The phosphoinositide 3-kinase (PI3-kinase)-protein kinase B (Akt) signaling pathway is essential in the induction of physiological cardiac hypertrophy. In contrast, protein kinase C β2 (PKCβ2) is implicated in the development of pathological cardiac hypertrophy and heart failure. Thus far, no clear association has been demonstrated between these two pathways. In this study, we examined the potential interaction between the PI3-kinase and PKCβ2 pathways by crossing transgenic mice with cardiac specific expression of PKCβ2, constitutively active (ca) PI3-kinase, and dominant-negative (dn) PI3-kinase. In caPI3-kinase/PKCβ2 and dnPI3-kinase/PKCβ2 double-transgenic mice, the heart weight-to-body weight ratios and cardiomyocyte sizes were similar to those observed in caPI3-kinase and dnPI3-kinase transgenic mice, respectively, suggesting that the regulation of physiological developmental hypertrophy via modulation of cardiomyocyte size proceeds through the PI3-kinase pathway. In addition, we observed that caPI3-kinase/PKCβ2 mice showed improved cardiac function while the function of dnPI3-kinase/PKCβ2 mice was similar to that of the PKCβ2 group. PKCβ2 protein levels in both dnPI3-kinase/PKCβ2 and PKCβ2 mice were significantly upregulated. Interestingly, however, PKCβ2 protein expression was significantly attenuated in caPI3-kinase/PKCβ2 mice. PI3-kinase activity measured by Akt phosphorylation was not affected by PKCβ2 overexpression. These data suggest a potential interaction between these two pathways in the heart, where PI3-kinase is predominantly responsible for the regulation of physiological developmental hypertrophy and may act as an upstream modulator of PKCβ2 with the potential for rescuing the pathological cardiac dysfunction induced by overexpression of PKCβ.
- transgenic mice
- constitutive active
- dominant negative
- protein kinase B
the phosphoinositide 3-kinases (PI3-kinases) generate lipids controlling an assortment of intracellular signaling pathways and having many diverse roles, including membrane trafficking along the endocytic pathway, cytoskeletal organization, adhesion, cell growth, and apoptosis (4, 8, 28, 29). Studies with transgenic (Tg) mice exhibiting heart-specific expression of constitutively active (caPI3-kinase) or dominant-negative (dnPI3-kinase) mutants have shown that PI3-kinase overexpression results in increased heart weight because of an increase in cardiomyocyte size, whereas dnPI3-kinase mice showed a decrease in heart size that correlated with a decrease in cardiomyocyte size (25). These data indicate that, for the heart, as in other systems, the PI3-kinase pathway is necessary for organ growth. PI3-kinase (p110α) has also been shown to be involved in the induction of exercise-induced physiological, but not pressure overload-induced pathological, cardiac hypertrophy (20, 22).
A pathway that is generally associated with pathological or maladaptive hypertrophy is protein kinase C (PKC), which consists of a family of serine-threonine kinases. To date, at least 11 isoforms of PKC have been discovered, and several of these isoforms, specifically PKCα, -β, -ε, and -ζ, have been widely studied for their mechanistic role in the cardiovascular system (23, 31). Heart failure, for example, which is often characterized by hypertrophy and a loss of cardiac function, is associated with an elevation in PKCα protein content and activity (1, 3). On the other hand, the role of PKCβ2 in cardiac hypertrophy and heart failure is not well defined. Overexpression of the PKCβ2 isoform in the Tg mouse heart leads to cardiac hypertrophy, fibrosis, and cell death, providing evidence that PKCβ2 is also involved in the structural and functional changes observed in pathological hypertrophy and heart failure (30). It has also been reported that overexpression of activated PKCβ2 causes induction of the promoters of β-myosin heavy chain (MHC) and skeletal α-actin, two marker genes for transcriptional signaling in hypertrophy (17, 18).
We therefore explored the potential interaction between the physiological PI3-kinase pathway and the pathological PKCβ2 pathway by crossing Tg mice with heart-specific PI3-kinase activation and PKCβ2 overexpression. We analyzed the effect of modulating the expression of these two pathways on the heart, specifically focusing on cardiomyocyte size and cardiac function.
MATERIALS AND METHODS
Generation of Tg mice.
caPI3-kinase and dnPI3-kinase Tg mice were produced by inserting the cDNA for iSH2p110 (with the Myc epitope tag) or a catalytically inactive p110 molecule (with a FLAG epitope tag) in the murine α-MHC promoter construct as described previously (25). The iSH2p110 gene is a chimeric molecule that contains the iSH2 domain of p85 fused to the NH2 terminus of bovine p110α by a flexible glycine linker that has been shown to function as a constitutively active molecule (9, 13). A catalytically inactive p110 molecule has a truncated p110 mutant that has p85 binding domains but lacks the kinase domain (p110kinase). It therefore competes with endogenous p110 for interaction with the p85 regulatory subunit and has an inhibitory effect on the function of the endogenous p110 molecule (33). PKCβ2 Tg mice were generated by inserting a 2.1-kb BamHI fragment of mouse PKCβ2 cDNA in the 3.3-kb EagI-SalI fragment of the plasmid containing rat MHC (α-MHC) promoter as previously described (30). Mice used in this study were of the FVB genetic background. All aspects of animal care and the experimental protocols used in this study were approved by the Institutional Animal Care and Use Committee of the Beth Israel Deaconess Medical Center.
The procedure was performed in 12- to 14-wk-old male mice as described previously (21, 22, 25). Briefly, after anesthesia with an isoflurane inhalant, echocardiography was performed with a Hewlett-Packard Sonos Agilent 5500 sector scanner equipped with a 15-MHz linear-array transducer. Anterior and posterior wall thickness and left ventricle (LV) internal dimensions were measured according to the leading-edge method of the American Society of Echocardiography (24).
Morphometric analysis of isolated cardiomyocytes.
Cardiomyocytes were enzymatically dissociated from the mouse heart according to the previously described protocol (21, 25). The hearts from 12- to 14-wk-old male Tg or wild-type (WT) mice were retrogradely perfused with 0.3% collagenase (Worthington). The dissociated cardiomyocytes were plated on laminin (10 μg/ml)-coated dishes. After 1 h of plating, unattached cells were removed by changing the media. Photographs were taken under a phase-contrast microscope, and cell surface area was determined by using the GNU Image Manipulation Program (GIMP 2.2.13) software (Open Source). Approximately 200 cells for each heart were measured, and the mean values for each heart were used for the statistical analysis.
Immunoblotting was performed as described previously (21, 22, 25). Protein concentration of whole heart tissue lysate was determined by the Bradford method (Bio-Rad), and glyceraldehyde-3-phosphate dehydrogenase (1:5,000) (RDI) was used as a loading control. Antibodies were diluted as follows: PKCβ2 (Santa Cruz), Akt (1:1,000; Cell Signaling), phospho-Akt (1:1,000; Cell Signaling).
Fractionation of whole heart tissue.
The hearts from 12- to 14-wk-old, male, WT, and Tg mice were snap-frozen in liquid nitrogen and homogenized in ice-cold 20 mM Tris·HCl (pH 7.5), 2 mM EDTA, 0.5 mM EGTA, 1 mM phenylmethlysulfonyl fluoride, 1 mM dithiothreitol, 0.3 M sucrose, and 25 μg/ml leupeptin (buffer A) with a Polytron electric homogenizer for 20 s and then with a Dounce homogenizer (60 strokes). The homogenates were centrifuged at 2,500 revolutions/min (rpm) for 10 min at 4°C, and the supernatant was ultracentrifuged at 40,000 rpm for 30 min at 4°C. The resulting supernatant was retained as the cytosolic fraction, and the pellets were resuspended with buffer A without sucrose (buffer B) and solubilized with 1% Triton X-100. After being rotated for 45 min at 4°C, soluble membrane fractions were obtained by ultracentrifugation at 40,000 rpm for 30 min. The membrane pellet was resuspended in buffer B by Dounce homogenization. PKCβ2 expression levels in the cytosol and membrane fractions were shown by Western blot.
Hearts were fixed in 10% formalin and paraffin-embedded. Sections were stained with hematoxylin and eosin and Masson Trichrome (MT) at the Histology Core facility at Beth Israel Deaconess Medical Center. Fibrosis was quantified at ×20 using a calibrated digital camera and software (DP 70 and DPController; Olympus America, Irving, TX) to survey entire heart sections. All areas of fibrosis were scored as either major (MT staining consisting of >10%/visual field) or minor (MT staining in <10%/visual field or presence of perivascular fibrosis).
RT-PCR analysis for PKCβ2 mRNA expression.
Total RNA was extracted from mouse ventricular tissue using TRIzol (GIBCO-BRL, Gaithersburg, MD). RT-PCR was performed by PCR amplification of cDNA reverse transcribed from mRNA using mouse PKCβ2 and 18S primers. The following primer sequences were used: PKCβ2 primer, forward 5′-ATTCCAGTGTCAAGTCTGCT and reverse 5′-CCCATGAAGTCATTCCTGCT; 18S primer, forward 5′-GTTATGGTTCC TTTGTCGCTCGCTC and reverse 5′-TCGGCCCGAGGTTATCTAGAGTCAC.
Adult rat cardiomyocyte culture.
Adult rat cardiomyocyte cultures were prepared from the hearts of female Sprague-Dawley rats using retrograde perfusion with 0.3% collagenase (Worthington Biochemical, Lakewood, NJ) as described previously (2, 11, 16).
All data were expressed as means ± SE. Between-group and among-group comparisons were conducted with unpaired Student's t-tests and ANOVAs, respectively. P values of <0.05 were considered significant.
PI3-kinase regulates heart weight-to-body weight ratio in double-Tg mice.
Tg caPI3-kinase mice showed a significant increase in the heart weight-to-body weight (HW/BW) ratio of ∼41% compared with WT, whereas dnPI3-kinase mouse hearts were significantly smaller (14%) than hearts of WT littermates (Fig. 1A and Table 1). PKCβ2 mice showed an increase (19%) in HW/BW ratio compared with WT controls. In crossing experiments, caPI3-kinase/PKCβ2 double-Tg mice exhibited an increase in HW/BW ratio similar to that of caPI3-kinase mice compared with WT mice, whereas dnPI3-kinase/PKCβ2 double-Tg mice displayed a 16% decrease in HW/BW ratio compared with WT mice. Thus the HW/BW ratios observed in double-Tg mice were similar to those of either caPI3-kinase or dnPI3-kinase, respectively (Fig. 1A and Table 1). The heart weight to tibia length ratios followed the same pattern as HW/BW ratios. However, as we have observed previously, the lung weight-to-body weight ratio did not differ significantly among the groups (22, 25). These findings suggest that the PI3-kinase pathway is dominant in the regulation of physiological developmental hypertrophy, whereas PKCβ2 does not appear to play a significant role in this process.
PI3-kinase pathway dominates in the regulation of cardiomyocyte size in the presence of PKCβ2.
To determine whether the differences in heart size in the Tg mice are related to changes in cardiomyocyte size, we measured the surface area of dissociated cells from each of the genetically altered mouse groups. We found that caPI3-kinase mice exhibit an increase (19%) whereas dnPI3-kinase mice exhibit a decrease (18%) in mean cell area compared with WT controls; these data are consistent with our previous report (Fig. 1, B and C, and Ref. 25). We also found that cardiomyocyte cell size was not increased in PKCβ2 mice, suggesting that the increase observed in the HW/BW ratio is most likely the result of an increase in fibrosis or nonmyocyte cellular proliferation. Among the double-Tg groups, caPI3-kinase/PKCβ2 mice showed a mean cell area comparable to caPI3-kinase littermates, and the mean cell area in dnPI3-kinase/PKCβ2 mice was comparable to dnPI3-kinase. From these findings, we conclude that the PI3-kinase pathway dominates in the regulation of physiological developmental hypertrophy by directly controlling the size of cardiomyocytes.
caPI3-kinase rescues cardiac function.
Next, we used echocardiography to assess cardiac function in double-Tg mice. There was a significant increase in intraventricular septum wall thickness in caPI3-kinase (27%) and caPI3-kinase/PKCβ2 Tg mice (40%) compared with the WT group (Table 2), but the difference between caPI3-kinase and caPI3-kinase/PKCβ2 Tg mice was not significant. Left ventricular posterior wall in diastole did not differentiate the groups either (0.84 mm for both), suggesting that there is no synergistic increase in LV wall thickness in double-Tg mice. LV systolic performance was measured by assessing the mean fractional shortening (FS) of each group. FS in caPI3-kinase and dnPI3-kinase mice did not differ from WT littermates. In contrast, PKCβ2 mice showed a decrease in cardiac function, as reported previously (30). Interestingly, cardiac performance in caPI3-kinase/PKCβ2 mice was improved significantly (44%) compared with PKCβ2 mice (Table 2). On the other hand, FS in dnPI3-kinase/PKCβ2 mice decreased (14%) and was similar to that of PKCβ2 mice. These findings indicate that the decreased heart function in PKCβ2 mice may be restored by an increase in the expression level of PI3-kinase in caPI3-kinase/PKCβ2 double-Tg animals.
Because the deterioration of cardiac function seen in PKCβ2 mice may be related to increased fibrosis, we investigated the presence of cardiac fibrosis using MT staining of the heart tissue sections. We found that PKCβ2 Tg mice showed a significant increase in MT staining, suggesting increased fibrosis (Fig. 2, A and B). In contrast, caPI3-kinase/PKCβ2 double-Tg mice showed a significant decrease in cardiac fibrosis compared with PKCβ2 Tg mice, suggesting that the overexpression of caPI3-kinase suppressed the cardiac fibrosis observed in PKCβ2 Tg mice.
PI3-kinase regulates PKCβ2 expression.
To determine the potential mechanism of improved cardiac function, we compared PKCβ2 expression levels among different Tg mice. PKCβ2 mice, as expected, exhibited a high expression level compared with WT, dnPI3-kinase, and caPI3-kinase mice (29-, 71-, and 59-fold increase, respectively). We also observed translocation of PKCβ2 to the membrane, suggesting PKCβ2 activation (Fig. 3A). In dnPI3-kinase/PKCβ2 mice, the expression level of PKCβ2 was comparable to the protein expression level found in PKCβ2 mice and significantly higher than in WT mice. Surprisingly, the PKCβ2 protein level was completely suppressed in caPI3-kinase/PKCβ2 mice, and it was similar to the baseline level found in WT mice (Fig. 3B). Furthermore, the mRNA expression level of PKCβ2 was also significantly decreased in caPI3-kinase/PKCβ2 double-Tg mice (Fig. 3C). Finally, we verified that PKCβ2 expression is regulated by PI3-kinase endogenously, using insulin to activate PI3-kinase in adult rat cardiomyocyte cultures. Activating PI3-kinase signaling with insulin significantly suppressed endogenous PKCβ2 (Fig. 3, D and E). These data support the previous results indicating that transcriptional repression of PKC expression in the caPI3-kinase/PKCβ2 double-Tg mouse heart significantly downregulates the expression of PKCβ.
PKCβ2 does not alter PI3-kinase activity.
To determine whether the overexpression of PKCβ2 in caPI3-kinase/PKCβ2 mice affects the PI3-kinase-protein kinase B (Akt) pathway, we measured the phosphorylation of Akt, which has been used extensively as evidence of PI3-kinase activity, in all studied groups using a phosphospecific antibody. There was no significant activation of Akt phosphorylation in PKCβ2 Tg mice, suggesting that PKCβ2 by itself does not alter PI3-kinase activity at baseline (Fig. 4, A and B). However, we observed, as expected, that the level of phosphorylated Akt was significantly higher in caPI3-kinase than in WT mice. This increased Akt phosphorylation was not changed significantly in caPI3-kinase/PKCβ2 mice compared with caPI3-kinase (Fig. 4, A and B). These data suggest that the protein level of Akt, a downstream target of PI3-kinase, was not altered significantly by overexpression of PKCβ2 in the double-Tg animals. Thus PKCβ2 does not appear to regulate PI3-kinase expression and is most likely a downstream target of PI3-kinase.
In this study, we examined potential interactions between the PI3-kinase and PKCβ2 pathways by crossing Tg mice with cardiac specific expression of caPI3-kinase, dnPI3-kinase, and PKCβ2. First, we found that the PI3-kinase pathway dominates in the regulation of physiological developmental hypertrophy by directly controlling the size of cardiomyocytes. Second, caPI3-kinase was able to rescue the cardiac dysfunction of PKCβ2 mice, suggesting that activation of the PI3-kinase pathway can overcome the deleterious effect of PKCβ2 in the heart. Finally, we show for the first time that there may be an interaction between the PI3-kinase and PKCβ2 pathways and that PI3-kinase is most likely an upstream regulator of PKCβ2.
Our study further supports the notion that class IA PI3-kinases are critical regulators for the developmental growth and physiological hypertrophy of the heart (20). Cardiomyocyte sizes in double-Tg caPI3-kinase/PKCβ2 and dnPI3-kinase/PKCβ2 mice were comparable to those found in caPI3-kinase or dnPI3-kinase littermates, respectively. However, despite increased heart weight, cardiomyocyte size in PKCβ2 mice was similar to that found in the WT control group. We thus infer that PKCβ2 is not involved in the regulation of physiological developmental hypertrophy. The observed increase in HW/BW ratio is most likely because of cardiac fibrosis. Indeed, Way et al. (32) reported an increase of connective tissue growth factor expression in the myocardium of diabetic PKCβ2 Tg mice, and associated this with the induction of cardiac fibrosis and dysfunction.
The PKC family is traditionally divided into the following three subgroups: the classic or conventional (α, β1, β2, γ), the novel (δ, ε, η, and θ) and the atypical (ζ, ν, μ, and ι). Several of these PKC isoforms may play a crucial role in the signaling pathways involved with both pathological and physiological hypertrophy (7). Members of the classic group, specifically PKCα and β, have been widely studied for their role in the cardiovascular system and act similarly by translocating from cytosol to membrane by the induction of ANG II, an oligopeptide known to cause vasoconstriction and increased blood pressure in vivo as well as cellular hypertrophy in vitro (14, 19, 26). PKCα, one of the more widely studied isoforms, has been identified as a regulator of cardiac contractility and calcium handling in cardiomyocytes, although it has not been shown to directly affect the hypertrophic growth response of the adult heart (3). We found that the decreased cardiac function exhibited by PKCβ2 mice was rescued when they were crossed with caPI3-kinase mice. In turn, crossing PKCβ2 with dnPI3-kinase mice (dnPI3-kinase/PKCβ2) did not improve cardiac function compared with PKCβ2 mice. Of note, the PKCβ2 mice used in this study originated in a Tg mouse line with a less significant phenotype than the Tg line originally reported (30). Nevertheless, this PKCβ2 Tg line is also characterized by a significant increase in cardiac fibrosis and cardiac dysfunction compared with WT mice. Our observations revealed that the cardiac dysfunction associated with increased PKCβ2 expression can be improved by an increase in PI3-kinase activity.
Our findings reveal the potential existence of an interaction between PI3-kinase and PKCβ2, either directly or indirectly, through a downstream target of the PI3-kinase pathway. We found that PKCβ2 expression is attenuated significantly by the activation of PI3-kinase, indicating that PI3-kinase may inhibit protein expression levels of PKCβ2, thus explaining the attenuation of the pathological characteristics of PKCβ2 mice in caPI3-kinase/PKCβ2 mice. Although an interaction of PKCζ and PI3-kinase has been demonstrated previously (27), the present study is the first to demonstrate that PKCβ2 is regulated by PI3-kinase activation. In fact, downregulation of PI3-kinase in dnPI3-kinase/PKCβ2 double-Tg mice did not affect the expression of PKCβ2 protein, suggesting that PI3-kinase is a negative regulator of PKCβ2.
Our data infer that PI3-kinase is an upstream regulator of PKCβ2, since PKCβ2 does not appear to change the protein levels of downstream targets of PI3-kinase. One of the possible downstream targets of PI3-kinase that may mediate its effect on PKCβ2 is Akt. Akt, which is activated by both receptor tyrosine kinases/cytokines (4, 8) and G protein-coupled receptors (5, 6), is a likely potential mediator of PKCβ2 protein level in mammalian cells, since it has been shown to regulate other downstream targets of PI3-kinase such as P70S6K1 and glycogen synthase kinase (GSK-3) (10, 15). In previously published data, Akt was found to regulate GSK-3β, which was identified as negatively modulating cardiomyocyte hypertrophy as well as mediating cardioprotection postischemia-reperfusion injury (12, 34). In turn, we found that Akt activity was increased in caPI3-kinase and caPI3-kinase/PKCβ2 mice.
Studying lipid kinases involved in signal-transduction cascades that mediate cardiac protection and cellular growth may provide insight into potential interventions for heart failure. We have demonstrated that cross talk between PI3-kinase and PKCβ2 pathways enables the pathological characteristics of PKCβ2 mice to be attenuated in caPI3-kinase/PKCβ2 mice. Further studies are required to determine the specific interactions between the PI3-kinase pathway and PKC isoforms, as well as the intermediary molecules involved.
This study was supported, in part, by National Heart, Lung, and Blood Institute Grant RO1-HL-65742 (P. M. Kang).
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.
- Copyright © 2009 the American Physiological Society