Effects of Gabaergic Phenols on Phospholipid Bilayers as Evaluated by 1H-NMR ()
1. Introduction
General anesthetics are substances that induce a reversible state of unconsciousness, characterized by amnesia and analgesia. Their exact action mechanism remains incompletely elucidated. They were originally believed to act via nonspecific interactions with the lipid bilayers, affecting membrane fluidity. More recently, general anesthetics have been shown to act by modulating ligandgated ion channels such as the GABAA receptor (GABAR) (see references in [1]).
The GABA-R, a ligand-gated ion channel, constitutes the main inhibitory receptor of the central nervous system. GABA-Rs, besides being activated by the GABA neurotransmitter, are modulated by numerous therapeutically important drugs, including barbiturates, anesthetics, benzodiazepines, neurosteroids and ethanol. These compounds are GABA-Rs allosteric modulators as they bind to distinct sites to potentiate GABA-evoked currents [1-5]. The phenols propofol and thymol, and lately carvacrol, eugenol and chlorothymol, have been shown to act as positive allosteric modulators on this receptor [6-8].
GABA-R is an intrinsic membrane protein which activity may be affected by surface-active compounds and by physical changes in the membrane [4,9-12]. Taking into account the lipophilicity of the above described phenols, their interaction with the lipid membrane phase, especially the lipids surrounding the receptor and a consequent non-specific receptor modulation cannot be discarded, justifying a detailed study of drug-membrane interaction.
Recently we determined several lipophilic parameters for these five gabaergic phenols. The results obtained, based on the octanol-water partition coefficient (logP o/w), retention data in high performance liquid chromatography (HPLC) using C18 and immobilized artificial membrane (IAM) columns at different temperatures, and partition coefficients determined with phospholipid liposomes, demonstrated the high capacity of all the compounds to interact with membrane phases [13]. In addition, by using Langmuir films and epifluorescence images, we described that all the compounds were able to diffuse into the membrane, placing themselves between phospholipid molecules probably at the head-group region [14]. Finally, by means of fluorescence anisotropy studies we have recently found that all five compounds were able to decrease the microviscosity of artificial membranes [8].
Altogether these results indicate that the phenols compounds interact with the lipid membrane phase, suggesting their anesthetic activity could be the combined result of their specific (with receptor proteins) as well as nonspecific interaction (with surrounding lipid molecules) modulating the supramolecular organization of the receptor environment.
One approach to investigate the interactions between drugs and lipid molecules is the use of 1H-NMR which could give information about changes on the membrane dynamics by the mapping of the different bilayer regions [15]. In the current study, by using 1H-NMR spectroscopy, we have investigated the effects of the insertion and the possible preferential location of the five phenol derivatives with GABAergic activity (propofol, thymol, carvacrol, eugenol and chlorothymol) on EPC membranes.
2. Experimental
2.1. Materials
Propofol (2,6-bis(isopropyl)-phenol), thymol (5-methyl- 2-isopropyl-phenol), carvacrol (2-methyl-5-isopropylphenol), eugenol (2-methoxy-4-prop-2-enyl-phenol) and chlorothymol (5-methyl-4-chloro-2-isopropyl-phenol) were obtained from Sigma Chemical Co. (St Louis, MO, USA), and used without further purification. Egg phosphatidyl choline (EPC) was from Avanti Polar Lipids (Alabaster, USA). Water was bidistilled in an all-glass apparatus (pH 6.5 ± 0.3). Other drugs and solvents used were of analytical grade.
2.2. Membrane Preparation
Liposomes were obtained by evaporating stock chloroform solutions of EPC under a stream of N2. The samples were left under vacuum for no less than 2 h to remove residual solvent. The lipids were then suspended in 0.05 M phosphate buffer solution, pH 7.4 and vortexed for 5 min to form large multilamellar vesicles (MLVs).
For NMR experiments, small unilamellar vesicles (SUV) were used. Briefly, MLVs, obtained as described above, but suspended in D2O, were sonicated until clear (ca. 15 min) in a Sonics and Materials equipment (Newtown, CT). During sonication the temperature was kept at 0˚C - 4˚C by intermittent (1 min) agitation cycles, in an ice-water bath.
2.3. Partition Coefficient Determination
Phenols concentrations inside the membrane, expressed as molar ratios with respect to EPC, were calculated from the membrane—buffer partition coefficient, P, of each compound. In turn P was determined by phase-separation between MLVs and buffer at pH 7.4, according to the Equation (1) [16]:
(1)
where n denotes the number of phenol moles, V is the volume, and the subscripts m and w refer to the membrane and aqueous phase, respectively. The volume of the membrane phase, Vm, was calculated assuming a lipid density of 1 g/mL [16]. The amount of each phenolic compound bound to the lipid phase was optically determined at their corresponding wavelengths of maximal absorption between 270 and 282 nm [13] after ultra-centrifugation at 120,000 × g for 2 h, by subtracting the supernatant concentration from the total drug concentration measured before phase mixing.
2.4. Nuclear Magnetic Resonance (NMR) Experiments
Spectra were collected in a Varian Innova 600 MHz (LNBio, Campinas, Brazil) equipment. The samples were degassed to avoid the interference of dissolved O2 with longitudinal relaxation times (T1) measurements. For 1H-NMR, a 90˚ pulse was typically 10 - 15 μs and the recycling time was set to 5 times the largest T1 (those of the aromatic protons of phenols), typically 6 s. T1 were obtained by the conventional inversion-recovery technique, at 37˚C. Using the determined partition coefficient values—see Section 2.3, all phenols were added to the sonicated vesicles up to 1:3 phenol:EPC molar ratio within the membrane.
3. Results and Discussion
Membrane-buffer partition coefficients were determined previously to the NMR experiments in order one could calculate the proper phenol amount to guarantee a 1:3 drug:lipid molar ratio in the membrane. PEPC/w (between egg-phosphatidylcholine liposomes and phosphate buffer, pH 7.4) values, follow a similar behavior to the partition coefficient determined in other comparable systems reported before [13] with a hydrophobic profile of: chlorotymol ≥ propofol > carvacrol ≥ thymol > eugenol.
1H-NMR spectra (600 MHz) of each compound, of EPC unilamellar vesicles and of samples containing vesicles in the presence of each compound, at a 1:3 (phenol:lipid) molar ratio, were collected. Typical spectra of phenol, EPC small vesicles and phenol in EPC vesicles (1:3 mole%) are shown in Figure 1, for propofol.
The assignments of EPC and phenols hydrogen signals are indicated in Figure 2, where capital letters refer to the phospholipid, and lower-case letters identify the nonequivalent resonance peaks of each phenol compound These assignments were in good agreement with those reported in the literature for EPC [15-17] and carvacrol, thymol and eugenol [18,19].
From the chemical shifts (C.S.) corresponding to the phenolic hydrogens in water or in EPC vesicles, changes in the chemical shifts between both systems (DC.S.) were calculated. The same procedure was applied to determine the DC.S. of EPC hydrogens in water or in the presence of each phenol. Table 1 shows the C.S. and DC.S. for
Figure 1. 1H-NMR spectra (600 MHz) of propofol in D2O, EPC small unilamellar vesicles and propofol in EPC vesicles (pD 7.4; 25˚C).
Tab
le 1. Partition coefficients (PEPC/w) of phenol compounds between egg-phosphatidylcholine liposomes and phosphate buffer, pH 7.4.
Figure 2. Hydrogen peaks assignments for each compound (lower-case letters) and EPC molecule (capital letters) in NMR spectra.
phenols and Figure 3 represent the DC.S. values determined for EPC.
Changes in the chemical shifts of hydrogens (DC.S. ≠ 0) would indicate variations in the chemical environment of the nuclei, being considered more significant those changes higher than 0.05 ppm [15]. All phenolic compounds assayed showed significant DC.S., especially in their aromatic hydrogens, indicating that, in the presence of EPC vesicles, they experience a different chemical environment, and confirming their interaction with the vesicles (Table 2).
Figure 3 shows the effect (DC.S.) of the five phenol compounds on chemical shift of EPC hydrogens. Eugenol, the less hydrophobic analog essentially affected hydrogen I (choline group nearby the phosphate atom of EPC), changing it downfield. All the other phenol compounds induced upfield shifts in EPC hydrogens around the glycerol backbone region. In the presence of thymol, the main DC.S. found in the EPC molecule was observed in hydrogens H, E and C. In the presence of propofol, a significant alteration was observed in hydrogen E and,