INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SIGNAL TRANSDUCTION proteins that may be involved in cardiac hypertrophy have been the focus of many studies in recent years. Among these proteins of interest are G_q, phospholipase C (PLC), and protein kinase C (PKC). It has been postulated that PKC plays a central role in the modulation of cardiac gene expression and hypertrophy (10, 14, 24). Once activated, PKC affects activity of transcriptional factors and gene expression (4, 13, 20), voltage-dependent calcium channels (25), Na⁺/H⁺ exchangers (21), sarcoplasmic reticular proteins (11), and myofilament proteins (22). Recently, a transgenic mouse with cardiac specific overexpression of the PKC-II isoform produced elevated total PKC activity, a hypertrophic phenotype, and cardiac failure. However, when this model was treated with oral administration of a specific PKC- inhibitor, the hypertrophic phenotype was attenuated (24). Alterations in PKC are also found in humans with end-stage heart failure as evidenced by increased translocation of PKC-, -I, and -II isoforms and elevated total PKC activity (3).

Upstream signals leading to PKC activation include heterotrimeric G protein (G_q)-coupled receptors and PLC-I. These elements integral to the PKC response may also participate in the hypertrophic response. Furthermore, the biochemical relationship that exists between them may affect downstream PKC responses. Factors that influence G_q activity include the regulator of G protein signaling (RGS) family of proteins, specifically RGS2, RGS3, and RGS4, which are activators of the intrinsic GTPase activity of G_q (9). Thus by modifying the function of G_q, RGS proteins can have the capacity to influence G_q effectors. The status of RGS2 during cardiac hypertrophy is presently unknown, whereas RGS3 and RGS4 are reported to be attenuated during cardiac hypertrophy in the spontaneously hypertensive, heart-failure prone (SHHF) rat (26).

Many of the components involved in signal transduction through the PKC pathway have been studied in neonatal cardiomyocytes and in in vitro models (27). However, are there little data available regarding the abundance and activity of proteins involved in this signal transduction pathway in the adult heart. Thus the purpose of this study was to use a well-defined animal model of cardiac hypertrophy to 1) characterize G_q, RGS2, RGS3, RGS4, PLC-I, and PKC protein abundance during compensated and decompensated cardiac hypertrophy, 2) assess whether and to what extent any alterations exist in the stoichiometric relationship among these proteins that may serve to explain any modulation of PKC activation, 3) characterize the enzymatic activity of PKC in this model, and 4) determine cardiac function during these hypertrophic conditions using an isolated heart preparation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animal modeling. All procedures were done in accordance with the University of Cincinnati animal care guidelines, which conform with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. Pressure-overload hypertrophy (POH) and decompensated congestive heart failure (CHF) were induced in male, adult, Hartley strain guinea pigs (age 8 wk, 250-300 g) by subtotal descending thoracic aortic banding as previously described (5, 15). Briefly, after anesthesia with pentobarbital sodium (25 mg/kg ip), a tracheal cut-down was performed and animals were intubated with a 20-gauge angiocatheter and ventilated by a Harvard rodent ventilator (Harvard Apparatus, model 683, South Natick, MA) at 25-55 breaths/min (tidal volume 1.5-2.0 ml). The descending thoracic aorta was exposed by an intercostal incision, and a uniform degree of constriction around the aorta was produced by tying a 2-0 surgical silk ligature tightly around a 6-mm length of hypodermic tubing having an external diameter of 1.25 mm. After ligature placement the hypodermic tubing was removed. Sham-operated controls underwent the same operation as banded animals except for permanent suture placement. All surgery was performed by the same investigator. Animals were housed and fed under identical conditions and killed, 4 or 8 wk after surgical modeling. After they were killed, animals that met three specific standards set a priori [lung-body weight 8.0 × 10², developed pressure  90 mmHg, and the maximal rate of pressure development (dP/dt_max)  1,500 mmHg/s] were classified as CHF. Animals manifesting left ventricular (LV) hypertrophy by having LV body weight  3.0 × 10³, yet with hemodynamic criteria matching sham-operated controls, were classified as compensated POH. After stratification of animals by these criteria, we observed that a majority of animals killed at 4 wk had POH and at 8 wk CHF.

Isolated perfused heart preparation and cardiac mechanics. Guinea pigs were anesthetized with 0.5 ml of a mixture containing ketamine (54 mg/kg), acepromazine maleate (1.8 mg/kg), and xylazine (10.9 mg/kg) and heparinized by injecting 200 U of heparin sodium (1,000 U/ml) into the abdominal aorta. Beating hearts were quickly excised, weighed, and then perfused using a modified Langendorff system with the ascending aorta terminally cannulated as previously described (15). The Krebs-Henseleit buffer containing (in mmol/l) 113.8 NaCl, 4.7 KCl, 1.10 MgSO₄, 0.12 KH₂PO₄, 23.6 NaHCO₃, 2.5 CaCl₂, 6.0 mannitol, and 11.0 glucose, pH 7.4-7.5, was saturated with 95% O₂-5% CO₂ and passed through a 0.45-mm aeration stone. An inline countercurrent heat exchanger was used to warm the buffer to 37°C. A water-filled latex balloon attached to the end of a 3-Fr Millar high-fidelity micromanometer catheter (Millar Instruments, Houston, TX) was inserted into the LV through the mitral valve orifice for pressure measurements. All hearts were paced with a Grass stimulator (model S88, Grass Instruments, Quincy, MA) to achieve a constant heart rate of 250-300 beats/min. The right ventricle was vented, and the LV balloon was inflated sufficiently to obtain a minimal diastolic pressure at which developed pressure was maximal and was kept isovolumic during initial perfusion. The hearts were perfused at a constant flow rate of 10 ml · g¹ · min¹ for the duration of the experiment.

Tissue protein extraction. Clear cell lysate was prepared from the guinea pig LV frozen at 70°C as detailed by the research applications manual from Santa Cruz Biotechnology (Santa Cruz, CA). Samples for PKC analysis were prepared separately. LV samples to be used for PKC activity were extracted and partially purified according to the method detailed by Wakasaki et al. (24). Briefly, all extraction procedures were performed at 4°C, and 0.3 g of the LV was excised and homogenized in 15 vols of ice-cold buffer A [Tris · HCl (pH 7.2) with 2 mM EDTA, 0.5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM dithiothreitol (DTT), 0.3 M sucrose, and 25 µg/ml leupeptin] and centrifuged at 800 g for 20 min. After homogenization, samples were subjected to ultracentrifugation at 100,000 g to separate cytosolic- and membranous-bound cell fractions. Membrane fractions were homogenized in buffer B (buffer A with 0.5% Triton X-100) and centrifuged at 100,000 g for 1 h. Both cytosolic and membranous bound fractions were passed through a DEAE Sephacel (Pharmacia) column equilibrated with Tris · HCl (pH 7.2) with 2 mM EDTA, 0.5 mM EGTA, 1 mM PMSF, 1 mM DTT, and 25 µg/ml leupeptin, the column was washed two times with equilibration buffer, and partially purified PKC was eluted out of the column with the same buffer supplemented with 200 mM NaCl.

PKC activity assay. Protein concentrations of samples were measured (Bio-Rad protein assay, Bio-Rad, Hercules, CA) using bovine serum albumin as a standard, samples were then stored in 50% glycerol, and PKC activity assays were performed within 24 h of extractions. PKC activity was assessed using the BioTrak kit from Amersham (Amersham Life Science, Arlington Heights, IL). This kit utilizes calcium, L--phosphatidyl-L-serine, and phorbol 12-myristate 13-acetate to stimulate PKC-mediated transfer of [³²P]ATP to a PKC-specific peptide substrate (Arg-Lys-Arg-Thr-Leu-Arg-Arg-Leu).

Preparation of immunoblots. Relative protein levels were estimated by quantitative Western immunoblots. Membrane- bound and cytosolic cell fractions from sham, POH, and CHF hearts were used to assess PKC isoforms , I, II, , , and . Whole tissue lysate was used for RGS2, RGS3, RGS4, G_q, and PLC-I immunoblots. In brief, samples were subjected to electrophoresis in 10% SDS polyacrylamide gels using a Bio-Rad mini gel apparatus (Bio-Rad, Hercules, CA) under denaturing conditions in 25 mM Tris, 250 mM glycine, and 0.1% SDS running buffer at pH 8.3. Optimal loading conditions were determined for each protein of interest on the basis of preliminary trials designed to optimize immunoreactivity; PKC- was loaded with 10 µg of lysate, PKC- with 8 µg of lysate, PKC- with 50 µg of lysate, G_q, RGS2, and PLC-I with 30 µg of lysate, and RGS3 and RGS4 with 50 µg of lysate. Gels were prepared in duplicate. After electrophoresis, one gel was stained with Coomassie blue R-250 to verify equal protein loading of all samples and the other gel was used for protein transfer to 0.45-mm supported nitrocellulose membrane overnight at 4°C using a Bio-Rad transfer apparatus. All membranes were also stained with Ponceau S to confirm equal transfer efficiency. Membranes were cut to allow for simultaneous probing of the protein of interest and for calsequestrin to confirm equal protein loading when appropriate (calsequestrin antibody was generously donated by Dr. Larry R. Jones, Indiana University School of Medicine). Calsequestrin was chosen as our housekeeping protein because it is accepted as being unchanged during cardiac hypertrophy (19). After a period of blocking using 5% nonfat milk-TBS solution (50 mM Tris · HCl, pH 7.4, 0.9% NaCl), membranes were incubated with appropriate primary antibody overnight at 4°C in a dilution appropriate to the protein of interest (1:1,000 for G_q, RGS2, PLC-I, PKC-, PKC-, and PKC-; 1:667 for RGS4; 1:1,500 for PKC- and RGS3; 1:2,000 for PKC-I, PKC-II, and calsequestrin). After three washes, membranes were incubated with secondary antibody for 1 h (goat anti-rabbit or rabbit anti-goat peroxidase labeled, KPL Laboratories, Gaithersburg, MD) in a dilution of 1:10,000 for PKC-I, PKC-II, PKC- and 1:20,000 for all others. Signals were visualized by enhanced chemiluminescence (Amersham Life Science). PKC isoforms, G_q, PLC-I, and RGS4 primary antibodies were purchased from Santa Cruz (Santa Cruz Biotechnology, Santa Cruz, CA). RGS2 antibody was custom made on the basis of the peptide sequence EDFKKTKSPQKLSSKARK by Research Genetics (Huntsville, AL). Anti-RGS3 primary antibody was a kind gift from Dr. Anthony Muslin (Washington University School of Medicine). Antibody specificity for all proteins was verified by attenuation or abolition of signal with isoform specific inhibitory peptides.

Statistical analysis. The data are expressed as the means ± SE. Statistical analyses were performed using one-way ANOVA followed by Bonferroni test when significant main effects were present. Significance was accepted at P  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

After they were killed, guinea pigs were stratified as compensated POH or CHF by gravimetric and LV functional data (see MATERIALS AND METHODS). Nine banded guinea pigs met criteria for classification as POH and thirteen for CHF as detailed on Table 1. All subsequent biochemical data reported here was organized based on this stratification.

Guinea pigs that were stratified as being in compensated POH had significantly hypertrophied LVs, but with no change in lung-body weight (Table 1) by experimental design. Animals in POH also displayed normal cardiac function as evidenced by indexes of developed pressure and cardiac contractility (dP/dt_max and dP/dt_min). In contrast, guinea pigs in CHF had a marked deterioration of cardiac performance with decreased developed pressure and impaired parameters of contractility and relaxation concomitant with pulmonary congestion and cardiac hypertrophy (Table 1).

Immunoblotting of PKC isoforms, PLC-I, G_q, RGS2, RGS3, and RGS4. Coomassie blue-stained gel and calsequestrin immunoblots were used to verify equal loading of protein samples and to normalize for any differences in protein loading when required (Fig. 1, A and B). Western blot analysis of G_q indicated no change in immunoreactivity among the three groups (Table 2). PLC-I was also found to be unchanged during POH and CHF compared with control (Fig. 1C, Table 2). In our study PLC-I was found to migrate as a double band with a minor band at 150 kDa (not shown) and a major band slightly smaller than 120 kDa. Data shown in Fig. 1C are representative of the major band found near 120 kDa. Because the accepted molecular weight for PLC-I is 150 kDa, we suspect that proteolysis or glycosylation was responsible for this doublet pattern. A peptide competitor used to verify the migration of PLC-I was found to compete out both bands at 150 and 120 kDa.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Representative immunoblots of calsequestrin, phospholipase C (PLC)-I and regulator of G protein signaling (RGS) 2 using left ventricular whole cell lysates from sham-operated control (Con), compensated pressure-overload cardiac hypertrophy (POH), and congestive heart failure (CHF). A: representative acrylamide gel stained with Coomassie blue to verify equal loading of protein. B: representative immunoblot of calsequestrin, which demonstrates no change during cardiac hypertrophy, indicates equal loading of protein on each lane. C: relative protein levels of PLC-I are unchanged during POH and CHF. D: RGS2, previously uncharacterized in the heart, is a 24-kDa protein that can be competed out using peptide competitor specific for the RGS2 primary antibody. Relative levels of RGS2, activator of intrinsic GTPase activity, are unchanged during POH and CHF.

View this table: [in this window] [in a new window]   Table 2.   Quantitative immunoblotting analysis of Gq, PLC-I, RGS2, RGS3, and RGS4 from left ventricle of guinea pigs banded at descending thoracic aorta

View this table: [in this window] [in a new window]   Table 3.   Quantitative immunoblotting analysis of unchanged PKC isoforms (II and ) from left ventricle of guinea pigs banded at descending thoracic aorta

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Protein kinase C (PKC) immunoreactivity during transition between hypertrophy and heart failure. Membranous and cytosolic fractions were prepared from guinea pig left ventricle. Representative immunoblots for PKC-, -, and - demonstrate greater immunoreactivity in membrane fraction during POH and CHF along with significant increase in translocation as evidenced by membrane-to-cytosol ratio (n = 6 for all groups).

PKC activity. PKC activity was measured in membranous and cytosolic extracts by counting transfer of phosphate from [-³²P]ATP to a PKC-specific peptide after stimulation with calcium, L--phosphatidyl-L-serine, and phorbol 12-myristate 13-acetate. Examination of the POH membrane fraction exhibited a reasonable trend for PKC activity to be increased but not significantly so (POH 7.54 ± 4.65 vs. control 3.83 ± 1.22 pmol · min¹ · µg¹, Fig. 3, P = 0.08). The lack of significance in membrane-bound activity, however, may be due to the high sample variation or small sample size. POH cytosolic PKC activity, however, was significantly greater compared with that of control animals (13.84 ± 3.71 vs. 8.61 ± 2.63 pmol · min¹ · µg¹, Fig. 3, P < 0.05). No statistical change was observed in the cytosolic or membranous PKC activity of the CHF group compared with controls.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Cytosolic (PKC-C) and membranous (PKC-M) activity during POH and CHF compared with sham-operated controls. PKC was partially purified from guinea pig left ventricle and separated into membranous and cytosolic fractions. PKC activity was assessed by measuring transfer of phosphate from [32P]ATP to a PKC-specific substrate peptide. PKC activity is increased in cytosolic fraction during POH but not CHF. * P  0.05 vs. control. Activity units are picomoles of phosphate transferred per minute per microgram of protein (n = 6 for each group).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this study we employed a well-characterized animal model that clearly displays compensated POH (LV hypertrophy, normal LV function, and no pulmonary congestion) and CHF (LV hypertrophy, depressed LV function, and pulmonary congestion) to examine the role of signal transduction proteins involved in PKC activation. The major findings are as follows: 1) RGS2, previously uncharacterized in the heart, was found to exist in left ventricular myocardium, yet the protein abundance was unchanged in control, compensated POH, and CHF guinea pigs; 2) no changes were found in protein abundance of G_q, PLC-I, RGS3, and RGS4 during POH and CHF; and 3) differential PKC isoform abundance and activity were observed during POH and CHF.

One of the goals of this study was to assess any alterations in the stoichiometry among G_q, RGS proteins, and PLC-I during the transition between POH and CHF. This is a potentially important issue because changes in the relative abundance of any of these molecules may modulate activation of downstream signaling proteins such as PKC. Furthermore, this group of signaling molecules function interdependently such that G_q half-life can be influenced by both PLC-I levels in the cell (2) and RGS molecules (9). Using reconstituted lipid vesicles, Berstein et al. (2) and others have demonstrated that the rate of GTP hydrolysis increases up to 50-fold after addition of PLC-I. The same study also found that PLC-I-mediated GTP hydrolysis was at one-half of maximal velocity when PLC-I levels were twofold greater than G_q and at maximal velocity when PLC-I levels were 20- to 30-fold greater than G_q (2). RGS molecules can also influence G activity by regulating the intrinsic GTPase activity. RGS2 has been found to specifically activate the GTPase activity of G_q (9), as has RGS4 (8). These studies provided us with a rationale for examining the levels of G_q, PLC-I, and RGS proteins during cardiac hypertrophy.

We examined a group of recently characterized signaling proteins, RGS2, RGS3, and RGS4, that are known to modulate the intrinsic GTPase activity of G_q. Any modulation of G_q could affect the response to a variety of known ligands, such as angiotensin II, endothelin I, and other -adrenergic agonists. To our knowledge, this is the first study to examine RGS2 status in in vivo hearts. Using a custom-made antibody, we found that RGS2 exists in the heart, but the protein abundance is unchanged during POH and CHF. Furthermore, we found no change in protein abundance of either RGS3 and RGS4 during POH and CHF. This is in contrast to a recent study done by Zhang et al. (26) in which a decrease in mRNA and protein levels of RGS3 and RGS4 were observed using whole homogenates of failing hearts from SHHF rats.

G_q has been implicated previously in the hypertrophic process, inasmuch as transgenic mice with cardiac specific overexpression of G_q display a phenotype of cardiac hypertrophy (6). Furthermore, the hypertrophic response to pressure overload has been found to be attenuated in transgenic mice that overexpress a peptide inhibitor specific to G_q (1). Despite transgenic studies that demonstrate G_q overexpression produces cardiac hypertrophy, our study did not reveal any changes in protein levels of G_q in this model of POH and CHF. To our knowledge PLC-I status is unknown during cardiac hypertrophy. A previous study examining PLC-I and G_q in scar tissue after myocardial infarction did report increases in expression of both of these signaling molecules at the mRNA and protein level (12). In our study we did not observe any variation in PLC-I protein level. On the basis of our data concerning PLC-I and G_q, we conclude that the stoichiometric relationship between these proteins was unchanged during the POH and CHF phases of cardiac hypertrophy. However, it is critical to point out that the question of enzymatic activity remains unanswered. Because changes in enzymatic activity of G_q and PLC-I in our model remain unknown at this time, it is possible that enzymatic activity can be modified without any alteration in protein abundance. We hypothesize that differential PKC activation may occur via an increase in activity of G_q and PLC-I rather than an upregulation of protein synthesis or via an unknown pathway. Further studies are required to assess the activity of G_q and PLC-I in our conventional models of POH and CHF.

Earlier studies examining aortic-banded rats indicated that PKC- isoforms and PKC- are significantly upregulated in the membrane and cytosol fraction and that this change was accompanied by an increase in Ca²⁺-dependent PKC activity in the same fractions (7). Similar results have been obtained in humans with end-stage heart failure as well (3). In the present study PKC-, -, and - protein abundance were augmented in a continuous fashion from compensated POH through heart failure, concomitant with a 61% increase in cytosolic PKC activity and a nonsignificant 97% increase in membranous PKC activity. The variable PKC isoform response observed among rats, guinea pigs, and humans during cardiac hypertrophy is suggestive of both a role for PKC during cardiac hypertrophy and a species-specific response of PKC isoforms. In contrast to chronic pressure overload, acute mechanical stretch of the myocardium activates only PKC- in the adult guinea pig heart (16).

It has been demonstrated that redistribution of PKC does not correlate in extent or duration with phosphorylation of PKC substrates, suggesting that translocation may not always equate to activity (23). When guinea pig hearts were subjected to oxidative stress using hydrogen peroxide, selective translocation of PKC isoforms was observed with no increase in PKC activity (unpublished data from our laboratory). It has also been reported that measurements of PKC activity are not sufficiently sensitive to detect the involvement of PKC in ischemic preconditioning (18). In the present study, we observed that animals during CHF demonstrate a greater abundance and translocation of PKC isoforms ,  and  but with little increase in PKC activity. Although the membranous protein expression of an individual PKC isoform is usually tightly coupled to its isoform-selective phosphorylation activity, the total activity of PKC measured in the present study assembles the phosphorylation activity of all isoforms expressed in the heart (17). Thus it is possible that because only the isoforms , , and  showed translocation, total PKC activity did not change in CHF. Another possibility exists in that PKC enzymatic activation may be an early event in response to pathological stimuli that leads to hypertrophic growth of the heart as seen in the transition between control and POH animals. The cessation of protein synthesis often observed during end-stage heart failure may be in part explained by the reduction of PKC activity to control levels in the present study, even though PKC isoforms are still abundant in quantity and translocation in CHF. Discrepancies between PKC translocation and activity may be a consequence of complicating effects of altered protein synthesis coupled with cellular redistribution of PKC isoforms. With respect to experimentally induced cardiac hypertrophy by pressure overload, it seems logical for peak levels of PKC activity to occur during POH because this is the most active period of cardiac remodeling.

In summary, we have found that differential PKC translocation and activation occur in pressure overload-induced left ventricular hypertrophy and failure, but the stimulus for this activation is still unclear.