Free
Meeting Abstracts  |   March 1995
The Membrane Lipid Cholesterol Modulates Anesthetic Actions on a Human Brain Ion Channel 
Author Notes
  • Received from the (Rehberg, Duch) Departments of Anesthesiology and Physiology Cornell University Medical College, New York, New York: and the Klinik fur Anasthesiologie, University of Bonn, Bonn, Germany. Submitted for publication June 1, 1991. Accepted for publication December 2, 1994. Supported by National Institutes of Health grant GM-41102 (D.S.D.). Barachotoxin was donated by Dr. J. Daly. Presented in part at the annual meeting of the American Society of Anesthesiologists, Washington, D.C. October 9–13, 1993.
  • Address reprint requests to Dr. Duch: Department of Anesthesiology, Room A-1030, Cornell University Medical College. 1300 York Avenue, New York, New York 10021.
Article Information
Meeting Abstracts   |   March 1995
The Membrane Lipid Cholesterol Modulates Anesthetic Actions on a Human Brain Ion Channel 
Anesthesiology 3 1995, Vol.82, 749-758. doi:
Anesthesiology 3 1995, Vol.82, 749-758. doi:
Key words: Anesthetics, general: mechanisms of action. Lipids: cholesterol. Theories of anesthetic action: lipid theories; protein theories; sodium channel.
DESPITE decades of experimental work, the physiologic and pharmacologic mechanisms that form the basis of clinical anesthesia remain elusive at virtually every level of examination. There is even a lack of consensus regarding the critical brain functions that are altered during anesthesia. [1 ] Nonetheless, it has been proposed [2,3 ] that anesthesia primarily results from the disruption of cellular membrane function, and most general theories of anesthesia propose that anesthetics cause changes in the functionalities of ionic channels, along with other possible interactions. [4 ].
Most molecular theories of anesthesia generally fit into two categories [2,3 ]: (1) Anesthetics alter protein function indirectly by causing primary changes in the physicochemical properties of the lipid membrane, and (2) anesthetics alter protein function directly through binding or other interactions. [5–7 ] Implicit in the lipid theory of anesthetic action is the assumption that membrane protein function depends, at least partly, on the properties of the surrounding lipids. Implicit in the protein theory is the assumption that only protein interactions are relevant to anesthetic action. [5,6 ] Underlying and complicating all these inquiries is a lack of fundamental information about how the lipid milieu influences membrane protein function. [8 ].
Experimental results have given mixed support to both opposing theories. While anesthetics have been shown to cause changes in the physicochemical properties of lipids, [2,5,9,10 ] these alterations do not always correlate with protein functional changes. [11 ] Although there is substantial evidence that anesthetics interact directly with proteins, [2,5,8 ] such interactions do not always fit the Meyer-Overton correlation, [12 ] do not describe all of the molecular anesthetic effects on these channels, [13 ] or do not follow known properties of anesthetic action such as pressure reversal. [14 ] To understand these relationships between anesthetics and lipid-protein systems, it is important that interactions among all three molecular components be examined separately and together under identical experimental conditions.
The lipid bilayer method is well suited to examine the involvement of lipids in anesthetic interactions. [15 ]. With this technique, various proteins can be incorporated into planar bilayer membranes with compositionally varied and controlled lipid compositions. In the current study, we examined the hypothesis that changes in the lipid composition of the membrane can alter anesthetic modification of membrane-bound proteins. Voltage-gated sodium channels from human brain were incorporated into planar lipid bilayers comprised of phosphatidylethanolamine (PE): phosphatidylcholine (PC; 4:1) mixtures with varying amounts of cholesterol. Previous work has indicated that sodium channels are an appropriate model for examining anesthetic interactions with voltage-gated ionic channels. [15,16 ] The suppression of sodium channel conductance by pentobarbital, a representative and well studied [17,18 ] anesthetic agent in this system, was correlated with membrane cholesterol content, which was systematically increased from 0 to 50%(weight/ weight) of the total lipid.
Materials and Methods
A detailed description of the methods used to incorporate and examine human brain sodium channels in planar lipid bilayers was described previously in this journal. [18 ] A summary of the experimental parameters and modifications is presented here.
Sodium Channel Preparation
With the approval of the CUMC Committee on Human Rights in Research, human brain cortical tissue samples were obtained as surgical waste during routine craniotomies. The tissue was immediately frozen at -80 degrees Celsius and stored at this temperature before and after preparation of the synaptosomal fractions. [19,20 ].
Planar Lipid Bilayer Technique
Lipid bilayers were formed across a 100–300 micro meter-diameter aperture in a Teflon partition separating two compartments of a standard Teflon recording chamber. [21 ] After painting the aperture with lipid, each compartment was filled with 4 ml 500 mM NaCl buffered to 7.35 pH with 10 mM HEPES (US Biochemical). Ag-AgCl electrodes were placed in both chambers of the bilayer cell. The lipids used were a 4:1 mix of PE and PC (Avanti Polar Lipids, Birmingham, AL) with cholesterol added to 0, 2%, 4%, 10%, 20%, or 50%(weight/ weight; Sigma, St. Louis, MO. This corresponds to molar fractions of cholesterol in the lipid mixture of 3.7%, 7.3%, 17.3%, 31.9% and 65.3%). Lipids were dissolved in 99.9% decane (0.05 mg lipid, including cholesterol, per microliter decane; Wiley Organics, Coshocton, OH). All experiments were conducted at room temperature (21–26 degrees Celsius, average 23.4 degrees Celsius); no corrections were made.
Membrane currents were recorded using an AXO-PATCH 200-amplifier (Axon Instruments, Foster City, CA), filtered at 50–200 Hz, digitized via TL-1 DMA interfaces (Axon Instruments), and written to hard disks on personal computers. Voltage protocols and current recordings were performed using pCLAMP and Axotape software (Axon Instruments). After formation of a lipid membrane, the background through the bi-layer was measured at all potentials used for channel characterization. Sodium channels from synaptosomal preparations were incorporated into the bilayers from the cis (front) chamber using 0.25 micro Meter batrachotoxin by gently blowing the channel preparation across the bilayer with a pipette. Batrachotoxin activates sodium channels, thereby aiding in their incorporation and allowing their observation in planar lipid bilayers. [22 ] Channel incorporation rates decreased with increasing cholesterol concentration (Table 1). If no incorporation occurred, a new membrane was formed after 20–60 min.
Table 1. Single-channel Properties of Channels before Addition of Pentobarbital in Bilayers with Differing Cholesterol Content
Image not available
Table 1. Single-channel Properties of Channels before Addition of Pentobarbital in Bilayers with Differing Cholesterol Content
×
Channel orientation was determined by channel-gating characteristics; the electrophysiologic sign convention is used in the presentation of all results. As channels could incorporate bidirectionally (i.e., with the cytoplasmic sides of different channels facing both chambers), only membranes with fewer than five channels incorporated in the same direction were used; membranes with larger numbers of channels were excluded. The use of membranes with fewer channels, incorporated unidirectionally into the bilayer membrane, eliminated complications and ambiguities in the interpretation of results. Control data were sampled for at least 15 min before addition of anesthetic to ensure that the channels and bilayer were stable, with no further channels incorporating in the bilayer. Pentobarbital in its acid form (Sigma) was added to the extracellular side of the channel from an ethanol stock solution; final concentrations ranged from 170 to 5,440 micro Meter pentobarbital. The final concentrations of ethanol reached in these experiments had no effect on sodium channel properties.*.
Data Analysis
Current traces were analyzed using pCLAMP-software (Axon Instruments). Single-channel slope conductances were obtained from linear regression fits of current-voltage curves. Open and closed times in the presence of pentobarbital could not be obtained by direct measurement because pentobarbital causes “flickery” closings of the channel [17 ] beyond the time-resolution of the recording system. The fractional open time was determined by averaging the current over time, subtracting the background conductance measured at each potential, and dividing by the single-channel conductance and the number of channels in the membrane.
Background conductances were obtained from current measurements through the bilayer before channel incorporation. To determine pentobarbital block at depolarized potentials (> -40 mV), depolarizing and hyperpolarizing sequences of 10 mV-steps in the voltage range of plus/minus mV were applied; each potential was held for 4 s.
Voltage-dependent steady-state activation characteristics were examined with a sequence of hyperpolarizing (-10 mV) voltage steps between + 50 and -110 mV. Currents were averaged for 4 s at each potential. Data were fitted to a two-level Boltzmann distribution. The function used is fo(V)= 1/(1 + exp(-zvF(V - V1/2)/RT)) with V being the membrane potential, V1/2the steady-state activation midpoint potential, Zvthe effective gating charge, F the Faraday constant, R the gas constant, and T the absolute temperature.
Statistics
Data are presented as mean plus/minus SEM unless noted otherwise. Error bars also indicate SEM. Curve-fitting procedures were based on the Marquardt-Levenberg algorithm. Concentration-response curves were fitted to simple hyperbolic functions; fits were weighted for the number of experiments per data point. Statistical significance was tested by unpaired t tests and single-factor analysis of variance (ANOVA; significance level P < 0.05) using the programs EXCEL (Microsoft, Redmond, WA) and INSTAT (Graphpad Software, San Diego, CA).
Results
Previous studies indicated that pentobarbital had two effects on human brain sodium channels: a voltage-independent reduction in the time-averaged single channel conductance and a concentration-dependent shift in the voltage dependence of fractional open time in the voltage region of channel activation gating. [17,28,23 ] We therefore examined the effects of cholesterol-containing membranes on these channel parameters in the presence and absence of pentobarbital.
Cholesterol Has No Effect on Control Channel Behavior
The addition of cholesterol to the lipid bilayer had no significant effects on the voltage-independent single-channel conductance, channel fractional open time at depolarized potentials (as measured by the time-averaged conductance), or activation gating parameters (Figure 1and Table 1). As shown in Figure 1(B), single-channel conductance was unaffected by up to 50% cholesterol added to the membrane (P = 0.19 for single-factor ANOVA with six groups and 41 df). The fraction of time that the channel remains open (fractional open time) was indistinguishable between control and cholesterol-containing membranes for all examined cholesterol concentrations (Figure 1(C), P = 0.93 for single-factor ANOVA with six groups and 41 total degrees of freedom). However, cholesterol reduced the number of single-channel incorporations (Table 1).
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
×
As described previously, [18 ] human brain sodium channels undergo transitions between the open, conducting state and a closed, resting state of the channel at hyperpolarized potentials. Channel activation is characterized by a midpoint of activation curve (the potential of half-maximal open time, V1/2) and the effective gating charge (Zv). Neither of these parameters changed significantly in response to the addition of cholesterol to the membrane (Table 1; P = 0.28 [V1/2] and 0.44 [Zv] for single-factor ANOVA with six groups and 41 df). At 50% cholesterol, the average midpoint of the activation curve is shifted toward hyperpolarized potentials, although this difference was not significant (additional unpaired t test, P = 0.43).
Cholesterol Inhibits Pentobarbital Suppression of Human Brain Sodium Channels
The most striking observation was a reduction of the pentobarbital block of sodium channels incorporated into cholesterol-containing membranes. As in previous studies without cholesterol, the dominant effect of pentobarbital on sodium channels was the concentration-dependent induction of a fast flicker as the channel rapidly switched between a fully open and a fully closed state of the channel [18,23 ](Figure 2(A)). This reduction in channel fractional open time (measured as a decrease in the time-averaged current through the channel) was independent of membrane potential at all voltages more positive than - 50 mV in all lipid mixtures (Figure 2(B)). As can be seen, however, cholesterol decreased pentobarbital suppression of fractional open time (i.e., cholesterol exerted a moderating effect on pentobarbital action). The reduction in time-averaged current induced by pentobarbital did not quantifiably change for recording times as long as 90 min per concentration in cholesterol-free membranes [18 ] or for more than 45 min in membranes containing up to 50%(weight/weight, or 65.3 mol%) cholesterol.
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
×
This moderating effect of cholesterol on pentobarbital action was examined with a Lineweaver-Burk transformation (double reciprocal plot). At cholesterol concentrations below 10%(weight/weight, or 17.3 mol%;Figure 3), cholesterol inhibition of pentobarbital block was consistent with a competitive interaction. It can be seen that increasing the cholesterol content of the lipid mixture increased the pentobarbital concentration needed to achieve a 50% suppression (IC50) of sodium channel currents. At higher cholesterol concentrations, however, this inhibitory effect saturated (Figure 4), indicating a noncompetitive interaction. In this latter analysis, less than 2%(weight/weight, or 3.7 mol%) cholesterol in the lipid mixture reduced the pentobarbital effect by half. These results indicate that the observed cholesterol inhibition of the anesthetic block of sodium channels by pentobarbital may occur through more than one type of molecular interaction or mechanism.
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
×
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
×
Voltage-dependent Anesthetics Effects Were Unaffected by Cholesterol
Pentobarbital causes a hyperpolarizing shift in the potential of half-maximal open time [18,23 ](Figure 5). When cholesterol was present in the membrane, this pentobarbital-induced shift appeared to become smaller, however, the difference to the shift in cholesterol-free membranes was not significant at any cholesterol concentration examined (Figure 5, P values are 0.20, 0.20, and 0.29 for 4%, 10%, and 50% cholesterol, respectively). As discussed elsewhere, [23 ] the inherent variability of channel fractional open time in the activation-gating region makes it difficult to assess the significance of relatively small shifts in the voltage-dependent properties of these channels.
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24 ]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24 ]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
×
Discussion
The experiments presented here indicate a correlation between increased membrane cholesterol concentration and decreased pentobarbital supression of ionic currents through sodium channels. This finding confirms the hypothesis that changes in membrane lipid composition can modulate or influence some anesthetic interactions and that, therefore, direct protein-anesthetic binding may not be the only type of interaction important to understanding molecular anesthetic mechanisms. A modulation of pharmacologic interactions by cholesterol has been proposed for ligand-operated channels, [24 ] as well as for calcium channels [25 ]; thus, cholesterol may be a common antagonist of some anesthetic actions. Although these results indicate an important role for cholesterol in the anesthetic modification of ionic channels, its nature needs to be determined.
Several points should be considered when interpreting these results. First, these experiments used planar lipid bilayers as a model system for examining the interactions of cellular membranes, proteins, and an anesthetic. Planar bilayers are the only controlled experimental system in which the lipid composition of the membrane can be systematically altered and examined. Planar lipid bilayers have been used extensively as model systems for more than 30 yr to understand molecular interactions relevant to membrane function. [26,27 ] Although there are differences in some of the properties of these bilayers and cell membranes, they provide a useful and relevant model for understanding the physicochemical properties of membranes and membrane-bound proteins, such as sodium channels.
In this regard, a considerable amount of research has focused on the functions of many voltage-gated and ligand-operated ionic channels in planar bilayers. [28,29 ] Planar bilayers have been used for more than a decade to extensively study the properties of sodium channels from many sources (e.g., rat, [30,31 ] dog, [32 ] and human brain [19 ]; toad, [33 ] rabbit, [31 ] rat, and lamb skeletal muscle [35 ]; bovine and sheep heart [35 ]; the electric organ of the electric eel [24 ]; lobster nerve [36 ] and squid optic nerve [37 ]). In all cases, sodium channel function is similar to that found in situ. These bilayer experiments have provided some of the first evidence that purified sodium channel preparations are functional and that a single polypeptide, in the absence of other subunits, can carry out all examined sodium channel functions. Furthermore, this model system has been used to examine and understand not only the ion conductance and gating properties of sodium channels [32,37,38 ] but also the interactions of sodium channels with various modifying agents and drugs. [18,34,35,39 ] Therefore, given the extensive knowledge gained about both bilayer properties and those of sodium channels, incorporation of sodium channels into planar lipid bilayers is arguably the best electrophysiologic method for examining the effects of membrane lipids on the electrophysiologic and pharmacologic functions of these proteins.
These experiments also use decane as a solvent to spread the lipids over the partition and help form bilayers. It is therefore important to consider the effect of decane in these experiments. When the decane-lipid mixture thins to a bilayer, most of the decane remains in the torus (the area surrounding the bilayer), and the proportion of decane present in lipid systems such as those used here is about 35% of the molecular composition of the bilayer. [26,27 ] However, decane at these concentrations has been found not to alter sodium channel functions. [40 ] The amount of decane present in the bilayer varies with the lipid composition of the mixture, and cholesterol decreases the decane content of the bilayer. [27 ] In this regard, however, it is significant that cholesterol had no effect on control channel properties in the absence of pentobarbital, and therefore, a change in decane content was unlikely to have been the underlying cause of the change in pentobarbital efficacy we observed with increasing cholesterol content. Hence, although this or some other type of cholesterol interaction with the bilayer that does not occur in biologic membranes cannot be ruled out, it is unlikely to be the mechanism underlying our experimental observations.
Second, the competitive effects of cholesterol on pentobarbital action occurred at cholesterol concentrations below those typically found in neuronal cells, [41 ] with an EC50of less than 2% cholesterol. This effect might occur at different cholesterol concentrations in other preparations but could be missed without complete concentration-response curves. The mechanistic implications of this competitive inhibition by low cholesterol concentrations are discussed below. At higher cholesterol concentrations (> 20%), additional inhibitory effects may occur (Figure 3and Figure 4), indicating that pentobarbital may alter sodium channel properties through more than one type of interaction involving the lipid bilayer.
Third, although pentobarbital suppression of sodium channel current was significantly altered by cholesterol, the effect of cholesterol on the voltage-dependent activation of the channel was not altered in these experiments. As previously discussed, [23 ] there is much variability in this property of sodium channels, between channels and between different measurements with the same channel. This variability makes it difficult to establish the significance of relatively small shifts in activation parameters. If cholesterol does not alter pentobarbital action at these potentials, while at the same time inhibiting pentobarbital suppression of sodium channel currents, there may be at least two different sites of pentobarbital action on the channel.
Finally, the concentrations of pentobarbital used in these experiments to achieve a 50% block of sodium channel function are substantially higher than the reported clinical concentrations. [42 ] Are the observed cholesterol-dependent changes in pentobarbital block of sodium channels then relevant to the anesthetic action of this drug? Whereas it generally is assumed that proteins with an EC50for functional modification (either block or activation of ionic channels) at or close to clinically used anesthetic concentrations are more relevant anesthetic targets than proteins with higher EC50S, this is only one factor in cellular anesthesia and needs to be considered in conjunction with other cellular properties. For example, variables such as nerve fiber diameter, ionic channel densities, physiologic frequencies of stimulation, and different channel subtypes play important physiologic [43 ] and pharmacologic [44,45 ] roles in determining cellular function. An illustration of the relevance of these other factors is given by evidence indicating that less than a 2% change in the open probability of resting sodium channels can cause myotonia, [46 ] that effective insecticides targeting sodium channels alter less than 1% of channels, [47 ] and that activation of less than 1% of ATP sensitive potassium channels results in the observed shortening of the action potential duration of ventricular myocytes during early hypoxia. [48 ] In this regard, pentobarbital at clinical concentrations blocks a significant, and conceivably crucial, proportion of sodium channels (Figure 4(B), IC10).
At least four possible mechanisms could explain this simple competitive interaction between cholesterol and anesthetic modification of the channel. (1) Anesthetics could alter protein function by changing the bulk physicochemical properties of the membrane (the lipid hypothesis discussed above). In this case, cholesterol would minimize anesthetic modifications of the bulk lipid properties and/or stabilize certain channel conformations to make them less sensitive to anesthetic-induced changes in the membrane. It is unclear, however, that this type of interaction would result in the competitive inhibition found in this study. Richards et al. [11 ] also argued against this theory. (2) Cholesterol could lower the effective anesthetic concentration at its site(s) of action. Miller and Yu [49 ] have shown that pentobarbital has a lower partition coefficient into cholesterol containing membranes, and therefore, the anesthetic concentration in the membrane necessarily would be lower. In this case, the anesthetic site of action most likely would have to include the lipid membrane, such as a site on the lipid-protein interface. A site on the protein interacting with anesthetics would be affected by anesthetic partitioning only if that site is solely accessible via the lipid membrane and not through the portions of the channel interfacing aqueous solutions (i.e., the intra- and extracellular faces of the channel, or its pore lining regions). However, partitioning was affected only slightly at concentrations where a half-maximal effect was seen in our system. (3) Following classic receptor interactions, an alternative or additional explanation would be that cholesterol could compete for a common binding site(s) on the channel. However, cholesterol binding then would appear to have no effect on measured sodium channel properties in the absence of pentobarbital. Additionally, this mechanism alone could not explain all of the interactions between cholesterol-containing bilayers and pentobarbital, especially because the inhibitory effect of cholesterol saturates at higher cholesterol concentrations. (4) Cholesterol could allosterically reduce the affinity of either a lipophilic or hydrophilic anesthetic binding site on the channel protein. Cholesterol has been shown to have binding sites in hydrophobic areas of membrane proteins. [50 ] Irrespective of which view proves correct, all interpretations imply a modulatory role for membrane lipids such as cholesterol in anesthetic interactions.
In summary, the effects of cholesterol on pentobarbital modification of the voltage-gated sodium channel presented here provide evidence that some anesthetic interactions with membranes are affected by the lipid composition of the cellular membrane. This work does not contradict experiments supporting a direct interaction between anesthetics and proteins but indicates that direct protein-anesthetic interactions, independent of the lipid membrane, are not the only type of interaction that occurs. It is more likely that both anesthetic-protein and anesthetic-lipid-protein interactions are important in anesthesia, or that lipids and proteins form functional complexes affected by anesthetics.
The authors thank Zita Dorner, Sabine Schmitz and Robert Silver, for technical assistance, and Dr. T.J. Blanck, Dr. H. Bonisch, Dr. M. Gothert, Dr. S. R. Levinson, and Dr. C. D. Richards, for discussions.
*Schmitz S. Urban BW: Unpublished data. 1994.
REFERENCES
Krnjevic K: Cellular mechanisms of anesthesia. Ann NY Acad Sci 625:1-16, 1991.
Miller KW: Molecular mechanisms by which general anesthetics act. Mechanisms of Drugs in Anaesthesia. Edited by Feldman S. Scurr CF, Paton W. London, Edward Arnold, 1993, pp 181-200.
Richards CD: The mechanisms of general anesthesia. Topical Reviews in Anaesthesia Edited by Norman J, Whitwam JG, Bristol, John Wright and Sons. 1980, pp 1-81.
Strichartz GR: Through the eye of the channel: Considerations of a reductionist view of general anesthesia {editorial} ANESTHESIOLOGY 69:155-156, 1988.
Franks NP, Lieb WR: Molecular mechanisms of general anaesthesia. Nature 300:487-493, 1982.
Franks NP, Lieb WR: Stereospecific effects of inhalational general anaesthetic optical isomers on nerve ion channels. Science 254:427-430, 1991.
Abraham MH, Lieb WR, Franks NP: Role of hydrogen bonding in general anesthesia. J Pharm Sci 80:7199-724, 1991.
Deuticke B, Haest CW: Lipid modulation of transport proteins in vertebrate cell membranes. Ann Rev Physiol 49:221-235, 1987.
Dluzewski Ar, Halsey MJ, Simmonds AC: Membrane interactions with general and local anaesthetics: A review of molecular hypotheses of anaesthesia. Mol Aspects Med 6:459-573, 1983.
Goldstein DB: The effects of drugs of membrane fluidity. Annu Rev Pharmacol Toxicol 24:43-61, 1984.
Richards CD, Martin K, Gregory S, Keightley CA, Hesketh TR, Smith GA, Warren GB, Metcalfe JC: Degenerate perturbations of protein structure as the mechanisms of anaesthetic action. Nature 276:775-779, 1978.
Armendi AJ, Tonner PH, Bugge B, Miller KW: Barbiturate action is dependent on the conformational state of the acetylcholine receptor. ANESTHESIOLOGY 79:1033-1041, 1993.
Dilger JP, Brett RS, Mody Hl: The effects of isoflurane on acetylcholine receptor channels: 2. Currents elicited by rapid perfusion of acetylcholine. Mol Pharmacol 44:1056-1063, 1993.
Moss GW, Franks NP, Lieb WR: Anesthetic inhibition of firefly luciferase, a protein model for general anesthesia, does not exhibit pressure reversal Biophys J 60:1309-1314, 1991.
Frenkel C, Duch DS, Urban BW: Effects of i.v. anaesthetics on human brain sodium channels. Br J Anaesth 71:15-24, 1993.
Urban BW: Differential effects of gaseous and volatile anaesthetics on sodium and potassium channels. Br J Anaesth 71:25-38, 1993.
Frenkel Ch, Duch DS, Recio-Pinto E, Urban BW: Pentobarbital suppresses human brain sodium channels. Mol Brain Res 6:211-216, 1989.
Frenkel Ch, Duch DS, Urban BW: Molecular actions of pentobarbital isomers on sodium channels from human brain cortex. ANESTHESIOLOGY 72:640-649, 1990.
Duch DS, Recio-Pinto E, Frenkel Ch, Urban BW: Human brain sodium channels in bilayers. Mol Brain Res 4:171-177, 1988.
Cohen RS, Blomberg F, Berzins K, Siekevitz P: The structure of postsynaptic densities isolated from dog cerebral cortex: 1. Overall morphology and protein composition. J Cell Biol 74:181-203, 1977.
Recio-Pinto E, Duch DS, Levinson SR, Urban BW: Purified and unpurified sodium channels from eel electroplax in planar lipid bilayers. J Gen Physiol 90:375-395, 1987.
Krueger BK, Worley JF, French RJ: Single sodium channels from rat brain incorporated into planar lipid bilayer membranes. Nature 303:172-175, 1983.
Rehberg B, Duch DS, Urban BW: The voltage-dependent action of pentobarbital on batrachotoxin-modified human brain sodium channels. Biochim Biophys Acta 1194:215-222, 1994.
Lechleiter J, Wells M, Gruener R: Halothane-induced changes in acetylcholine receptor channel kinetics are attenuated by cholesterol. Biochim Biophys Acta 856:640-645, 1986.
Mason PR: Membrane interaction of calcium channel antagonists modulated by cholesterol. Biochem Pharm 45:2173-2183, 1993.
Fettiplace R, Haydon DA: Water permeability of lipid membranes. Physiol Rev 60:512-550, 1980.
Fettiplace R, Gordon GM, Hladky SB, Requena J, Zingsheim HP, Haydon DA: Techniques in the formation and examination of “black” lipid bilayer membranes. Methods Membr Biol 4:1-75, 1975.
Latorre R. Alvarez O: Voltage-dependent channels in planar lipid bilayer membranes. Physiol Rev 61:77-150, 1981.
Montal M: Reconstitution of channel proteins from excitable cells in planar lipid bilayer membranes. J Membr Biol 98:101-115, 1987.
Krueger BK, Worley JF, French RJ: Single sodium channels from rat brain incorporated into planar lipid bilayer membranes. Nature 303:172-175, 1983.
Hartshorne RP, Keller U, Talvenheimo JA, Catterall WA, Montal M: Functional reconstitution of the purified brain sodium channel in planar lipid bilayers. Proc Natl Acad Sci USA 82:210-214, 1985.
Chabala LD, Urban BW, Weiss LB, Green WN, Andersen OS: Steady-state gating of Batrachotoxin-modified sodium channels. J Gen Physiol 98:197-224, 1991.
Naranjo D, Latorre R: Ion conduction in substates of the batrachotoxin-modified Na-channel from toad skeletal muscle. Biophys J 64:1038-1050, 1993.
Garber SS, Miller C: Single Sodium sup + channels activated by veratridine and batrachotoxin. J Gen Physiol 89:159-180, 1987.
Zamponi GW, Doyle DD, French RJ: Fast lidocaine block of cardiac and skeletal muscle sodium channels: One site with two routes of access. Biophysical J 65:80-90, 1993.
Castillo C, Villegas R, Recio-Pinto E: Alkaloid-modified sodium channels from lobster walking leg nerves in planar lipid bilayers. J Gen Physiol 99:897-930, 1992.
Behrens MI, Oberhauser A, Bezanilla F, Latorre R: Batrachotoxin-modified sodium channels from squid optic nerve in planar lipid bilayers. Ion conduction and gating properties. J Gen Physiol 92:23-41, 1989.
Cukierman S, Zinkand WC, French RJ, Krueger BK: Effects of membrane surface charge and calcium on the gating of rat brain sodium channels in planar lipid bilayers. J Gen Physiol 92:431-417, 1988.
Zamponi GW, Doyle DD, French RJ: State-dependent block underlies the tissue specificity of lidocaine action on batrachotoxin-activated cardiac sodium channels. Biophysical J 65:91-100, 1993.
Haydon DA, Urban BW: The action of hydrocarbons and carbon tetrachloride on the sodium current of the squid giant axon. J Physiol (Lond) 338:435-450, 1983.
Hamberger A, Svennerholm L: Composition of gangliosides and phospholipids of neuronal and glial cell enriched fractions. J Neurochem 18:1821-1829, 1971.
Firestone L.I., Miller JC, Miller KW: Tables of physical and pharmacological properties of anesthetics. Molecular and Cellular Mechanisms of Anesthesia. Edited by Roth MJ, Miller KW. New York, Plenum, 1986.
Mandel G: Tissue specific expression of the voltage-sensitive sodium channel. J Membr Biol 125:193-205, 1992.
Butterworth JF, Raymond SA, Roscoe RF: Effects of halothane and enflurane on firing threshold of frog myelinated axons. J Physiol 411:493-516, 1989.
Raymond SA, Shin HC: A role for changes in axonal excitability in general anesthesia. Molecular and Cellular Mechanisms of Alcohol and Anesthetics. Edited by Rubin E, Miller KW, Roth SH, New York, Annals of the New York Academy of Science, 1991.
Cannon SC, Brown RH, Corey DP: Theoretical reconstruction of myotonia and paralysis caused by incomplete inactivation of sodium channels. Biophys J 65:270-288, 1993.
Narahashi T: Nerve membrane Sodium channels as targets of insecticides. Trends Pharmacol Sci 13:236, 1992.
Weiss JN, Venkatesh N, Lamp ST: ATP-sensitive Potassium sup + channels and cellular Potassium sup + loss in hypoxic and ischaemic mammalian ventricle. J Physiol (Lond) 447:649-673, 1992.
Miller KW, Yu SC.: The dependence of the lipid bilayer membrane: Buffer partition coefficient of pentobarbitonic on pH and lipid composition. Br J Pharmacol 61:57-63, 1977.
Jones OT, McNamee MG: Annular and nonannular binding sites for cholesterol associated with the nicotinic acetylcholine receptor. Biochemistry 27:2364-2374, 1988.
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
Figure 1. Comparison of sodium channel properties in the absence and presence of cholesterol. (A) Current traces in control (phosphatidylethanolamine (PE) and phosphatidylcholine (PC), 4:1 ratio) lipids and with 4% or 50%(weight/weight, corresponding to 7.3 and 65.3 mol%, respectively) cholesterol added. Synaptosomal fractions of human brain cortex were prepared, incorporated into planar bilayers in the presence of 250 nM batrachotoxin, and voltage-clamped (40 mV, filtered at 200 Hz). (B) Current-voltage plots for single sodium channels. Current amplitudes are averages of open-closed-open transitions measured manually. Circles = 4PE:1PC (no cholesterol), n = 19; triangles = 4%(weight/weight) cholesterol, n = 6; squares = 50%(weight/weight) cholesterol, n = 5. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.1 plus/minus 0.5, 25.5 plus/minus 0.3, and 24.6 plus/minus 0.2 pS). Data for 2%, 10%, and 20% cholesterol (not shown) yielded 27.6 plus/minus 1.1, 26.5 plus/minus 0.6, and 26.0 plus/minus 0.6 pS, respectively. Differences are not significant (values are within 95% confidence intervals). (C) Current through the sodium channels was averaged over time and plotted versus membrane potential. Circles = 4PE:1PC (no cholesterol), n = 6; triangles = 4%(weight/weight) cholesterol, n = 4; squares = 50%(weight/weight) cholesterol, n = 4. Error bars are SEM. Straight lines are linear regression fits of the data (slope conductances are 26.0 plus/minus 0.4, 27.9 + 1.0, and 26.1 plus/minus 0.8 pS); differences are not significant (values are within 95% confidence intervals).
×
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
Figure 2. Effect of cholesterol on pentobarbital-induced sodium channel suppression. (A) Current traces as in Figure 1A, except in the presence of 680 micro Meter pentobarbital (same channels as in Figure 1A). (B) Time-averaged current through the same channels as in Figure 1C, after addition of 680 micro Meter pentobarbital; curves were significantly different (slope conductances are 13.7 plus/minus 0.2, 17.4 plus/minus 0.3, and 20.9 plus/minus 0.2 pS, respectively).
×
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
Figure 3. Lineweaver-Burk transformation of the effect of cholesterol on pentobarbital action. Data were recorded as described in methods. Suppression is measured as percent decrease in time-averaged current from control. Plotted are the data for channels in 4PE:1PC-membranes (0% cholesterol; circles = IC50of 845 plus/minus 72 micro Meter); 4%(weight/weight, corresponding to 7.3 mol%; squares = IC50of 1,451 plus/minus 97 micro Meter); and 10% cholesterol (weight/weight, corresponding to 17.3 mol%; triangles = IC50of 1,950 plus/minus 242 micro Meter). Linear regression fits have r2values of 0.96, 0.99, and 0.96, respectively. Data indicate a competitive inhibition of pentobarbital action by cholesterol at low cholesterol concentrations (below 10%).
×
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
Figure 4. Decrease in potency of pentobarbital induced by increasing cholesterol concentration. (A) Increased in IC50: values were obtained from linear regression fits to the double reciprocal plots of the concentration-response data as shown above. Data were fitted with a rectangular hyperbola; Bmax= 1,949 plus/minus 154 (SEM) micro Meter pentobarbital; EC50- 1.9 plus/minus 1.4 (SEM)% cholesterol (weight/weight, corresponding to 3.5 mol%). (B) IC10values calculated for the same data as in Figure 4(A).
×
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24 ]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
Figure 5. Effect of cholesterol on sodium channel activation. Filled circles denote data before and open squares after addition of 680 micro Meter pentobarbital. Data were calculated by averaging the averaged data from each membrane. Curves represent least-squares fits of a Boltzmann function to the averaged data (solid line - controls; dotted line - pentobarbital). Error bars denote SEM. (A) 4PE:1PC; nine experiments and 11 channels. Fits yielded potentials of half-maximal fractional open time (midpoint potentials) of 77.6 and 93.8 mV for controls and pentobarbital, respectively (shift of 16.2 mV). [24 ]. (B) 4% cholesterol; four experiments and five channels. Midpoint potential was shifted 6.1 mV by pentobarbital. (C) 10% cholesterol; three experiments and seven channels. Midpoint potential was shifted -13.1 mV by pentobarbital. (D) 50% cholesterol; four experiments and eight channels. Midpoint potential was shifted -11.0 mV by pentobarbital.
×
Table 1. Single-channel Properties of Channels before Addition of Pentobarbital in Bilayers with Differing Cholesterol Content
Image not available
Table 1. Single-channel Properties of Channels before Addition of Pentobarbital in Bilayers with Differing Cholesterol Content
×