High-affinity blockade of voltage-operated skeletal muscle and neuronal sodium channels by halogenated propofol analogues

Abstract

Background and purpose: Voltage-operated sodium channels constitute major target sites for local anaesthetic-like action. The clinical use of local anaesthetics is still limited by severe side effects, in particular, arrhythmias and convulsions. These side effects render the search for new local anaesthetics a matter of high interest.

Experimental approach: We have investigated the effects of three halogenated structural analogues of propofol on voltage-operated human skeletal muscle sodium channels (Na(V)1.4) and the effect of one compound (4-chloropropofol) on neuronal sodium channels (Na(V)1.2) heterologously expressed in human embryonic kidney cell line 293.

Key results: 4-Iodo-, 4-bromo- and 4-chloropropofol reversibly suppressed depolarization-induced whole-cell sodium inward currents with high potency. The IC(50) for block of resting channels at -150 mV was 2.3, 3.9 and 11.3 microM in Na(V)1.4, respectively, and 29.2 microM for 4-chloropropofol in Na(V)1.2. Membrane depolarization inducing inactivation strongly increased the blocking potency of all compounds. Estimated affinities for the fast-inactivated channel state were 81 nM, 312 nM and 227 nM for 4-iodopropofol, 4-bromopropofol and 4-chloropropofol in Na(V)1.4, and 450 nM for 4-chloropropofol in Na(V)1.2. Recovery from fast inactivation was prolonged in the presence of drug leading to an accumulation of block during repetitive stimulation at high frequencies (100 Hz).

Conclusions and implications: Halogenated propofol analogues constitute a novel class of sodium channel-blocking drugs possessing almost 100-fold higher potency compared with the local anaesthetic and anti-arrhythmic drug lidocaine. Preferential drug binding to inactivated channel states suggests that halogenated propofol analogues might be especially effective in suppressing ectopic discharges in a variety of pathological conditions.

Figure 1

Structures of the halogenated propofol analogues used in this study (top) and, for comparison, the structure of lidocaine. The substituted benzene ring is a common feature of many local anaesthetic drugs.

Figure 2

Block of resting channels at −150 mV holding potential. Left: concentration dependence of block of sodium inward current by 4-iodo-, 4-bromo- and 4-chloropropofol in Na_V1.4 and 4-chloropropofol in Na_V1.2 at −150 mV holding potential. Each symbol represents the mean value of the residual sodium currents in the presence of drug (I_norm, mean±s.d.), derived from at least five independent experiments for each concentration tested. The solid lines are Hill fits to the averaged data yielding a half-maximum blocking concentration EC₅₀, of 2.3 μM for 4-iodopropofol, 3.9 μM for 4-bromopropofol and 11.3 μM for 4-chloropropofol in Na_V1.4, and 29.2 μM for 4-chloropropofol in Na_V1.2, respectively. Right: representative current traces following a 40 ms depolarization from −150 to 0 mV in the absence (control and washout) and presence of 3 μM 4-chloropropofol in Na_V1.2.

Figure 3

Affinity of 4-iodo-, 4-bromo and 4-chloropropofol to fast-inactivated channels derived from the concentration dependence of the shifts in the voltage-dependence of channel availability. (a) Top: representative current traces illustrating the increase in blocking potency of 3 μM (left) and 10 μM (right) 4-chloropropofol in Na_V1.2 neuronal sodium channels when depolarizations were started from more positive holding potentials—note that the current inhibition is complete at a potential near the potential for half-maximum channel inactivation (−50 mV) in control conditions. Bottom: steady-state availability curves assessed by a two-pulse protocol in the absence (control and washout) and presence of (from left to right) 3, 10 and 30 μM 4-chloropropofol in Na_V1.2 sodium channels. Each symbol represents the mean fractional current derived from seven different experiments, elicited by a 4 ms test pulse to 0 mV, following a 100 ms inactivating prepulse from −150 mV to the indicated prepulse potential. Currents were normalized to maximum value (in each series at −150 mV pre-potential); solid lines represent the best Boltzmann fit (Equation 2, Materials and methods section) to the data. The indicated errors are standard deviations. In the presence of drug, currents were normalized either to maximum value in the presence of drug (filled symbols) or to maximum value in the controls (empty symbols). Vertical arrows illustrate the increase in peak current suppression induced by the test drug at more depolarized potentials compared with −150 mV prepulse potential. This increase in the peak current suppression induced by the test drug at more depolarized prepulse potentials results in a voltage shift in the midpoints of the availability curve (ΔV_0.5) illustrated by the vertical arrows. This voltage shift is concentration dependent; the concentration dependence of this effect is depicted in Figure 4. (b) The effects of 1 μM 4-iodopropofol, 3 μM 4-bromopropofol or 10 μM 4-chloropropofol on steady-state availability curves in Na_V1.4 sodium channels, assessed by a similar two-pulse protocol. Membrane depolarization increases blocking potency in all compounds, revealed by a voltage shift of the respective availability curve in the direction of negative pre-potentials.

Figure 4

Model-dependent assessment of the affinity to fast-inactivated channels. Concentration dependence of the drug-induced negative shifts in the midpoints (ΔV_0.5) of the steady-state availability plots relative to the starting values (from top to bottom) for 4-iodo, 4-bromo- and 4-chloropropofol in Na_V1.4, and for 4-chloropropofol in Na_V1.2, respectively. Each symbol represents the mean value derived from at least four different experiments each, error bars are standard deviations. The solid line is a least-squares fit of Equation 3 (Materials and methods section) to the averaged data. Parameters inserted as constant factors into the equation are printed in italics, the slope factor k was derived from Boltzmann fits to the control data and the ECR₅₀ is derived from concentration–response curves at −150 mV. The estimated concentrations for half-maximum effect on inactivated channels (ECI₅₀) derived from that fit were 81, 312 and 227 nM for 4-iodopropofol, 4-bromopropofol and 4-chloropropofol in Na_V1.4, respectively, and 450 nM for 4-chloropropofol in Na_V1.2. The Hill coefficients for inactivated state binding ranged between 1.4 and 1.0.

Figure 5

Prolongation of the time course of recovery from fast inactivation and use-dependent block induced by 4-chloropropofol in Na_V1.2 (a) and by 4-iodo-, 4-bromo- and 4-chloropropofol in Na_V1.4 (b). (a) Upper panel, use-dependent block: representative current traces showing the first trace (control and washout) and the first and the last three traces in the presence of 4-chloropropofol out of a series of 10 depolarizations from −100 to 0 mV applied at 100 Hz. The current traces show a moderate use dependence of the effect of 4-chloropropofol revealed by an increase in the blocking effect from the first pulse to the last pulse of approximately 10% in 10 μM 4-chloropropofol (right row of traces). Lower panel: prolongation of recovery from sodium channel inactivation induced by 4-chloropropofol assessed by a two-pulse protocol at −100 mV membrane potential in Na_V1.2 in the controls and in the presence of either 3 μM (left diagram) or 10 μM (right diagram) 4-chloropropofol. The diagrams show the mean fractional current after recovery (ordinate) from five experiments each, plotted against the recovery time interval between inactivating prepulse and test pulse (abscissa) on a logarithmic scale. Error bars are standard deviations. Filled triangles represent the recovered current in the presence of drug normalized to the prepulse current in the presence of drug; empty triangles represent the recovered current in the presence of drug normalized to the prepulse current in the controls. Solid lines are exponential fits (Equation 4) to the data yielding the respective time constants of recovery τ_rec. 4-Chloropropofol induces a prolonged recovery from inactivation which shows concentration dependence. (b) Prolongation of recovery from sodium channel inactivation induced by 3 μM concentrations of 4-iodopropofol (top), 4-bromopropofol (bottom, left) and 4-chloropropofol (bottom, right) at −100 mV membrane potential in Na_V1.4. The diagrams show the mean fractional current after recovery (ordinate) from 4–5 experiments each, plotted against the recovery time interval between inactivating prepulse and test pulse (abscissa) on a logarithmic scale. Error bars are standard deviations. Filled triangles represent the recovered current in the presence of drug normalized to the prepulse current in the presence of drug, empty triangles represent the recovered current in the presence of drug normalized to the prepulse current in the controls. Solid lines are exponential fits (Equation 4) to the data yielding the respective time constants of recovery τ_rec. All compounds prolong recovery from inactivation in Na_V1.4. The insets show representative current traces showing the first trace out of a series of 10 in the absence of drug (control and washout) and the first and the last three traces in the presence of 3 μM 4-chloropropofol. The current traces show a moderate use dependence of the blocking effect of 4-chloropropofol during 100 Hz trains revealed by an additional fall in the last pulse relative to the first pulse in the presence of drug of around 15%.