 1 mM N-α-p-tosyl-l-arginine methyl ester, 1 mM DTT, 10 μg/ml leupeptin, and 30 mM Tris·HCl (pH 7.5). Anti-GFP antibodies conjugated with protein A beads were added to perform immunoprecipitation. The immunoprecipitated samples were assayed for PKC activity in a buffer containing 0.4 mg/ml histone III, 25 mM PIPES, 1 mM EGTA, 0.05 M KCl, 1 mM MgCl2, 0.1 mg/ml phosphatidyl serine, and 0.05 mM [γ-32P]ATP (pH 7.0) at 25°C in the presence and absence of 50 ng/ml phorbol 12-myristate 13-acetate (PMA). The kinase activity is reported as micromoles of 32P incorporated per minute per milligram of protein.

Immunocytochemistry and imaging.

Transfected and nontransfected ventricular myocytes were cultured for 4 h on poly-l-ornithine-coated glass coverslips to allow attachment. After attachment, the myocytes were exposed to the A1R agonist chlorocyclopentyladenosine (CCPA) and/or the A1R antagonist dipropylcyclopentylxanthine (DPCPX) as described in the appropriate figures. Myocytes were fixed in 4% formaldehyde for 10 min and permeabilized with 0.1% Triton X-100 in suffusion solution (SS). SS contained (in mM) 136.4 NaCl, 4.7 KCl, 1.0 CaCl2, 10 HEPES, 1.0 NaHCO3, 1.2 MgSO4, 1.2 KH2PO4, 10 glucose, 0.6 ascorbate, and 1.0 pyruvate. After being blocked with 3% BSA in SS at room temperature for 1 h, the preparations were incubated with rabbit antibodies against PKCε, PKCδ (Calbiochem; San Diego, CA), and GFP (Medical & Biological Laboratories; Watertown, MA) at 4°C for 12–16 h. These antibodies were used to assess the presence of endogenous PKCε and PKCδ and the level of PKCε transfection in the myocytes. After being washed with SS three times, they were incubated with fluorescence-labeled Cy5- or FITC-conjugated anti-rabbit secondary antibodies (Jackson ImmunoResearch; West Grove, PA, or Molecular Probes; Eugene, OR, respectively) at 37°C for 1 h. Excess secondary antibody was removed, and the samples were mounted in 3% 1,4-diazabicyclo(2.2.2)octane (DABCO, Sigma)-90% glycerol in SS. The immunostained samples were viewed using a Leica DM IRBE inverted microscope equipped with a TCS SP2 confocal system, a 65-mW argon laser, two helium/neon lasers (1.2 and 10 mW), and differential interference contrast accessories (Leica Microsystems; Heidelberg, Germany). Images were acquired and analyzed with LCS software and Adobe Photoshop 6.0 software (Adobe Systems; San Jose, CA).

PAGE and immunoblotting.

Ventricular myocytes were dissolved and subjected to PAGE and immunoblotting as previously described (14). Primary rabbit PKCε and PKCδ antibodies were used and visualized with a secondary anti-rabbit antibody conjugated to horseradish peroxidase.

PKC-ε inhibitor peptide.

Ventricular myocytes were rendered permeable to PKCε inhibitory peptide (Glu-Ala-Val-Ser-Leu-Lys-Pro-Thr, EMD Biosciences; La Jolla, CA) by exposure to a permeabilization buffer (PB) consisting of 20 mM HEPES (pH 7.4), 10 mM EGTA, 140 mM KCl, 150 μg/ml peptide, 6 mM ATP, 2 μM β-escin, and 0.02% DMSO for 10 min at room temperature. After the permeabilization, the myocytes were returned to SS by exposing the cells to gradually increasing concentrations of Ca2+ stepwise every 2 min from 0, 50, 100, 300, and 500, to 1,000 μM in SS containing 6 mM ATP and no initially added Ca2+. The myocytes were suffused with SS for 20 min and stimulated to contract at 12 beats/min.

Myocyte contractile function for overexpression studies.

Transfected or nontransfected ventricular myocytes contained in 35-mm dishes with 3.0 ml of SS were placed on an inverted fluorescent microscope (Diaphot, Nickon) equipped with a SIT camera (VE 1000 SIT, DAGE MTI). A 35-mm plastic ring equipped with platinum wire electrodes was used to initiate myocyte contraction every 10 s. In addition, the ring also possessed an inflow and aspiration outflow for continual suffusion of the myocytes with SS at 0.4 ml/min. The SIT camera and a JVC CR-6004 videocassette recorder (Technical Video Resources; Fairfield, CT) were used to record myocyte shortening with each contraction. After the transfected and nontransfected myocytes were delineated in a fluorescent field, a visual field was used to assess shortening of the myocytes. The extent of myocyte shortening was determined by playback and printing of the relaxed and fully shortened images of each myocyte using a Sony UP-811 video graphic printer. The difference in the myocyte length between the two images provided the extent of shortening. Shortening is expressed in micrometers.

Protocol for assessing myocyte contractile function.

The ventricular myocytes were equilibrated for 30–40 min without electrical stimulation and continually suffused with fresh SS (Fig. 1). A train of four contractions 10 s apart was then recorded as control shortenings. Suffusion was continued with isoproterenol (Iso), a β1-adrenergic receptor agonist, at 0.4 μM present in the SS for 10 min, and a train of four contractions was recorded for the first sequence. In the second sequence, suffusion was continued with the Iso together with the A1R agonist CCPA present at a concentration of 2 μM in the SS for 10 min. A train of four shortenings was recorded. In the third sequence, Iso, CCPA, and the A1R antagonist DPCPX (2 μM) were included together in the SS for 10 min. A final train of four shortenings was recorded.

Fig. 1.

Protocol for exposure of rat ventricular myocytes to isoproterenol (Iso), chlorocyclopentyladenosine (CCPA), and dipropylcyclopentylxanthine (DPCPX). The equilibration time ranged from 30 to 40 min before the initiation of each experiment at time 0. The extent of myocyte shortening was recorded at 0, 10, 20, and 30 min (*) just before and after the addition of Iso (0.4 μM), Iso + CCPA (2 μM), and Iso + CCPA + DPCPX (2 μM).

Myocyte contractile function in the presence of PKCε inhibitor peptide.

Myocyte contractile shortening was determined using an IonOptix System (Milton, MA) in a continuous suffusion chamber as previously described (13). Briefly, myocytes were stimulated to contract at 12 beats/min and exposed to 0.4 μM Iso, 2 μM CCPA, or a combination of these two agents for 4 min in either the absence or presence of a PKCε inhibitor peptide. Myocyte cell shortening was recorded and expressed in micrometers.

Statistical methods.

All data are presented as means ± SE. ANOVA was performed with additional testing using the Student-Newman-Keuls test (Statmost, DatAxiom Software; Los Angeles, CA). A probably (P value) of <0.05 was accepted as indicating a statistically significant difference.


Buffer salts acids and general laboratory reagents were obtained from Fisher Scientific (Medford, MA). Iso, ascorbic acid, CCPA, DPCPX, HEPES, DMSO, BSA, poly-l-ornithine, β-escin (aescin), and formaldehyde were purchased from RBI/Sigma Chemical (St. Louis, MO). The anti-PKCε and PKCδ rabbit polyclonal and goat anti-rabbit antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). The anti-GFP rabbit polyclonal antibodies were obtained from MBL (Ina, Japan). Triton X-100 was from RPI (Mt. Prospect, IL) and MEM was from GIBCO/BRI (Rockville, MD). Iso (10 mM) was prepared in 0.1% Na2S2O5 fresh daily and, upon final dilution, is a solution of 1 mM ascorbic acid. CCPA and DPCPX (10 mM) were prepared in DMSO fresh daily.


Overexpression of GFP-tagged PKCε in cardiomyocytes.

To observe the effects of PKCε on the antiadrenergic action of A1R stimulation, GFP-tagged PKCε was overexpressed in isolated ventricular myocytes using the mammalian expression vector pEGFPN1/PKCε. The GFP tagging makes it possible to identify the transfected cells during contraction monitoring. To ensure that GFP signals observed in cells represent GFP-PKCε but not the degradation products, the transfected cells were subjected to Western blot analysis using anti-GFP antibodies (Fig. 2). We recognized the expressed GFP-tagged PKC protein as a single band at the expected molecular weight of the GFP-PKCε chimera, and there were no degradation products having the GFP moiety. The results indicate that the GFP signals in cells solely represent GFP-tagged PKCε.

Fig. 2.

Western blot of green fluorescent protein (GFP)-PKCε-transfected cells. Cos7 cells transfected with the GFP-PKCε expression vector or pEGFPN1 were harvested at 12 h after transfection, and the total cell homogenates were subjected to Western blotting using anti-GFP antibodies as described in methods. A single molecular mass band of GFP-PKCε (116 kDa) was observed. No degradation products having a GFP signal were detected.

To examine whether COOH-terminal GFP hampers PKCε function, PKCε-GFP was extracted from transfected Cos7 cells and immunoprecipitated, and the precipitate was assayed for PKC activity as described in methods. The kinase activities of PKCε-GFP in the presence and absence of PMA were 3.2 and 0.25 μmol·min−1·mg−1, respectively. The activity was significantly regulated by PMA, suggesting that the PKCε-GFP chimera maintains the authentic PMA-dependent kinase activity.

Detection of endogenous PKCε as revealed using a PKCε antibody indicates that upon exposure of the nontransfected myocyte to CCPA, the enzyme translocates to t-tubular-like structures in the cell (Fig. 3, a and b). Preincubation of nontransfected myocytes with the A1R antagonist DPCPX prevented the CCPA-induced translocation of the enzyme (Fig. 3c). This is not unusual because PKCε has been reported to translocate to membrane caveolae upon activation in rat cardiomyocytes (48). In contrast, distribution of endogenous PKCδ as revealed using a PKCδ antibody indicates that CCPA does not cause this isoform to translocate (Fig. 3, g and h). The present results suggest that A1R stimulation using CCPA elicits translocation of PKCε to t-tubular-like structures of the ventricular myocytes. Immunoblotting of endogenous PKCε and PKCδ revealed that both isoforms were present in the myocytes. Although there appeared to be more PKCδ compared with PKCε, the former isoform did not translocate to t-tubular-like structures with CCPA stimulation of the myocytes.

Fig. 3.

Effect of CCPA and DPCPX on the location of endogenous PKCε and PKCδ in rat isolated ventricular myocytes. The myocytes were immunostained with either PKCε or PKCδ rabbit primary antibodies, and the endogenous enzymes were detected by Cy5-conjugated anti-rabbit IgG secondary antibody. For PKCε, the Cy5 images (a–c) and differential interference contrast (DIC) images (d–f) were captured in the absence [control (a and d)] and presence of a 10-min incubation of 2 μM CCPA (b and e) or an incubation for 20 min with 2 μM DPCPX. The final 10 min of the latter incubation contained 2 μM CCPA (c and f). For PKCδ, the Cy5 images (g–i) and DIC images (j–l) were captured in the absence [control (g and j)] and presence of a 10-min incubation of 2 μM CCPA (h and k) or an incubation for 20 min with 2 μM DPCPX. The final 10 min of the latter incubation contained 2 μM CCPA (i and l). Bar = 10 μm.

In transfected myocytes, GFP-tagged PKCε also moved to t-tubular-like structures with CCPA (Fig. 4, a and b), and its translocation was hampered by DPCPX (Fig. 4c). However, the total fluorescent signal in the transfected myocytes was significantly higher than in the nontransfected cells. These results demonstrate that the CCPA-induced translocation of PKCε determined by anti-PKCε antibody staining is not a result of nonspecific signals and that GFP tagging did not interfere with the CCPA-induced translocation of PKCε.

Fig. 4.

Effect of PKCε overexpression on the location of PKCε in rat isolated ventricular myocytes in the absence and presence of CCPA. The myocytes were immunostained with a rabbit GFP primary polyclonal antibody, and the transfected enzyme was detected by FITC-conjugated anti-rabbit IgG secondary antibody. The FITC images (a–c) and DIC images (d–f) were captured in the absence [control (a and d)] or presence (b and e) of a 10-min incubation of 2 μM CCPA or with cells (c and f) incubated for 20 min with 2 μM DPCPX with the final 10 min of incubation in combination with 2 μM CCPA. Bars = 10 μm.

Expressed GFP-tagged PKCε was estimated to be approximately fivefold (5.5 ± 0.2; n = 3 experiments) greater in the transfected myocytes compared with endogenous PKCε in nontransfected myocytes. This comparison was obtained by determining the mean value of the total fluorescent intensities of anti-PKCε antibody signals for nontransfected and transfected myocytes (n = 10 cells). It is reasonable to assume that the overexpression of PKCε should enhance the activation of relevant downstream pathways compared with responses in the absence of overexpression.

To more clearly delinate the structure to which PKCε translocated, we compared the localization of the GFP signal and the brightfield image. The localization of GFP was coincident with the I band of the myocyte (Fig. 5). Because t tubules are located at the center of the I-band in myocytes, the results suggest that A1R stimulation, using CCPA, elicits translocation of PKCε to t-tubular-like structures of the ventricular myocytes.

Fig. 5.

Effect of CCPA-induced translocation of GFP-tagged PKCε to the I-band in rat isolated ventricular myocytes. The myocytes were stimulated with CCPA as described in methods and Fig. 1. Top, GFP fluorescence signals were detected as described in Fig. 4. a: GFP-tagged PKCε; b, DIC; c, merged image of a and b. Bar = 20 μm. Middle, fluorescence and DIC image distribution of the longitudinal section (76 μm in length as shown in the top) of the images (a–c). Green trace, GFP signals; black trace, DIC signals. Bottom, magnification of the longitudinal section of the image (c). Note that the fluorescence signals of GFP-tagged PKCε were well synchronized with that of the I band.

Cardiomyocyte contractile activity.

Overexpression of PKCε in ventricular myocytes increases the A1R-mediated reduction of the β1-adrenergic elicited contractile response. The average resting myocyte length was 109.5 ± 3.1 μm. The myocytes shortened by ∼10% upon electrical stimulation. With Iso administration at 0.4 μM in the SS, the extent of shortening increased to ∼30%. Overexpression of PKCε by transfection in myocytes had no significant effect on the basal extent of cell shortening (Fig. 6). A 10-min exposure to Iso caused a 28.9 ± 3.4% and 36.4 ± 3.7% increase in the extent of myocyte shortening above the basal levels of normal and transfected myocytes, respectively. The Iso-induced increase in the shortening was not significantly different between normal and transfected myocytes. In the presence of 2 μM CCPA, an A1R agonist, the Iso-produced increase in shortening was reduced to only a 12.9 ± 0.7% increase above the basal level of shortening in normal myocytes. However, in transfected myocytes, the CCPA totally inhibited the Iso-elicited increase in the extent of myocyte shortening.

Fig. 6.

Effect of PKCε overexpression on ventricular myocyte shortening in the presence of Iso (0.4 μM), CCPA (2 μM), and DPCPX (2 μM) as described in methods and Fig. 1. Values are means ± SE for 12 nontransfected (normal) and 8 transfected (+PKCε) myocytes. *Significant difference from the values without Iso, P < 0.05; †significant difference from the Iso values, P < 0.05; **significant difference from the comparable normal value; ‡significant difference from the Iso + CCPA values, P < 0.05.

To verify that the CCPA inhibition of the Iso-elicited contractile response was due to activation of the A1R, the A1R antagonist DPCPX was used in both normal and transfected myocytes. DPCPX restored the Iso-induced increase in myocyte shortening in both normal and transfected myocytes to levels not significantly different from the increase in shortening resulting from the administration of Iso alone. These data suggest that the CCPA-induced reduction of the Iso-induced increase in shortening is due to A1R stimulation.

Further experiments were conducted by transfecting myocytes with PKCε-free vector (null vector). The presence of the expression vector devoid of PKCε in the transfected myocytes did not influence the antiadrenergic action of CCPA. Iso increased shortening by 30.6 ± 1.9% in normal and 32.8 ± 2.5% in null vector-containing myocytes (Fig. 7). The increase in shortening caused by Iso in the presence of CCPA was 13.1 ± 2.3% and 14.7 ± 2.8% for normal and null vector-containing myocytes, respectively. These reductions in the Iso responses were not significantly different. DPCPX restored the Iso-produced increase in shortening in the presence of CCPA. These results indicate that presence of the null expression vector did not influence either the antiadrenergic or Iso-induced contractile response.

Fig. 7.

Effect of overexpression of the pEGFP-N1 construct devoid of +PKCε (null vector) on ventricular myocyte shortening in the presence of Iso (0.4 μM), CCPA (2 μM), and DPCPX (2 μM) as described in methods and Fig. 1. Values are means ± SE for 10 nontransfected (normal) and 6 transfected (null vector) myocytes. *Significant difference from the values without Iso, P < 0.05; †significant difference from the Iso values, P < 0.05; ‡significant difference from the Iso + CCPA values, P < 0.05.

Comparison of the antiadrenergic effects observed in normal, PKCε-transfected, and null vector-containing myocytes indicates that myocytes in which PKCε is overexpressed display a greater antiadrenergic response (Fig. 8). The antiadrenergic effect was approximately twofold greater in the PKCε-overexposed myocytes, suggesting that PKCε plays an important role in the antiadrenergic action of A1R stimulation.

Fig. 8.

Effect of CCPA inhibition of the Iso-elicited contractile response in nontransfected (normal) myocytes or myocytes transfected with PKCε (+PKCε) or the pEGFP-N1 construct devoid of PKCε (null vector). The percent inhibition is the reduction cause by CCPA (2 μM) in the presence of Iso (0.4 μM). Values are means ± SE for 12 nontransfected (normal) myocytes, 8 transfected (+PKCε) myocytes, and 6 myocytes transfected with the empty construct (null vector). *Significant difference from both normal and null vector values, P < 0.05.

The reduction of the Iso-elicited contractile response caused by CCPA was prevented by using a PKCε inhibitory peptide. Ventricular myocytes rendered permeable to the peptide, as outlined in methods, displayed normal contractile function. Upon permeablization, myocytes that were exposed to the peptide shortened similarly to those that were not exposed to peptide (Fig. 9). Iso at 0.4 μM increased shortening by 94.7 ± 19.6% and 141.2 ± 15.8% in myocytes exposed and not exposed to the inhibitory peptide, respectively. Although the peptide tended to increase the Iso contractile response, it was not significant. The inhibitory peptide was without effect on myocyte shortening in the presence of 2 μM CCPA alone. CCPA in the presence of Iso caused only a 46.2 ± 8.9% increase in shortening when myocytes were not exposed to the peptide. However, in myocytes exposed to the peptide, CCPA plus Iso increased shortening by 117.5 ± 14.3%. Thus the PKCε inhibitory peptide prevented the antiadrenergic action of CCPA in Iso-stimulated myocytes.

Fig. 9.

Effect of PKCε inhibitor peptide (PKCε IP) on ventricular myocyte shortening in the presence of Iso (0.4 μM) and CCPA (2 μM) as described in methods. Values are means ± SE for 6 myocytes. *Significant difference from the values without Iso, P < 0.05; †significant difference from Iso values, P < 0.05.


The present study indicates that A1R stimulation results in a translocation of endogenous and overexpressed PKCε to t-tubular-like structures of ventricular myocytes. Furthermore, compared with nontransfected ventricular myocytes and myocytes transfected with PKCε-free or null vector, overexpression of PKCε in myocytes potentiates the antiadrenergic action of A1R stimulation in isoproterenol-stimulated contracting myocytes. Overexpression of PKCε in ventricular myocytes has no effect on basal and Iso-elicited contractions but exerts an effect upon stimulation with the A1R agonist CCPA. Although PKCδ was present in the ventricular myocytes, CCPA did not cause translocation of this isoform of PKC. In addition, a PKCε inhibitory peptide prevented the antiadrenergic effect caused by CCPA in nontransfected myocytes without having a significant influence on basal or Iso-elicited contractions. The results also appear to indicate that ventricular myocyte PKCε requires A1R stimulation to manifest an effect. Overall, these findings suggest that PKCε may play an important role in mediating the adenosinergic effects of A1R stimulation in cardiac muscle.

Over the past several years, the importance of PKC in signaling pathways modulating heart function has become apparent. Studies have shown that PKC activation influences cardiac contractility (6, 31, 55), gene expression (37), the development of hypertrophy (1, 52), and the manifestation of preconditioning of the ischemic myocardium (2, 54). The PKC isoform PKCε has been consistently detected in adult ventricular myocytes (23, 24, 41) and is observed endogenously in the present study using an antibody to this isoform of the enzyme. Generally, activation of PKC by extracellular agonists elicits the translocation from the cytosol to the membrane fraction of cells. In this study, PKCε appears to translocate to t-tubular-like structures with A1R stimulation. This finding is in agreement with previous reports revealing that PKCε translocates to striated structures thought to be the t tubules of the cardiomyocyte as evidenced by the colocalization of the PKC-ε-GFP construct with α-actinin, a structural protein marker for the Z line of the sarcomere (41). However, the importance of this translocation is not fully understood.

In the myocardium, t-tubular membranes penetrate the cardiomyocytes near the Z lines (20). L-type Ca2+ channels, inducible nitric oxide (NO) synthase (NOS2) (5), RACK2, the anchor protein for PKCε, and A-kinase anchor protein (AKAP) (26) have been found to be localized in the t tubules. Such a position adjacent to sarcoplasmic reticular ryanodine receptors and myofibrils creates the potential for an intricate modulation of excitation-contraction coupling via the regulation of intracellular Ca2+ release and myofilament function (53). With specific reference to PKA, the close proximity of AKAP to the L-type Ca2+ channel ensures that the activation of PKA via extracellular β-adrenergic receptor agonists results in the phosphorylation and activation of the channel (21). The ensuing increase in Ca2+ current (ICa) that elicits an increase in cardiomyocyte contractility would be terminated by the dephosphorylating activity of protein phosphatase 2A also associated with the Ca2+ channel (8).

Gi protein-coupled receptors such as those stimulated by adenosine may potentially reduce β-adrenergic-stimulated (phosphorylated) Ca2+ channel activity and the resulting augmentation of contractility via several mechanisms. A1R activation has been reported to reduce the high-affinity binding of agonist to myocardial β1-receptors (44) as well as reduce adrenergic stimulation of adenylyl cyclase (30) and cAMP formation (10, 51). Each action alone would result in a reduced adrenergic-stimulated PKA activity (12) and phosphorylation state of many proteins associated with enhanced contractile function, including the L-type Ca2+ channel, (18). Second, adenosine has been found to manifest antiadrenergic actions via carboxymethylation of protein phosphatase 2A (34). The resulting increase in enzyme activity would potentially result in the dephosphorylation of myocardial proteins manifesting the increase in contractility resulting from β1-adrenergic stimulation. Third, adenosine has been found to enhance the release of NO from endothelial cells (32) and cardiomyocytes (25). NO has been reported to play an important role in attenuating the responsiveness of the heart to adrenergic stimulation (28); however, this concept is controversial (27).

It has recently been reported that PKC plays a role in the adenosine-induced modulation of cardiomyocyte function (31). In the present study, adenosine was observed to stimulate the translocation of PKCε to the membrane fraction of isolated cardiomyocytes, perhaps as a result of phospholipase D activation (16, 39). In addition, PKC inhibitors have been found to antagonize adenosine receptor agonist-induced decreases in cardiomyocyte shortening velocity (31). In the present study, it has been documented that PKCε is translocated to the t tubules of the rat ventricular myocytes in response to A1R stimulation. Furthermore, overexpression of PKCε resulted in an augmentation of the antiadrenergic action of adenosine compared with the nontransfected ventricular myocyte. This functional manifestation of PKCε translocation appears to be complex. As a result of the colocalization of RACK2 and L-type Ca2+ channels in the t-tubular membrane, it is plausible that the translocation of PKCε to the t tubules results in the phosphorylation of L-type Ca2+ channels (26). It has recently been reported that an attenuation of ICa in rat ventricular myocytes results from the activation of PKCε (23). Thus it may be presumed that the translocation of PKCε is a normal component of the antiadrenergic action of A1R stimulation that is amplified by enhanced levels of PKCε. Additional complexity may result from the considerable cross-talk between various signaling pathways in the cell (26). For example, ICa is independently increased by α1- and β1-adrenergic receptors that involve PKC and PKA signaling pathways, respectively. However, α1-agonists attenuate the ability of β1-agonists to enhance ICa (3). With respect to contractile activity, similar interactions have been found to exist between A1R and the adenosine A2 receptor (38). Perhaps other mechanisms in addition to PKCε translocation are at work influencing the magnitude of ICa in the β1-adrenergic stimulated heart. In addition to the apparent importance of PKCε in the antiadrenergic and cardioprotective actions of A1R stimulation, other isoforms of PKC involving G protein-coupled receptors may also be important in cardiac myocytes. For example, there are findings supporting a role of PKCδ in the preconditioning and long-term protective actions of adenosine and other agonists of G protein-coupled receptors (33, 56). It is also of interest that PKCε deletion eliminates the preconditioning reduction of infract size in the mouse heart (50). PKC-α may be an important determinant of myocyte contractility (4).

In summary, the present studies indicate that A1R stimulation results in the translocation of PKCε to t-tubular-like structures of ventricular myocytes. Overexpression of PKCε in ventricular myocytes potentates the antiadrenergic action resulting from A1R stimulation in contracting myocytes subjected to β1-adrenergic stimulation. Furthermore, an inhibitory peptide of PKCε prevented the antiadrenergic action of A1R stimulation. These results suggest that PKCε may play an important role in the adenosinergic effects in cardiac muscle resulting from A1R activation.


This study was made possible by National Institutes of Health (NIH) Grants HL-66045 and AG-11491 (to J. G. Dobson, Jr.) and HL-61426 (to M. Ikebe). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.


The authors thank Lynne G. Shea, Kris Morrill, and Christine Taylor for preparing the isolated adult rat ventricular myocytes and assistance with the experiments.


  • 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.


View Abstract