PKC alpha regulates the hypertrophic growth of cardiomyocytes through extracellular signal-regulated kinase1/2 (ERK1/2)

Abstract

Members of the protein kinase C (PKC) isozyme family are important signal transducers in virtually every mammalian cell type. Within the heart, PKC isozymes are thought to participate in a signaling network that programs developmental and pathological cardiomyocyte hypertrophic growth. To investigate the function of PKC signaling in regulating cardiomyocyte growth, adenoviral-mediated gene transfer of wild-type and dominant negative mutants of PKC alpha, beta II, delta, and epsilon (only wild-type zeta) was performed in cultured neonatal rat cardiomyocytes. Overexpression of wild-type PKC alpha, beta II, delta, and epsilon revealed distinct subcellular localizations upon activation suggesting unique functions of each isozyme in cardiomyocytes. Indeed, overexpression of wild-type PKC alpha, but not betaI I, delta, epsilon, or zeta induced hypertrophic growth of cardiomyocytes characterized by increased cell surface area, increased [(3)H]-leucine incorporation, and increased expression of the hypertrophic marker gene atrial natriuretic factor. In contrast, expression of dominant negative PKC alpha, beta II, delta, and epsilon revealed a necessary role for PKC alpha as a mediator of agonist-induced cardiomyocyte hypertrophy, whereas dominant negative PKC epsilon reduced cellular viability. A mechanism whereby PKC alpha might regulate hypertrophy was suggested by the observations that wild-type PKC alpha induced extracellular signal-regulated kinase1/2 (ERK1/2), that dominant negative PKC alpha inhibited PMA-induced ERK1/2 activation, and that dominant negative MEK1 (up-stream of ERK1/2) inhibited wild-type PKC alpha-induced hypertrophic growth. These results implicate PKC alpha as a necessary mediator of cardiomyocyte hypertrophic growth, in part, through a ERK1/2-dependent signaling pathway.

Figure 1.

Adenoviral-mediated gene transfer of wild-type PKCα, βII, δ, and ɛ in cardiomyocytes. (A) Western blot analysis for antibody cross-reactivity from PKCα, βII, δ, and ɛ from AdPKCα, AdPKCβII, AdPKCδ, AdPKCɛ, or control Adβgal-infected neonatal cardiomyocytes. (B) Western blot analysis of cytosolic and particulate subcellular protein distribution of each PKC isoform at baseline or after stimulation with PMA. Distribution of both endogenous and overexpressed protein are shown. (C) Histogram representation of PKC particulate (open area) and soluble (dark area) protein levels demonstrates a 5–6-fold increase in protein expression (P < 0.05). Results were averaged from three independent experiments.

Figure 2.

Adenoviral-mediated gene transfer of PKCα, βII, δ, and ɛ in neonatal cardiomyocytes demonstrates distinct subcellular localization and translocation after agonist treatment. (A) Immunocytochemical analysis of PKC antibody specificity from AdPKCα-, AdPKCβII-, AdPKCδ-, and AdPKCɛ-infected cardiomyocytes. Antibodies are shown on the top while viral infection is shown on the left. (B) PKC isozyme distribution was characterized before and after PMA or PE treatment. Similar results were obtained in three independent experiments. The arrows show areas of significant redistribution after agonist stimulation. (C) PKCα shows efficient translocation to the particulate fraction only in previously hypertrophied cardiomyocytes.

Figure 2.

Adenoviral-mediated gene transfer of PKCα, βII, δ, and ɛ in neonatal cardiomyocytes demonstrates distinct subcellular localization and translocation after agonist treatment. (A) Immunocytochemical analysis of PKC antibody specificity from AdPKCα-, AdPKCβII-, AdPKCδ-, and AdPKCɛ-infected cardiomyocytes. Antibodies are shown on the top while viral infection is shown on the left. (B) PKC isozyme distribution was characterized before and after PMA or PE treatment. Similar results were obtained in three independent experiments. The arrows show areas of significant redistribution after agonist stimulation. (C) PKCα shows efficient translocation to the particulate fraction only in previously hypertrophied cardiomyocytes.

Figure 3.

Adenoviral-mediated gene transfer of wild-type PKCα is sufficient for cardiomyocyte hypertrophy. (A) α-actinin (orange) or ANF (green) coimmunostained cardiomyocyte cultures reveal myocyte morphology, sarcomeric organization, and hypertrophic ANF expression (arrows show perinuclear ANF protein expression). (B) Cardiomyocyte cell surface areas were quantified from serum-free, Adβgal-, Adβgal +PE–, AdPKCα-, AdPKCβII-, AdPKCδ-, AdPKCɛ-, and AdPKCζ-infected cultures. α-Actinin–stained cells were imaged and surface areas were calculated with NIH Image software (n = 100 cells each). (C) Percentage of cardiomyocytes expressing ANF protein. (n = 25 fields at 400× magnification). (D) Quantification of relative [³H] leucine incorporation per μg protein normalized as percentage of control. (E) PKC-specific kinase assay shows a significant increase in activity from overexpression of each isoform. Results were averaged from three independent experiments for A–D, and a representative experiment is shown for E. *P < 0.05 vs. Adβgal, ^# P < 0.05 vs. Adβgal + PE, ^† P < 0.05 vs. Adβgal + PMA.

Figure 3.

Adenoviral-mediated gene transfer of wild-type PKCα is sufficient for cardiomyocyte hypertrophy. (A) α-actinin (orange) or ANF (green) coimmunostained cardiomyocyte cultures reveal myocyte morphology, sarcomeric organization, and hypertrophic ANF expression (arrows show perinuclear ANF protein expression). (B) Cardiomyocyte cell surface areas were quantified from serum-free, Adβgal-, Adβgal +PE–, AdPKCα-, AdPKCβII-, AdPKCδ-, AdPKCɛ-, and AdPKCζ-infected cultures. α-Actinin–stained cells were imaged and surface areas were calculated with NIH Image software (n = 100 cells each). (C) Percentage of cardiomyocytes expressing ANF protein. (n = 25 fields at 400× magnification). (D) Quantification of relative [³H] leucine incorporation per μg protein normalized as percentage of control. (E) PKC-specific kinase assay shows a significant increase in activity from overexpression of each isoform. Results were averaged from three independent experiments for A–D, and a representative experiment is shown for E. *P < 0.05 vs. Adβgal, ^# P < 0.05 vs. Adβgal + PE, ^† P < 0.05 vs. Adβgal + PMA.

Figure 3.

Adenoviral-mediated gene transfer of wild-type PKCα is sufficient for cardiomyocyte hypertrophy. (A) α-actinin (orange) or ANF (green) coimmunostained cardiomyocyte cultures reveal myocyte morphology, sarcomeric organization, and hypertrophic ANF expression (arrows show perinuclear ANF protein expression). (B) Cardiomyocyte cell surface areas were quantified from serum-free, Adβgal-, Adβgal +PE–, AdPKCα-, AdPKCβII-, AdPKCδ-, AdPKCɛ-, and AdPKCζ-infected cultures. α-Actinin–stained cells were imaged and surface areas were calculated with NIH Image software (n = 100 cells each). (C) Percentage of cardiomyocytes expressing ANF protein. (n = 25 fields at 400× magnification). (D) Quantification of relative [³H] leucine incorporation per μg protein normalized as percentage of control. (E) PKC-specific kinase assay shows a significant increase in activity from overexpression of each isoform. Results were averaged from three independent experiments for A–D, and a representative experiment is shown for E. *P < 0.05 vs. Adβgal, ^# P < 0.05 vs. Adβgal + PE, ^† P < 0.05 vs. Adβgal + PMA.

Figure 4.

Adenoviral-mediated gene transfer of PKCα in rat neonatal cardiomyocytes colocalizes with α-tubulin. (A) Immunocytochemical analysis of PKCα translocation at progressive times after PMA stimulation. (B) Costaining for PKCα (green) and α-tubulin (red) demonstrates coincident immunoreactivity. (C) As a control, neither AdPKCα, Adβgal, nor PMA altered the normal pattern of α-tubulin localization. Similar results were observed in three independent experiments.

Figure 5.

Adenoviral-mediated gene transfer of multiple PKC isozymes does not alter PKCα subcellular localization. (A) Immunocytochemical analysis of PKCα translocation in cardiomyocytes coinfected with AdPKCα in conjunction with either AdPKCβII, AdPKCδ, or AdPKCɛ. (B) Western blot analysis of PKCα protein from similarly infected cardiomyocytes reveals no significant disruption of endogenous PKCα subcellular localization by overexpression of other PKC isozymes. (C) Histogram representation of PKC particulate (open area) and cytosolic (dark area) protein levels (n = 3 experiments). Adenoviral-mediated expression and redistribution of PKCα was not reduced by coexpression of PKCβII, δ, or ɛ (P < 0.05).

Figure 5.

Adenoviral-mediated gene transfer of multiple PKC isozymes does not alter PKCα subcellular localization. (A) Immunocytochemical analysis of PKCα translocation in cardiomyocytes coinfected with AdPKCα in conjunction with either AdPKCβII, AdPKCδ, or AdPKCɛ. (B) Western blot analysis of PKCα protein from similarly infected cardiomyocytes reveals no significant disruption of endogenous PKCα subcellular localization by overexpression of other PKC isozymes. (C) Histogram representation of PKC particulate (open area) and cytosolic (dark area) protein levels (n = 3 experiments). Adenoviral-mediated expression and redistribution of PKCα was not reduced by coexpression of PKCβII, δ, or ɛ (P < 0.05).

Figure 6.

Adenoviral-mediated gene transfer of dominant negative PKCα, βII, δ, and ɛ in neonatal cardiomyocytes. (A) Western blot analysis of protein expression levels from AdPKCαdn-, AdPKCβIIdn-, AdPKCδdn-, or AdPKCɛdn-infected cardiomyocytes. (B) Histogram representation of total PKC isozyme protein levels demonstrates a 7–9-fold increase in protein after adenoviral infection compared with endogenous wild-type protein levels (n = 3).

Figure 7.

Assessment of hypertrophy in cardiomyocytes expressing dominant negative PKC isozymes. (A) α-Actinin (orange) and ANF (green) coimmunostained cardiomyocyte cultures. (B) Cardiomyocyte cell surface area was quantified from serum-free, Adβgal-, Adβgal + PE–, AdMKP-1–, AdMKP-1 + PE–, AdPKCαdn + PE–, AdPKCβIIdn + PE–, AdPKCδdn + PE–, and AdPKCɛdn + PE–infected cultures. α-Actinin–stained cells were imaged with confocal microscopy and digitized, and surface areas were calculated with NIH Image software (n = 100 cells each). (C) Percentage of cardiomyocytes expressing ANF protein (n = 25 fields at 400× magnification). (D) Quantification of [³H] leucine incorporation per μg protein. All results were obtained in three independent experiments. *P < 0.05 vs. Adβgal. ^† P < 0.05 vs. Adβgal + PE.

Figure 7.

Assessment of hypertrophy in cardiomyocytes expressing dominant negative PKC isozymes. (A) α-Actinin (orange) and ANF (green) coimmunostained cardiomyocyte cultures. (B) Cardiomyocyte cell surface area was quantified from serum-free, Adβgal-, Adβgal + PE–, AdMKP-1–, AdMKP-1 + PE–, AdPKCαdn + PE–, AdPKCβIIdn + PE–, AdPKCδdn + PE–, and AdPKCɛdn + PE–infected cultures. α-Actinin–stained cells were imaged with confocal microscopy and digitized, and surface areas were calculated with NIH Image software (n = 100 cells each). (C) Percentage of cardiomyocytes expressing ANF protein (n = 25 fields at 400× magnification). (D) Quantification of [³H] leucine incorporation per μg protein. All results were obtained in three independent experiments. *P < 0.05 vs. Adβgal. ^† P < 0.05 vs. Adβgal + PE.

Figure 8.

Overexpression of PKCα and ɛ induces ERK1/2 phosphorylation. (A) Western blot analysis for ERK1/2, p38, and JNK wild-type and phosphorylated forms in AdPKC-infected cardiomyocytes. Anisomycin was used as a control for JNK and p38 activation only. (B) Quantitation from three independent experiments demonstrates significant ERK1/2 activation induced by AdPKCα and ɛ infection compared with Adβgal infection. *P < 0.05 vs. Adβgal.

Figure 9.

AdPKCαdn inhibits PMA-induced ERK1/2 phosphorylation in neonatal cardiomyocytes. (A) Western blot analysis of phosphorylated ERK1/2 from AdPKCαdn-, AdPKCβIIdn-, AdPKCδdn-, and AdPKCɛdn-infected cardiomyocytes before and after PMA treatment. The amount of total ERK1/2 did not vary. (B) Histogram showing a significant decrease in phosphorylated ERK1/2 only with AdPKCαdn infection (n = 3). *P < 0.05 vs. Adβgal; ^† P < 0.05 vs. Adβgal + PMA.

Figure 10.

Adenoviral-mediated gene transfer of dominant negative MEK1 attenuates PKCα-induced neonatal cardiomyocyte hypertrophy. (A) α-Actinin (orange) and ANF (green) coimmunostained cardiomyocyte cultures reveal myocyte-specific morphology and sarcomeric organization (nuclei are light blue). AdMEK1dn infection, but not AdMKK3dn, attenuates morphologic hypertrophy. (B) Cardiomyocyte cell surface areas were imaged and quantified using NIH Image software (n = 100 cells each). (C) Quantitation of the percentage of ANF-positive cardiomyocytes (n = 25 fields at 400× magnification). (D) Quantification of relative [³H] leucine incorporation per μg protein normalized as percentage of control. All results were quantified from three independent experiments. *P < 0.05 vs. Adβgal; ^† P < 0.05 vs. AdPKCα + Adβgal.

Figure 10.

Adenoviral-mediated gene transfer of dominant negative MEK1 attenuates PKCα-induced neonatal cardiomyocyte hypertrophy. (A) α-Actinin (orange) and ANF (green) coimmunostained cardiomyocyte cultures reveal myocyte-specific morphology and sarcomeric organization (nuclei are light blue). AdMEK1dn infection, but not AdMKK3dn, attenuates morphologic hypertrophy. (B) Cardiomyocyte cell surface areas were imaged and quantified using NIH Image software (n = 100 cells each). (C) Quantitation of the percentage of ANF-positive cardiomyocytes (n = 25 fields at 400× magnification). (D) Quantification of relative [³H] leucine incorporation per μg protein normalized as percentage of control. All results were quantified from three independent experiments. *P < 0.05 vs. Adβgal; ^† P < 0.05 vs. AdPKCα + Adβgal.