INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

HUMAN DIABETES MELLITUS is associated with altered cardiac functions independent of vascular complications (14, 20). A great deal of evidence for diabetes-related cardiac dysfunction has been accumulated in experiments on animal models (26). One of the important features found in experimental diabetic hearts is the diminished inotropic response to -adrenoceptor stimulation. This has been clearly demonstrated in isolated hearts (30) as well as atrial (10, 21) and ventricular (11, 35) muscles. In accordance with the diminished inotropic responsiveness, a reduction in the number of myocardial -adrenoceptors has been reported (1, 11, 19, 21, 22, 33). A defect in the coupling of myocardial -adrenoceptors to adenylate cyclase has been also reported (1, 9, 19, 32). However, it has been pointed out that the decrease in the number of myocardial -adrenoceptors does not necessarily result in an altered -adrenoceptor-mediated response of diabetic myocardium (5). Furthermore, unaltered coupling of myocardial -adrenoceptors to adenylate cyclase in diabetes has been suggested (12, 23, 29). Thus whether altered -adrenoceptor expression and signal transduction are truly responsible for the diminished functional responses to stimulation of the receptors in diabetic hearts is not certain.

Stimulation of myocardial -adrenoceptors leads to activation of a guanine nucleotide-binding protein (G protein), termed G_s, that triggers activation of adenylate cyclase and in turn increases cAMP. As a result of increased levels of cAMP, protein kinase A (PKA) is activated and it phosphorylates a variety of regulatory proteins, including sarcolemmal L-type Ca²⁺ channels (27) and phospholamban in the sarcoplasmic reticulum (SR) (15). Phosphorylation of Ca²⁺ channels results in an increased opening of the channels and augments Ca²⁺ influx during the cardiac action potential (27). Phosphorylation of phospholamban results in increased Ca²⁺ uptake by SR Ca²⁺-ATPase (25). The increase in the SR Ca²⁺ uptake rate is expected to lead to a large amount of Ca²⁺ sequestrated by the SR that would be available for release. Thus, when myocardial -adrenoceptors are stimulated, more Ca²⁺ enters the cell via Ca²⁺ channels and a larger amount of Ca²⁺ is loaded into the SR. The process is followed by greater Ca²⁺ release from the SR and hence an increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) during systole, leading to positive inotropy. We have recently shown that the -adrenoceptor-G protein-adenylate cyclase system is fully functional but cAMP-dependent phosphorylation of phospholamban is blunted in hearts from Wistar rats with 4-6 wk of streptozotocin (45 mg/kg)-induced diabetes (8). This suggests that the lack of cAMP-dependent phospholamban phosphorylation may be associated with a decrease in the rate of Ca²⁺ release and/or uptake by the SR, thereby contributing to the diminished inotropic response to -adrenoceptor stimulation. However, the question remains as to whether the impaired responsiveness to myocardial -adrenoceptor stimulation in diabetes is largely caused by altered SR functions of uptake and release of Ca²⁺ or changes in the phosphorylation process of other regulatory proteins such as Ca²⁺ channels are also involved in this impairment.

In an attempt to gain insight into the primary defect in the diminished responsiveness to myocardial -adrenoceptor stimulation in diabetes, we compared the effects of isoproterenol on cell length, [Ca²⁺]_i transient, L-type Ca²⁺ current (I_Ca), and SR Ca²⁺ content in ventricular myocytes isolated from streptozotocin-induced diabetic rat hearts with those of age-matched control rats. The Ca²⁺ fluoroprobe indo 1 was used to measure changes in [Ca²⁺]_i, and caffeine-induced Ca²⁺ release was used as an index of SR Ca²⁺ content. We also investigated the effects of forskolin and dibutyryl cAMP (DBcAMP) on these variables to assess whether diabetes-induced changes in the isoproterenol responses are the result of a defect at the level of -adrenoceptors or at a level distal to the receptors.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Induction of diabetes. Male Wistar rats (8 wk old, 180-220 g) were anesthetized with diethyl ether and received a single intravenous injection of streptozotocin (45 mg/kg; Sigma Chemical, St. Louis, MO) into the tail vein. Streptozotocin was dissolved in a citrate buffer solution (0.1 M citric acid and 0.2 M sodium phosphate, pH 4.5). Control rats received an equivalent volume of the citrate buffer solution alone. Control and diabetic rats were caged separately but housed under similar conditions. Both groups of animals were fed the same diet and water ad libitum until they were used 4-6 wk later. This period of diabetes was chosen because our previous study characterized alterations in the myocardial -adrenoceptor-G protein-adenylate cyclase system during this period (8). Some control groups received water ad libitum but only enough food per day to maintain body weight in the same range as the diabetic rats (~5-10 g/day). This group was designated the food-restricted, age-matched control group. Five diabetic rats were treated daily with subcutaneous injection of Ultralente insulin (8 U/day; Novo Nordisk, Copenhagen, Denmark). Insulin therapy was begun 1 day after streptozotocin injection and was continued up to the day before the animals were killed. On the day of the experiment, a blood sample was collected and serum glucose level was determined by Rapid Blood Analyzer Super using Uni-Kit (Chugai, Tokyo, Japan). All animals injected with streptozotocin developed severe diabetes, as indicated by increased serum glucose levels (range 433-680 mg/dl).

Isolation of ventricular myocytes. Single ventricular myocytes of the rats were obtained by essentially the same technique as described previously (18). Briefly, the heart was quickly dissected from a open-chest rat that was ventilated with an artificial respirator. The rat was cannulated and perfused by a Langendorff apparatus. The heart was retrogradely perfused with a modified Tyrode solution at a temperature of 36°C until its beating rate became stable. The composition of the solution (pH 7.4) was (in mM) 143 NaCl, 5.4 KCl, 1.3 CaCl₂, 0.5 MgCl₂, 0.33 NaH₂PO₄, 5.0 HEPES, and 5.5 glucose. The perfusate was changed to a nominally Ca²⁺-free solution for 5 min, resulting in cessation of the heartbeat. The quiescent heart was perfused with a nominally Ca²⁺-free Tyrode solution containing collagenase [0.03-0.05% (wt/vol); Wako Pure Chemical, Osaka, Japan] for 40-60 min. The collagenase solution was washed out with a Kraftbrühe (KB) solution that contained (in mM) 70 KOH, 50 L-glutamic acid, 40 KCl, 20 taurine, 20 KH₂PO₄, 3.0 MgCl₂, 10 glucose, 0.5 EGTA, and 10 HEPES (pH 7.4). The ventricular tissue was cut into small pieces, agitated gently in a small beaker with KB solution, and then filtered through a 100-mm stainless steel mesh.

Simultaneous measurement of length and indo 1 fluorescence. Single myocytes bathed in KB solution were loaded with the fluorescent Ca²⁺ probe indo 1 by incubation with indo 1-AM (5 µM; Dojin, Kumamoto, Japan) and 0.02% Pluronic F-127 (Molecular Probes, Eugene, OR) for 10 min at room temperature, followed by washing with KB solution for 60 min. Small aliquots of loaded myocytes were placed in the experimental chamber filled with Tyrode solution, allowed to settle for 5 min, and superfused with Tyrode solution for at least 15 min. Myocytes were then field stimulated at a rate of 0.5 Hz by a pair of platinum electrodes connected to an electronic stimulator (SEN-7203, Nihon Kohden, Tokyo, Japan) through an isolation unit (SS-104J, Nihon Kohden).

Assessment of SR Ca²⁺ content. Rapid application of a caffeine-containing solution induces a [Ca²⁺]_i transient and contracture in cardiomyocytes, and the amplitude can be used as an index of SR Ca²⁺ content (3). The indo 1-AM-loaded myocytes were stimulated at 0.5 Hz until the [Ca²⁺]_i transients became stable. Stimulation was then stopped, and a rapid pulse of 10 mM caffeine (<1 s) was applied to the myocyte via a micropipette (500-mm tip) at the end of an ~10-s rest interval. Our estimates for caffeine-induced [Ca²⁺]_i transients after rest intervals of 5-120 s confirmed that there was little difference in the caffeine responses elicited after these different intervals. The micropipette was placed downstream with respect to the superfusate flow, and this position prevented caffeine leaking out of the pipette from diffusing to the myocyte. After control measurements, the myocytes were exposed to 1 nM isoproterenol, 1 µM forskolin, or 1 mM DBcAMP and stimulated continuously at 0.5 Hz until a steady-state effect of each drug was achieved. Postrest caffeine protocol was then repeated.

Measurement of I_Ca. Membrane currents were recorded in the whole cell voltage-clamp mode of the patch-clamp technique, using glass patch electrodes with a resistance of 1-2 M. The electrode was connected to the input of a patch-clamp amplifier (CEZ 2300, Nihon Kohden). The signals were displayed on an oscilloscope (2221A, Sony-Tektronix, Tokyo, Japan) and were simultaneously fed to a data recorder system, consisting of a videocassette recorder (NV-F1, National, Osaka, Japan) and a PCM converter system (PCM-501ES, Sony, Osaka, Japan) as a backup. The current and voltage signals were filtered at 1 kHz, digitized by a analog-digital converter (ADX-98, Canoopus Electronics, Kobe, Japan) at 2 kHz, and stored in the 20-Mb hard disk of a personal computer (PC-98RL, NEC, Tokyo, Japan) for later analysis.

Statistical analysis. All values are expressed as means ± SE. Student's t-test was used to make comparisons between control and diabetic groups. Whenever the necessary prerequisites for the condition of a parametric test were met, a nonparametric Mann-Whitney U-test was applied. P values <0.05 were taken as significant.

Drugs. The following compounds were used: l-isoproterenol hydrochloride (Sigma), forskolin (Research Biochemicals, Natick, MA), DBcAMP (Daiichi Pharmaceutical, Tokyo, Japan), and caffeine (Nacalai Tesque, Kyoto, Japan). Isoproterenol and DBcAMP were dissolved in distilled water. Forskolin and caffeine were dissolved in absolute ethanol and dimethyl sulfoxide, respectively, before dilution in distilled water. Further dilutions to the desired concentrations were made with suitable buffer solution.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

General features of animals. Four to six weeks after injection, all rats treated with streptozotocin exhibited severe hyperglycemia, and their serum glucose levels (583 ± 11 mg/dl, n = 30) were ~3.5-fold (P < 0.001) higher than the levels of age-matched control animals (167 ± 3 mg/dl, n = 24). Furthermore, the body weights of diabetic rats (179 ± 4 g, n = 30) were significantly lower than those of control rats (315 ± 4 g, n = 24; P < 0.001). These data are in good agreement with previous work done in our laboratory (8). Food intake by the food-restricted group was reduced so that their loss of body weight (153 ± 4 g, n = 5) was similar to the diabetic group, but serum glucose (152 ± 4 mg/dl, n = 5) was maintained at the level of the age-matched control group. Serum glucose levels (144 ± 15 mg/dl, n = 5) and body weight (288 ± 6 g, n = 5) were significantly improved in insulin-treated diabetic rats.

Morphological parameters of isolated ventricular myocytes. When ventricular myocytes in KB solution were taken into the experimental chamber filled with the Ca²⁺-containing Tyrode solution, cell viability of control myocytes showed a relatively constant level of 50-70%. A similar percentage of viable myocytes was maintained in food-restricted control and insulin-treated diabetic groups. On the other hand, diabetic myocytes were very intolerant to Ca²⁺ and the yield of Ca²⁺ tolerant cells was quite different from experiment to experiment: the percentage of viable myocytes ranged from 20 to 70%. However, the morphological parameters in diabetic myocytes isolated successfully by our enzymatic digestion did not differ from those in control myocytes. Thus, when the length and width of viable myocytes, defined by the rod shape of the cell, were determined in one series of experiments, length was 123.6 ± 2.8 and 116.7 ± 2.5 µm and width was 28.5 ± 0.9 and 27.6 ± 0.7 µm in control and diabetic groups, respectively (n = 30 cells from 3 rats for each). There was no statistically significant difference in these parameters between the two groups.

Changes in myocyte contractility and [Ca²⁺]_i transients. Diabetic myocytes showed cell shortening qualitatively similar to that of controls when stimulated electrically at 0.5 Hz. Thus cell shortening (change in length/resting length × 100) was 2.04 ± 0.12% (n = 63) for control and 2.04 ± 0.17% (n = 61) for diabetic myocytes. In addition, diastolic [Ca²⁺]_i and peak systolic [Ca²⁺]_i, as monitored by the ratio of fluorescence at 410 nm to that at 485 nm, did not differ between control (0.306 ± 0.004 and 0.381 ± 0.007, n = 63) and diabetic (0.318 ± 0.007 and 0.384 ± 0.008, n = 61) myocytes.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Representative tracings showing effect of 10 nM isoproterenol on cell shortening and intracellular Ca2+ concentration ([Ca2+]i; expressed as indo 1 fluorescence) transients in indo 1-AM-loaded myocytes obtained from control (A) and diabetic (B) hearts.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Concentration-response curves for effects of isoproterenol on cell shortening (A) and [Ca2+]i transients (B) in indo 1-AM-loaded myocytes obtained from control () and diabetic () hearts. Points (means ± SE; n = 5-9 cells) represent net increase in parameters expressed as percentage of basal values recorded before application of isoproterenol. * P < 0.05, ** P < 0.01 compared with corresponding control values.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Effects of 1 µM forskolin and 3 mM dibutyryl cAMP (DBcAMP) on cell shortening (A) and [Ca2+]i transients (B) in indo 1-AM-loaded myocytes obtained from control (open bars) and diabetic (hatched bars) hearts. Columns (means ± SE; n = 6-8) represent net increase in parameters expressed as percentage of basal values recorded before application of agents. * P < 0.05, ** P < 0.01 compared with corresponding control values.

Changes in I_Ca. In each voltage-clamp experiment, membrane capacitance was measured immediately after disruption of the membrane patch. Membrane capacitance of diabetic myocytes (123.9 ± 6.4 pF, n = 30) did not significantly differ from that of controls (132.1 ± 8.3 pF, n = 30). Typical tracings of I_Ca elicited by a depolarizing pulse from a holding potential of 40 mV to 0 mV in control and diabetic myocytes are shown in Fig. 4A. The net I_Ca obtained in diabetic myocytes was apparently the same as that in controls. The estimated density of I_Ca was identical in control and diabetic myocytes (14.0 ± 1.0 vs. 12.4 ± 1.0 pA/pF, n = 30 for each group). The current-voltage relationships obtained from the cell types were similar (Fig. 4B). Thus in both control and diabetic myocytes, depolarizations less negative than 30 mV elicited an inward current, and I_Ca density reached a maximum at 0 mV.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of 10 nM isoproterenol on Ca2+ current (ICa) in control and diabetic myocytes. A: typical tracings of ICa elicited by a depolarizing pulse from a holding potential of 40 mV to 0 mV before (open circles) and after (closed circles) treatment with 10 nM isoproterenol. B: current-voltage relations for control and diabetic myocyte before (open symbols) and after (closed symbols) treatment with 10 nM isoproterenol. Peak amplitudes of currents elicited by increments of 10-mV depolarizing pulses from 30 to +50 mV from a holding potential of 40 mV (10-s intervals) are plotted.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Concentration-response curves for effect of isoproterenol on ICa in control () and diabetic () myocytes. ICa was elicited by a depolarizing pulse from a holding potential of 40 mV to 0 mV. Points (means ± SE; n = 3-5 cells) represent percent increases in ICa over basal levels.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   Effects of 10 µM forskolin and 3 mM DBcAMP on ICa in control (open bars) and diabetic (hatched bars) myocytes. ICa was elicited by a depolarizing pulse from a holding potential of 40 mV to 0 mV. Data (means ± SE; n = 3-5 cells) represent percent increases in ICa over basal levels.

Changes in SR Ca²⁺ content. Caffeine-induced Ca²⁺ release was used as an indirect estimate of the relative size of the SR Ca²⁺ pool. Rapid application of 10 mM caffeine induced a large rapid increase in [Ca²⁺]_i (see Fig. 7). The amplitude of [Ca²⁺]_i initiated by this maneuver was 150 ± 10% (n = 5) of the amplitude of the steady-state [Ca²⁺]_i transients in control and 105 ± 9% (n = 5; P < 0.05) in diabetic myocytes. Thus the changes were smaller in diabetic myocytes compared with controls, indicating that SR Ca²⁺ content in cardiomyocytes is decreased in diabetes. The diminished caffeine-induced [Ca²⁺]_i transient was partially improved by insulin therapy (125 ± 14%, n = 5).

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Representative tracings showing effect of 1 nM isoproterenol on [Ca2+]i change induced by rapid application of 10 mM caffeine in indo 1-AM-loaded myocytes obtained from control (A) and diabetic (B) hearts. Recordings of steady-state and caffeine-induced [Ca2+]i transients before and after isoproterenol stimulation are shown.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Over the last 20 years many studies from several laboratories have indicated some mechanisms for the diminished cardiac responsiveness to -adrenoceptor stimulation in experimental diabetes; however, the conclusions from these studies have not always been consistent (26). The reasons for this controversy are not apparent but may be related to differences in experimental models used, i.e., differences in species, strains, ages of animals, and duration of diabetes. The present study was performed using isolated cardiomyocytes prepared from rats with 4- to 6-wk streptozotocin-induced diabetes. There may be a concern about representativeness of diabetic myocytes as to whether the surviving myocytes still present features of diabetic myocardium, because diabetic myocytes appeared to be more sensitive to the disruption of the isolation process. However, the inotropic response of the myocytes to -adrenoceptor stimulation was consistent and mimicked the change found using isolated cardiac muscles in our laboratory (8). The principal findings of this work are that 1) the increasing effects of -adrenoceptor stimulation on cell shortening and [Ca²⁺]_i transient were markedly depressed in diabetic myocytes; 2) -adrenoceptor stimulation was equally effective in increasing I_Ca in control and diabetic myocytes; and 3) the enhancement by -adrenoceptor stimulation of the caffeine-induced [Ca²⁺]_i transient was blunted in diabetic myocytes. The altered responses to -adrenoceptor stimulation observed in diabetic myocytes were prevented by insulin therapy. Furthermore, we found no changes in -adrenoceptor-mediated responses in myocytes from food-restricted rats in which mean body weight was similar to that of diabetic rats. Therefore, the alterations in -adrenoceptor responsiveness of diabetic myocytes shown in this study are a consequence of the resulting diabetic state and independent of direct cardiac effects of streptozotocin or malnutrition.

The present observation that the effect of isoproterenol on cell shortening was markedly reduced in ventricular myocytes isolated from streptozotocin-induced diabetic rat hearts compared with myocytes from age-matched control rat hearts confirms and extends previous studies from our and other laboratories showing a decreased inotropic response to -adrenoceptor stimulation in multicellular myocardial tissues taken from diabetic rats (8, 10, 11, 21, 30, 35). Additionally, diabetic myocytes showed reductions in the cell-shortening responses to forskolin and DBcAMP. The diminished inotropic responses to these agents have been also observed by us in papillary muscles isolated from diabetic rats (8). These findings are consistent with the view that even when -adrenoceptors and adenylate cyclase are bypassed, diabetes still causes an impairment of the inotropic response. There have been many reports of diminished numbers of myocardial -adrenoceptors in diabetic rats (1, 11, 19, 21, 22, 33). However, our previous and present data with forskolin and DBcAMP suggest that a potential defect in the inotropic responsiveness to -adrenoceptor stimulation may reside at a level distal to -adrenoceptors rather than at the level of the receptors.

It has been shown that an increased amplitude of the [Ca²⁺]_i transients is a mechanism for the positive inotropic action of myocardial -adrenoceptor stimulation (6). We found that the increase in the amplitude of the [Ca²⁺]_i transients produced by isoproterenol was significantly attenuated in diabetic myocytes. This agrees with previous results (36). Thus the decreased [Ca²⁺]_i response is associated with the inotropic response to -adrenoceptor stimulation in diabetic myocytes. However, the decreased [Ca²⁺]_i response involved a decrease in sensitivity to isoproterenol as estimated by the pD₂ value despite the lack of difference in contractile sensitivity to isoproterenol between control and diabetic myocytes, implying that there may be a dissociation of the contractile response from the [Ca²⁺]_i response in diabetic myocytes. Because -adrenoceptor stimulation results in increasing I_Ca and promoting sequestration of Ca²⁺ by the SR (25, 27), the observed increase in the amplitude of the [Ca²⁺]_i transients by isoproterenol can be interpreted as the increase in Ca²⁺ influx via Ca²⁺ channels and the release of Ca²⁺ from the SR. In this study diabetic myocytes were less responsive to forskolin and DBcAMP with respect to amplitude of the [Ca²⁺]_i transients. Yu et al. (36) have also demonstrated a depressed maximum [Ca²⁺]_i response to 8-bromo-cAMP in diabetic myocytes. These data with cAMP-generating or -mimetic agents further strengthen the argument that the diminished number of myocardial -adrenoceptors does not totally account for the decreased inotropic response to -adrenoceptor agonists in diabetes. Indeed, our previous study has shown that the level of stimulation of adenylate cyclase activity with -adrenoceptor stimulation is preserved well in diabetic myocardium (8). This suggests that a defect beyond the level of adenylate cyclase is associated with the impaired functional responsiveness in diabetes.

To further explore a possible defect at the level beyond the steps of activation of adenylate cyclase, the ability of isoproterenol to increase I_Ca was examined in control and diabetic myocytes using the whole cell patch-clamp technique. Basal I_Ca density was not altered in diabetic myocytes compared with age-matched control myocytes. Jourdon and Feuvray (13) have also demonstrated that ventricular myocytes from streptozotocin-induced diabetic rats (3-4 wk) exhibited no change in I_Ca density with no alteration either in the current inactivation constants or in the voltage dependence of both activation and steady-state inactivation, although long-term streptozotocin-induced diabetes (24-30 wk) has been reported to show a decrease in I_Ca (31). We found that the increasing effect of isoproterenol on I_Ca did not differ between control and diabetic myocytes. In contrast, the increase in I_Ca produced by isoproterenol has been shown to be less in ventricular myocytes from genetically diabetic rats at the age of 19 mo but not at the age of 8 mo (28). The reason for this discrepancy is not clear but may be related to different characteristics of the diabetic animal model used. The present findings of no difference in the I_Ca-increasing effect of isoproterenol are in accordance with our previous biochemical study, which found that -adrenoceptor-adenylate cyclase coupling is not impaired in diabetic myocardium (8). Furthermore, our results indicate that phosphorylation of cardiac Ca²⁺ channels through activation of PKA is unaltered in diabetes. This is also supported by the finding that forskolin and DBcAMP increased I_Ca similarly in control and diabetic myocytes. The results obtained using forskolin and DBcAMP also exclude the possibility that the adenylate cyclase-independent action of isoproterenol on Ca²⁺ channels via the direct effect of G_s on the channel gating mechanism (34) might have contributed to the isoproterenol response of diabetic myocytes being apparently normal. However, it is not completely ruled out that alterations in extra- and intracellular environments might change these responses of diabetic myocytes, because the internal and external solutions impose identical environments on both control and diabetic myocytes in our experimental conditions. This point remains to be further studied. Nevertheless, on the basis of our estimates of effects of -adrenoceptor stimulation and related interventions on I_Ca, it seems reasonable to conclude that the decreased inotropic and [Ca²⁺]_i responses to -adrenoceptor stimulation in diabetic myocytes are unlikely to be caused by alterations in PKA-dependent phosphorylation of Ca²⁺ channels and consequently increased open probability of the channels.

With respect to SR Ca²⁺ content, it has been reported that the magnitude of rapid cooling contracture, which can be used as an indirect index of the SR Ca²⁺ available for release, is diminished in diabetic papillary muscles (4). This observation has subsequently been confirmed in isolated ventricular myocytes (37). As previously demonstrated (16, 37), the present study showed that the rise of [Ca²⁺]_i in response to rapid caffeine application was also significantly decreased in diabetic myocytes. Taken together, these results suggest that SR Ca²⁺ content is reduced in diabetes. The reduction in cardiac SR Ca²⁺ content may stem from depressed SR Ca²⁺-ATPase activity and Ca²⁺ uptake into the SR (17). The amount of Ca²⁺ released from the SR is believed to be graded by the amount of trigger Ca²⁺ entering the cell (2). We showed that basal I_Ca density was unaltered in diabetic myocytes. Our finding may imply that the main source of trigger Ca²⁺, Ca²⁺ influx through Ca²⁺ channels, is not impaired in diabetes. Alternatively, Ca²⁺ release from the SR may be caused by a voltage-sensitive Ca²⁺ release mechanism distinct from the Ca²⁺ channel-mediated Ca²⁺ entry, as proposed by Ferrier and Howlett (7). On the other hand, a decrease in ³H-labeled ryanodine binding sites in diabetic myocardium has been reported (37), suggesting that the density of SR Ca²⁺-release channels is reduced in diabetes. Thus decreased SR Ca²⁺ content and/or reduced SR Ca²⁺-release channels may lead to an impairment of the Ca²⁺ release process of the SR. However, the basal [Ca²⁺]_i found in electrically stimulated myocytes showed that there was no difference between control and diabetic groups, which is consistent with a previous report (36). Therefore, the expression of an impaired Ca²⁺ release from the SR in diabetes may become manifest only when myocytes are challenged with the interventions that result in increased Ca²⁺ to release from the SR.

Isoproterenol enhanced the rise of [Ca²⁺]_i in response to rapid caffeine application in control myocytes more markedly than in diabetic myocytes. Similar findings were obtained with forskolin and DBcAMP instead of isoproterenol. These results suggest that the increase in SR Ca²⁺ uptake via PKA-dependent phosphorylation of phospholamban may be impaired in diabetes. In support of this, we have recently shown that phosphorylation of phospholamban in response to isoproterenol and forskolin is blunted in diabetic rat hearts (8). Therefore, it would be logical to conclude from the present results that the decreased responsiveness to myocardial -adrenoceptor stimulation in diabetes results largely from the impairment of SR functions of uptake and release of Ca²⁺.

In summary, the cell-shortening effect and the increase in [Ca²⁺]_i transients produced by -adrenoceptor stimulation were markedly impaired in ventricular myocytes isolated from streptozotocin-induced diabetic rats. Activation of Ca²⁺ channels with -adrenoceptor stimulation was not altered in diabetic myocytes. The functions of the SR appeared to be basically depressed in diabetes. Furthermore, the impairment of PKA-dependent regulation of Ca²⁺ uptake and release by the SR, which is one of the downstream events activated in response to -adrenoceptor stimulation, might have occurred. These alterations in the SR may primarily contribute to the impaired inotropic response to -adrenoceptor stimulation in diabetes.