We have examined the changes in quantity and activity of cardiac sarcolemmal (SL) phosphoinositide-phospholipase C (PLC)-β1, -γ1, and -δ1 in a model of congestive heart failure (CHF) secondary to large transmural myocardial infarction (MI). We also instituted a late in vivo monotherapy with imidapril, an ANG-converting enzyme (ACE) inhibitor, to test the hypothesis that its therapeutic action is associated with the functional correction of PLC isoenzymes. SL membranes were purified from the surviving left ventricle of rats in a moderate stage of CHF at 8 wk after occlusion of the left anterior descending coronary artery. SL PLC isoenzymes were examined in terms of protein mass and hydrolytic activity. CHF resulted in a striking reduction (to 6–17% of controls) of the mass and activity of γ1- and δ1-isoforms in combination with a significant increase of both PLC β1 parameters. In vivo treatment with imidapril (1 mg/kg body wt, daily, initiated 4 wk after coronary occlusion) improved the contractile function and induced a partial correction of PLCs. The mass of SL phosphatidylinositol 4,5-bisphosphate and the activities of the enzymes responsible for its synthesis were significantly reduced in post-MI CHF and partially corrected by imidapril. The results indicate that profound changes in the profile of heart SL PLC-β1, -γ1, and -δ1 occur in CHF, which could alter the complex second messenger responses of these isoforms, whereas their partial correction by imidapril may be related to the mechanism of action of this ACE inhibitor.
- myocardial infarct
- signal transduction
- phospholipase C isoenzymes
- angiotensin-converting enzyme inhibition
- phosphatidylinositol 4,5-bisphosphate
phosphoinositide-phospholipase C (PLC) is an inositol phospholipid phosphodiesterase that is involved in numerous transmembranal signals (reviewed in Ref. 41). Its most common physiological substrate, phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2], is converted into two messenger molecules, inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] andsn-1,2-diacylglycerol (DAG), which participate in many different physiological processes. The membrane level of PtdIns(4,5)P2 is also an important signaling factor, as this phosphoinositide is a membrane-attachment site and/or an essential requirement for the function of several proteins (28). The three known classes of mammalian PLC (β, γ, and δ) comprise at least 10 isoforms (41) and display differences in structure, function, and activating mechanisms in response to stimulation of specific cell-surface receptors (24, 28, 41,43). ANG II, α1-adrenergic agonists, and endothelin-1 are relevant stimulants of PLC isoenzymes via the α-subunits of the heterotrimeric Gq subfamily (41). Binding of polypeptide growth factors to their receptors with intrinsic or associated tyrosine kinase activity activates PLC-γ isoenzymes, whereas receptor-initiated events for the activation of PLC-δ isoenzymes are presently unclear (41).
Although little is known about the number and characteristics of the PLC isoforms in normal cardiac cells, δ1 and γ1 seem to be the predominant forms expressed in adult ventricular cardiomyocytes compared with β1 and β3 (15, 44, 53). Binding sites for Ins(1,4,5)P3 and its phosphorylated derivative inositol 1,3,4,5-tetrakisphosphate have been found at the cardiac sarcoplasmic reticular (SR) level (20, 26) and may serve to enhance SR Ca2+ release and uptake, respectively (12, 39). These messenger-mediated SR Ca2+ movements may modulate the inotropic response of the cardiac muscle to agonists (5, 12). Immunolocalization of Ins(1,4,5)P3receptors at the fascia adherens of the intercalated disks may suggest a possible role of these receptors in local Ca2+ entry or in intercellular signaling between cardiomyocytes (27). Among the biological functions of DAG (38), one of the most significant is the activation of most protein kinase C isoenzymes, which phosphorylate several cardiac ion channels and are involved in myocyte hypertrophy via downstream signaling mechanisms (5, 37).
The molecular events underlying the contractile dysfunction in congestive heart failure (CHF) after a large myocardial infarct (MI) are incompletely defined. In particular, there is little information on the status of PLC isoenzymes of the cardiac sarcolemma (SL), which is the site where the surface signals are transduced to the interior of the cell. Our studies in 8-wk post-MI hearts indicated a depression in the total PLC activity of the SL membranes from surviving left ventricular (LV) tissue (31), whereas an elevated expression of PLC-β1 and -β3 proteins and increased PLC-β1 activity were noted in crude membrane fractions (25). The latter changes were intensified in infarct scar tissue and in the remnant myocardium, which borders the site of infarction (25). Such findings suggest an abnormal abundance and/or function of specific SL PLC isoforms in CHF, and this warrants investigation. In fact, this event may impact negatively on the complex second messenger response of PLC-linked receptors and thereby contribute to the pathogenesis of heart failure. In addition to investigating this event, we wanted to test our recent hypothesis that myocardial PLC is a pharmacological target in post-MI failing hearts (25), insofar as this enzyme might be a mechanism whereby angiotensin-converting enzyme (ACE) inhibitors exert their therapeutic action. Indeed ANG II, which is considered a factor in triggering the onset of pathological hypertrophy and subsequent development of CHF (34, 55), acts through PLC-β (9, 41) and -γ (14, 50) isoenzymes. Furthermore, ANG II selectively activates the PLC-β1 isoform in vascular smooth muscle cells (43).
The present study was conducted on surviving LV tissue of rats at 8 wk after occlusion of the left anterior descending coronary artery when the animals were in a moderate stage of CHF (25, 31). The primary focus was on the possible changes in quantity and activity of SL PLC-β1, -γ1 and -δ1 isoforms, which are the most relevant and well-characterized variants of PLC in mammalian cells (41), and on the effect of an in vivo late treatment with imidapril, a long-lasting ACE inhibitor (56). The abundance of cytosolic PLC isoforms available for their receptor-initiated translocation to the plasma membrane and the SL levels of PtdIns(4,5)P2 and the phosphoinositide kinases responsible for its synthesis were also assessed.
MATERIALS AND METHODS
All experimental protocols for animal studies were approved by the Animal Care Committee of the University of Manitoba, following the guidelines established by the Canadian Council on Animal Care. MI was produced in male Sprague-Dawley rats (weighing 175–200 g) by surgical occlusion of the left arterior descending coronary artery, as described previously (25, 31). The animals were first anesthetized with 5% isoflurane in a flow rate of oxygen (2 l/min). Then, after the thoracic fur was shaved, an incision was made along the left sternal border, the fourth rib was cut proximal to the sternum, and retractors were inserted. The pericardial sac was pierced so that the heart could be exteriorized through the intercostal space, and the left anterior descending coronary artery was ligated 2–3 mm from its origin with a suture of 6-0 silk. The heart was repositioned in the chest, and the incision was closed with a purse-string suture. Throughout the operative procedure, the rats were maintained on a positive-pressure ventilation system delivering 2.5% isoflurane in 2 l/min of oxygen. The mortality of the experimental animals operated on in this manner was ∼40% within the first 48 h after surgery. Age-matched, sham-operated animals served as controls and were treated similarly, except that the suture around the coronary artery was not tied. The animals were allowed to recover and were maintained on food and water ad libitum for a period of 8 wk before cardiac function and biochemical assessment. Imidapril was administered (1 mg/kg body wt, oral gavage, daily) to some randomly chosen infarcted animals for the last 4 wk. As in previous studies (25), animals with large transmural infarcts (≥40% of the LV free wall) were used.
The LV function of randomly selected animals from each of the three groups was assessed (25). Rats were anesthetized by an injection of ketamine-xylazine (100:10 mg/kg ip). After intubation of the trachea to maintain adequate ventilation, the right carotid artery was exposed and a micromanometer-tipped catheter (2-0; Millar SPR-249) was inserted and advanced into the LV. The catheter was secured with a silk ligature around the artery, and, after a 15-min stabilization of the heart function, LV pressures and maximum rates of isovolumic pressure development (+dP/dt max) and decay (−dP/dt max
Preparation of cardiac cytosolic and SL fractions.
Sham-operated and experimental animals were killed by decapitation, and the hearts were quickly excised and immersed in ice-cold 0.6 mol/l sucrose-10 mmol/l imidazole, pH 7.0 (buffer A). The atrial, macrovascular, connective, and, in particular, scar tissue were carefully removed, and the right ventricle was separated. The viable LV tissue (including intraventricular septum) from three to five hearts was pooled to prepare cytosolic and SL membrane fractions. Briefly, the tissue was washed, minced, and homogenized in 3.5 ml of buffer A per gram of tissue with a Polytron (6 × 10 s, setting 5). Large particles were removed by centrifugation at 12,000g (30 min, 4°C). A small aliquot of the first supernatant was centrifuged at 110,000g (30 min, 4°C), and the resulting supernatant was frozen and stored (−80°C) as the soluble cytosolic fraction. The rest of the first supernatant was diluted with 300 mmol/l KCl buffer to solubilize accessorial proteins and then further processed for the preparation of SL membranes according to the method used previously (31). The final pellet was resuspended in 0.25 mol/l sucrose-10 mmol/l histidine (pH 7.4), frozen in liquid N2, and stored at −80°C until assayed. All the above steps were carried out at 0–4°C. Markers enzymes (31) were assessed in this SL fraction (n = 3–4). In particular, the relative specific activity (specific activity in SL/specific activity in the homogenate) of K+-p-nitrophenol phosphatase (SL marker) was similar in 8-wk control and experimental preparations (15.9 ± 0.6, 15.7 ± 1.1, and 16.2 ± 1.2 in sham, MI, and MI-treated groups, respectively), indicating an equal degree of enrichment of the SL membrane in control and experimental groups. The relative specific activity of rotenone-insensitive NADPH-cytochrome c reductase (SR marker) was 0.40 ± 0.07, 0.43 ± 0.05, and 0.44 ± 0.05 and that of cytochrome c oxidase (mitochondrial marker) was 0.54 ± 0.07, 0.59 ± 0.09, and 0.57 ± 0.10 in sham, MI and MI-treated groups, respectively. These results appear to indicate that the SL fractions under study were relatively pure and had only a minimal but equal amount of contamination from other subcellular organelles. Protein concentrations were determined by the method of Lowry et al. as indicated elsewhere (31).
Total PLC assay.
The total PLC activity associated with the SL and cytosolic fractions was determined as already described (31). Briefly, the substrate was prepared by mixing an aliquot of [3H]PtdIns(4,5)P2with an aliquot of unlabeled PtdIns(4,5)P2. This mixture was dried under a stream of N2 and redissolved in 0.1 g/ml (232 mmol/l) sodium cholate. The substrate solution was kept under N2 gas overnight at 4°C and was diluted to 112 mmol/l sodium cholate shortly before addition to the incubation mixture. Typically, reactions were carried out at 37°C in a mixture containing 30 mmol/l HEPES-Tris (pH 7.0), 100 mmol/l NaCl, 2 mmol/l EGTA, 3.13 mmol/l CaCl2 (to generate a free Ca2+ concentration of 1.13 mmol/l according to the MaxChelator computer program, see Ref. 29), 15 μg cytosolic or SL proteins, 14 mmol/l sodium cholate, and 20 μmol/l3H-labeled substrate (20–30 dpm/pmol). Unless otherwise indicated, the reactions were terminated after 2.5 min by the addition of 144 μl of ice-cold CHCl3/CH3OH/HCl (1:2:0.2, vol/vol), followed by 48 μl of 2 mol/l KCl and 48 μl CHCl3. Conditions for blanks were identical, except that protein was added after the reaction was stopped. Phase separation was facilitated by mixing and centrifugation, and the resulting aqueous upper phase was aspirated and applied to a 500-μl Dowex AG1-X8 microcolumn (formate form, 100–200 mesh). After the column was rinsed with water and with borax in sodium formate, inositol mono-, bis-, and trisphosphates were eluted each with 1 ml of 0.1 mol/l formic acid containing 0.2, 0.4, and 1.0 mol/l formate, respectively. Quantitation was done by liquid scintillation counting in 10 ml of cytoScint. Ins(1,4,5)P3 was the primary product of PtdIns(4,5)P2hydrolysis, as already reported (32).
Immunoprecipitation of PLC-β1, -γ1, and -δ1 and assay for their activity.
These procedures have been already reported (25). SL membrane proteins were extracted using buffer containing 1% wt/vol sodium cholate, 50 mmol/l HEPES (pH 7.2), 200 mmol/l NaCl, 2 mmol/l EDTA, 10 μg/ml phenylmethylsulfonyl fluoride, and 10 μg/ml leupeptin by rotation for 2 h at 4°C. The samples were then centrifuged (280,000g for 25 min), and the supernatant was recovered as the solubilized membrane fraction. The membrane extract was incubated overnight at 4°C (rotation) with monoclonal antibodies to PLCs [anti-bovine PLC-β1, mixed monoclonal antibodies (no. 05-164); anti-bovine PLC-γ1, mixed monoclonal IgG antibodies (no. 05-163); anti-bovine PLC-δ1, mouse monoclonal antibodies (no. 05–343); all from Upstate Biotechnology, Lake Placid, NY] (5 μg of antibody to 350 μg membrane extract, i.e., a ratio of 1:70 μg/μg). All the antibodies cross-react with their corresponding PLC isoenzymes but not with the other two isoenzymes (48). The immunocomplex was captured by adding 100 μl of washed protein G Sepharose bead slurry (50 μl packed beads) at 4°C by rotation for 2 h. The agarose beads were collected by pulse centrifugation (5 s) at 10,000 g and assayed for the activity of PLC isoenzymes. The hydrolysis of [3H]PtdIns(4,5)P2was measured basically according to the method described by Wahl et al. (51). Briefly, the reaction was performed in the presence of 30 mmol/l HEPES (pH 6.8), 70 mmol/l KCl, 100 mmol/l NaCl, 0.8 mmol/l EGTA, 0.8 mmol/l CaCl2 (free Ca2+, 23.3 μmol/l), 20 μmol/l [3H]PtdIns(4,5)P2(20–30 dpm/pmol) dissolved in 14 mmol/l sodium cholate overnight, and an aliquot (10 μl) of immunoprecipitate suspension. The reaction was carried out at 37°C for 2.5 min and then stopped by trichloroacetic acid precipitation. Precipitates were removed by centrifugation at 10,000 g for 5 min, and the supernatant was collected for quantification of inositol phosphates by liquid scintillation counting. The efficiency of the immunoprecipitation of each isoenzyme was ascertained by determining any residual PLC isoenzyme activity in the 10,000g supernatant after capturing the immunocomplex by protein G Sepharose. The supernatant was concentrated to 100 μl by using microconcentrators (Centricon-3, Amicon Canada, Oakville, ON) and then tested for PLC isoenzyme activities. The immunoprecipitation was complete, as PLC-dependent [3H]PtdIns(4,5)P2hydrolysis of any immunoprecipitated isoenzyme could not be detected in the supernatant. For control experiments, immunoprecipitation and subsequent activity measurements were conducted with nonimmune mouse IgG.
Western blot of PLC isoenzymes.
High-molecular-weight markers (Bio-Rad, Hercules, CA) and 20 μg of SL or cytosolic proteins were separated on SDS-PAGE. Separated proteins were transferred onto 0.45-μm polyvinylidene difluoride (PVDF) membrane. PVDF membrane was blocked overnight at 4°C in Tris-buffered saline with 0.1% Tween 20 (TBST) containing 5% skim milk and probed with primary PLC isoenzyme antibodies. Primary antibodies were diluted in TBST (1:2,000). Horseradish peroxidase-labeled anti-rabbit IgG was diluted 1:3,000 in TBST and used as secondary antibody. PLC-β1, -γ1, and -δ1 were visualized by enhanced chemiluminescence according to the manufacturer’s instructions (Boehringer Mannheim, Laval, PQ). Band intensities of the Western blot were quantified using a charge-coupled device camera imaging densitometer (Bio-Rad GS 670).
SL PtdIns(4,5)P2 content.
SL PtdIns(4,5)P2 content was determined using the Biotrak RIA kit (Amersham Life Science). The manufacturer’s instructions modified according to the method of Chilvers et al. (4) were followed. Briefly, the SL levels of PtdIns(4,5)P2 were quantified by conversion of PtdIns(4,5)P2 in lipid extracts into Ins(1,4,5)P3by alkaline hydrolysis. Extracts were then neutralized and assayed for Ins(1,4,5)P3 as already indicated (25).
Phosphatidylinositol kinase and phosphatidylinositol 4-phosphate kinase assay.
The activities of phosphatidylinositol kinase and phosphatidylinositol 4-phosphate [PtdIns(4)P] kinase were assayed as described previously (29). Thirty milligrams of SL proteins were preincubated in a solution mixture containing 40 mmol/l HEPES-Tris (pH 7.4), 5 mmol/l MgCl2, 2 mmol/l EGTA, 1 mmol/l dithiothreitol, and 30 μg alamethicin for 30 min at 30°C. PtdIns(4)P kinase was assayed in the presence of 25 μmol/l PtdIns(4)P. The phosphorylation was started by adding [γ-32P]ATP in a final concentration of 1 mmol/l (0.16 Ci/mmol). The reaction was terminated after a 1-min incubation by the addition of methanol-10 N HCl (100:1, vol/vol) followed by the addition of 2.5 N HCl and chloroform. After centrifugation, the aqueous phase was discarded and the organic phase was washed once with chloroform-methanol-0.6 N HCl (3:48:47, vol/vol). Aliquots of the combined organic phases were used for the analysis of phosphoinositides by TLC. The solvent for the separation of phosphoinositide species contained chloroform-acetone-methanol- glacial acetic acid-water (40:15:13:12:8, vol/vol). The 32P-labeled phospholipid spots were visualized by overnight autoradiography using X-Omat-R X-ray film. PtdIns(4)P and PtdIns(4,5)P2 were scraped from the plates, and the radioactivity in each fraction was determined by liquid scintillation counting.
All values are expressed as means ± SE. The differences between two groups were evaluated by Student’st-test. The data from more than two groups were evaluated by one-way ANOVA followed by Duncan’s multiple comparison test. A probability of 95% or more was considered significant.
General characteristics and LV function.
Coronary occlusion resulted in the presence of reproducible transmural infarcts in the LV. The remnant heart muscle of the experimental animals underwent significant hypertrophy during 8 wk post-MI, as indicated by an increase of viable LV weight and by the augmented ratio of LV weight to body weight compared with control values (Table1). A significant increase in the wet-to-dry weight ratio of the lungs evidenced the presence of pulmonary edema in post-MI animals. These parameters were partially corrected by treating the infarcted animals with imidapril. The increase in LV end-diastolic pressure and the concomitant loss of contractile function (±dP/dt max) observed in the MI group were almost completely normalized by the imidapril treatment (Table 1). These results are consistent with earlier observations in this model, indicating that the experimental untreated animals were in a stage of moderate CHF (25, 31).
Total PLC activity of SL and cytosolic fractions.
Results in Table 2 show that the in vivo treatment with imidapril significantly improved the depression of total SL PLC activity already observed in 8 wk post-MI rats (31). As already reported (32), Ins(1,4,5)P3 was the primary product of PtdIns(4,5)P2 hydrolysis by PLC, and its alterations reflected those of total PLC (Table 2). Preliminary experiments seem to suggest that the drug had no direct effect on the enzyme. In fact, in vivo treatment of sham control animals with imidapril did not modify the total SL PLC activity (12.37 ± 0.22 and 13.03 ± 0.29 nmol inositol phosphates formed ⋅ min−1 ⋅ mg protein−1 for untreated and treated sham controls, respectively; n= 5). Likewise, in vitro exposure of SL membranes from sham and failing hearts to 10−8–10−4M imidapril did not affect the total PLC activity in both groups (data not shown). However, the mass/activity profile of PLC isoenzymes in a sham-operated, drug-treated group was not examined. This leaves open the possibility of direct effects of imidapril on some isoforms, which may be reciprocally compensated, leaving the total PLC activity unchanged. The PLC activity of the cytosol was similar in all three animal groups (Table 2). A reduction in the total activity of SL PLC was detectable at 1 wk following coronary ligation and progressively intensified, becoming significant at 2 and 4 wk (Table3). The finding that the first signs of CHF in this model occur at 4 wk after MI (45) suggests that the effect of the delayed ACE inhibitor therapy (Table 1), which started at the fifth week after the induction of MI (30), is that of correcting, and not of preventing, a preexisting lesion. Sham control and MI groups at 1, 2, and 4 wk post-MI did not show differences in total cytosolic PLC (Table3), and their activity values were similar to those seen at 8 wk (Table2). It should be noted that, unlike the cytosolic enzyme, an effect of age on SL PLC activity was detected (Table 3), which emphasized the need for internal sham controls at each of the different time periods. The enzymatic synthesis of SL phosphatidyl-N-monomethylethanolamine displayed similar age-related changes during the same life span (30).
Protein mass and functional activity of SL PLC isoenzymes.
Previous findings (25, 31) suggested an abnormal abundance and/or function of SL PLC isoenzymes in CHF that could be ameliorated by ACE inhibition. Accordingly, PLC-β1, PLC-γ1, and PLC-δ1, the most well-characterized variants of PLC in mammalian cells (41), were examined in purified SL membranes from viable LV of 8-wk post-MI animals without or with in vivo treatment with imidapril, a long-lasting ACE inhibitor (56). Western analysis with monoclonal antibodies that discriminate among the PLC isoenzymes under study was used to determine the immunoreactive PLC-β1, -γ1, and -δ1 protein bands and showed that the three forms are present in rat heart SL with typical molecular masses (Fig.1 C) (40,47). Wolf (52) found that only PLC-δ1 was associated with a SL preparation from canine LV myocardium, without excluding the existence of additional isoenzymes. Although each antibody cross-reacts with its corresponding PLC isoenzyme but not with the other two isoenzymes (48), other proteins have been shown to coimmunoprecipitate with PLC-β1, -γ1, and -δ1, and they may vary depending on the tissue type/cell type being probed (34). Thus the multiple immunoreactivity of our immunoblots (Figs.1 C and2 B) is in agreement with what has been previously detected using antibodies from our or a different commercial source (34, 47, 48).
The rank order of the isoenzymes’ hydrolytic activity toward PtdIns(4,5)P2 in control SL was PLC-δ1 > PLC-γ1 > PLC-β1, whereas that of the isoforms’ immunoreactivity was PLC-γ1 > PLC-δ1 > PLC-β1 (Fig. 1,A andB). CHF resulted in a radical reduction vs. controls of the mass and activity of γ1 and δ1 isoforms in combination with a significant increase of both PLC-β1 parameters (Fig. 1,A andB). The imidapril treatment induced a partial but significant correction of PLC-γ1 and -δ1 activity and normalized PLC-β1 activity, whereas the protein masses of the γ1 and δ1 isoforms were significantly above normal values and no change in PLC-β1 mass was observed relative to the CHF group (Fig. 1, Aand B). Thus it is evident that profound changes in the profile of heart SL PLC isoenzymes occur in CHF, in that PLC-β1 becomes the most prominent form, whereas the presence of PLC-γ1 and -δ1 is minimized. Of further note, in the imidapril-treated group, the protein levels did not correlate with the measured activities and that was particularly clear in the case of PLC-β1 (Fig. 1,A andB). Discrepancies between protein mass and activity have been already reported in the case of human heart transglutaminase II (21).
Abundance of PLC isoenzymes in cytosol.
PLC isoenzymes are present in both the membrane and cytoplasmic compartments of the cells. Physiological concentrations of monomeric PtdIns(4,5)P2 dissolved in the cytosol are negligible (23). Thus, in vivo, cytosolic PLC isoenzymes must migrate to the membranes where their lipid substrate resides, to catalyze the production of second messengers in stimulated cells (23). We assessed the amount of cytosolic PLC isoforms that could bind to membranes in response to stimuli and their potential pathophysiological shifts in CHF. Densitometric analysis of band intensity revealed that, in control heart, PLC-γ1 was the most abundant form followed by PLC-δ1, whereas PLC-β1 was barely detectable (Fig.2 A). The alterations in post-MI CHF were different from those observed in SL; in fact, only a significant decrease of PLC-δ1 level was noticed. Imidapril treatment was associated with a significant increase of PLC-β1 and -δ1 and a decrease of PLC-γ1 vs. controls (Fig.2 A). Apart from the imidapril-related γ1 form, the SL and cytosolic alterations were not reciprocally compensated in the experimental groups, suggesting changes in protein abundance of PLC isoenzymes (13). In fact, the SL-to-cytosol ratio of the isoenzymes’ protein mass, taken as an index of distribution, showed a 9.8-fold increment of PLC-β1 in post-MI CHF vs. control, suggesting an elevated compartmentalization of this form in the SL membrane of failing hearts, whereas the decreased ratio of PLC-γ1 and PLC-δ1 suggested their predominant localization in CHF cytosol (Table4). The ACE inhibition therapy was associated with elevated amounts of all the three isoenzymes at the SL level, with PLC-β1 being prominent (Table 4).
SL PtdIns(4,5)P2 content and phosphoinositide kinase activity.
PtdIns(4,5)P2, which is synthesized in the SL membrane by the coordinated and successive action of phosphatidylinositol kinase and PtdIns(4)P kinase (29), serves as a lipid substrate for all PLC isoenzymes and also has a high-affinity membrane-anchoring site for PLC-δ1 (49). Thus abnormal SL PtdIns(4,5)P2 levels may compromise the activity of PLC isoenzymes as well as the binding of PLC-δ1 to the membrane in stimulated cells. We examined the SL PtdIns(4,5)P2 content and found it to be significantly decreased in post-MI CHF in comparison to controls, as well as partially but significantly corrected by the imidapril treatment (Table 5). Both kinases were depressed to a similar extent in the SL membranes from failing hearts, but only PtdIns(4)P kinase activity was substantially reenhanced by the ACE inhibitor regimen (Table 5).
The rat infarct model employed in this and other recent studies in our laboratory (25, 31) results in a form of CHF that resembles that occurring in humans after a large transmural MI (18), with the development of CHF mimicking that of the clinical condition (58). Our major finding was the overabundance and hyperactivity of the SL PLC-β1 isoenzyme in failing hearts, in direct contrast with the drastic reduction of PLC-γ1 and -δ1 activity (11.1% and 14.5% of control, respectively) and protein mass (6.6% and 17.8% of control, respectively). These changes translate into an amplification of the PLC-β1-dependent function with almost complete loss of the PLC-γ1 and -δ1 functions in post-MI CHF. Of note, PLC-β1 mRNA is expressed in human myocardium (44).
SL-to-cytosol ratio of PLC-β1immunoreactive protein mass showed a 9.8-fold increment in post-MI CHF, which suggests an enhanced compartmentalization of this PLC variant in the SL membrane of failing hearts. Such an increase could be the consequence of elevated gene expression (25) in combination with agonist-evoked recruitment of PLC-β1 to the plasma membrane. The early and sustained high levels of plasma and myocardial ANG II (36, 45) may play a role. In fact, ANG II acts through the PLC-β class (9, 41) via ANG II type I (AT1) receptors (50), which have an increased expression in the surviving tissue of post-MI rat hearts (40). The sympathetic nervous system is also activated after MI and in CHF with augmented levels of circulating and myocardial catecholamines (10, 42), which may contribute to the high SL PLC-β1. In support of this possibility, infusion of the rat kidney with norepinephrine showed an increase in membranal PLC-β1activity and protein expression without changes in the γ1 form (57). The relative abundance of PLC-γ1 and -δ1 in the cytosol of failing hearts contrasts with their low levels in SL preparations. The precise reasons for these findings are presently unknown. However, it should be noted that the NH2-terminal part of the pleckstrin homology domain of PLC-δ1 possesses a critical region rich in basic amino acid residues that bind with high affinity to the polar head of PtdIns(4,5)P2(49, 54). This property confers to the δ1 isoenzyme a unique capacity of association with the plasma membrane that could be reduced by the diminished amount of PtdIns(4,5)P2in the SL of failing hearts.
PLCs of the β, γ, and δ classes display differences in terms of structure, activating mechanisms, and functions (24, 41, 46, 54). For example, phosphatidylinositol 3,4,5-trisphosphate can activate PLC-γ but not PLC-β and PLC-δ isoenzymes because it binds to the SH2 domains (1), which are unique to PLC-γ (41). Thus specific responses may occur depending on the type, quantity, and activity of the isoenzymes present in the membrane. Our studies indicate that the functional responses of SL PLC in the failing heart may be distinctively due to the upregulated PLC-β1 isoform and, perhaps, to PLC-β3, for which we have reported an increased cardiac expression (25). The distinct functions of each PLC isoenzyme in the myocardium and the extent of their overlap have not yet been established. Moreover, assigning functions to a redundant signaling isoenzyme may be complicated by the connection of its pathway to other signaling pathways and by the possibility that its increased level may lead to unpredicted changes in other signaling cascades. Nonetheless, the limited activity of SL PLC-γ1 and -δ1 and the augmented activity of PLC-β1 suggest that specialized responses exist in experimental post-MI CHF. For example,1) it is known that PLC-β1 activation by the α-subunit of the Gq subfamily is terminated by the intrinsic GTPase activity of Gqα that hydrolyzes bound GTP to GDP. PLC-β1 has the distinct feature of stimulating the intrinsic GTPase activity of Gqα (23, 41). Thus PLC-β1 regulates the termination of its own activation and of the signal; this event may be anticipated in failing hearts by the elevated PLC-β1 activity.2) The increase in myocardial catecholamines (10), the high density of α1-adrenoceptors (6, 22), the normal Gqα level (25), and the high mass/activity of PLC-β1, all of which have been observed in the surviving LV tissue of post-MI CHF animals, are consistent with an activation of the α1/Gqα/PLC-β1pathway and may explain the augmented responsiveness of the failing hearts to α1-agonists in this model of CHF (6). High plasma and myocardial catecholamines selectively downregulate β1-adrenoceptors in the failing heart, leading to subsensitivity of the β1 agonist-mediated biochemical and mechanical responses (2). In this context, the α1/Gqα/PLC-β1pathway may serve as an efficient source of cardiac-positive inotropy in our model of CHF. However, only a detailed examination of all the components of the pathway will ascertain the relevance of this possibility in human post-MI CHF. 3) An upgrade of the biological functions of ANG II and of the other agonists that operate mainly (if not exclusively) via Gqα/PLC-β1may also be expected.
Many functions may be severely impaired in CHF as a direct consequence of the radical reduction of PLC-γ1 and -δ1, which are prominent isoforms in normal heart SL. In fact,1) a significant attenuation of the myocardial responsiveness to polypeptide growth factors, which activate downstream PLC-γ1 as a specific effector enzyme (41), may be expected. The possible action of ANG II via this isoenzyme in the heart (14, 41, 50) may also be downgraded, such that ANG II would act only through the hyperactive AT1/Gqα/PLC-β1axis. 2) The stimulation of PLC-γ1 by intramembranal signaling lipid molecules [e.g., phosphatidic acid (formed by phospholipase D), arachidonic acid (released by phospholipase A2), and phosphatidylinositol 3,4,5-trisphosphate (41)] would be limited.3) The striking decrease of SL PLC-δ1, which is also stimulated by phosphatidic acid (52), may preclude the possibility of valid interactions between SL phopholipase D and PLC in CHF.4) Because Gh (transglutaminase II) seems to transfer the signal from α1-adrenoceptors to PLC-δ1 (22), this signal would be markedly depressed in post-MI failing hearts.
We detected a diminished amount of PtdIns(4,5)P2 in the SL membrane of decompensated hearts. This seemed to be due, at least partially, to its decreased synthesis by phosphatidylinositol and PtdIns(4)P kinases, as previously reported (33) and also observed in this study. The lack of PtdIns(4,5)P2 substrate would be an additional factor in attenuating the PLC-dependent generation of Ins(1,4,5)P3 and DAG and would reduce the formation of another membrane-delimited messenger, phosphatidylinositol 3,4,5-trisphosphate, by phosphatidylinositol 3-kinase (7). The diverse biochemical events that are regulated by PtdIns(4,5)P2 and that could be affected by the altered concentration of this lipid in the membrane have been recently reviewed (16, 28). Of note, the decreased number of SL PtdIns(4,5)P2 molecules could compromise the contractile performance of the heart independently of PLCs, by directly causing a depression of the inward rectifier K+ channels (19), as well as of the SL Na+/Ca2+exchange and Ca2+ pump activities (3, 17).
The present study was conducted with the view that the therapeutic action of the ACE inhibitor imidapril could be associated with the functional recovery of phosphoinositide-PLC. Although the profile of PLC isoenzymes at prefailure stages requires further investigation, we have provided evidence of a progressive decline of total PLC activity that occurred soon after MI (Table 2) and well before the first signs of CHF (45). This indicates that SL PLC is already abnormal during the hemodynamically compensatory stages of cardiac hypertrophy that precede CHF and suggests that PLC dysfunction may play a role in the pathogenesis of CHF after MI. Late imidapril monotherapy, instituted 4-wk post-MI (30), partially corrected the total SL PLC activity and isoenzymes’ parameters of the 8-wk post-MI failing heart, as well as the synthesis and content of membrane PtdIns(4,5)P2. These positive effects were accompanied by the amelioration of LV function in experimental animals, which may indicate a causal relationship between PLC function and imidapril therapy. It remains to be established if PLC correction should be ascribed to an attenuation of the activity of the renin-angiotensin system or to the hemodynamic effects following the ACE inhibitor treatment. The changes in protein levels of the SL and cytosolic isoenzymes of the MI group treated with imidapril are difficult to explain at present. In particular, we observed inconsistency between relative protein mass and activity in SL membranes. The possibility that imidapril per se may affect these isoenzymes cannot be excluded. In this context, examination of the isoforms’ mass and activity in the heart SL and cytosol of sham-operated, imidapril-treated animals may provide some insights. It may also be possible that the 4-wk therapy of the diseased animals had induced defects in the protein structure of the cardiac isoenzymes, and this could have affected their catalytic activity and/or their binding to the substrate. Indeed, PLC-δ1activity is depressed by site-directed mutagenesis as well as chemical modification of histidine residues within a highly conserved sequence of the X region (8), whereas relocation of the NH2 terminus might affect the binding to PtdIns(4,5)P2 (23). Alterations of the physicochemical characteristics of the membrane environment of the isoenzymes may also have contributed to the discrepancies between the mass and activity of PLCs (23). Specific studies should examine these possibilities.
In conclusion, the results of this study show that profound changes in the protein mass/activity profile of SL PLC-β1, -γ1, and -δ1 occur in post-MI CHF, which could alter the second messenger responses of these isoenzymes. Their partial reversibility by imidapril may confer pathophysiological significance to phosphoinositide-PLC isoenzymes and may be related to the mechanism of action of this ACE inhibitor. The early post-MI occurrence of PLC dysfunction and its improvement by delayed drug therapy suggest that greater benefits (perhaps prevention of PLC changes) may be obtained with an early treatment of the infarcted animals.
We thank Dr. I. M. C. Dixon for critical reading of the manuscript.
Address for reprint requests and other correspondence: V. Panagia, Institute of Cardiovascular Sciences, St. Boniface General Hospital Research Centre, 351 Tache Ave., Winnipeg, Manitoba, Canada R2H 2A6 (E-mail:).
This study was supported by a grant (to V. Panagia) from the Medical Research Council of Canada (MRC Group in Experimental Cardiology). V. Panagia was a Senior Investigator of the Medical Research Council of Canada.
Presented in part at the XX Annual Meeting of the American Section of the International Society for Heart Research, Ann Arbor, MI, August 9–12, 1998.
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. §1734 solely to indicate this fact.
- Copyright © 1999 the American Physiological Society