| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
We investigated the direct effects of midazolam and diazepam on cardiac excitation-contraction coupling in adult rat ventricular myocytes. Freshly isolated rat ventricular myocytes were loaded with fura-2/AM and field-stimulated at 28°C. Intracellular Ca2+ transients (340:380 ratio) and myocyte shortening (video edge detection) were simultaneously monitored in individual cells. Midazolam (3100 µM) caused a dose-dependent decrease in both peak intracellular Ca2+ and cell shortening. Diazepam (30 and 100 µM) increased myocyte shortening and peak Ca2+ concomitant with a decrease in time to peak Ca2+. A larger concentration of diazepam (>300 µM) nearly abolished intracellular Ca2+ and cell shortening. Midazolam (100 µM) and diazepam (300 µM) decreased the amount of Ca2+ released from intracellular stores in response to caffeine. Diazepam (30 µM), but not midazolam (10 µM), caused a downward shift in the dose-response curve to extracellular Ca2+ for shortening, with no concomitant effect on peak intracellular Ca2+ transient. These results indicate that midazolam and diazepam have different inotropic effects on cardiac excitation-contraction coupling at the cellular level, which is mediated by altering the availability of intracellular-free Ca2+. However, the benzodiazepines have no direct influence on excitation-contraction coupling in rat ventricular myocytes, except at very large doses. Inhibition of Ca2+ release from caffeine-sensitive intracellular Ca2+ stores may play some part in myocardial depression at the larger concentrations of benzodiazepines. Diazepam, but not midazolam, decreased myofilament responsiveness to Ca2+. IMPLICATIONS: Midazolam and diazepam differentially alter the cardiac excitation-contraction coupling at the cellular level, which is mediated by altering the availability of intracellular free Ca2+ in adult rat ventricular myocytes. In addition, diazepam, but not midazolam, decreases myofilament Ca2+ sensitivity. However, the benzodiazepines have no direct influence on excitation-contraction coupling, except at very large doses.
Midazolam and diazepam are frequently used as an adjunctive drug during general anesthesia or as a sedative for critically ill patients (1,2). Although these benzodiazepines are characterized by their relatively minor alterations of hemodynamic variables, they cause a decrease in arterial blood pressure in humans and animals (13). However, because of concomitant changes in preload, afterload, baroreflex activity, and central nervous system activity after the induction, the direct effects of anesthetics on intrinsic myocardial contractility are difficult to assess in vivo (4). In vitro studies provide a more direct approach to examine the specific effects of anesthetics on myocardial contractility. Midazolam and diazepam have a negative inotropic effect in isolated heart preparations (58). However, the mechanism underlying the myocardial depressant effects of these benzodiazepines are not fully understood. Inhibition of the L-type Ca2+ channel is one possible mechanism for this myocardial depressant effect because Ca2+ influx across the sarcolemma via L-type Ca2+ channel is the trigger for Ca2+ release from the sarcoplasmic reticulum (SR) in cardiac myocytes. Midazolam reduces the L-type Ca2+ channel current (ICa) in a dose-dependent manner in isolated canine ventricular myocytes (9). Nakae et al. (10) reported that diazepam and midazolam cause a direct myocardial depression mainly by the inhibition of sarcolemmal L-type Ca2+ channel in cultured neonatal rat ventricular myocytes. However, the immature myocardium has different characteristics compared with adult (mature) myocardium (11). In addition to these studies of the sarcolemmal L-type Ca2+ channel, it is equally important to evaluate the direct effect of anesthetics in intact cells in their normal physiological environment. Therefore, it is important to know the direct effects of the benzodiazepines on excitation-contraction coupling and Ca2+ signaling at the cellular level in adult rat ventricular myocytes. Our objective was to determine whether midazolam or diazepam alters cardiac excitation-contraction coupling at the cellular level using freshly isolated, individual, field-stimulated adult rat ventricular myocytes. This experimental model allowed us to simultaneously measure changes in the amplitude and timing of intracellular-free Ca2+ concentration ([Ca2+]i) transients and myocyte shortening independent of any neural, humoral, or locally derived factors. We also assessed the effects of these benzodiazepines on SR Ca2+ handling and responsiveness of contractile protein to Ca2+.
This study was approved by The Cleveland Clinic Foundations Animal Care and Use Committee. Isolated adult ventricular myocytes from rat hearts were obtained as previously described (12,13). In brief, the hearts were excised, cannulated via the aorta, attached to a modified Langendorff perfusion apparatus, and perfused with oxygenated (95% O2 and 5% CO2) Krebs-Henseleit buffer (KHB; 37°C) containing the following (in mM): 118 NaCl, 4.8 KCl, 1.2 MgCl2, 1.2 KH2PO4, 1.2 CaCl2, 37.5 NaHCO3, and 16.5 dextrose, with a pH value of 7.35. After a 5-min equilibration period, the perfusion buffer was changed to a Ca2+-free KHB containing 30 mg of collagenase type II (Worthington Biochemical Corp, Freehold, NJ, Lot # M6C152; 347 U/mL). After collagenase digestion (20 min), the ventricles were minced and shaken in KHB, and the resulting cellular digest was washed, filtered, and resuspended in phosphate-free HEPES-buffered saline (HBS) containing the following (in mM): 118 NaCl, 4.8 KCl, 1.2 MgCl2, 1.25 CaCl2, 11 dextrose, 25 HEPES, and 5 pyruvate, with a pH value of 7.35 and vigorously bubbled immediately before use with 100% O2. Typically, 68 x 106 cells per rat heart were obtained using this procedure. Viability, as assessed by the percentage of cells retaining a rod-shaped morphology with no blebs or granulations, was routinely between 80% and 90%. Myocytes were suspended in HBS (1 x 106 cells/mL) and stored in an O2 hood until used. For simultaneous measurement of contraction and [Ca2+]i, ventricular myocytes (0.5 x 106 cells/mL) were incubated in HBS containing 2 µM of fura-2/AM at 37°C for 20 min, as described previously (12). Fura-2loaded ventricular myocytes were placed in a temperature regulated (28°C) chamber (Bioptechs, Inc, Butler, PA) mounted on the stage of an Olympus IX-70 (Olympus America, Lake Success, NY) inverted fluorescence microscope. The volume of the chamber was 1.5 mL. The cells were superfused continuously with HBS at a flow rate of 2 mL/min and field stimulated via bipolar platinum electrodes at a frequency of 0.3 Hz and duration of 5 ms using a Grass SD9 stimulator (Grass-Telefactor, West Warwick, RI). Myocytes were chosen for study according to the following criteria: (a) rod-shaped appearance with clear striations and no membrane blebs, (b) a negative staircase of twitch performance on stimulation from rest, and (c) the absence of spontaneous contractions. Fluorescence measurements were performed on single ventricular myocytes using a dual-wavelength spectrofluorometer (Deltascan RFK6002, Photon Technology International, South Brunswick, NJ) at excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm. The cells were also illuminated with red light at a wavelength more than 600 nm for simultaneous video edge detection. An additional postspecimen dichroic mirror deflects light at wavelengths more than 600 nm into a charge-coupled device video camera for measurement of myocyte shortening and relengthening. The fluorescence sampling frequency was 100 Hz, and data were collected using a software package from Photon Technology International (FelixTM). [Ca2+]i was estimated by comparing the cellular fluorescence ratio with fluorescence ratios acquired using fura-2 (free acid) in buffers containing known Ca2+ concentrations. Simultaneous measurement of cell shortening was monitored using a video edge detector (Crescent Electronics, Sandy, UT) with 16-ms temporal resolution. The video edge detector was calibrated using a stage micrometer so that cell lengths during shortening and relengthening could be monitored. Lab ViewTM (National Instruments, Austin, TX) was used for data acquisition of cell shortening using a sampling rate of 100 Hz. Fluorescence data for the [Ca2+]i were imported into Lab ViewTM where both the [Ca2+]i and myocyte contractile responses were synchronously and simultaneously analyzed, as described previously (12). The following variables were calculated for each individual contraction: diastolic [Ca2+]i and cell length, systolic [Ca2+]i and cell length, change in [Ca2+]i and twitch amplitude, time to peak (Tp) for [Ca2+]i and shortening, and time to 50% recovery (Tr) for [Ca2+]i and shortening. Variables from 15 contractions ([Ca2+]i and shortening) were averaged to obtain mean values at baseline and in response to the various interventions. Averaging the variables over time minimizes beat-to-beat variation. Myocyte length (micrometers) in response to field-stimulation was measured and is expressed as the change from resting cell length (twitch amplitude). Changes in twitch amplitude in response to the interventions are expressed as a percentage of baseline shortening. Changes in the timing variables were measured in milliseconds and were normalized to changes in amplitude. Changes in [Ca2+]i were measured as the change in the 340:380 ratio from baseline. Changes in the 340:380 ratio in response to the interventions were expressed as a percentage of the control response in the absence of any intervention. Protocols were designed such that each cell could be used as its own control.
Protocol 1: Dose-Dependent Effects of Midazolam and Diazepam on [Ca2+]i and Myocyte Shortening
Protocol 2: Effect of Midazolam and Diazepam on SR Ca2+ Stores
Protocol 3: Effect of Midazolam and Diazepam on Myofilament Ca2+ Sensitivity Collagenase type II was obtained from Worthington Biochemical (Freehold, NJ). Midazolam (Versed, Roche, Basel, Switzerland) and diazepam (Valium, Roche) were obtained as the commercial products from the Cleveland Clinic Pharmacy. Caffeine was purchased from Sigma Chemical Co (St Louis, MO). Each experimental protocol was performed on multiple myocytes from the same heart and repeated in at least four hearts. Results obtained from myocytes in each heart were averaged so that all hearts were weighted equally. The dose-dependent effects of midazolam or diazepam on myocyte shortening and [Ca2+]i were assessed using one-way analysis of variance with repeated measures and the Bonferroni/Dunn post hoc test. Comparisons between groups were made by two-way analysis of variance. Results are expressed as mean ± SEM. Differences were considered statistically significant at P < 0.05.
A total of 147 cells were used for the study. Baseline [Ca2+]i was 83 ± 8 nM, and the diastolic cell length was 112 ± 4 µm. Peak [Ca2+]i was 360 ± 30 nM. Twitch amplitude was 10% (11.7 ± 0.5 µm) of the baseline diastolic resting cell length. Tp [Ca2+]i and shortening were 159 ± 7 ms and 177 ± 5 ms, respectively. Tr for [Ca2+]i and shortening were 322 ± 5 ms and 289 ± 5 ms, respectively. Figure 1A demonstrates that the addition of midazolam (100 µM) to an individual, field-stimulated ventricular myocyte results in marked inhibition of myocyte shortening and a concomitant decrease in peak [Ca2+]i, whereas smaller concentrations had no effect. The myocardial depressant effect of midazolam was completely reversed after washout. Individual contractions and [Ca2+]i transients are shown in Figure 1B. Midazolam had no effect on resting [Ca2+]i or cell length. The summarized data are shown in Figure 1C. Midazolam had no effect on [Ca2+]i or shortening at concentrations up to 30 µM, whereas 100 µM of midazolam caused profound myocardial depression.
Midazolam had no significant effect on Tp or Tr for shortening or [Ca2+]i at concentrations up to 30 µM (Fig. 2A). Because midazolam (100 µM) nearly abolished the field stimulated contraction of the myocyte, only 34 cells could be obtained for analyzing the changes in Tp or Tr. At the largest concentration (100 µM), midazolam prolonged Tp and Tr for myocyte shortening and [Ca2+]i.
Figure 3A demonstrates that the addition of diazepam (30 and 100 µM) to an individual, field-stimulated ventricular myocyte results in a small increase in myocyte shortening and peak [Ca2+]i. However, myocyte shortening and peak [Ca2+]i were completely depressed at the largest concentration of diazepam (300 µM). The myocardial depressant effect was readily reversible after washout of diazepam. The vehicle for diazepam, an ethanol:propylene glycol mixture, had no significant effect on myocyte shortening or peak [Ca2+]i. Individual contractions and [Ca2+]i transients are illustrated in Figure 3B. Diazepam had no effect on resting [Ca2+]i or cell length. The summarized data are shown in Figure 3C. Diazepam exerted a dose-dependent, biphasic effect on myocyte shortening and [Ca2+]i. Increases in myocyte shortening and peak [Ca2+]i were observed with 30 and 100 µM diazepam, whereas marked myocardial depression was observed at the largest concentration tested (300 µM).
Diazepam (30 and 100 µM) decreased Tp for [Ca2+]i without a concomitant effect on shortening (Fig. 2B). Tr for cell shortening and [Ca2+]i were unaffected by diazepam at concentrations up to 100 µM. Because diazepam (300 µM) virtually abolished the field-stimulated contraction, only three to four cells could be analyzed for changes in Tp or Tr. Diazepam (300 µM) prolonged Tp and Tr for myocyte shortening and [Ca2+]i. We assessed the extent to which midazolam or diazepam altered the amount of Ca2+ released from the SR in response to caffeine (20 mM). Only the largest doses of midazolam (100 µM) and diazepam (300 µM) suppressed the caffeine-induced [Ca2+]i transient compared with control (Fig. 4).
The effect of [Ca2+]o on myocyte shortening and peak [Ca2+]i were examined in the presence and absence of midazolam or diazepam. Increasing [Ca2+]o from 1 to 4 mM (control, without benzodiazepine) resulted in a dose-dependent increase in shortening and a concomitant increase in peak [Ca2+]i (Fig. 5). Midazolam (10 µM) did not alter the dose-response curve to [Ca2+]o for myocyte shortening or [Ca2+]i as compared with control. However, diazepam (30 µM) caused a downward shift in the dose-response curve to [Ca2+]o for shortening without a concomitant effect on peak [Ca2+]i.
In this study, we investigated the effects of midazolam and diazepam on simultaneous measurements of cell shortening and [Ca2+]i in individual, field-stimulated adult rat ventricular myocytes. This is the first systematic study reporting the direct effects of benzodiazepines on cardiac excitation-contraction coupling, SR Ca2+ handling, and myofilament Ca2+ sensitivity at the cellular level in adult ventricular myocytes. We found that midazolam decreased shortening and peak [Ca2+]i in response to field stimulation at the largest concentration tested. In contrast, diazepam elicited a biphasic inotropic response with a modest increase in shortening and peak [Ca2+]i at smaller concentrations and a potent myocardial depressant effect and marked decrease in peak [Ca2+]i at the largest concentration tested. The myocardial depressant effects of both benzodiazepines were readily reversible after washout. Midazolam and diazepam also prolonged the timing variables for cell shortening and [Ca2+]i at the largest concentrations examined. Both benzodiazepines decreased the amount of Ca2+ released from the SR in response to caffeine at the largest concentrations. Furthermore, diazepam, but not midazolam, caused a downward shift in the dose-response curve to [Ca2+]o for shortening without any concomitant effect on peak [Ca2+]i. Although the hemodynamic effects of midazolam and diazepam have been studied extensively in vivo (13,14), there have been several studies that described the direct inotropic effects of these benzodiazepines in vitro (5,6,8,10). Reves et al. (6) reported the depression of the maximum first derivative of left ventricular pressure with respect to time (dP/dtmax) by midazolam and diazepam in the isolated rat heart. They found the 50% effective dose for myocardial depression was 15 µg/mL (42 µM) with midazolam and 23 µg/mL (80 µM) for diazepam. Stowe et al. (8) observed that 121 µM of midazolam reduced the left ventricular pressure by 50% in the isolated guinea pig heart. A similar dose of midazolam (145 µM) caused a 50% depression in isometrically contracting human atrial muscle (5). In cultured neonatal rat ventricular myocytes, both diazepam (100 µM) and midazolam (100 µM) decreased contractility by more than 50% (10). These observations are consistent with the reduction in myocyte shortening by midazolam and diazepam in the present study. We also observed a small positive inotropic effect with small concentrations of diazepam. Because the vehicle for diazepam, an ethanol:propylene glycol mixture, had no significant effect on myocyte shortening or peak [Ca2+]i, we believe that diazepam has a positive inotropic effect at the smaller concentrations. Although the precise mechanism is not clear from the present study, an increase in Ca2+ influx across the sarcolemma and/or a sensitization in SR Ca2+ release (15) would be associated with the diazepam-induced positive inotropic effect. The direct effects of benzodiazepines on myocardial contractility are evident in vitro; however, the positive inotropic effect may have been masked in multicellular preparations or in vivo because of the benzodiazepines acting at multiple sites (central and local), thereby counteracting the direct cardiac effects. A blockade of sarcolemmal L-type Ca2+ channel is one potential mechanism for benzodiazepine induced myocardial depression because a small Ca2+ influx via this channel can trigger the further Ca2+ release from the SR. We have reported the direct myocardial depressant effects of midazolam and diazepam in cultured neonatal rat ventricular myocytes (10). In the previous study, the myocardial depressant effects of the benzodiazepines were improved in the presence of the L-type Ca2+ channel opener Bay K 8644. However, care must be taken when the results from neonatal myocardium are used as an explanation for the myocardial depressant mechanism because the neonatal (immature) myocardium has different characteristics compared with adult (mature) myocardium (11). For instance, immature myocardium generates less tension than the adult myocardium because of the differences in contractile protein. In addition, neonatal myocardium depends predominantly on transsarcolemmal Ca2+ influx to produce contraction rather than on Ca2+ from the SR. Nevertheless, Buljubasic et al. (9) reported that 60 µM of midazolam decreased peak ICa by 47% in canine myocardial cells using the whole-cell voltage-clamp technique. These findings suggest that the negative inotropic effects of these benzodiazepines are related, at least in part, to decreased ICa. However, it is unlikely that a blockade of sarcolemmal L-type Ca2+ channel plays a major role in benzodiazepine-induced myocardial depression in adult rat ventricular myocytes because a positive inotropic effect of an increase of extracellular Ca2+ concentration was unaffected by relatively smaller doses of midazolam (10 µM) or diazepam (30 µM) (Fig. 5). We also examined whether benzodiazepines cause changes in the timing of myocyte shortening and/or [Ca2+]i. Changes in the timing variables would reflect alterations in the rate of Ca2+ influx, SR Ca2+ cycling, and/or myofilament Ca2+ sensitivity (12,16). Diazepam accelerated Tp for [Ca2+]i with no concomitant effect on myocyte shortening at small concentrations, although this effect was small. This result suggests that the rate of Ca2+ influx and/or release of Ca2+ from the SR may be faster in the presence of small concentrations of diazepam. Alternatively, and perhaps more likely, the decrease in myofilament Ca2+ sensitivity could also explain the decrease in Tp for [Ca2+]i while masking the decrease in Tp for shortening. In contrast, both diazepam and midazolam prolonged Tp and Tr for [Ca2+]i and shortening at the largest concentrations tested, which suggests that the benzodiazepines may exert negative inotropic affects by altering SR Ca2+ dynamics. The effects observed here are similar to those associated with inhibition of the SR Ca2+ pump with thapsigargin (16), which also results in a depression in peak [Ca2+]i and SR Ca2+ stores, as well as an increase in Tp and Tr for [Ca2+]i. Because a decrease in peak [Ca2+]i could also result from a decrease in the amount of Ca2+ released from the SR, we used caffeine, which triggers Ca2+ release from cardiac SR, to investigate potential interactions of the benzodiazepines on SR Ca2+ stores. Rapid application of caffeine (20 mM) to quiescent myocytes resulted in a [Ca2+]i transient similar in magnitude to that observed with field stimulation (12). Midazolam and diazepam decreased the amplitude of the caffeine-induced [Ca2+]i transient at the largest concentrations tested. These results suggest that midazolam and diazepam decrease SR Ca2+ content and/or modulate the process of caffeine-induced Ca2+ release itself. From these results concerning the effects of benzodiazepines on SR Ca2+ handling, the largest doses of midazolam and diazepam exerted negative inotropic effects, at least in part, by altering SR Ca2+ release. Because alterations in myofilament Ca2+ sensitivity could be another potential mechanism for the negative inotropic property of the benzodiazepines, we introduced the dose-response curves to [Ca2+]o to examine whether alterations in myofilament Ca2+ sensitivity played a role in the actions of anesthetics (12). Diazepam, but not midazolam, caused a downward shift in the dose-response curve to [Ca2+]o for shortening without a concomitant effect on [Ca2+]i. These results indicate that diazepam, but not midazolam, may decrease the maximal response of the myofilaments to Ca2+ as [Ca2+]i increases. The results of this study need to be interpreted in the context of the experimental conditions (low temperature, 28°C, and low frequency of stimulation, 0.3 Hz). Results obtained here may differ from those under physiologic conditions in vivo in animals or in humans. However, these conditions are required to maintain myocyte viability throughout the time course of the experiments. Moreover, all anesthetics bind to plasma protein, which decreases the effective circulating concentration of these anesthetics. Therefore, the actual free plasma concentration of anesthetic available to bind to tissues will be a fraction of the total plasma concentration. Free peak plasma concentrations during the induction of anesthesia are reported to be approximately 0.5 µM for midazolam (96% protein binding) (1) and 0.7 µM (96% protein binding) for diazepam (17). Thus, it is unlikely that the cardiovascular effects of clinically relevant doses of midazolam and diazepam in vivo are mediated by their direct action on cardiac contractility. In conclusion, the negative inotropic effects of midazolam and diazepam seem to be mediated mainly by a decrease in the availability of intracellular-free Ca2+. Diazepam caused a positive inotropic effect at smaller doses, whereas a potent negative inotropic effect was observed at the largest concentration tested. Midazolam and diazepam decreased the caffeine-releasable SR Ca2+. Diazepam, but not midazolam, decreased myofilament responsiveness to Ca2+. Nevertheless, the negative inotropic effects of the benzodiazepines are not apparent at clinically relevant doses.
Supported, in part, by National Heart, Lung and Blood Institute (NHLBI) Grants HL-38291, HL-40361, and HL-65701 and an American Heart Association Beginning Grant-in-Aid (9806210). Dr Kanaya was also supported by Professor Akiyoshi Namiki, MD, PhD, Department of Anesthesiology, Sapporo Medical University School of Medicine, Japan. The authors would like to thank Cindy Shumaker for her excellent technical support and Cassandra Talerico for outstanding work in preparing the manuscript.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|