INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

WE HAVE RECENTLY ESTABLISHED that an autocrine release of angiotensin II (ANG II) plays a central role in attenuating a transient and a sustained K⁺ current in single myocytes isolated from (Type 1) diabetic rats (45). The in vitro blockade of ANG II action or formation (for >5 h) augments both currents. This effect is blocked by cycloheximide, suggesting that new protein synthesis mediates current augmentation after removal of ANG II-related current inhibition. This protein synthesis may be of the pore-forming -subunits of the channels, but not necessarily so. There is currently an increasing awareness of the complexity of channel function, with key roles established for -subunits as well as various accessory and chaperone proteins. These can interact with the pore-forming subunit of the channel and modulate its surface expression (3, 27, 40, 50). Thus changes in current magnitude may reflect alterations in these accessory proteins as well as changes in channel processing or in posttranslation modification (e.g., 38). It is therefore essential to directly establish whether an augmentation in the -subunit of the channel protein can in fact be identified after interference with ANG II action.

Several additional issues relating to the autocrine/paracrine control of ion channels arise from the preceding work. ANG II has multiple and diverse cellular actions (11). A key modulator of cellular function, protein kinase C (PKC), is a major target of ANG II (11). PKC is activated in diabetes, with several isoforms affected (22, 29). The -isoform of PKC is of particular interest, because it is involved in several cardiac pathologies and plays a key role in cardioprotection (10). Furthermore, PKC- also interacts with K⁺ channels (33, 44). Diabetes leads to a change in the subcellular distribution of PKC-, so that a greater proportion is found in membrane fractions (29). This is prevented by ANG II receptor blockade (29). The acute effects of PKC- activation on K⁺ currents are altered in diabetes as well (44). Alterations in K⁺ currents can lead to severe cardiac arrhythmias (19, 27, 48). Thus it is of importance to establish some of the mediating mechanisms, such as whether the suppression of the two K⁺ currents by ANG II is associated with PKC- activation and translocation.

Other pathological conditions that are associated with PKC- activation and subcellular translocation also result in attenuated K⁺ currents. These include hypoxia and ischemia (16) as well as cardiac hypertrophy (17). The same transient and sustained currents are attenuated in ventricular myocytes from hypothyroid rats (46), in which PKC- is also activated (42). It was thus of interest to establish whether ANG II is also a mediator of K⁺ current suppression under conditions other than diabetes. Finally, there have been reports indicating that after stress there is enhanced production and/or release of other autocrine/paracrine agents in the heart. For example, endothelin (ET)-1 is released (12), possibly from cardiac myocytes as well as from other cell types (41).

The aims of this study were therefore to extend earlier work and establish the following: 1) Does restoration of current magnitudes after ANG II blockade involve the synthesis of new channels? 2) Does autocrine/paracrine regulation of K⁺ currents occur in other pathologies such as under hypothyroid conditions? 3) Do other agents such as ET-1 play a role in the chronic suppression of K⁺ currents? 4) Is PKC- involved in the chronic suppression of K⁺ currents and, more specifically, does the restoration of current magnitudes involve a return of PKC distribution to the normal (prediabetic, euthyroid) state?

The present work suggests that the answer to all of these questions is affirmative.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All of the experiments were done according to guidelines established and approved by the Animal Care Committee of the University of Calgary. Rats were anesthetized by methoxyflurane or CO₂ inhalation and killed by cervical dislocation. Hearts were removed, and the aortas were cannulated for coronary perfusion on a Langendorff apparatus.

Animals. Ventricular cells or tissue were obtained from three groups of adult Sprague-Dawley male rats (200-300 g): untreated control rats, rats made diabetic with a single intravenous injection of streptozotocin (100 mg/kg) 8-14 days before the experiments, and rats made hypothyroid by thyroidectomy 4-5 wk before the experiments.

Cell isolation. Current recordings were done using isolated cells from the free wall of the right ventricle, obtained by enzymatic dispersion using collagenase (Yakult; Tokyo, Japan) and protease (Sigma type XIV). Details can be found in an earlier publication (44).

Current recordings. These were done using the whole cell suction electrode method. Low-resistance electrodes (2-4 M) were used to minimize series resistance, which was also compensated for electronically. Pipette solutions contained (in mM) 120 K-aspartate, 30 KCl, 4 Na₂ATP, 10 EGTA, 1 CaCl₂, 1 MgCl₂, and 10 HEPES (pH brought to7.2 with KOH). The perfusing solution contained (in mM) 150 NaCl, 5.4 KCl, 1 CaCl₂, 1 MgCl₂, 5.5 glucose, 5 HEPES, and 0.3 CdCl₂ to block L-type Ca²⁺ current. Two currents were recorded: the calcium-independent transient outward current (34) and a sustained delayed rectifier current. The transient current was measured as the peak outward current (I_peak) and not as the difference between the peak and steady-state currents, because the steady-state current was also subject to differential changes in these experiments. The sustained current activates considerably slower than I_peak (4) and was measured at the end of 500-ms pulses, at which time I_peak is completely inactivated. The late current is referred to as I_late. Recent analysis by Himmel et al. (21) has shown that a mixture of four currents is present in rat ventricular myocytes, and the effects of the angiotensin-converting enzyme (ACE) inhibitor quinapril were analyzed in detail using their methods (see below) to establish which components are affected.

Data analysis. Currents were recorded at 20-22°C (digitized at 2 kHz using a DT2821 analog-to-digital board) and stored for subsequent analysis. Cell capacitance was measured for each cell by integrating current traces (digitized at 10 kHz) obtained in response to 5-mV steps from a holding potential of 80 mV. Dividing current magnitudes by capacitance gave current densities.

Reagents. Saralasin was obtained from Calbiochem, and PD142893 and the ET-converting enzyme (ECE) inhibitor (fragment 16-38 of Big ET-1) were from Sigma. Quinapril was a gift from Dr. A. Gillis (Foothills Hospital, Calgary, Alberta, Canada). The PKC-related peptides were obtained from Dr. D. Mochly-Rosen, Stanford University.

Western blotting. The channel proteins encoded by Kv4.2 and Kv4.3 underlie the transient outward current in rat ventricular cells (14, 34). The sustained current has several components (21) and is the complex expression of several channel proteins (34). In our experiments, antibodies directed against Kv4.2, Kv1.2 (Alomone), and Kv4.3 (Alomone and an antibody kindly provided by Dr. J. Nerbonne) were used. We were able to obtain consistent and reliable data using the Kv4.2 and Kv1.2 antibodies. However, both antibodies directed against Kv4.3 gave bands of poorer quality, and the results could not be used. However, the results with the Kv4.2 and Kv1.2 antibodies were sufficient to establish that current changes are paralleled by changes in protein density (see below).

Quantification of Kv4.2 and Kv1.2. SDS-PAGE gels used for transfer to nitrocellulose and Western blotting were run simultaneously with gels loaded with 100 µg protein and stained with Coomassie brilliant blue R-250 (Bio-Rad). Densitometry was performed on the immunoreactive bands as well as on an arbitrary band on the Coomassie blue protein gels to quantitatively normalize to protein loading levels in all lanes. A Sharp JX-330 Scanner and Image Master 1D Software (Pharmacia Biotech) were used. Each gel had two to three samples from each experimental group, which allowed for variations in optical density (OD) due to variations in film development. Channel protein in the different groups (in OD units) was normalized to the OD of the Coomassie blue band or to protein loading levels (100 µg protein) and averaged for the different groups. Both methods of normalization gave similar results.

Statistics. An unpaired Student's t-test or (mostly) ANOVA with a Student-Newman-Keuls multiple-comparison post hoc test were used to establish whether experimental groups differed (with P < 0.05 indicating a significant difference).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The first set of results, summarizing and extending earlier work, demonstrates that the exposure of myocytes from diabetic rats to the ACE inhibitor quinapril for 6 h or longer augments the peak (transient) and late outward currents. Figure 1 shows sample current traces (A) as well as the full current-voltage relationships (B) for I_peak (left) and I_late (right). Mean (±SE) current densities are plotted against membrane potentials in the absence and presence of quinapril (6- to 9-h incubation). I_peak was significantly augmented by quinapril (P < 0.0005 for membrane potentials between 30 and +50 mV). I_late in this potential range was also significantly enhanced (P < 0.005). There were no effects of quinapril on I_late in the negative potential range (50 to 110 mV), in which the inward rectifying background current is activated.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effects of an angiotensin-converting enzyme (ACE) inhibitor on K+ currents in ventricular myocytes from diabetic rats. A: sample current traces obtained in myocytes from a diabetic rat in response to 500-ms depolarizing steps from 80 mV to membrane potentials ranging from 0 to +50 mV. Left, traces in an untreated cell; right, membrane potentials in a cell exposed to 1 µM quinapril for 7 h. The line on the left indicates the zero-current level. B: current-voltage relationships showing the mean (±SE) densities for peak currents (Ipeak; left) and late currents (Ilate; right) in the absence of quinapril (circles) or after 6-9 h in 1 µM quinapril (squares). The differences for both currents were highly significant between 30 and +50 mV (P < 0.0005 for Ipeak; P < 0.005 for Ilate).

We subsequently attempted to analyze more precisely which current components are modulated by quinapril. We repeated the analysis of Himmel et al. (21) and analyzed the inactivation of the total outward currents. Prepulses (2,000 ms) were given to potentials ranging from 140 to 20 mV in 10-mV steps (5-mV steps between 60 and 20 mV), followed by a 500-ms pulse to +50 mV. Current amplitudes were normalized to maximal currents, and the data were fit to a sum of two Boltzmann functions with the residual component r. The analysis was done separately for I_peak and I_late at the end of the test pulses. The different parameters (see METHODS for the equation) were compared in cells from diabetic rats in the absence (n = 25) or presence (n = 21) of quinapril (6-9 h). The mean and SE values, shown in Table 1, show that both a and b (for I_peak) are significantly augmented by quinapril, with no effect on the residual current r or on half-inactivation voltages. The slope factor k was also unchanged (not shown for clarity). This suggests that both a transient and a delayed rectifier current are enhanced by ACE inhibition. We then proceeded to investigate whether this could be correlated to an augmentation in the expression of proteins underlying these currents.

The increase in current magnitude was shown earlier to be sensitive to cycloheximide, suggesting that protein synthesis is involved. The transient outward current in rat ventricular cells is associated with the channel proteins Kv4.2 and Kv4.3 (14, 18, 34). The molecular correlate of I_late is unclear and is presumed to be a mixture of several isoforms, including Kv1.2 (34).

Nishiyama et al. (35) reported that under diabetic conditions, there is a decrease in Kv4.2 message and protein levels. Interestingly, hypothyroid conditions also resulted in a reduction in message levels of Kv4.2 (36). Qin et al. (39) also showed a reduction in message and protein levels of Kv4.2 in diabetic conditions. These changes in Kv4.2 presumably underlie the attenuation in the transient outward current in diabetic and hypothyroid conditions (26, 46).

We set out to establish whether the augmentation in I_peak by the ACE inhibitor quinapril is paralleled by an increase in Kv4.2. We performed Western blots on homogenized ventricular tissue obtained from hearts perfused for 7-9 h. We used control and diabetic rat hearts, with the latter divided into untreated hearts and hearts perfused with 1 µM quinapril. We confirmed that Kv4.2 density is attenuated under our experimental conditions as well, although in this group the reduction in Kv4.2 density in diabetic hearts was not significant. Nevertheless, Kv4.2 protein in diabetic hearts was significantly (P < 0.03) augmented after perfusion with quinapril compared with untreated diabetic hearts. Figure 2, top left, shows sample blots showing results from the three experimental groups. HEK-293 cells, transfected with cDNA for Kv4.2, were also used as a control to ensure the correct identification of the band corresponding to Kv4.2. A further control was obtained by incubating tissue with the peptide used to generate the antibody. This eliminated the signal at the appropriate location on the gels (Fig. 2, top right). Densitometry scans of the gels were used to quantify the signals, with the mean (±SE) data (normalized OD) shown in Fig. 2, bottom.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Western blots obtained with antibodies directed against Kv4.2. Top: two gels from ventricular extracts obtained from control (Cont) and streptozotocin (STZ)-treated rats in the absence of quinapril or after 9-h perfusion with 1 µM quinapril (Q). The left shows one lane in which extracts of human embryonic kidney (HEK)-293 cells, transfected with Kv4.2, were run in parallel to the ventricular extracts. The right shows that incubation with the control peptide used to generate the antibodies (4 blank lanes) prevents labeling of the band that corresponds to Kv4.2 (although other bands were also blocked). In each gel, extracts from 2 different rats are shown for each group. Bottom: summary data obtained by densitometry. Optical density (OD)/100 µg protein is plotted for the 3 groups, showing that in STZ-treated rats there is a significant (* P < 0.03) increase in Kv4.2 density after quinapril treatment. OD normalized to scanned bands on Coomassie blue-stained gels gave similar results.

This result shows that the augmentation of the transient outward current, which occurs after a reduction in the formation of angiotensin, is associated with an enhanced synthesis of Kv4.2 channels.

Western blot using the Kv1.2 antibody showed that this protein is reduced in diabetic hearts compared with control and, more importantly, that Kv1.2 is significantly (P < 0.02) augmented after 6-9 h in quinapril. This result is shown in Fig. 3. Blotting with this antibody consistently gave two bands, which may represent two forms of the channel (e.g., one phosphorylated). The identity of the band for Kv1.2 (used in densitometry) was confirmed in a control experiment using the competing peptide. This abolished the lower of the two bands, as shown in Fig. 3.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Western blot using an antibody directed against Kv1.2. Top: gels with the band for Kv1.2 using cardiac extracts from control (C) rats, STZ-treated rats, and STZ-treated rats after perfusion for 8-9 h with 1 µM quinapril (Quin). Two bands were labeled by this antibody, but the competing peptide only blocked the lower one, as shown in the two lanes on the right. Extracts from 2 control rats in the absence and presence of the peptide identify the lower band as corresponding to Kv1.2. Bottom: mean ± SE values of the OD (normalized to an arbitrary band on a Coomassie blue-stained gel with the same samples). Kv1.2 is significantly decreased in diabetic hearts, but quinapril significantly enhances the abundance of this protein. * P < 0.05.

The results with Western blot show that the increase in transient and sustained (late) currents is paralleled by an increase in channel protein density. These results correlate to the increase we found in parameters a and b (for the early current) in the inactivation process (see above) and thus provide strong validation for the analysis of Himmel et al. (21), who suggested that a and (early) b correspond to a sustained and a transient current, respectively.

Hypothyroid conditions. Because the local cardiac renin-angiotensin system is activated in a variety of pathological conditions (11), we proceeded to examine whether similar effects could be demonstrated in situations other than diabetes. We have previously extensively studied the effects of altered thyroid status on the same K⁺ currents, demonstrating a reduction in both transient and sustained currents in hypothyroid conditions (46). Thus, in the next set of experiments, we studied whether ACE inhibition could restore the attenuated currents in myocytes from hypothyroid rats, as is the case in diabetic conditions. Similar protocols were followed, with exposure of myocytes to 1 µM quinapril for 6 h or longer. This procedure significantly enhanced both the transient and sustained outward currents, measured at +50 mV. Quinapril increased I_peak from 15.3 ± 0.8 (40 cells) to 19.7 ± 1.0 (47 cells) pA/pF (P < 0.003) and I_late from 5.9 ± 0.3 to 7.3 ± 0.3 pA/pF (means ± SE, P < 0.003).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Effects of an ACE inhibitor on K+ currents in myocytes from hypothyroid rats. A: sample current traces obtained in response to pulses from 80 mV to potentials ranging from 0 to +50 mV in the absence of quinapril (left) or after 6.5 h in 1 µM quinapril (right). B: current-voltage relationships for Ipeak (left) and Ilate (right) in the absence of quinapril (; n = 27) or after 6-9 h in 1 µM quinapril (; n = 32). Mean (±SE) current densities are shown. For Ipeak, the differences were significant (P < 0.003) between 30 and +50 mV (asterisks left out for clarity). For Ilate, the differences were significant (P < 0.003) between 20 and +50 mV.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   A: Western blots obtained with antibodies directed against Kv4.2. Blots were obtained from ventricular tissue in control, hypothyroid [thyroidectomized (T)], and hypothyroid hearts perfused for 9 h with 1 µM quinapril (2 rats from each group). B: histogram showing mean (±SE) values of the OD/100 µg protein for each group. Kv4.2 is significantly (P < 0.05) reduced in hypothyroid conditions. After perfusion with quinapril, Kv4.2 density is significantly (P < 0.007) enhanced. * P < 0.05.

Involvement of ET-1. We next examined whether other autocrine/paracrine factors could be implicated in suppressing the two K⁺ currents in the diabetic and hypothyroid states. Of particular interest was the peptide ET-1. This peptide, which may be involved in cardioprotection and/or arrhythmogenesis, is released in the heart under conditions of stress (12, 41). Importantly, there is evidence that ET-1 is released from the myocytes themselves (1, 23). ET-1 was also of particular interest because among its multiple actions is the suppression of several K⁺ currents in rat and human ventricular myocytes (7, 24, 28; see also DISCUSSION) as well as an activation and selective translocation of PKC- (25, 43, 47).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Effects of endothelin (ET)-1 receptor blockade on K+ currents in diabetic myocytes. A: sample current traces (same voltage protocol as above) obtained from an untreated cell (left) and from a cell incubated for 6 h with 0.5 µM PD142893 (right), a blocker of (types A and B) ET-1 receptors. B: current-voltage relationships for Ilate (left) and Ipeak (right) in the absence of PD14289 (; n = 17) or after >6-h exposure to PD142893 (; n = 11). The currents were significantly enhanced by PD142893 between 30 and +50 mV for Ilate and between 40 and +50 mV for Ipeak. C: summary data for Ilate (left) and Ipeak (right) densities (at +50 mV) showing increases in Ilate and Ipeak (see text for values). ** P < 0.005.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Effects of inhibition of ET-converting enzyme (ECE) on outward currents in cells from STZ-induced diabetic rats. A: sample current traces from an untreated cell (left) and from a cell incubated for 7 h with 5 µM of the 16-38 fragment of Big ET-1, an ECE inhibitor (right). Current traces are in response to 500-ms pulses from 80 mV to potentials ranging from 20 to +50 mV. B: current-voltage relationships for Ipeak (left) and Ilate (right) in the absence () or presence () of the ECE inhibitor. The effects of the ECE inhibitor were highly significant (P < 0.001) for both currents between 20 and +50 mV.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Effects of blocking ET-1 receptors on the density of Kv4.2 in thyroidectomized rats. Top: Western blot using antibodies for Kv4.2 obtained from extracts from hearts perfused for >7 h. The bands are from a control rat, thyroidectomized rats, and thyroidectomized rats perfused with 0.27 µM PD142893 (missing bands in middle were from a different protocol). Bottom: mean normalized (to a band on a Coomassie blue-stained gel) OD values in thyroidectomized rats in the absence (open bars; n = 6) or presence (hatched bars; n = 7) of 0.27 µM PD142893. The density of Kv4.2 was significantly (* P < 0.04) enhanced by blocking ET-1 receptors.

Involvement of PKC. We subsequently attempted to gain further insight into the cellular mechanisms involved in these autocrine/paracrine effects. ANG II has a multitude of intracellular effectors (11). One target of interest is PKC, which is a key regulator of many cellular functions (20). It was hypothesized that if the activation of PKC by ANG II (or by ET-1) mediates the attenuation of the K⁺ currents under investigation, then exposure of cells to a PKC inhibitor would at least partially restore current density.

View larger version (29K): [in this window] [in a new window]   Fig. 9.   Effects of protein kinase C (PKC) inhibition on K+ currents in cells from STZ-induced diabetic rats. A: sample current traces in response to pulses ranging from 10 to +50 mV shown in the absence of inhibitor (left) or after 7-h incubation with 100 nM bisindolylmaleimide, a PKC inhibitor. B: summary data of Ipeak (left) and Ilate (right) densities (at +50 mV) in the absence of inhibitor and after 6-9 h with the PKC inhibitor. ** P < 0.005.

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Effects of quinapril in the presence of a diacylglycerol analog. A: current traces (in response to steps from 80 mV to potentials ranging from 10 to +50 mV) obtained in cells from diabetic rats. The traces shown are from cells without quinapril (left), after 6 h in 1 µM quinapril (middle), and after 6 h in quinapril and 20 µM dioctanoyl-rac-glycerol (DiC8; added 1 h before quinapril; right). B: current-voltage relationships for Ipeak (left) and Ilate (right) corresponding to the 3 experimental conditions. Mean ± SE values are shown. DiC8 completely prevented the enhancement of the transient current by quinapril but only partially prevented the effect on Ilate.

View larger version (33K): [in this window] [in a new window]   Fig. 11.   Enhancement of Kv4.2 density by ACE inhibition is blocked by PKC activation. Top left: Western blots with bands obtained from a control rat, 2 thyroidectomized rats, 3 thyroidectomized rats perfused with 1 µM quinapril for >7 h, and 3 thyroidectomized rats perfused (7 h) with quinapril and 20 µM DiC8 (added 30 min before quinapril). Right, bands from 2 STZ-induced diabetic rats, 3 STZ-induced diabetic rat hearts perfused with quinapril for 8-9 h, and 3 bands from diabetic hearts perfused with quinapril and DiC8. The augmentation of Kv4.2 density by quinapril in both hypothyroid and diabetic conditions is clearly blocked by DiC8; 100 µg protein was loaded in all lanes. Bottom: summary data (means ± SE), with band densities normalized to an arbitrary band on a Coomassie blue-stained gel, obtained in parallel to the Western blot. Left, data from hypothyroid hearts (n = 8, 0.06 < P < 0.07); right, data from diabetic hearts (n = 3, P < 0.05).

View larger version (34K): [in this window] [in a new window]   Fig. 12.   Effects of the ANG II receptor blocker saralasin in the presence of a specific PKC- translocation agonist peptide. A: current traces obtained in cells from diabetic rats after incubation for 7 h with 1 µM saralasin (left) or with 1 µM saralasin and 1 µM peptide added 1 h before saralasin (right). B: summary data showing mean (±SE) current densities (at +50 mV) of Ipeak (left) and Ilate (right) in the absence of saralasin, in the presence of saralasin, and with saralasin and the PKC- agonist peptide. * P < 0.01; ** P < 0.001.

Control myocytes. It has been suggested that some of these autocrine/paracrine factors may be released under basal conditions, with an increase occurring under stress situations such as ischemia (12). In the final set of experiments, we tested whether quinapril or PD142893 would affect these K⁺ currents in cells from control rats. We found that an exposure of 6-9 h of cells from control rats to either 1 µM quinapril (21 cells) or 0.5 µM PD142893 (23 cells) significantly augmented I_late but had no effect on I_peak. In control cells (n = 24), the mean densities (at +50 mV) for I_peak and I_late were 22.8 ± 1.3 and 6.2 ± 0.3 pA/pF, respectively. The corresponding values after ACE inhibition by quinapril were 20.1 ± 1.5 (P > 0.05) and 7.3 ± 0.3 pA/pF (P < 0.02). After exposure to the ET-1 receptor inhibitor, the mean densities were 21.3 ± 2.1 (P > 0.05) and 7.4 ± 0.2 pA/pF (P < 0.003).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Summary of results. The results presented above provide several novel findings. Transient and sustained K⁺ currents in rat ventricular myocytes are attenuated under both diabetic and hypothyroid conditions. We showed previously that these currents are augmented after the inhibition of ANG II action or formation in diabetic conditions (Fig. 1). We now show that this is the case in hypothyroid rats as well (Fig. 4) and demonstrate that the enhancement of the transient current involves an augmentation of Kv4.2, one of the channel proteins underlying this current (Figs. 2 and 5). In diabetic conditions, we also showed (Fig. 3) that quinapril augments Kv1.2, which underlies some of I_late in rat ventricular myocytes (34). Furthermore, we found that under diabetic and hypothyroid conditions, an autocrine/paracrine release of ET-1 modulates these K⁺ currents. The in vitro blockade of ET-1 receptors or the inhibition of ET-1 formation in myoctes from diabetic or hypothyroid rats also leads to enhancement of these K⁺ currents (Figs. 6 and 7) associated with enhanced Kv4.2 protein (Fig. 8). The results also show that part of the mechanism(s) mediating the changes in current density by both ANG II and ET-1 depend(s) on a subcellular redistribution of PKC-. The attenuation of K⁺ currents is mediated by PKC activation, because a PKC inhibitor can reverse this attenuation (Fig. 9). This augmentation apparently depends on a recovery of PKC- distribution from the pathological toward the normal pattern (Figs. 10-12). Thus maintaining PKC- in an activated/translocated state prevents the enhancement of I_peak by ANG II and ET-1 inhibition. The enhancement of I_late is apparently less affected by PKC-related mechanisms than I_peak. Further work is required to establish the contribution of Kv1.2 to the b component of voltage-dependent inactivation of I_peak.

Interpretation and significance. The simplest interpretation of these results is consistent with the reported activation of the local cardiac renin-angiotensin system (11) and the autocrine release of ANG II by the myocytes or a paracrine release from nonmyocytes. In parallel, or as a result of ANG II action (1, 23), there is also an activation and autocrine/paracrine release of ET-1. Both ANG II and ET-1 contribute to suppression of the transient and sustained outward currents, the major repolarizing K⁺ currents in ventricular cells. The suppression of K⁺ currents by exogenous ET-1 when added acutely in vitro has been shown previously (7, 24). We have shown (44) that an acute activation of PKC by DiC8, which normally suppresses K⁺ currents, does not have this action in diabetic and hypothyroid conditions, presumably because PKC is already activated in these pathologies. Thus acute actions of ET-1 released by autocrine mechanisms would presumably not be present (although this was not tested directly). Our present data suggest an endogenous chronic release under diabetic and hypothyroid conditions that attenuates currents at least partially by inhibiting synthesis of the channel -subunits.

Limitations of study. There are several limitations to this study, although these were considered not to affect the major conclusions. First, it was beyond the scope of the work to directly measure the release of ANG II and ET-1 from the myocytes to establish whether there are true autocrine mechanisms involved. However, such release has been demonstrated previously (1, 11). The combined use of multiple interventions (saralasin, quinapril, PD142893, and the ECE inhibitor in both diabetic and hypothyroid conditions) strongly supports our working hypothesis and rules out possible nonspecific effects of any one of the agents used. It was beyond the scope of this work to fully analyze changes in other channel proteins. However, the changes measured in Kv4.2 and Kv1.2 strongly support the working hypothesis, whereby ANG II and ET-1 suppress channel protein synthesis. The signals for Kv4.3 were considerably less satisfactory despite the use of two different antibodies and different methods of membrane purification. There was some disadvantage to the Western blotting experiments in that ventricular tissue rather than isolated cells were used. This, however, is common practice (e.g., 36) and was done here to obtain sufficient protein. Although other cell types are abundant in ventricular tissue, the predominant tissue mass is provided by the myocytes. Other cells such as fibroblasts lack Kv4.2. In addition, the parallel changes in I_peak and Kv4.2 that were obtained also strongly support our conclusions. A final limitation lies in the fact that no direct evidence was obtained for a maintained pathological subcellular distribution of PKC with DiC8 and the peptide agonists. This too was beyond the scope of the present work, but it should be emphasized that Malhotra et al. (29) showed that ANG II receptor blockers prevented PKC- redistribution in cardiac cells, consistent with our findings.