INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SEVERE CARDIAC COMPLICATIONS occur in both patients with and animal models of chronic diabetes mellitus (7). Various disorders in subcellular organelles also occur during diabetes, but their contribution to the overall depression of heart function remains ill defined. Also, the signals responsible for impaired subcellular structures of cardiac myocytes are not fully understood. For example, a marked depression of sarcolemmal Na⁺-K⁺-ATPase activity is well documented in experimental diabetes with insulin deficiency (12, 16-18), but the mechanism for this depression is poorly understood. The sarcolemmal Na⁺-K⁺-ATPase is a key component in regulation of the ion homeostasis and resting potential in cardiac myocytes. It consists of - and -subunits. The catalytic -subunit contains the binding site for ATP, Na⁺, K⁺, and ouabain. The -subunit is a glycoprotein that is necessary for the functional activity of the Na⁺-K⁺-ATPase (26).

Diabetes is associated with elevated plasma levels of free fatty acids (FFAs) and a marked increase in their oxidation (5, 20, 23). Thus excessive utilization of FFAs by the diabetic myocardium could represent a signal for the defects in subcellular organelles. To support this suggestion, several lipid-lowering agents are known to exert beneficial effects on the diabetic heart (5, 20, 23). We previously reported that the carnitine palmitoyltransferase I (CPT I) inhibitor etomoxir (10) partially normalized the myosin isozyme population in diabetic rat hearts (23). Also, bypassing the CPT I block via administration of dietary medium-chain fatty acids did not blunt the normalization of myosin isozymes (23). Thus the major effect of etomoxir may be due to its lipid-lowering action that results from inhibition of de novo fatty acid synthesis (2, 27). Similarly, changes in the sarcoplasmic reticulum Ca²⁺-pump ATPase activity were also prevented after treatment of diabetic rats with etomoxir (23).

The present study examined whether etomoxir treatment can reverse diabetes-induced alterations in sarcolemmal function (Na⁺-K⁺-ATPase activity), plasma lipids, and heart function in rats with streptozotocin (Stz)-induced diabetes. Several recent studies reported changes in Na⁺-K⁺-ATPase subunit expression in pathological conditions such as pressure-overload hypertrophy (4) and diabetes (26). Ng et al. (16) found that the ₁-subunit of the Na⁺-K⁺-ATPase was unaltered in diabetic rat hearts, but both the ₂- and ₁-subunits were decreased. Thus diabetes may cause a differential regulation of Na⁺-K⁺-ATPase subunits similar to that for myosin isozymes. Accordingly, this study utilizes [³H]ouabain binding and Western blot analysis to examine the diabetes-induced alterations in Na⁺-K⁺-ATPase subunits and the ability of etomoxir to reverse these changes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental model and hemodynamic measurements. Diabetes was produced in male Sprague-Dawley rats (175-200 g) via injection of Stz (65 mg/kg body wt) (18, 21, 23) into the tail vein, and sham-treated rats served as controls. Two weeks after injection, control rats and diabetic rats received either tap water or water containing (+)etomoxir (sodium salt; RBI, Natick, MA) for 4 wk. The etomoxir intake was adjusted to 8 mg/kg body wt by monitoring the water intake.

Measurements of Na⁺-K⁺-ATPase and [³H]ouabain binding. Cardiac sarcolemmal membrane was isolated as previously described (19). Three hearts were pooled for each experiment. Marker enzyme activities (18) revealed that the membrane preparations from control, diabetic, and etomoxir-treated hearts contained minimal (3-5%) cross contamination with other subcellular organelles. The activity of Na⁺-K⁺-ATPase was assayed as hydrolysis of ATP as described previously (18). The K⁺-dependent p-nitrophenylphosphatase (KpNPPase) activity was measured as hydrolysis of p-nitrophosphate (18).

Serum parameters. Serum concentrations of glucose and triglycerides were measured by using enzymatic, colorimetric kits (kits 16-UV and 336-20, Sigma Chemical, St. Louis, MO). Nonesterified FFAs were also determined with a colorimetric kit (Wako, Osaka, Japan). The 3,3',-5-triiodo-L-thyronine (T₃) levels were assessed by fluoroimmunoassay (Delfia; Pharmacia, Fairfield, NJ) and insulin was assayed by a rat insulin RIA kit (Linco Research, St. Louis, MO).

Western blot analysis. Relative Na⁺-K⁺-ATPase content was measured by SDS-PAGE. The sarcolemmal proteins were electroblotted to Immobilon-P transfer membrane (Millipore). Monoclonal anti-₁-subunit of Na⁺-K⁺-ATPase mouse IgG (1:10,000), polyclonal anti-₂-subunit rabbit IgG (1:2,000), polyclonal anti-₃-subunit rabbit IgG (1:2,000), or polyclonal anti-₁-subunit rabbit IgG (1:2,000) from Upstate Biotechnology (Lake Placid, NY) was used to detect subunits. The membranes were subsequently incubated for 1 h with biotinylated anti-mouse IgG (1:1,000) for ₁-subunit and biotinylated anti-rabbit IgG antibody for ₂-, ₃-, and ₁-subunits (Amersham, Arlington Heights, IL). Finally, the membranes were incubated with streptavidin-conjugated horseradish peroxidase (1:5,000) and processed for chemiluminescence (ECL Kit, Amersham) by using Hyperfilm-ECL (Amersham).

Statistics. All values are presented as means ± SE. Statistical analysis was performed by Student's t-test or one-way ANOVA followed by Scheffé's test. A P < 0.05 was considered significant. Estimates of kinetic parameters, such as dissociation constant or maximal intensity of binding sites (B_max), were derived from Scatchard plot analysis of [³H]ouabain-binding data. The [³H]ouabain-binding data were analyzed with the LIGAND program of McPherson (14), which is based on the F-test for deriving the best fit. Regression analysis was checked by a manual method.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Injection of Stz resulted in a general decrease in growth characteristics of diabetic rats after 6 wk (Table 1). Body weight decreased by 25%, and heart weight decreased by 16%, thereby increasing the heart weight-to-body weight ratio by 12%. Diabetic rats also exhibited increased serum concentrations of glucose, triglycerides, and FFAs, whereas insulin and T₃ concentrations were decreased (Table 1). Treatment of diabetic and control rats with etomoxir did not influence body weight but increased heart weight by 16 and 27%, respectively. In diabetic rats, etomoxir significantly decreased serum triglycerides and FFA concentrations, but it did not alter the levels of insulin, glucose, or T₃ (Table 1). The etomoxir treatment reduced serum triglycerides and FFAs in control rats, but the decrease was not statistically significant (Table 1). Etomoxir treatment increased the yield of sarcolemma expressed as milligram sarcolemmal protein per gram heart weight by 193% in diabetic rats and 227% in control rats (Table 1).

In untreated diabetic rats, heart rate, left ventricular systolic pressure, +dP/dt, and dP/dt were decreased, whereas left ventricular end-diastolic pressure was increased (Table 2). Treatment with etomoxir partially prevented the depression in left ventricular systolic pressure, +dP/dt, and dP/dt. Moreover, elevation of the left ventricular end-diastolic pressure was completely prevented by etomoxir. The etomoxir did not, however, reverse the decreased heart rate in diabetic rats. Etomoxir treatment had no significant effects on hemodynamic parameters in control rats (Table 2).

Diabetes produced a significant depression of sarcolemmal Mg²⁺-ATPase activity (Fig. 1). Etomoxir treatment exerted no significant effect on Mg²⁺-ATPase of diabetic rats. Diabetic rats also exhibited a significant depression of Na⁺-K⁺-ATPase activity (Fig. 2). Etomoxir treatment of diabetic rats prevented the decrease in Na⁺-K⁺-ATPase activity when expressed per gram heart weight but not when expressed per milligram sarcolemmal protein. Similarly, diabetic rats exhibited decreased KpNPPase activity (Fig. 3). Etomoxir reversed the depression of KpNPPase activity in diabetic rats when expressed per gram heart weight but not when expressed per milligram sarcolemmal protein. In control rats, etomoxir treatment decreased Mg²⁺-ATPase and Na⁺-K⁺-ATPase activity when expressed per milligram sarcolemmal protein but not when expressed per gram heart weight. In contrast, etomoxir increased KpNPPase activity in the control rats when expressed per gram of heart weight.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Effect of etomoxir treatment on cardiac Mg2+-ATPase of control and diabetic rat hearts. Activity expressed per mg sarcolemmal (SL) protein (A) and per g heart wt (B). Values are means ± SE of 4 experiments. * P < 0.05 compared with control rat hearts.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Effect of etomoxir treatment on Na+-K+-ATPase of control and diabetic hearts. Activity expressed per mg SL protein (A) and per g heart wt (B). Values are means ± SE of 4 experiments. * P < 0.05 compared with control rat hearts.  P <0.05 compared with diabetic rat hearts.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Effect of etomoxir treatment on K+-dependent p-nitrophenylphosphatase of control and diabetic hearts. Activity expressed per mg sarcolemmal protein (A) and per g heart wt (B). Values are means ± SE of 4 experiments. * P < 0.05 compared with control rat hearts.  P <0.05 compared with diabetic rat hearts.

Scatchard plot analysis of specific [³H]ouabain binding to sarcolemmal membranes is shown in Fig. 4. The B_max of the high-affinity binding sites but not of low-affinity sites was decreased in untreated diabetic rats. Etomoxir treatment normalized B_max expressed per gram heart weight in diabetic rats but had no effect in control rats (Table 3). However, B_max was not normalized by etomoxir when expressed per milligram sarcolemmal protein (Table 3). Etomoxir treatment increased B_max of the low-affinity binding sites in both diabetic and control rats, even though B_max (expressed per g heart wt) was not significantly depressed in the diabetic rats. To examine any acute effects of etomoxir, its action on sarcolemmal enzymes was tested in vitro. Etomoxir did not affect Na⁺-K⁺-ATPase but decreased Mg²⁺-ATPase activity at all concentrations tested (1-100 µM; Table 4).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Scatchard plot analysis of specific [3H]ouabain binding to sarcolemmal preparation. Results with high concentrations of [3H]ouabain (low-affinity sites) are shown in A, whereas those with low concentrations (high-affinity sites) are shown in B. Values are taken from typical experiments involving 4 control () and 4 diabetic rat heart () preparations.

In a separate set of experiments, we measured the relative protein content of the Na⁺-K⁺-ATPase subunits by Western blotting (Fig. 5). Diabetes markedly decreased the content of the ₂- and ₃-subunit content without appreciable changes in the ₁- or ₁-subunit content. Etomoxir treatment of diabetic rats produced a further decrease in the ₂- and ₃-subunit content but also reduced the ₁-subunit content. However, treatment of control rats with etomoxir also induced significant changes in the Na⁺-K⁺-ATPase subunit distribution. Etomoxir treatment decreased the ₂- and ₁-subunit content in control rats without any change in the ₁- or ₃-subunit content. These findings are particularly puzzling in light of the large increase (227%) in apparent sarcolemmal protein yield induced by etomoxir treatment in control rats.

View larger version (109K): [in this window] [in a new window]   Fig. 5.   Representative Western blots depicting expression of 1-, 2-, 3-, and 1-subunits of rat Na+-K+-ATPase in sarcolemmal preparations isolated from control (C), diabetic (D), diabetic plus etomoxir (D + E) or control plus etomoxir (C + E) rat hearts. All lanes were loaded with 20 µg of sarcolemmal protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrates that diabetic rat hearts exhibit a marked depression of left ventricular function involving both systolic pressure development and relaxation. The impaired heart function is associated with a reduction in Na⁺-K⁺-ATPase activity, a decrease in ₂- and ₃-subunit content of the enzyme, and a depression of B_max for the high-affinity ouabain- binding sites but not for the low-affinity sites. Because the high-affinity site is affected by the ₂- and ₃-subunits and the low-affinity site by the ₁-subunit (16), our data are consistent with those of Ng et al. (16) indicating that Stz-induced diabetes results in a decrease in the ₂-subunit but not in the ₁-subunit. To examine possible metabolic links between the diabetic state and functional disorders, we treated diabetic rats with the CPT I inhibitor etomoxir. Inhibition of CPT I reduces FFA utilization and increases glucose oxidation in a compensatory manner (21). However, plasma FFAs are not elevated but are reduced on a long-term basis by CPT I inhibitors, which inhibit acetyl-CoA carboxylase (2, 27), the rate-limiting enzyme of de novo fatty acid synthesis. This mixed profile of decreasing both FFA utilization and FFA supply should exert beneficial effects on the deranged metabolic state of insulin-dependent diabetes mellitus.

We previously reported that etomoxir partially improves the population of myosin isozymes and prevents changes in the sarcoplasmic reticulum Ca²⁺-pump ATPase activity in Stz-induced diabetes and that a close correlation exists between myosin V₃ and plasma concentrations of FFAs and triglycerides (23). The present data show that cardiac function is also partially normalized by etomoxir; this agrees with the findings of Schmitz et al. (25). It is noteworthy that the depressed function of the diabetic rat heart was improved, although hypoinsulinemia and the associated hyperglycemia were unaffected. Furthermore, etomoxir did not affect the reduced thyroid influence arising from decreased T₃ levels, which depress heart function. Although CPT I inhibition should enhance glucose utilization, a previous study involving CPT I bypass with dietary medium-chain fatty acids suggests that FFA plays a major role in the metabolic disturbances of Stz-induced diabetes (23). A reduction in the elevated circulating FFA levels should blunt the inhibitory effect of FFA on glucose oxidation, which would already be depressed due to hypoinsulinemia (28).

One might also argue that the etomoxir-induced alterations in acylderivatives may mediate improved cardiac function. Etomoxir treatment is expected to reduce long-chain acylderivatives, which can cause heart dysfunction and decrease Na⁺-K⁺-ATPase activity (1, 6). However, Lopaschuk et al. (13) claimed that the protective effect of etomoxir in reperfused ischemic myocardium arises from enhanced glucose utilization and not from changes in long-chain acylderivatives levels. Other evidence also suggests that the improvement in heart function is attributable at least in part to the lipid-lowering action of etomoxir. The antihypertensive compound hydralazine also improves the heart function in diabetic rats by reducing the serum lipid concentrations (20). Our study supports the therapeutic concept of normalization of serum FFAs whenever glucose utilization of heart muscle is depressed. Relief from FFA-induced inhibition of glucose utilization in the heart will increase the formation of glycolytic-ATP. Glycolytic ATP or membrane-bound ATP is essential for maintenance of the function of membrane-bound enzymes such as Na⁺-K⁺-ATPase (3, 15). Thus glycolytic-ATP may contribute to the hemodynamic improvement seen in etomoxir-treated diabetic hearts.

Another finding in the present study is that etomoxir treatment increased the yield of sarcolemma isolated from control and diabetic rat hearts. Our sarcolemmal preparation is expected to also contain some transverse tubular system (9). The increase in density of sarcolemmal/T-tubular membrane yield after etomoxir treatment may reflect the hypertrophic effects of etomoxir on myocyte volume. There was no change in cross contamination of the sarcolemmal preparation after etomoxir treatment, based on the activity of marker enzymes. Thus the increase in sarcolemmal protein yield was not due to increased contamination by sarcoplasmic reticulum membrane, for example, although etomoxir also increases the yield of this membrane structure in rat hearts (22). The fact that the increased sarcolemmal protein yield was accompanied by a decrease in overall Na⁺-K⁺-ATPase subunit content suggests that etomoxir may decrease the density of Na⁺-pump molecules per unit sarcolemmal area or alter the relative distribution of Na⁺-K⁺-ATPase subunits.

Diabetes was associated with a decrease in sarcolemmal protein and a depression in Na⁺-K⁺-ATPase activity and high-affinity ouabain-binding sites. These effects of diabetes could be explained by the large reduction in ₂- and ₃-subunit content seen in these hearts. Etomoxir treatment of diabetic rat hearts produced a further reduction in the ₂- and ₃-subunits as well as a decrease in ₁-subunit. However, etomoxir treatment increased sarcolemmal protein yield by 193% in diabetic rat hearts, which would counter the effects of the decreased subunit content on enzyme activity and high-affinity ouabain binding sites when expressed per gram heart weight. Etomoxir treatment also increased sarcolemmal protein in control hearts (227%) but did not reduce ₃-subunit content. It did, however, produce a marked reduction in ₁-subunit content. Changes in ₁-subunit content would indirectly influence Na⁺-K⁺-ATPase activity because the ₁-subunit is required for functional integrity of the enzyme (26). Thus we believe that the changes in enzyme activity and ouabain binding in etomoxir-treated hearts may reflect overall changes in the subunit composition of the enzyme rather than relative changes in a specific subunit.

In the present study, etomoxir treatment markedly reversed the depression of Na⁺-K⁺-ATPase and KpNPPase activity when expressed per gram heart weight. Moreover, etomoxir normalized the depression of B_max for the high-affinity ouabain binding sites (expressed per g heart wt). Although B_max for the low-affinity sites (expressed per gram heart weight) was not significantly depressed in diabetic rats, etomoxir treatment increased B_max in diabetic as well as control rats. Therefore, etomoxir may exert differential effects on high-affinity vs. low-affinity binding sites in control rats. Sahin-Erdemli et al. (24) also postulated a differential regulation of high- and low-affinity binding sites. These authors showed that deoxycorticosterone acetate increased Na⁺-K⁺-ATPase activity by an increase in protein content of the ₁-subunit, whereas the ₂- and ₃-subunits were not affected. Our approach to express Na⁺-K⁺-ATPase data per gram heart weight is based on the findings of Gick et al. (11), who suggested that "expression of enzyme activity per unit protein is rendered difficult if protein content per unit weight is not constant among the tissues examined." In our study, protein content of the tissue examined (heart) was increased in etomoxir-treated vs. untreated rats. Thus expression of enzyme activity per gram tissue weight may represent a more valid basis for intratissue comparisons under treatment conditions that alter tissue protein content.

In summary, our results indicate that the depressed cardiac function in Stz-induced diabetes in rats is associated with a decrease in Na⁺-K⁺-ATPase activity and high-affinity binding sites. Etomoxir partially improves the depressed heart function and increases the activity of Na⁺-K⁺-ATPase (expressed per g heart wt) as well as B_max for the high-affinity sites for ouabain. We conclude that improved glucose utilization or reduced FFA utilization associated with etomoxir treatment may play a role in maintenance of the long-term activity of Na⁺-K⁺-ATPase.