Propofol may exert negative inotropic and chronotropic actions in the heart. Single-channel studies show that propofol affects the kinetics of opening and closing of cardiac L-type calcium channels (ICa(L)) without altering channel conductance. The aim of this study was to investigate the mechanisms of depressant effects of propofol on cardiac whole-cell ICa(L).
Single ventricular myocytes were freshly dissciated from guinea pig hearts using enzymatic isolation. One-suction electrode voltage-clamp technique (whole-cell mode) was used. LCa(L) was separated from other contaminated ionic currents. Propofol was applied in the commercial 10% Intralipid emulsion formula (Zeneca, UK).
In isolated cardiomyocytes, propofol significantly inhibited whole-cell ICa(L) in a concentration-dependent manner (K D = 52.0 microM; Hill coefficient = 1.3). The solvent (Intralipid) did not affect ICa(L). Propofol decreased ICa(L) at all potentials tested along the voltage axis and reduced the slope conductance. The threshold potential for activation and the peak potential of the current-voltage relationship were not changed by propofol. The steady-state activation curves overlapped in the absence and the presence of 56 microM propofol. In contrast, the steady-state inactivation curve was shifted in the hyperpolarizing direction. The time course of the recovery from inactivation was delayed by 56 microM propofol. The blocking action on ICa(L) of propofol shows marked resting block and use-dependent block. Propofol caused more pronounced inhibition at a higher stimulation frequency. The effect of propofol on the inactivation process was even more clear on ICa(L).
The authors conclude tha propofol, at supratherapeutic concentrations, inhibits cardiac ICa(L). This inhibition is mainly due to a shift of inactivation curve and a reduction in slope conductance.
ALTHOUGH the lipid solubility and potency of anesthetics have been correlated linearly, the cellular and subcellular mechanisms responsible for the actions of general anesthetics remain uncertain. Recently, the transmembrane ionic channels have been suggested as plausible targets or sites of the action. [2,3]For example, acetylcholine receptor channels, calcium channels, [5,6]and potassium channels are affected directly by general anesthetics. However, whether all the general anesthetics share a common, unitary cellular mechanism of anesthetic action remains to be explored.
Most general anesthetics exert negative chronotropic and inotropic actions in the heart. The cellular mechanisms of the inhibition of nodal cells and the depression of contractility [10,11]by those anesthetics are rather complex, from actions on transmembrane ionic flow to those on intracellular calcium [13,14]and contractile machinery. [15,16]For example, a reduction in L-type calcium current contributes to the negative inotropic actions. [17–19].
Propofol is a widely used intravenous anesthetic that may also cause bradycardia and hypotension, which in turn may be caused by a direct negative inotropic action in the heart. [20–22]Evidence from both in whole-cell and single-channel levels suggests that L-type calcium current can be inhibited by propofol. [5,6]The purpose of this study was to further explore the mechanisms responsible for the inhibitory effects of propofol on L-type calcium channels in guinea pig ventricular myocytes.
Materials and Methods
Isolation of Single Ventricular Myocytes
The investigation conformed with the Guiding Principles in the Care and Use of Animals as approved by the Council of the American Physiologic Society.
Single cardiac ventricular myocytes were isolated from guinea pig hearts. The enzymatic dissociation method used was similar to that reported by Mitra and Morad. The animal was killed by a blow on the neck and the heart was excised immediately. The aorta was cannulated with a polyethylene tube filled with oxygenated Tyrode solution, which contained (in mM): 145 NaCl, 5.4 KCl, 1.8 CaCl2, 0.5 MgCl2, 5 glucose, and 11.6 HEPES (pH was adjusted to 7.40 with NaOH). The heart was then mounted quickly onto a Langendorff apparatus with retrograde perfusion by gravity. The heart was first perfused with nominally Calcium sup 2+-free solution for 5 min and followed by a solution containing 40 mg collagenase (type A, Boehringer, Mannheim, Germany) and 4 mg protease (type XIV, Sigma, St. Louis, MO) for 5 to 10 min. Subsequently, the heart was perfused with enzyme-free low-Calcium2+(0.18 mM) Tyrode solution for 5 min. Finally, the heart was removed from the Langendorff apparatus and the ventricles were chopped into small pieces from which the single myocytes were liberated by gentle agitation in low-Calcium2+ Tyrode solution. The cells settled at the bottom of the container by gravity and the supernatant of the cell suspension was removed by replacing it with normal Tyrode solution. High cell yield was usually obtained and the cells were rectangular, striated, and quiescent. The Calcium2+-tolerant cells were stored at room temperature and were used for experiments within 10 h.
The Tyrode solution for isolation of single ventricular myocytes contained (in mM): NaCl 145, KCl 5.4, MgSO41, KH2PO41.2, glucose 5.5, and HEPES 6.0 (pH was adjusted to 7.4 with NaOH). Calcium2+ concentration in Tyrode solution was either nominally free, 0.18 mM, or 1.8 mM. Normal Tyrode solution contained (in mM): NaCl 145, CsCl 5.4, CaCl sub 2 (or BaCl2) 5.4, MgCl20.5, glucose 5.5 and HEPES 11.8 (pH was adjusted to 7.4 with NaOH). Extracellular Sodium sup +-free Tyrode solution was prepared simply by replacing NaCl with equimolar Tris-Cl. Tetrodotoxin (1–10 micro Meter) was applied to block fast Sodium sup + channels in some experiments. The pipette solution contained (in mM): CsCl 140, MgCl22, tetraethylammonium chloride 20, Na2ATP 5, EGTA 20, and HEPES 5 (pH was adjusted to 7.4 with KOH). All electrolytes and substances including BaCl2, CsCl, adenosine triphosphate, tetraethylammonium chloride, EGTA and tetrodotoxin were purchased from Sigma (St. Louis, MO). Propofol (Diprivan, 1% w/v, emulsion form in Intralipid) was purchased from Zeneca (United Kingdom). Propofol was diluted in Tyrode solution to final concentration and directly applied to cardiomyocytes by superfusion. The highest concentration of propofol in emulsion form in this study was 560 micro Meter. The control experiment of the solvent was using Intralipid (10%), which was purchased from Kabi Pharmacia (Stockholm, Sweden). Intralipid contained 100 g purified soybean oil, 12 g purified egg phospholipids, 22.5 g glycerol USP, and 1,000 ml water. Intralipid was diluted to final concentration in Tyrode solution. All experiments were performed at room temperature.
Data Acquisition and Analysis
The isolated single ventricular myocytes were placed in the recording chamber (2-ml in volume), which was mounted on an inverted microscope (IMT-2, Olympus, Japan), and perfused with Tyrode solution with different composition as indicated. Whole-cell L-type calcium current was recorded using patch-clamp technique. The suction pipette was made of borosilicate (2.00 mm OD and 1.25 mm ID, Jencons, Leighton Buzzard, UK) and pulled in a Flamming/Brown micropipette puller (model P-87, Sutter Instrument Company, Novato, CA). The tip of electrode was fabricated with a fire polisher (Narishige, Japan). The pipette tip resistance when filled with the pipette solution was 2–5 M Omega for whole-cell mode. Current was recorded using a patch clamp amplifier (Axopatch 1-D, Axon Instruments, Inc., Foster City, CA). Data were sampled at 2 kHz and filtered at 1 kHz (-3dB cutoff frequency) with an 8-pole Bessel-type filter (VBF/8.03, Kemo, Beckenham Kent, UK). Digitized traces were stored on an MO (magneto-optical devices) disk unit (RMO-S550, Sony, Japan) and analyzed using pCLAMP software (6.0, Axon Instruments, Inc.). Data were expressed as mean+/-SEM. The significance of difference between groups of control and propofol was tested by paired Student's t test. Probability values of less than 0.05 were regarded as statistically significant. The concentration-response curve was fitted using the Hill equation. The steady-state activation and inactivation curves were fitted with Boltzmann function.
Actions of Propofol on L-type Calcium Current
(Figure 1) demonstrates the results of an experiment in which 56 micro Meter propofol significantly inhibited transmembrane L-type calcium current (ICa(L)) in single guinea pig ventricular myocytes. The original tracings of ICa(L) are shown in the absence, the presence of 56 micro Meter propofol, and after washout of propofol (Figure 1(A)). The membrane potential was held at -50 mV and stepped to +10 mV for 300 ms of duration at frequency of 0.2 Hz. The time of recording of ICa(L) corresponded to that indicated in Figure 1(B), which shows the time course of the effect of 56 micro Meter propofol on ICa(L). The cell was dialyzed for 20 min after breaking the patch of the cell membrane. Then propofol was added in the bathing solution. During 5 min of perfusion of propofol, ICa(L) was decreased progressively to a steady state. This inhibition by propofol was reversible after washout from the superfusate. The extent of inhibition of ICa(L) by propofol was clearly separated from that of spontaneous rundown of ICa(L) per se by the slope of the development of the blocking action.
It is necessary to demonstrate that propofol, but not its solvent, Intralipid, inhibited ICa(L). Figure 2shows the time course of ICa(L) in the absence and the presence of Intralipid or propofol. Immediately after breaking the cell patch, ICa(L) was monitored for 10 min during dialysis of pipette solution into the cell. During the first part of the experiment, ICa(L) was elicited by the double-pulse voltage protocol. The holding potential was held at -90 mV and then the membrane potential was depolarized to -40 mV for a short time before a further depolarization step to +10 mV. L-type calcium current was therefore elicited and clearly separated from the contamination of fast Sodium sup + current. Subsequently, superfusion containing 1% dilution of Intralipid (10%) was applied for 5 min. There were no significant changes of ICa(L) by Intralipid (Figure 2(a)). This was followed by application of 56 micro Meter propofol (0.1% dilution of Diprivan, 1% w/v) for 10 min. To observe any resting block of propofol, the depolarizing pulses were not delivered for the first 5 min of propofol perfusion. After this period of electrical quiescence (without any stimulation), in the presence of 56 micro Meter propofol, the first record of ICa(L) was registered and compared with control (Figure 2(a)). Figure 2(b) shows ICa(L) at 10 min after perfusion of 56 micro Meter propofol. The inhibitory effect of propofol on ICa(L) was reversible when propofol was washed out in the bathing solution containing Intralipid. Total block on ICa(L) by 56 micro Meter propofol when ICa(L) was elicited by double-pulse voltage protocol was 52.6%+/-2.9% of control (n = 5, P < 0.05). In the same cell, 56 micro Meter propofol was reapplied. But this time, ICa(L), in the absence (Figure 2(c)) and the presence of propofol (Figure 2(d)), was elicited directly from holding potential of -40 mV to test potential of +10 mV. Resting block (the first ICa(L) after 5 min of quiescence) and frequency-dependent block (at 0.5, 1, and 2 Hz) were observed in the presence of propofol. Again, this inhibitory effect was reversible. Similarly, 56 micro Meter propofol decreased ICa(L) by 58.8%+/-4.3%(n = 5) when calcium current was elicited from -90 mV to +10 mV. Fast Sodium sup + current was inhibited by 1 micro Meter tetrodotoxin and Sodium sup +-free medium (replaced by equimolar Tris-Cl).*.
Because the sensitivity of ventricular L-type calcium channels to propofol is unknown, we measured the concentration-dependence of its block on ICa(L). In Figure 3, the normalized ICa(L) was plotted as a function of different concentrations of propofol (solid circles). The I sub Ca(L) was obtained from holding potential of -50 mV to +10 mV in the absence and the presence of propofol. The data could be fitted with a sigmoid curve based on the Hill equation: Y = 1/(1 +(([X]/Kp) sup ^ - n)). The value of KDwas 52.0, micro Meter and Hill coefficient was 1.3. Three different concentrations of Intralipid were tested (open circles) to eliminate the possible action of solvent on ICa(L). It is clear that even at the highest concentration (1% dilution of 10% Intralipid in bathing solution), similar to that in Diprivan, Intralipid did not affect ICa(L).
Effects of Propofol on Steady-state Activation and Inactivation of I sub Ca(L)
(Figure 4)(A) shows the current-voltage relationship in the absence and the presence of 56 micro Meter propofol. The current density of peak ICa(L)(ordinate) was plotted as a function of membrane potential (abscissa) averaged from seven cells. After superfusing, bathing solution containing 56 micro Meter propofol for 10 min, ICa(L) at the peak of the current-voltage relationship (+10 mV) was suppressed by 53%+/- 4%(n = 7, p < 0.05). There was no marked shift of the threshold potential or the peak along the voltage axis. The slope conductance from the current-voltage relationship in the absence and the presence of 56 micro Meter propofol was 0.17+/-0.002 mS/cm2and 0.11+/- 0.001 mS/cm2, respectively (n = 7, P < 0.05).
(Figure 4)(B) shows the steady-state activation curve of ICa(L) in the absence and presence of propofol. The activation curve was obtained from dividing the peak inward currents at potentials negative to +50 mV (Figure 4(A)) by the fully activated current, which was estimated by extrapolation of a linear regression through the currents at potentials between +10 mV and +30 mV. The activation curve was fitted with the Boltzmann function: I = Imax/(1 + exp [-(V - V1/2)/ s]). In control, V1/2 (potential for half maximal activation) was -1.1 +/-0.3 mV and s (slope parameter) was 6.6+/-0.2 mV. In the presence of 56 micro Meter propofol, V1/2 was -1.2+/-0.3 mV and s was 6.2+/-0.2 mV (n = 6, P > 0.05).
(Figure 5) demonstrates the effects of propofol on the steady-state inactivation curve of ICa(L). The inactivation curve was determined by measuring ICa(L) at a test potential of +10 mV after 30 s prepulses. Namely, the membrane potential was first clamped at -90 mV and then stepped to different potentials ranging from -60 mV to -10 mV (Figure 5(A)). Representative current tracings of ICa(L) obtained from four different preconditioned potentials are given in Figure 5. Figure 5(B) shows the inactivation curve in the absence and the presence of 56 micro Meter propofol. The normalized curves were fitted with the Boltzmann equation: I = Imax/(1 + exp[(V - V1/2)/ s]). After addition of 56 micro Meter propofol in the bathing solution for 10 min, the inactivation curve was shifted to more hyperpolarized potential. In control, V1/2 (potential for half maximal inactivation) was -24.6 +/-0.2 mV, and s (slope parameter) was 4.4+/-0.3 mV. In the presence of 56 micro Meter propofol, V1/2 was -33.7+/- 0.3 mV and s was 4.7+/-0.3 mV. While the slope factor was not significantly affected by propofol, the difference in half maximal potential (V1/2) was significant (n = 4, P < 0.05).
Frequency-dependent Effects of Propofol
(Figure 6) displays the results of an experiment designed to test for frequency-dependent (use-dependent) effects of propofol. In this experiment, 300-ms voltage-clamp pulses from a holding potential of -50 mV to a test potential of +10 mV were applied at a frequency of 0.5, 1, and 2 Hz in the absence and presence of 56 micro Meter propofol. Under control conditions, the magnitude of the peak ICa(L) declined slightly after a train of 6 pulses at 0.5 Hz. The frequency dependence became more clear at 1 Hz and 2 Hz. However, in the presence of 56 micro Meter propofol there were dramatic changes in this frequency-response relationship. First, there was a marked resting block (tonic block) by propofol (Figure 6(A), c1 to d1). d1 was measured as the first elicited ICa(L) 5 min after the application of propofol. This resting block phenomenon was shown even more clearly in Figure 6(B). Second, a decline in the peak ICa(L) during the train-of-six depolarizing pulses was dramatic in the presence of 56 micro Meter propofol (Figure 6(A), d2-d6). The development of relative block was even more pronounced at 1 Hz and 2 Hz (Figure 6(B and C)). Similar phenomena of resting block and use-dependent block on ICa(L) by propofol were observed in six different cells.
A standard double-pulse protocol was used to measure the rate of recovery from inactivation of ICa(L) in the absence and presence of propofol. A 300-ms prepulse to +10 mV from holding potential of -50 mV was applied, followed by a variable recovery time at -50 mV. Then, a test pulse to +10 mV was applied to determine the relative amount of current present (Figure 7(A)). Under control conditions (in the absence of propofol), the recovery of the ICa(L) could be well fitted by a single exponential, whereas in the presence of 56 micro Meter propofol the recovery process was better described by a single exponential with a longer time constant as displayed in Figure 7(B). This experiment is representative of four such experiments, and the average time constant of recovery was 263+/-26 ms under control conditions. In the presence of 56 micro Meter propofol, the average time constant of recovery was 498+/-86 ms (n = 4, P < 0.05).
Effect of Propofol on I sub Ba(L)
There are some differences between calcium and barium divalent ions acting as permeant ions through L-type calcium channels. For example, the conductance of IBa(L) is larger and the inactivation process is slower than those of ICa(L). Figure 8(A) shows this discrepancy between ICa(L) and IBa(L) via L-type calcium channel, which was completely blocked by 1 mM cadmium ions. Because we have demonstrated that propofol affected inactivation of ICa(L), it is interesting to see whether this also is true for IBa(L). Figure 8(B) displays a superposition of tracings of IBa(L) in the absence and presence of 56 micro Meter propofol. Similarly, propofol decreased the peak amplitude of IBa(L)(56.4%+/-4.0% of the control, n = 9, p < 0.05) and accelerated the inactivation of IBa(L)(from 74.9 ms+/-3.8 ms to 49.2 ms+/-3.7 ms, n = 9, P < 0.05). A family of IBa(L) tracings at different membrane potentials in the absence and presence of propofol were demonstrated in Figure 8(C).
The current experiments have revealed that L-type calcium channels in cardiac myocytes were inhibited by propofol (KD= 52.0 micro Meter). The solvent Intralipid alone did not affect ICa(L). This inhibitory effect of propofol on ICa(L) was reversible and was mainly owing to a reduction in slope conductance and a shift of steady-state inactivation curve. Both marked resting block and frequency-dependent block on ICa(L) by propofol have been demonstrated. The rate of recovery from the inactivated state to resting state of L-type calcium channels was decreased by propofol. Propofol also inhibited IBa(L) and accelerated its inactivation process when Barium2+ ions were the main charge carrier via L-type calcium channels.
Concentration Dependence of Propofol Action
Propofol caused dose-dependent depression on electromechanical activities in different methods and cardiac preparations. [5,6,20–22]The concentration of propofol used in those experiments varied, from therapeutic (3–4 micro gram/ml or 17–22 micro Meter) to supratherapeutic (300 micro Meter). To our knowledge, there is no information about the concentration-response relationship of the inhibitory action of propofol on cardiac L-type calcium current. From our experimental results, we provided a value of KDof 52.0 micro Meter. This half maximal inhibitory concentration of propofol on cardiac ICa(L) is not much greater than therapeutic plasma concentration (4–20 micro Meter). However, it should be mentioned that, due to its highly lipid-soluble and protein-binding (> 90%) properties, the concentration of free propofol in the plasma is probably much different from that in the Tyrode solution. Above all, the concentration of free propofol at the site of the action is more critical than that in the plasma. In addition, the value of therapeutic plasma concentration simply indicates the effect of propofol on central neurons (anesthetic or sedative actions in central nervous system), instead of on cardiomyocytes (negative chronotropic or inotropic actions in the heart). Therefore, one should be cautious in extrapolating the data from in vitro study to those from in vivo conditions. The above argument probably can, at least in part, explain why propofol did not depress cardiac functions in most cases, except in some geriatric patients or those with compromised cardiovascular functions.
A lack of negative inotropic effect of propofol has been found in several animal species, such dog, rabbit, rat, and hamster. In contrast, propofol indeed depressed cardiac functions in guinea pig and ferret. The exact reasons for this discrepancy remain unclear, although it has been suggested that different cardiac cellular electrophysiologic properties in various animal species might contribute to this differential actions of propofol in the heart. In this study, we demonstrated that propofol inhibited cardiac ICa(L) in guinea pig ventricular myocytes, which is in agreement with a previous report. Together with the findings of the inhibitory effects on the action potentials (depressed plateau and shortened the action potential duration) and contractile force, a reduction of transmembrane L-type calcium current may contribute to the negative inotropic action of propofol. Recently, the depressant effects of anesthetic agents on various types of calcium channels (e.g., L, T, N, P) in the central nervous system have been thought to play a role in their anesthetic actions. Although P-type calcium channels are relatively insensitive to propofol, propofol inhibits T-type and L-type calcium currents in neuronal cells. The inhibitory effect of propofol on L-type calcium current seems to be similar in both cardiac and neuronal cells. Whether this is true for other excitable cells containing ICa(L) needs further investigation.
Propofol-induced Changes in g sub max and Gating Shift
There is little information about the effects of propofol on transmembrane ionic currents or for cardiac L-type calcium current. In guinea pig ventricular myocytes, 100 micro Meter propofol reduced inward calcium current. Nevertheless, further detail about the cellular mechanisms responsible for this inhibition was lacking. The same research group later studied the effect of propofol on a single calcium channel in isolated guinea pig ventricular myocytes. They concluded that 100 micro Meter propofol altered only the kinetics of opening and closing of calcium channels without appreciable effect on channel conductance. Nevertheless, detailed information about the effect of propofol on averaged macroscopic IBa(L), latency to first opening, and interchange among three modes of openings has not been provided. Therefore, it is hard to interpret their single-channel data while comparing our whole-cell experimental results. Our principal findings show that propofol exerted the following effects:(1) marked shifting of the steady-state inactivation curve, but not the steady-state activation curve;(2) a dramatic faster inactivation of ICa(L) and IBa(L); and (3) a reduction in slope conductance from the current-voltage relationship. A shift of the steady-state inactivation curve indicates that available L-type calcium channels were reduced by propofol. When Barium2+ ions were the main charge carrier, propofol also inhibited IBa(L), as it did on ICa(L). This indicates that propofol directly affected the L-type calcium channels instead of permeant ions per se. When extracellular Calcium2+ ions were substituted by Barium2+, the Calcium2+i-dependent inactivation mechanism was eliminated. Under these conditions, propofol accelerated the inactivation process of IBa(L). This result indicates that propofol affected the intrinsic inactivation mechanism of L-type calcium channels. The slope conductance measured from the current-voltage curve of ICa(L), but not the threshold potential or peak potential, was reduced by propofol. A slight left shift of reversal potential of the current-voltage relationship in the presence of propofol was probably caused by an increase in a leak current, instead of a change in the ionic selectivity of L-type calcium channels. Recently, it also was demonstrated that propofol inhibited T- and L-type calcium currents in chick dorsal root ganglion neurons by shortening the inactivation time constant and affecting the slope conductance. .
Frequency-dependent Effects of Propofol
Some calcium antagonists block L-type calcium channels of the myocardium in a voltage- and use-dependent fashion. This phenomenon is related to the preferential binding of calcium antagonists to the inactivated state of the calcium channels. Similarly, sodium channel blockers act via hydrophilic and hydrophobic regions of the lipid membrane, therefore, show use-dependent blocking action and shifting the steady-state inactivation curve to more negative potentials. [35,36]In this study, we report that propofol also blocked ICa(L) in a use-dependent manner (more blocking at higher stimulation frequency). This effect is mainly a result of more channels becoming inactivated and a prolongation of recovery from the inactivation state to resting state of L-type calcium channels. Namely, more calcium channels would stay in the inactivated state in the presence of propofol. This suggests that propofol preferentially binds to the inactivated state of L-type calcium channels. To test whether propofol affected ICa(L) to a similar extent when the inactivation gating was removed, holding potential was changed to -90 mV. To eliminate the contamination of fast Sodium sup + inward current, either a double-pulse voltage protocol or removal of extracellular (replaced by equimolar Tris-Cl) and/or use of tetrodotoxin (a specific Sodium sup + channel blocker) was applied. Under these experimental conditions, the use-dependent block on ICa(L) by propofol became less prominent. Resting block was observed clearly when the holding potential was set at -90 mV or -40 mV. Therefore, based on the modulated receptor model, propofol may preferentially bind to the resting state and inactivated state of L-type calcium channels. Finally, owing to the highly lipophilic property of propofol, it is reasonable to believe that propofol may affect the gating mechanism via a hydrophobic pathway. However, the possibility of a hydrophilic pathway is not completely excluded.
In conclusion, propofol inhibits cardiac L-type calcium current at supratherapeutic concentrations. This inhibitory effect of propofol is caused by a hyperpolarizing shift of steady-state inactivation curve and may contribute to its negative inotropic action.
The authors thank the National Science Council and the National Health Research Institute, Taipei, Taiwan for their grant support.