INTRODUCTION

Top Abstract Introduction Materials Results Discussion References

IN HUMANS type 1 insulin-dependent diabetes mellitus (IDM) is associated with the development of cardiomyopathy, which is independent of the myocardial disease produced by coronary atherosclerosis (13, 15, 38, 44). Diabetic cardiomyopathy appears to be responsible for some of the excess cardiovascular morbidity and mortality that occurs in diabetic patients (14, 38, 44). Many biochemical and metabolic defects have been observed in the myocardium and skeletal muscle of animals and patients with IDM as well as type 2 diabetes, including diabetic cardiomyopathy, increased -myosin heavy chain (-MHC) and ventricular myosin light chain 2 (vMLC-2), impaired Ca²⁺-activated actinomyosin adenosinetriphosphatase (ATPase) activity of the ventricular myocytes, increased secretion of atrial natriuretic peptide (ANF) and angiotensin II-converting enzyme (ACE), skeletal muscle weakness, impaired utilization of glycogen stores and glucose in skeletal muscle, and a decrease in the number of glucose transporters (GLUT-4) in skeletal muscle membranes (4, 6, 38, 44). Nevertheless, the basic biochemical and molecular biological defects responsible for the transformation of normal to diseased myocardium and skeletal muscle remain undefined (2, 13, 14, 38, 44). Recent studies suggest that IDM may result from an overactive immune response of T-lymphocytes and macrophages, which invade the -pancreatic cells and release cytokines and free radicals, including nitric oxide, which then destroy these cells, impairing their ability to manufacture and secrete insulin (4, 6, 7, 29). This concept was based on the findings in the BB/Wor rat and NOD mouse, which develop IDM similar to that seen in humans and in which cyclosporin, an immunosuppressant, can prevent the onset of diabetes (4, 6, 29). However, human clinical trials with cyclosporin have met with only limited success because of the nephrotoxicity of this drug and the severity of the pancreatic damage at the time of diagnosis of IDM in humans (4, 29). Thus the BB/Wor diabetic rat can be used as a model for human type 1 IDM-mediated cardiomyopathy and skeletal muscle myopathy.

Protein kinase C (PKC) is a family of enzymes involved in the phosphorylation of enzymatic and regulatory protein substrates (21, 23). Phosphorylation and dephosphorylation of enzymes are important physiological mechanisms utilized for the activation and deactivation of enzymatic activity (15, 21, 23). Twelve isozymes of PKC have been identified and subsequently subdivided into three major categories: 1) Ca²⁺-dependent and phospholipid- and diacylglycerol (DAG)-activated PKC isozymes (cPKC), 2) Ca²⁺-independent and phospholipid- and DAG-activated PKC (nPKC), and 3) atypical PKC isozymes (aPKC) that are Ca²⁺ independent and activated by phosphatidylcholine or phosphatidylethanolamine (15, 21, 23, 36). Activation of PKC may result from mechanical and physical deformation of the myocardial cell membrane (e.g., stretching or distension), the interaction of agonists with their myocardial cell membrane receptors, or elevated intracellular concentrations of the substrates that activate PKC, including DAG, phosphatidylcholine, phosphatidylethanolamine, and Ca²⁺. Mechanical distension or the inter- action of agonists with their myocardial membrane receptors activates phospholipase C, resulting in the hydrolysis of phosphatidyl 4,5-bisphosphate into inositol (1,4,5)-trisphosphate [Ins(1,4,5)P₃] and DAG. Ins(1,4,5)P₃ releases intracellular Ca²⁺, which then combines with cytosolic cPKC. This aids in the binding of an inactive form of cPKC to the phosphatidylserine (PS) residues of the cell membrane. The binding of DAG to the membrane-bound inactive form of cPKC results in its activation and subsequent ability to phosphorylate enzymes or receptor proteins. Alternatively, DAG can bind to inactive nPKC. This allows the binding of the DAG-nPKC complex to PS residues in the membrane and results in the subsequent activation of nPKC (15, 21, 23, 36). An alternate pathway for activation of PKC also exists. Activation of phospholipase D hydrolyzes phosphatidylcholine to phosphatidic acid, which is further metabolized to DAG and to phosphatidylethanol. Both DAG and phosphatidylethanol can subsequently activate PKC (15, 21, 36). A similar mechanism exists in skeletal muscle (15, 23, 36).

Phosphorylation of myocardial and skeletal muscle enzymes and proteins by PKC isozymes can affect cardiac rhythm and cardiac and skeletal muscle contractility, gene expression, and growth (2, 12, 15-17, 21, 23, 24, 35, 36, 40, 42). Protein kinase C-mediated phosphorylation of troponin T, troponin I, troponin-tropomyosin complex, and troponin-C protein in isolated myocardial and skeletal muscle cells is associated with inhibition of Ca²⁺-activated myofibrillar actomyosin MgATPase activity and contractility (21, 23, 24). The PKC- isozyme can increase Ca²⁺ influx through L-type Ca²⁺ channels (35), which are also present in the membrane of ventricular myocytes and skeletal muscle (15, 21, 23, 29). Activation of cPKC- and nPKC- is associated with myocardial ventricular hypertrophy, including overexpression of -MHC, vMLC-2, ANF, and ACE (36).

Recent studies in vascular smooth muscle and in cardiac myocytes in culture suggest that elevated plasma levels of glucose, the hallmark of both type 1 and type 2 diabetes mellitus, may be causal to elevated tissue levels of PKC and thereby may play a pivotal role as a mediator of the progression from normal to diseased vasculature and myocardium (1, 8, 16, 25, 26, 39, 42). Hyperglycemia may increase the DAG content of the rat myocardium (16, 25, 26). Total PKC activity also increased in the myocardium of diabetic rats (37, 43). Persistent upregulation of PKC activity may lead to alterations in myocardial contractility and cell growth (2, 13, 15-17, 21, 24, 25, 36, 37, 40, 42), which may compensate for the potential decrease of protein synthesis resulting from impaired sensitivity of the diabetic myocardium to this action of adenosine 3',5'-cyclic monophosphate (cAMP) (32). Moreover, sustained elevations of PKC activity can increase the activity of serum and tissue levels of ACE in the diabetic rat (39).

Increased ACE activity can increase sera and tissue levels of angiotensin II, which can cause myocardial remodeling and hypertrophy (35, 38). We previously reported that treatment of rats with the ACE inhibitor benazepril prevented the development of myocardial dysfunction in streptozocin (STZ)-induced diabetic cardiomyopathy (9). These data all suggest that dysregulation of PKC may play a pivotal role in the development of cardiac and skeletal muscle myopathy in IDM. However, if changes in PKC are causal to the development of diabetic myopathy, we hypothesize that alterations in PKC must be present early in its development at a time when functional abnormalities would be minimal or absent. We tested this hypothesis in BB/Wor diabetic (D) and diabetic-resistant (DR) rats by measurement of the activity and total quantity of PKC and PKC isozymes in ventricular tissue, gracilis muscle, and circulating neutrophils and by measurement of cardiac structure and function in vivo, using hemodynamic measurements and echocardiography, and direct evaluation of myocardial function in vitro, utilizing the isolated working heart preparation.

	METHODS AND MATERIALS

Top Abstract Introduction Materials Results Discussion References

General experimental protocol. Male BB/Wor D rats (diabetic for 30-41 days) and age-matched DR rats were obtained from the National Institute of Diabetes and Digestive and Kidney Diseases breeding and maintenance facility at the University of Massachusetts. The D rats were maintained on daily subcutaneous injections of 0.6-3.0 U/rat of a long-acting (once a day) protamine, zinc, and insulin suspension (protamine, zinc, and Iletin I, Eli Lilly, Indianapolis, IN) before their feeding period on the basis of their urinary excretion of glucose from the first appearance of glucosuria. DR rats were given equivalent subcutaneous injections of vehicle. Daily injections of insulin were necessary because the D rats, a model for IDM, die within a few days without insulin maintenance. All animals were housed on a 12:12-h light-dark cycle with access to water and standard rat chow ad libitum. All animals were anesthetized intramuscularly with a solution consisting of (in mg/kg) 50 ketamine-4 xylazine before echocardiography and Doppler flowmetry or surgery for collection and measurement of blood for serum glucose concentrations using a standard glucose monitor (Boehringer Ingelheim, Ridgefield, CT). The D and DR rats were subdivided into two cohorts. One cohort was subjected to echocardiography and Doppler flowmetry followed by invasive measurement of cardiovascular dynamics. Their blood was removed and the neutrophils were isolated on a Ficoll-Percoll gradient as described previously (20). The hearts were also removed and the ventricles were rapidly frozen in liquid nitrogen. The second cohort of rats was subjected to echocardiography and Doppler flowmetry and the hearts were then removed and used for in vitro perfusion while the gracilis muscle was removed and frozen in liquid nitrogen. The ventricles, gracilis muscles, and neutrophils were assayed for PKC activity and isozyme content as described below. The experimental protocol (1224) was approved by the Louisiana State University Institutional Animal Care and Use Committee.

Echocardiographic analysis of cardiac performance. M-mode and two-dimensional echocardiograph and Doppler flow recordings were made using a Toshiba model 270 echocardiography instrument with a 7-MHz transducer. The transducer was calibrated with phantoms before use. Two-dimensional echocardiograms were recorded from both parasternal long- and short-axis views, apical four-chamber views and suprasternal views of the aortic valve, and from both ascending and descending aorta. M-mode recordings were obtained of the left ventricle (LV) at the level of the mitral papillary muscle and at the level of the mitral valve in the parasternal view using two-dimensional echocardiographic guidance in both the short- and long-axis views. Aortic valve M-mode recordings from the parasternal long-axis and suprasternal views were obtained. Pulsed-wave Doppler was used to examine mitral diastolic inflow from the apical four-chamber view and tricuspid diastolic inflow from the apical four-chamber and parasternal short-axis views. Aortic valve flow and isovolumic relaxation time were calculated from the apical five-chamber view using pulsed-wave Doppler. Color Doppler studies were performed for evaluation of LV length from the apical four-chamber view. Color-flow imaging was also used to determine flow in both ascending and descending aorta from the suprasternal view. Data were recorded on Super VHS 0.5-in. tape for playback. Six consecutive cardiac cycles for each view and parameter were digitized from tape onto a Freeland digital acquisition system, and the average value was calculated. Echocardiographic measurements included M-mode interventricular septal thickness, LV internal dimension at the end of diastole, and M-mode posterior LV wall thickness using the leading edge-leading edge method as recommended by the American Society of Echocardiography (22). Mitral Doppler peak E and A wave velocities and E:A ratio, tricuspid Doppler E and A wave velocities and E:A ratio (both minimum and maximum), and the difference between maximum and minimum E wave Doppler velocities (E Vel_maxmin) were obtained. Tricuspid E Vel_maxmin was also calculated.

Determination of LV performance in vivo. The anesthetized rats were placed on a blanketed, temperature-controlled surgical table (37°C). A polyethylene catheter (PE-30) connected to a Gould P23D pressure transducer was advanced into the LV via the right carotid artery. The position of the catheter in the LV was confirmed by the LV pressure (LVP) tracing. After LVP and heart rate (HR) stabilized, HR, LVP, and the positive and negative rate of change of LVP (+LV dP/dt and LV dP/dt) were continuously recorded and continuously logged on an Apple 650 Quadra computer. The data for 1-min increments were obtained at 300 Hz, and the average values were calculated. +LV dP/dt and LV dP/dt were obtained by differentiation of the LVP signals.

Determination of cardiac performance in vitro. Hearts were perfused on a standard working heart apparatus as described previously in detail (9). Heparin sodium (200 U iv) was administered to the rats immediately before they were killed, and 50 U/ml were added to the initial perfusate, Krebs-Henseleit buffer aerated with a 95% O₂-5% CO₂ mixture, to maintain a pH of 7.4. The perfusate was filtered using a Whatman Polycap HD filter (10-µm pore size) to prevent cells and debris from clogging the coronary circulation. All hearts were paced at 250 beats/min and allowed to stabilize before the start of the experiment. Aortic pressure, measured with a Statham P23Db transducer positioned at the level of the aortic valve, was set at 65 mmHg. LVP was measured with a Statham P23Db pressure transducer connected to a 26-gauge needle inserted through the ventricular wall at the "dimple" located at the apex of the heart. The LVP transducer was positioned at a level corresponding to midventricle. Positive and negative LV dP/dt were obtained by differentiating the LVP signal. Left atrial pressure (LAP) was measured via an atrial cannula connected to a Statham P23Db pressure transducer positioned at the level of the left atrium and was varied during the experiment by adjusting the height of the atrial reservoir. Initial LAP was set at 10 cmH₂O. After equilibration of the heart, LAP was reduced to 5 cmH₂O, increased in 5-cmH₂O increments to a maximum of 20 cmH₂O, and then reduced to 10 cmH₂O at the end of the experiment. The response of the heart to each increase in LAP was allowed to stabilize before the recording of the data. Pressures and derived indexes were recorded on a Grass model 79 polygraph. An Apple Macintosh Quadra 650 computer sampled and digitized the data from the polygraph for storage and subsequent analysis. The cardiac performance parameters measured were maximum developed LVP (LVP_max), +LV dP/dt, LV dP/dt, and LAP.

Specificity of assay for PKC isozymes. Recent data suggest the possibility that some antibodies specific for PKC isozymes cross-react with other PKC isozymes to give false indexes of the identity and quantity of PKC isozyme present in various tissues (30). To ascertain the existence or absence of constitutive PKC isozymes and the specificity of the antibodies used to test for the presence of the PKC antibody-specific proteins, we analyzed the ventricular homogenate for the presence of PKC isozyme mRNA and protein and subsequently performed PKC antibody isozyme neutralization using authentic PKC isozyme peptides as antigens.

Determination of cDNA for PKC isozymes. Transcripts for PKC isozymes were measured by reverse transcriptase-polymerase chain reaction (RT-PCR) in whole homogenates of ventricular myocardium as previously described for alveolar macrophages (10, 11, 20). Briefly, the total RNA of the homogenized ventricle was isolated using Trizol reagent (GIBCO, Gaithersburg, MD). Total cDNA was obtained by reverse transcription of total RNA and was labeled with [³²P]dCTP. Total cDNA (10 ng) was amplified by using the specific PKC isozyme primers and [³²P]dCTP. Primer sequences for the PKC isozymes are shown in Table 1. The relative amounts of PKC isozyme cDNA were determined by phosphorimager scan and quantitation of the smear and signal bands, normalized to that of -actin (n = 8) as described previously in detail (10, 11, 20). Whole homogenate PKC isozyme content was determined with Western blot analyses.

Western blot analyses of PKC isozymes. Western blot analyses were performed as described previously for the assay of nitric oxide synthase protein (10, 11) but were modified for the myocardium and gracilis muscle and for extraction and detection of PKC isozymes. Aliquots of frozen ventricle, gracilis muscle, or neutrophils obtained from BB/Wor homozygous D and DR rats were quantitatively pulverized and homogenized in homogenization buffer I [20 mM tris(hydroxymethyl)aminomethane (Tris) · HCl, pH 7.5, 0.25 M sucrose, 2 mM EDTA, 2 mM ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 0.02% leupeptin, 1 mM phenylmethylsulfonyl fluoride (PMSF), and 0.1% Triton X-100] in a cold room. The homogenates were incubated for 1 h at 4°C and subsequently centrifuged (17,500 g for 30 min at 4°C). The supernatants were then centrifuged at 37,000 g to isolate the mitochondria, the resulting supernatant was centrifuged at 100,000 g at 4°C, and the particulate (membrane) and soluble fractions (cytosol) were stored at 20°C until assayed for protein with the bicinchoninic acid (BCA) method (10, 11). The protein samples (50 or 100 µg) were then separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a 10% (wt/vol) acrylamide separating gel and a 4% (wt/vol) acrylamide stacking gel. The protein was then electrophoretically transferred to nitrocellulose using a Semi-Dry transfer cell (Bio-Rad, Hercules, CA) and a transfer buffer consisting of Tris · HCl (48 mM) and 39 mM glycine (pH 9.2) containing 0.037% (wt/vol) SDS and 20% (vol/vol) methanol.

Determination of specificity of antibodies for PKC isozymes. Protein extracts (50 and 100 µg) of ventricular myocardium prepared as described in Determination of cDNA for PKC isozymes were subjected to SDS-PAGE electrophoresis as described in Western blot analyses of PKC isozymes. Immunoblots were prepared as described in Western blot analyses of PKC isozymes with PKC isozyme-selective antibodies (0.5 µg/ml) that had been preincubated (+) or had not been preincubated () with the isozyme-specific antigen peptide (1 µg/ml), which was used as a specific antibody antagonist (Transduction Laboratories). The PKC isozyme-specific antibody that is preincubated with its corresponding antigen peptide should lose its ability to bind to the antibody-selective PKC isozyme because its binding sites should now be occupied by the antigen peptide. If a signal band was selectively blocked by incubation of the antibody with the corresponding antigen peptide, it was defined as a specific signal that contained the same structure as the corresponding antigen peptide.

Measurement of PKC activity. Ventricles, gracilis muscle, and neutrophils were rapidly removed from the D and DR rats utilized in the in vivo studies described in Determination of LV performance in vivo. The pieces of tissue were washed in PBS to remove residual blood and were then frozen in liquid nitrogen. Portions of the frozen ventricles, gracilis muscle, and neutrophils were quantitatively pulverized and then placed in homogenization buffer I (20 mM Tris · HCl, pH 7.5, 0.25 M sucrose, 2 mM EDTA, 2 mM EGTA, 0.02% leupeptin, and 1 mM PMSF) and homogenized in a cold room with a Polytron set at 7 for 20 s, followed by homogenization for 60 strokes using a Dounce homogenizer. The homogenates were centrifuged as described in Western blot analyses of PKC isozymes for the PKC isozyme assay, and the particulate and soluble fractions were obtained for measurement of PKC activity. The pellets were washed and resuspended in buffer II (20 mM Tris · HCl, pH 7.5, 2 mM EDTA, 2 mM EGTA, 0.02% leupeptin, 1 mM PMSF, and 0.1% Triton X-100) and again homogenized. The homogenates were solubilized in buffer II without Triton X-100. After a 45-min incubation period at 4°C, the soluble fraction was obtained by ultracentrifugation at 100,000 g for 30 min, and the pellet was retained as the particulate or membrane fraction. Both membrane and soluble fractions were passed through 0.5 ml DEAE columns (Pharmacia, Gaithersburg, MD), washed twice with buffer II (2 ml), and then eluted with 0.4 ml of phosphate buffer containing 200 mM NaCl (37). PKC activity was determined with an Amersham kit for PKC activity (Amersham, Arlington Heights, IL) by measurement of the phosphorylation of the specific substrate octapeptide (RKRTLRRL) in the presence of Ca²⁺, PS, and DAG using [³²P]ATP. (Amersham Searle, Waltham, MA). The protein content of the fractions was measured by the BCA method as described previously in detail (10, 11). The enzymatic data were expressed as picomoles per milligram protein per minute after comparison to a standard curve and correction for the nonspecific kinase activities found in the absence of Ca²⁺, PS, and DAG.

Statistical analysis of data. Each experiment contained four to six animals per treatment group. Data were analyzed with analysis of variance (ANOVA) for a randomized complete block or completely random sample design. Differences between and among means were analyzed with Dunnett's and Duncan's tests. Biochemical data were analyzed with multiple ANOVA and means were compared with Newman-Keuls test; P < 0.05 was accepted for statistical differences between and among means.

	RESULTS

Top Abstract Introduction Materials Results Discussion References

Body weights, blood glucose concentration, and cardiac measurements. Body weights did not differ among the DR and D rats (Table 2). The D rats demonstrated glucosuria between 28 and 49 days after their birth and were diabetic, as evidenced by a glucosuria of +2 to +4, for 6-8 wk before they were killed. At the time of death blood glucose concentrations were increased in D rats compared with those of DR rats (P < 0.05) (Table 2). Although total heart weight and LV mass did not differ among the DR and D rats, there was a small but insignificant increase in LV mass (P = 0.076) and a significant increase in LV mass index (LV wt/body wt; P = 0.013) in the LV obtained from the BB/Wor D rats compared with the LV mass index obtained from the DR rats. When total heart weight was expressed as a percentage of total body weight (heart weight index), no significant difference was observed between DR and D rats (Table 2). Interventricular septal weight was increased in the hearts obtained from the BB/Wor D rats compared with that obtained from the DR rats (P = 0.03).

Echocardiographic measurements. Analysis of M-mode echocardiographic recordings revealed an increase in interventricular septal thickness in D rats compared with that obtained from DR rats (P = 0.05), whereas no significant difference in LV internal dimension or LV posterior wall thickness was evident between the two groups of rats (Table 3). Doppler patterns of mitral inflow did not differ between the DR and D rats. However, the E wave peak velocity of tricuspid inflow (both minimum and maximum) was increased in the D rats compared with that of the DR rats (P = 0.03) (Table 3, Fig. 1).

View larger version (63K): [in this window] [in a new window]   Fig. 1.   Typical echocardiogram showing increased tricuspid E wave and decreased A wave velocities in the BB/Wor diabetic (D) rat (A) and diabetes-resistant (DR) rat (B).

Determination of LV performance in vivo and in vitro. LVP_max did not differ between the BB/Wor DR and D rats (Table 4). Moreover, LVEDP was not significantly elevated in the D rats compared with that of the DR rats (Table 4). In contrast, myocardial contractility as reflected by +LV dP/dt (P < 0.05) and the rate of relaxation of the LV as reflected by LV dP/dt (P = 0.08) were lower in the BB/Wor D rats compared with these parameters in the BB/Wor DR rats (Table 4). However, no significant differences in myocardial function existed in vitro in the paced, isolated hearts obtained from the D and DR rats when maintained at an LAP of 5 or 20 cmH₂O (data not shown) and at 10 cmH₂O (Table 4).

PKC isozyme mRNA, specificity of antibody, and linearity of PKC isozyme content and enzymatic activity. The constitutive mRNAs present in the ventricle of the D rats as determined by RT-PCR were PKC- >>> PKC- > PKC- > PKC- = PKC- >>> PKC- (Fig. 2). Similarly, the PKC isozymes detectable in the whole homogenate of ventricle were PKC- >> PKC- = PKC- (Fig. 3). When the ventricular homogenate was separated into the soluble (cytosolic) and particulate (membrane) fractions, the PKC isozymes detectable in the membrane fraction of the ventricle were PKC- >> PKC- = PKC-, whereas those of the cytosol were PKC- > PKC- > PKC- > PKC- > PKC- > PKC-. PKC- and PKC- were absent in two of five ventricles despite the presence of their mRNA (Fig. 3). Each of the PKC isozyme-specific antibodies elicited a signal when reacted with its authentic isozyme and with the brain extract at a molecular weight that corresponded to the authentic PKC isozyme. PKC- was detectable in the cytoplasm of the unstimulated ventricle, PKC- and PKC- were barely detectable in the cytoplasm of the adult rat ventricle, and PKC-, PKC-, and PKC- were found in both the cytoplasm and membrane extracts of the adult rat ventricle. Because the antibodies for PKC-, PKC-, and PKC- failed to give any significant signal in the range of 72 to 90 kDa in the particulate fraction that contained significant amounts of PKC- and PKC-, it is unlikely that any significant cross-reactivity existed between these antibodies for PKC isozymes and PKC- and PKC-. Moreover, because PKC- and PKC- did not give any signals at the molecular weight of each other, it is also unlikely that cross-reactivity exists between these isozymes. Finally, preincubation of the PKC isozyme-specific antibodies with their antigen-specific proteins eliminated the detection of a band at the corresponding molecular weight of the antibody-specific isozyme (Fig. 3). Similar results were obtained in the skeletal muscle extracts (data not shown). Thus, under the conditions of the experiment and with these tissues, the antibodies used appeared to be isozyme specific.

View larger version (76K): [in this window] [in a new window]   Fig. 2.   Typical reverse transcriptase-polymerase chain reaction (RT-PCR) gel of protein kinase C (PKC) mRNA (A) and Western blot of particulate protein of PKC isoforms (B) found in ventricle of adult D rat. A: , , , , , , and  indicate PKC isozyme cDNA that is being tested; M column represents no. of base pairs (bp). Bottom bands in A represent those of -actin. Constitutive PKC isozyme mRNAs present in DR rat ventricles are PKC- >>> PKC- > PKC- > PKC- = PKC-. For details, see METHODS AND MATERIALS.

View larger version (69K): [in this window] [in a new window]   Fig. 3.   Effect of antigen-specific neutralization of PKC isozyme antibodies on detection of PKC isozymes by Western blot in homogenates of rat brain and in cytosolic and membrane fractions of rat heart (H). , PKC isozyme was detected in presence of antibody alone; +, PKC isozyme was tested for with antibody preincubated with specific antigen peptide. In this animal PKC- and PKC- isozymes are barely detectable. Note the absence of any significant signals after antigen-specific antibody neutralization (for details, see METHODS AND MATERIALS).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Relationship between optical density (gel density units/mm2) of Western blot of PKC isozymes as a function of protein applied to gels in ventricles obtained from D rat. A: soluble fraction. B: particulate fraction. Each mean value represents n = 5 rats.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   PKC activity of particulate, soluble, and total homogenate of ventricle expressed as function of amount of ventricular protein used in assay (pg of substrate/20 min). For details, see METHODS AND MATERIALS. Each mean and SD represent responses from 5 animals.

Ventricular PKC isozyme content and PKC activity. PKC-, PKC-, PKC-, and PKC- were the predominant PKC isozymes in the soluble fraction of the rat heart, with PKC- and PKC- in most abundance (Figs. 6 and 7). In contrast, PKC-, PKC-, and PKC- were the only PKC isozymes found of those tested in the particulate fraction of the rat ventricles obtained from D and DR rats in vivo, as identified by Western blot analysis and the enhanced chemiluminescence technique (Figs. 6 and 7). The PKC-II and PKC- isozymes were present in low amounts in the soluble fraction of the ventricles obtained from the D and DR rats but were absent from the particulate fraction of the ventricles obtained from these groups of rats (Figs. 6 and 7). The content of PKC- protein increased by 89.4 and 38.6% in the particulate and soluble fractions, respectively, of the hearts obtained from the BB/Wor D rats compared with the content of this PKC isozyme in these fractions of hearts obtained from the BB/Wor DR rats (Fig. 7). Although the concentration of PKC- protein was increased by 24.2% in the particulate fraction of the hearts obtained from the D rats compared with that in the ventricles obtained from the DR rats, the amount of this PKC isozyme in the soluble fractions of the ventricles obtained from these rats did not differ (Figs. 6 and 7). The concentrations of both PKC-II and PKC- did not differ in the soluble and particulate fractions, respectively, of the ventricles obtained from the BB/Wor D and DR rats (Figs. 6 and 7). Total PKC activity in the ventricles obtained from D rats was (mean ± SD) 985 ± 52 pg · mg protein¹ · min¹ (n = 4), whereas that of the DR rats was 768 ± 53 pg · mg protein¹ · min¹ (n = 4, P < 0.05). The PKC activity increased by 44.4 and 18.4% in the particulate and soluble fractions, respectively, obtained from the ventricles of the BB/Wor D rats compared with the PKC activity of the hearts obtained from the BB/Wor DR rats (Fig. 8).

View larger version (76K): [in this window] [in a new window]   Fig. 6.   Typical Western blot of PKC-, PKC-II, PKC-, PKC-, and PKC- in whole brain homogenate (B) used as a standard tissue containing multiple PKC isozymes and in soluble (cytosol) and particulate fractions (membrane) of heart ventricles obtained from BB/Wor rats with insulin-dependent diabetes mellitus (IDM; D) and BB/Wor rats without IDM (C). Molecular masses of isozymes (in kDa) are at left (for details, see METHODS AND MATERIALS).

View larger version (28K): [in this window] [in a new window]   Fig. 7.   Changes of isozyme content of PKC-, PKC-, PKC-, PKC-, and PKC- in soluble (A) and membrane fractions (B) of ventricles of hearts obtained from BB/Wor diabetic (D) and diabetes-resistant (DR) rats. Each mean ± SD represents the responses from 5 animals/group. Data are expressed in gel density units. * Response significantly different from that of DR rats (P < 0.05).

View larger version (29K): [in this window] [in a new window]   Fig. 8.   Changes in the enzymatic activity of total PKC activity and PKC activity in cytosolic (soluble) and membrane (particulate) fractions of ventricles obtained from BB/Wor D and diabetes-resistant rats. Means ± SD represent responses from 4 rats/group. Data are expressed in pg phosphorylated substrate · mg protein1 · min1. * Response significantly different from that of DR rats (P < 0.05).

Gracilis muscle PKC isozyme content and PKC activity. PKC- and small amounts of PKC-II were the predominant PKC isozymes in the soluble fraction of the freshly frozen gracilis muscle, whereas PKC- >> PKC- >> PKC- in the particulate fraction (Fig. 9). The gel density of both PKC isozymes increased in the gracilis muscle obtained from the D rat compared with the DR rat. However, PKC- increased by 438% (P < 0.05), whereas that of PKC- increased by 92% (P < 0.05) (Fig. 10A). PKC activity in the particulate fraction of the gracilis muscle was greater than that of the ventricle and was increased in the gracilis muscle obtained from D rats compared with that obtained from DR rats (Fig. 10A).

View larger version (74K): [in this window] [in a new window]   Fig. 9.   Typical Western immunoblot of isozyme content of PKC-, PKC-, PKC-, PKC-, and PKC- in the brain (B) and soluble (cytosol) and particulate (membrane) fractions of the gracilis muscle obtained from BB/Wor DR rats (C) and BB/Wor D rats (D). Note presence of PKC-, PKC-, and PKC- in particulate fraction and the increase in PKC- and PKC- in the gracilis muscle of the D rat compared with the control DR rat (for details see METHODS AND MATERIALS).

View larger version (26K): [in this window] [in a new window]   Fig. 10.   PKC activity (bars at left; nmol · mg protein1 · min1 for 20-min period) and PKC isozyme content (PKC-, PKC-, PKC-, or PKC-; gel density units/mm2) determined by Western immunoblot of freshly isolated and frozen gracilis muscle (A) and neutrophils (B) from BB/Wor diabetic rats and BB/Wor diabetes-resistant rats (control). Means ± SD represent responses from 5 animals/group. * Response significantly different from that of control rats (P < 0.05).

Neutrophil PKC isozyme content and PKC activity. The neutrophil content of the fractionated phagocytes was 99.9 ± 0.1% with 98 ± 0.4% viability based on exclusion of Evans blue dye. This was in agreement with previous findings (20). Small amounts of PKC- and PKC-II and large amounts of PKC- were the predominant PKC isozymes in the soluble fraction of the freshly frozen neutrophils (data not shown), whereas PKC- and PKC- were predominant in the particulate fraction of the freshly isolated neutrophils. (Fig. 10B). The gel density of the PKC isozymes did not increase in the soluble (data not shown) or particulate (Fig. 10B) fractions of the neutrophils obtained from the D rats compared with the DR rats. However, basal neutrophil PKC activity, which was the highest when compared with that of the ventricle and gracilis muscle, increased by 23 ± 24% (P > 0.05) in the particulate fraction of the neutrophils obtained from D rats compared with that obtained from DR rats (Fig. 10B).

	DISCUSSION

Top Abstract Introduction Materials Results Discussion References

This study demonstrates with the use of echocardiography, Doppler flowmetry, and hemodynamic measurements that during the early incipient stages of IDM in the BB/Wor D rat interventricular septal thickness, the E wave peak velocity of tricuspid inflow, and LV weight index were increased, and myocardial contractility and the rate of relaxation were decreased compared with these parameters in age-matched DR rats. Thus, in vivo, the early diabetic cardiomyopathy of the D rat is characterized by a less compliant LV and a restrictive right ventricle (RV) with increased septal thickness. Although we did not measure skeletal muscle function in the D or DR rats, it is known that skeletal muscle dysfunction occurs during the progression of IDM, as well as enhanced susceptibility to infections (4, 6, 19). The in vivo changes in cardiac function and structure reported herein were minimal and before the onset of fully expressed diabetic cardiomyopathy or impaired skeletal muscle function (33) were accompanied by selective increases in particulate-bound PKC- and PKC- isozymes, with undetectable changes in these or the PKC-II or PKC- isozymes in the soluble fraction. Moreover, they also occurred at a time when neutrophil PKC isozymes did not differ between the D and DR rats. This increase in PKC isozyme content was accompanied by increases of the basal total, soluble, and particulate PKC activity of the ventricles and the particulate fraction of the gracilis muscle, whereas that of the neutrophil did not change when obtained from the BB/Wor D rats compared with that of the BB/Wor DR rats. These findings lend support to the hypothesis that sustained elevations of ventricular and gracilis muscle particulate (membrane) PKC activity and isozyme content occur before, and may be responsible for, the early transition from normal to diseased myocardium and skeletal muscle during the development of diabetic myopathies.

Cardiac function and structure. Most studies of diabetic cardiomyopathy in the BB/Wor diabetic rat model and humans have focused on LV abnormalities, the most consistent of which have been reduced LV compliance and diastolic dysfunction (4, 13, 14, 22, 26, 31, 38, 44). We are unaware of any reported echocardiographic studies in diabetic rats. However, echocardiographic evidence for abnormalities in LV function, diastolic function, and compliance in diabetic humans has been demonstrated by several investigators (13, 22, 28, 31). Interestingly, although both +LV dP/dt and LV dP/dt (relaxation) were depressed in the BB/Wor D rat compared with these parameters in the DR rat, the major functional and structural alteration in the BB/Wor D rat was detected in the interventricular septum and the RV when assessed by echocardiography and Doppler flowmetry. The increased E wave velocity present in the tricuspid inflow pattern as recorded by Doppler sonography may be indicative of a restrictive pattern of ventricular filling (13, 22, 28, 31). Although an increase in tricuspid E wave velocity may also reflect an increase in venous return (preload), it is unlikely that cardiac output was increased in the BB/Wor diabetic rats because similar increases of E wave velocity recorded from the mitral valve were not forthcoming. Evidence for early RV involvement in experimental diabetic cardiomyopathy is also forthcoming from the histopathological evidence of more severe cardiomyopathy in the RV compared with the LV in rats with STZ-induced diabetes (9, 38, 44). This is consistent with the experience of most investigators, who find that ~12 wk of hyperglycemia are necessary to produce typical findings of LV dysfunction and overt diabetic cardiomyopathy in BB/Wor D rats (4, 31). Thus the slight increase in interventricular septal thickness and the trend toward increased LV mass associated with normal LV systolic pressure, reduced positive and negative LV dP/dt, and slightly impaired diastolic function as determined in vivo by echocardiography and cardiac pressure measurements without in vitro functional changes in the isolated working heart are consistent with evidence that the BB/Wor D rats were studied at the incipient stages of diabetic cardiomyopathy in vivo.

PKC and diabetic cardiomyopathy. Elevated plasma levels of glucose have been shown to increase PKC activity in two major organ systems that are targets for diabetes-mediated injury. These include the retinal capillary endothelial cells in cell culture (33), cultured aortic smooth muscle and endothelial cells (1, 15, 17, 35, 39, 40), and cardiac myocytes (2, 8, 16, 17, 25, 37). Diabetes also increases the PKC content and activity in the hearts and vasculature of experimental animals and humans (1, 2, 8, 15-17, 25, 36, 37, 41). Many factors have been suggested to play a role in the transition from normal to diseased myocardium in IDM. Among these are alterations in intracellular pH (40); derangements of intracellular metabolism of Ca²⁺, Ca²⁺ transport, and transmembrane permeability to Ca²⁺ (43, 44); changes in the isozyme patterns of myosin (13, 43); impaired mitochondrial function (4, 6, 19, 44); altered utilization of glucose and fatty acids (6, 31, 41, 43); decreased activity of cAMP (33); and increased activity of PKC (8, 16, 25, 26, 37, 42, 43). PKC may be the most important of these diabetogenic factors because the obligatory hyperglycemia of IDM can upregulate PKC (6, 16, 26, 41) and the PKC- isozyme of PKC can modulate intracellular pH and alter intracellular and transmembrane Ca²⁺ metabolism in part by increasing Ca²⁺ influx through L-type Ca²⁺ channels (35) present in ventricular myocyte and skeletal muscle membranes (15, 21, 23, 29). Activation of cPKC- and nPKC- is also associated with myocardial ventricular hypertrophy, including overexpression of -MHC, vMLC-2, ANF, and ACE (36, 39), and impairment of myocardial contractility consequent to stimulation of -MHC and vMLC-2 and phosphorylation of troponin and the troponin-tropomyosin complex (15, 17, 21, 23, 24, 27, 36, 40). Thus many of the changes in myocardial structure and cardiac and skeletal muscle function in IDM may result from the combined activation and suppression of the actions of the PKC isozymes on various biochemical reactions of myocardial and skeletal muscle cells and their surrounding supporting cells (18, 25, 27, 41). In support of this speculation are the findings that chronic administration of a selective inhibitor of PKC- to diabetic rats decreased the vascular and cardiac changes associated with STZ-induced diabetes (1, 17). Our finding of an increase in particulate but not soluble PKC- and PKC- protein in the ventricle and gracilis muscle, and increases in total, particulate, and soluble PKC activity in these same tissues when obtained in vivo from the BB/Wor D rat, whereas neutrophil PKC protein and activity did not differ between the D and DR rat, is consistent with the reported increases of DAG and particulate and soluble PKC activity in cardiac ventricular myocytes obtained from STZ-induced diabetic and BB/Wor D rats compared with that of control rats (8, 16, 25, 26, 37, 42). Moreover, it suggests that these changes may be related to progression of the disease rather than to a genetic predisposition for increased PKC in the BB/Wor D rat inasmuch as the neutrophil was refractory to the changes. Also, the finding of overexpression of PKC- in the heart of the BB/Wor D rat may have clinical significance because PKC- has been shown to modulate cardiac L-type Ca²⁺ channels expressed in Xenopus oocytes (34). Intracellular Ca²⁺ is increased and Ca²⁺ sequestration decreased in the myocardium of BB/Wor D rats, STZ-induced diabetic animals, and humans with IDM and type II diabetes mellitus (2, 44). If the changes in PKC- are found to precede the increase of intraventricular and skeletal muscle Ca²⁺, then the changes in the PKC isozyme would be clearly important relative to the transition of normal myocardium and skeletal muscle to diabetes-induced cardiomyopathy and skeletal muscle myopathy.

Limitations of study. A major limitation of this study is that we measured whole ventricle PKC activity and we only measured five PKC isozymes, whereas previous studies measured the PKC activity and isozyme profile of pure ventricular or atrial myocytes in cell culture (6, 16, 18, 21, 25, 42). We were able to detect increases in several PKC isozymes as well as measure the PKC-II isozyme in the soluble fraction and PKC- isozyme in the particulate fraction of the hearts and gracilis muscle obtained from both BB/Wor D and DR rats. Thus it is unlikely that the small amount of protein derived from the arteries and endothelium within the ventricles or within the gracilis muscle could contribute significantly to the PKC activity or mask the PKC isozyme profile of the larger amount of muscle. In support of this conclusion is the finding that the relative density of the constitutive mRNA for PKC-, PKC-, PKC-, and PKC- paralleled the relative density of the PKC isozyme proteins as determined by Western blot analyses.

Conclusion. At an early stage during the development of diabetic cardiomyopathy (which is characterized by a restrictive RV, small decreases of +LV dP/dt and LV dP/dt, and an increase in the interventricular septal thickness and mass), total, particulate, and soluble PKC activity of the whole heart is also increased. This is accompanied by elevations in the particulate fraction of PKC- and PKC- isozymes. Similar results were observed in gracilis muscle but not in the circulating neutrophil. Because these isoforms of PKC can produce changes in intracellular pH, Ca²⁺, contractility, and cell growth characteristic of IDM, the data support the hypothesis that increased PKC activity and changes of the PKC isozyme profile may play an important role in the transition from normal to abnormal myocardium and skeletal muscle in type I diabetes mellitus.