Characterization of the binding of the local anesthetics procaine and tetracaine to model membranes of phosphatidylethanolamine: a deuterium nuclear magnetic resonance study

The interaction of the local anesthetics tetracaine and procaine with multilamellar dispersions of phosphatidylethanolamine has been investigated by using 2H NMR of specifically deuterated anesthetics. Tetracaine was found to partition more strongly than procaine into the lipid. The 2H NMR spectra showed a quadrupole doublet and a narrow line, with the former corresponding to membrane-bound anesthetic and the latter to anesthetic free in solution. The integrated areas of the narrow line and of the doublet correspond to the concentrations of free and bound anesthetic predicted from the Kp values. There is no strong pH dependence for the quadrupole splittings of tetracaine, suggesting a similar depth of penetration into the lipid bilayer over the entire pH range. The data are consistent with a model in which tetracaine acts as a wedge to stabilize the phosphatidylethanolamine bilayer against transition to a hexagonal structure. Procaine is proposed to sit higher in the phosphatidylethanolamine bilayer than does tetracaine. The T1 values were generally shorter in the membrane than in solution, suggesting slower motions, particularly for the aromatic ring of tetracaine.     

A 2H-NMR study on the interactions of the local anesthetic tetracaine with membranes containing phosphatidylserine

The interaction of the local anesthetic tetracaine with phosphatidylserine-containing model membranes has been studied by 2H-NMR. Charged tetracaine exhibited an unusually large partition coefficient into multilamellar dispersions of phosphatidylserine. The 2H-NMR spectra consisted of a Pake doublet and a narrow line, with the former corresponding to tetracaine in the bilayer and the latter to tetracaine free in solution. A strong pH dependence of the quadrupole splittings indicated different membrane locations for charged and uncharged tetracaine. In equimolar mixtures of phosphatidylserine and phosphatidylcholine the partition coefficients and 2H-NMR spectra were much more like those observed in neat phosphatidylcholine than in neat phosphatidylserine. Dilution studies at pH 5.5 indicated that in phosphatidylserine/phosphatidylcholine mixtures tetracaine experiences a three-site exchange similar to that found earlier for tetracaine in phosphatidylcholine. Tetracaine is in fast exchange between sites weakly bound to membrane and free in solution, and in slow exchange with a strongly bound site in the membrane.     

Interactions of the local anesthetic tetracaine with membranes containing phosphatidylcholine and cholesterol: a 2H NMR study

The interactions of the local anesthetic tetracaine with multilamellar dispersions of 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) and cholesterol have been investigated by deuterium nuclear magnetic resonance of specifically deuteriated tetracaines, DMPC and cholesterol. Experiments were performed at pH 5.5, when the anesthetic is primarily charged, and at pH 9.5, when it is primarily uncharged. The partition coefficients of the anesthetic in the membrane have been measured at both pH values for phosphatidylcholine bilayers with and without cholesterol. The higher partition coefficients obtained at pH 9.5 reflect the hydrophobic interactions between the uncharged form of the anesthetic and the hydrocarbon region of the bilayer. The lower partition coefficients for the DMPC/cholesterol system at both pH values suggest that cholesterol, which increases the order of the lipid chains, decreases the solubility of tetracaine into the bilayer. For phosphatidylcholine bilayers, it has been proposed [Boulanger, Y., Schreier, S., & Smith, I. C. P. (1981) Biochemistry 20, 6824-6830] that the charged tetracaine at low pH is located mostly at the phospholipid headgroup level while the uncharged tetracaine intercalates more deeply into the bilayer. The present study suggests that the location of tetracaine in the cholesterol-containing system is different from that in pure phosphatidylcholine bilayers: the anesthetic sits higher in the membrane. An increase in temperature results in a deeper penetration of the anesthetic into the bilayer. Moreover, the incorporation of the anesthetic into DMPC bilayers with or without cholesterol results in a reduction of the lipid order parameters both in the plateau and in the tail regions of the acyl chains, this effect being greater with the charged form of the anesthetic.     

A new look at the hemolytic effect of local anesthetics, considering their real membrane/water partitioning at pH 7.4

The interaction of local anesthetics (LA) with biological and phospholipid bilayers was investigated regarding the contribution of their structure and physicochemical properties to membrane partition and to erythrocyte solubilization. We measured the partition into phospholipid vesicles—at pH 5.0 and 10.5—and the biphasic hemolytic effect on rat erythrocytes of: benzocaine, chloroprocaine, procaine, tetracaine, bupivacaine, mepivacaine, lidocaine, prilocaine, and dibucaine. At pH 7.4, the binding of uncharged and charged LA to the membranes was considered, since it results in an ionization constant (pK_app) different from that observed for the anesthetic in the aqueous phase (pK_w). Even though it occurred at a pH at which there is a predominance of the charged species, hemolysis was greatly influenced by the uncharged species, revealing that the disrupting effect of LA on these membranes is mainly a consequence of hydrophobic interactions. The correlation between the hemolytic activity and the LA potency shows that hemolytic experiments could be used for the prediction of activity in the development of new LA molecules.     

Lipid-protein interactions and effect of local anesthetics in acetylcholine receptor-rich membranes from Torpedo marmorata electric organ

The selectivity of lipid-protein interaction for spin-labeled phospholipids and gangliosides in nicotinic acetylcholine receptor-rich membranes from Torpedo marmorata has been studied by ESR spectroscopy. The association constants of the spin-labeled lipids (relative to phosphatidylcholine) at pH 8.0 are in the order cardiolipin (5.1) approximately equal to stearic acid (4.9) approximately equal to phosphatidylinositol (4.7) > phosphatidylserine (2.7) > phosphatidylglycerol (1.7) > G(D1b) approximately equal to G(M1) approximately equal to G(M2) approximately equal to G(M3) approximately equal to phosphatidylcholine (1.0) > phosphatidylethanolamine (0.5). No selectivity for mono- or disialogangliosides is found over that for phosphatidylcholine. Aminated local anesthetics were found to compete with spin-labeled phosphatidylinositol, but to a much lesser extent with spin-labeled stearic acid, for sites on the intramembranous surface of the protein. The relative association constant of phosphatidylinositol was reduced in the presence of the different local anesthetics to the following extents: tetracaine (55%) > procaine (35%) approximately benzocaine (30%). For stearic acid, only tetracaine gave an appreciable reduction (30%) in association constant. These displacements represent an intrinsic difference in affinity of the local anesthetics for the lipid-protein interface because the membrane partition coefficients are in the order benzocaine > tetracaine approximately procaine.     

Effects of local anesthetics and hemicholinium-3 on 45-Ca efflux in barnacle muscle fibers.

Benzocaine, which occurs in the uncharged form in the physiological range of pH, caused inhibition of 45-Ca efflux in branacle muscle fibers. By contrast, in the presence of a low external Ca-2+ concentration it produced stimulation of the efflux. Both the inhibitory and stimulatory actions of benzocaine appeared to be less potent than those of procaine. Hemicholinium-3 (HC-3), on the other hand, which exists only in the charged form, caused a large stimulation of the 45-Ca efflux following microinjection, and the potency of this action was found to be at least 10 times greater than that of procaine. External application of HC-3 produced inhibition occasionally. Effects of tetracaine were similar to those produced by procaine; however, its inhibitory action was greater in more alkaline solution, which is the opposite of that observed with procaine. Lidocaine produced a less consistent effect than procaine; the inhibitory action of the former was less potent but the stimulatory action of the two anesthetics were comparable, p-Aminobenzoic acid was without effect on 45-Ca efflux. These results indicate that both the charged and uncharged forms of local anesthetics are capable of causing stimulatory and inhibitory effects on 45-Ca efflux in barnacle muscle fibers, and that the inhibition produced is the result of action on the CA-Ca exchange system whereas the stimulation is the result of release of Ca from internal storage sites.     

Charge and pH dependent drub binding to model membranes: A 2H-NMR and light absorption study

The binding of the local anesthetics tetracaine and procaine to model membranes of egg phosphatidylcholine and bovine phosphatidylserine has been studied by ²H-NMR and light absorption. Dispersions of drug-lipid mixtures in 0.1 M NaCl were centrifuged and the concentration of drug in the supernatant was measured by ultraviolet light absorption. Several freeze-thaw cycles of the sample were used before centrifugation to facilitate equilibration of the drug between the bilayers. Binding curves for the drug were obtained as a function of pH. The results were simulated by a theoretical model based on the Gouy-Chapman theory, in which both the charged and the uncharged forms of the drug, and the equilibrium between them, were included. Two deuterated forms of the drugs, [²H₆]tetracaine and [²H₄]procaine, were used for the ²H-NMR experiments. In most cases the ²H-NMR spectrum contained a broad central resonance and an underlying quadrupolar pattern. However, after five freeze-thaw cycles only a single broad resonance was observed under most conditions. Particle size measurements showed that freeze-thawing resulted in a more uniform population of liposomes of smaller average diameter than those obtained by simple vortex mixing. The single broad resonance observed in both cases is interpreted as due to rapid exchange of the anesthetic between lipid and bulk solution. In the absence of freeze-thawing, the quadrupolar pattern is attributed to anesthetic species in exchange with only a limited amount of water. The data suggest that a true equilibrium between lipid, water and anesthetic is only attained after freeze-thawing.     

Local anesthetic inhibition of pancreatic phospholipase A2 action on lecithin monolayers.

Using quantitative data previously reported for the penetration of local anesthetics into lecithin monolayers, the effects of surface and subphase concentrations of anesthetics on the inhibition of pancreatic phospholipase A2 action on didecanoyl phosphatidylcholine monolayers was investigated. Inhibition as a function of subphase concentration of anesthetic was in the order: dibucaine greater than tetracaine greater than butacaine greater than lidocaine = procaine. Inhibition as a function of surface concentration showed no obvious correlation; procaine inhibited at a very low surface concentration, followed by lidocaine at a somewhat higher concentration, and tetracaine, butacaine and dibucaine only at rather high concentrations. Ultraviolet difference spectroscopy indicated an interaction between lidocaine and enzyme in the subphase. Fluorescence studies showed that lidocaine is a competitive inhibitor of enzyme-lipid interface interaction. It is proposed that the more surface-active anesthetics inhibit by surface effects while the less surface-active anesthetics (lidocaine and procaine) inhibit by interaction with the enzyme in the subphase, which prevents enzyme penetration at the monolayer interface.     

Membrane-buffer partition coefficients of tetracaine for liquid-crystal and solid-gel membranes estimated by direct ultraviolet spectrophotometry

The membrane-buffer partition coefficient of tetracaine was measured by direct ultraviolet spectrophotometry in dimyristoylphosphatidylcholine unilamellar liposomes at temperatures above and below the main phase transition. The partition coefficients of uncharged tetracaine to solid-gel (18 degrees C) and liquid-crystal (30 degrees C) membranes were 6.9 x 10(4) and 1.2 x 10(5), respectively. Despite the general assumption that local anesthetic binding to the solid membrane is negligible, this study showed that the solid membrane binding amounts to 57.5% of the liquid membrane binding. Binding of the charged form to the liquid or solid membrane was not detectable under the present experimental condition of 0.03 mM tetracaine bulk concentration. The present method measures metachromasia of local anesthetics when bound to lipid membranes. Its advantage is that the separation of the vesicles from the solution is not required. A linearized equation is presented that estimates the partition coefficient or binding constant graphically from a linear plot of the absorbance data. The method is applicable for estimation of drug partition when a measurable spectral change occurs due to complex formation.     

Effect of lipid membranes on the apparent pK of the local anesthetic tetracaine. Spin label and titration studies

Electrometric titrations and spin label data demonstrate changes in the experimentally determined apparent pK of an ionizable drug in the presence of membranes. This effect is attributed to the difference in partition coefficients for the charged and uncharged forms of the drug. Investigation of the binding of a local anesthetic, tetracaine, to egg phosphatidylcholine membranes indicates that the drug apparent pK decreases in the presence of membranes, the decrease being a function of membrane concentration. The agreement between titration and spin label studies is very good and could be simulated by calculating membrane-bound and free populations of charged and uncharged tetracaine from the independently-measured partition coefficients for the two forms.     

Drug-induced zeta potential changes in liposomes studied by laser Doppler spectroscopy

The technique of laser Doppler spectroscopy is used to measure the electrophoretic mobility of liposomes under the influence of one beta-blocking agent and three local anesthetics. All four drugs decrease the mobility (i.e., the zeta potential) of negatively charged phospholipids (soybean lipids, phosphatidylserine and cardiolipin). The mobility of electrostatically neutral pure phosphatidylcholine (zero mobility under control conditions at pH 7 and 4) is increased linearly with the logarithm of drug concentration, indicating binding and incorporation of positively charged drug molecules. The sequence of strength of activity, measured by zeta-potential changes, corresponds to that found in biological tissues: propranolol greater than tetracaine greater than lidocaine greater than procaine. For purely negatively charged lipids (phosphatidylserine, cardiolipin) the activity of the drug is higher at acidic pH, (pH 4), while for electrostatically neutral (phosphatidylcholine) or partly neutral (soybean) lipid liposomes drug activity is about the same at pH 9, 7 and 4. A Hill plot of the data reveals noncooperative drug binding. From the line width of the scattering power spectrum the mean particle radius and the average interparticle distance in the samples are determined.     

Effect of lipid membranes on the apparent pK of the local anesthetic tetracaine spin label and titration studies

Electrometric titrations and spin label data demonstrate changes in the experimentally determined apparent pK of an ionizable drug in the presence of membranes. This effect is attributed to the difference in partition coefficients for the charged and uncharged forms of the drug. Investigation of the binding of a local anesthetic, tetracaine, to egg phosphatidylcholine membranes indicates that the drug apparent pK decreases in the presence of membranes, the decrease being a function of membrane concentration. The agreement between titration and spin label studies is very good and could be simulated by calculating membrane-bound and free populations of charged and uncharged tetracaine from the independently-measured partition coefficients for the two forms.     

Effect of local anesthetics on the electrical characteristics of an excitable model membrane composed of dioleyl phosphate I. Static and steady-state response

The local anesthetics, tetracaine, procaine and lidocaine, interacted with a negatively charged lipid membrane composed of dioleyl phosphate (DOPH), which exhibited a self-sustained oscillation of the membrane potential. The anesthetics depolarized the membrane potential when present in increasing concentrations, whereas they increased the membrane resistance at low concentrations and decreased it at high concentrations. The above results were analyzed on the basis of electrochemical theory taking into account ion flux across the membrane. The electrical characteristics are affected by both the hydrophobicity and the diffusion constant of local anesthetics within the membrane.     

Spin label study of local anesthetic-lipid membrane interactions. Phase separation of the uncharged from and bilayer micellization by the charged form of tetracaine

The interaction between tetracaine and egg phosphatidylcholine (egg PC) multibilayers was examined. ESR spectra of an ester spin label indicate that at low uncharged anesthetic: lipid ratios, membrane organization decreases. At higher ratios, saturation and phase separation occur, as suggested by a second spectral component which appears when the water solubility of tetracaine is reached. However, experiments with the drug in the absence and in the presence of membranes, making use of a phospholipid spin label, suggest that the new phase does not consist of solid tetracaine alone. Location of the new phase in the membrane would require a change in partition coefficient, while its location outside would imply a mechanism whereby the anesthetic would come off the membrane as aggregate containing spin probe and phospholipid. Charged tetracaine forms micelles which disrupt-unilamellar egg PC vesicles (Fernandez, M.S. (1981) Biochim. Biophys. Acta 646, 27–30). Micellar tetracaine added to bilayers containing a PC spin probe changes the spectrum from one typical of a bilayer into one typical of micelles, indicating the formation of a tetracaine-egg PC mixed micelle. The effect is reversible upon dilution to concentrations below the critical micelle concentration of tetracaine. When membranes are prepared in the presence of a water-soluble spin label, TEMPOcholine, ascorbate destroys the signal of untrapped label; when mixed phospholipid-tetracaine are formed by addition of micellar tetracaine, this leads to a complete loss of the ESR signal. High drug concentrations are often used for anesthesia and could be related to morphological nerve damage caused by large doses of anesthetics.     

Spin label study of local anesthetic-lipid membrane interactions. Phase separation of the uncharged form and bilayer micellization by the charged form of tetracaine

The interaction between tetracaine and egg phosphatidylcholine (egg PC) multibilayers was examined. ESR spectra of an ester spin label indicate that at low uncharged anesthetic: lipid ratios, membrane organization decreases. At higher ratios, saturation and phase separation occur, as suggested by a second spectral component which appears when the water solubility of tetracaine is reached. However, experiments with the drug in the absence and in the presence of membranes, making use of a phospholipid spin label, suggest that the new phase does not consist of solid tetracaine alone. Location of the new phase in the membrane would require a change in partition coefficient, while its location outside would imply a mechanism whereby the anesthetic would come off the membrane as an aggregate containing spin probe and phospholipid. Charged tetracaine forms micelles which disrupt-unilamellar egg PC vesicles (Fernandez, M.S. (1981) Biochim. Biophys. Acta 646, 27-30). Micellar tetracaine added to bilayers containing a PC spin probe changes the spectrum from one typical of a bilayer into one typical of micelles, indicating the formation of a tetracaine-egg PC mixed micelle. The effect is reversible upon dilution to concentrations below the critical micelle concentration of tetracaine. When membranes are prepared in the presence of a water-soluble spin label, TEMPOcholine, ascorbate destroys the signal of untrapped label; when mixed phospholipid-tetracaine are formed by addition of micellar tetracaine, this leads to a complete loss of the ESR signal. High drug concentrations are often used for anesthesia and could be related to morphological nerve damage caused by large doses of anesthetics.     

Locations and dynamical perturbations for lipids of cationic forms of procaine, tetracaine, and dibucaine in small unilamellar phosphatidylcholine vesicles as studied by nuclear Overhauser effects in 1H nuclear magnetic resonance spectroscopy

Locations and dynamical perturbations for lipids of local anesthetics (procaine . HCl, tetracaine . HCl, and dibucaine . HCl) in sonicated egg yolk phosphatidylcholine (PC) vesicles have been studied by 1H-1H nuclear Overhauser effect (NOE) measurements. It was found that tetracaine and dibucaine bind much strongly to the neutral lipids than does procaine and that their mobilities are lowered to such an extent that spin diffusion is transmitted (i.e., omega 2 tau c2 much greater than 1). The intermolecular NOEs between drugs and PC were more effective in the case of dibucaine than with tetracaine, indicating that dibucaine binds to the lipids more strongly than tetracaine; this order agrees well with that of anesthetic potency. However, it was only tetracaine that gave any appreciable dynamical perturbation to the PC vesicles when they were monitored by the extent of transfer of the negative NOE from alpha-methylene protons to choline methyls, olefinic methines, acyl methylenes and terminal methyl protons. This finding was interpreted as being due to the differences in the locations of these drugs in small unilamellar vesicles: (1) procaine interacts with lipids very weakly at the outer surface of the vesicles; (2) tetracaine binds to the lipids both at the outer and inner halves of the bilayer, inserting its rod-like molecule in a forest of acyl chains of PC; (3) dibucaine binds tightly to the polar head-group of PC, which resides only at the outer half of the bilayer vesicles. It was concluded that the relative order of anesthetic potency within these drugs can be correlated not with the ability to affect membrane fluidity but with the ability to bind to lipids at the polar head-group of the bilayer vesicles.     

Molecular dynamics of the local anesthetic tetracaine in phospholipid vesicles

Upon introduction into phosphatidylcholine vesicles, the 13C magnetic resonance peaks of the aromatic resonances of tetracaine are broadened while the T1 relaxation times show little change. Addition of tetracaine to vesicles containing 30% cholesterol produces a similar broadening in the 13C NMR spectrum of tetracaine. Nuclear magnetic resonance parameters of phosphatidylcholine in vesicles which are unchanged by the addition of equimolar tetracaine include 13C T1 relaxation time and 31P linewidth, T1 relaxation time, and nuclear Overhauser effect enhancement. These results are interpreted as indicating a hydrophobic interaction between hydrocarbon portions of the anesthetic and phospholipid bilayer. The rotational correlation time of tetracaine about its long axis in the vesicles has been calculated from the 13C NMR spin lattice relaxation times to be about 10(-10.3) s and is unchanged by incorporation into the phospholipid bilayer. The positively charged ammonium group of tetracaine interacts with the negatively charged phosphate group of the vesicle lipids. Using shift reagents and 31P NMR, tetracaine has been shown to displace cations from the bilayer surface, and does not undergo fast flip-flop across the vesicle bilayer.     

Partition coefficients (n-octanol/water) of N-butyl-p-aminobenzoate and other local anesthetics measured by reversed-phase high-performance liquid chromatography

For the determination of the logarithmic partition coefficients between n-octanol and water (log P_o/w) of local anesthetics, the pH of the aqueous phase needs to be adjusted to high values to ensure that the local anesthetics are in the unionized form. Using the shake-flask or the stir-flask method, this high pH may catalyze hydrolysis, leading to increasing amounts of impurities in time. These impurities exclude non-selective quantification methods like UV spectrometry and require repetitive quantitative analysis of both liquid phases resulting in a tedious and time-consuming method. A rapid reversed-phase HPLC method was developed to measure log P_o/w of the local anesthetics N-butyl-p-aminobenzoate, methyl-p-aminobenzoate, benzocaine, procaine, mepivacaine, prilocaine, lidocaine, bupivacaine, etidocaine, tetracaine and oxubuprocaine.     

Inhibition of synaptosomal enzymes by local anesthetics

The effects of tertiary amine local anesthetics (procaine, lidocaine, tetracaine and dibucaine) and chlorpromazine were investigated for three enzyme activities associated with rat brain synaptosomal membranes, i.e., (Na⁺ + K⁺)-ATPase (ouabain-sensitive), Mg²⁺-ATPase (ouabain-insensitive) and acetylcholinesterase. Approximately the same concentrations of each agent gave 50% inhibition of both ATPase, for example 7.9 and 10 mM tetracaine for Mg²⁺-ATPase and (Na⁺ + K⁺)-ATPase, respectively; these concentrations are 10-fold higher than required for inhibition of mitochondrial F₁-ATPase. The relative inhibitory potency of the several agents was proportional to their octanol/water partition coefficients. Acetylcholinesterase was inhibited by all agents tested, but the ester anesthetics (procaine and tetracaine) were considerably more potent than the others after correction for partition coefficient differences. For tetracaine, 0.18 mM gave 50% inhibition and showed competitive inhibition on a Lineweaver-Burk plot, but for dibucaine a mixed type of inhibition was observed, and 0.63 mM was required for 50% inhibition. Tetracaine evidently binds at the active site, and dibucaine at the peripheral or modulator site, on this enzyme.     

Inhibition of fast axonal transport in vitro by tetracaine: an increase in potency at alkaline pH, and no change in potency in calcium-depleted nerves

Some of the present in vitro experiments compare the degree of inhibition of fast axonal transport produced by tetracaine at neutral and at alkaline pH. In desheathed spinal nerves from bullfrog, 0.5 mM tetracaine reduced the quantity of [3H]leucine-labeled proteins which were transported to a ligature by 43% at pH 7.2 and by 96% at pH 8.2; separate experiments established that transport was not affected by the pH change in the absence of tetracaine. The relationship between pH and transport-blocking potency of tetracaine (pKa 8.2) is such that the local anesthetic is more potent when more uncharged form of the molecule is present; this may reflect the easier penetration across the axonal plasma membrane by the uncharged form of the tetracaine molecule. The axonal smooth endoplasmic reticulum has been attributed the function of a calcium reservoir, and it appeared possible that local anesthetics could block axonal transport by releasing calcium from this structure. However, the inhibition of transport produced by 1 mM tetracaine (pH 7.1) in sheathed nerves was approximately 80% both in nerves with a lower than normal calcium content (47% of normal) and in nerves with a normal calcium content; this result does not support the hypothesis that inhibition of axonal transport by local anesthetics is mediated by an increase in intracellular free Ca2+, but does not rule out the hypothesis either.