Journal of Photochemistry
and Photobiology, B: Biology, 7 (1990)
THE LOCAL ANAESTHETIC QUENCHER OF PERYLENE MICELLES MARTA S. FERNtiDEZ
75-86
75
TETRACAINE AS A FLUORESCENCE IN
and ESTELA CALDER6N
Departamento de Bioquimica, Centro C&T Investigacibn y de Estudias Avanzados de1 Institute Politecnico National, Apartada Postal 14-740, 07000 M&ico, D.F. (Medico) (Received
November
Keywords. micelles.
10, 1989;
accepted
March 21, 1990)
Tetracaine, local anaesthetic, perylene, fluorescence quenching,
At neutral pH the local anaesthetic tetracaine hydrochloride quenches the fluorescence of the lipophik dye perylene incorporated into non-ionic micelles. The process follows the Stern-Volmer equation, suggesting that quenching occurs through encounter of fluorophore and quencher. As the pH is lowered from 5 to 1, the apparent quenching constant decreases sigmoidally, the midpoint of the curve being at pH 2.3, close to the pK value characterizing the ionization of the anaesthetic aromatic butylamino group. Quenching is completely reversed below pH 1. These results show that the ability of tetracaine to quench the fluorescence of perylene incorporated into micelIes depends on the absence of charge on its aromatic amine. Quenching was also studied in homogeneous dioxane-water solution. In this system the quenching constant also decreases sigmoidally as the pH is lowered. The inflection point of the curve is nearly coincident with the pK of tetracaine butylamino group in the same partially non-aqueous medium. Protonation of this group induces 60% reversal of the quenching, suggesting that the main mechanism of fluorescence extinction could be the electron transfer from unprotonated tetracaine aromatic amine to perylene in the excited state. However, an additional process which remains operative even when such an amino group is positively charged must also be involved. It can be concluded that the complete reversal of tetracaine quenching of perylene fluorescence in micelles induced by low pH is due to the inability of the anaesthetic to become partitioned into micelles upon protonation of its aromatic amine. In contrast, at neutral pH the local anaesthetic is able to reach the micelIe non-polar core where perylene is located. This is consistent with the models, suggesting that the membrane-bound tetracaine assumes a rod-like configuration parallel to the surface normal with the aromatic butylamino group located into a highly hydrophobic region.
loll-1344/90/$3.50
0 Elsevier
Sequoia/Printed
in The Netherlands
76
1. Introduction
The local anaesthetic tetracaine hydrochloride interacts both hydrophobically and electrostatically with membranes [ 11. The ability of this drug to establish hydrophobic interactions has been demonstrated by its micelleforming capacity [Z], the consequence of which is a disrupting effect on liposomes occurring when the vesicles are exposed to anaesthetic concentrations above the critical micellar value (approximately 60 mM) [3, 41. As for the electrostatic interaction of tetracaine with membranes, it gives rise to interesting effects concerning the displacement of ions from the membrane-water interface. The cationic form of tetracaine can, for instance, neutralize the phosphate group of phosphatidylcholine and displace P?’ from the surface of liposomes, as judged by the reversal of the shift that the paramagnetic cation induces in the nuclear magnetic resonance signal of the phospholipid trimethylammonium protons [5]. At neutral pH the tetracaine molecule has a clear-cut amphipathic character due to the positively charged aliphatic amine [6] and to a hydrophobic moiety equivalent to a decyl chain [ 21. Such a molecule may become inserted in membranes with its long axis parallel to the hydrocarbon chains and with the cationic dimethylamino group exposed to the lipid-water interface [5, 7,8]. This location, which has been supported by a number of studies [g-13], is also consistent with the extreme sensitivity of the acid-base dissociation of tetracaine dimethylamine to the Gouy-Chapman diffuse double-layer electrostatic potential as well as to the polarity of the interfacial region, which can be detected when the anaesthetic interacts with micelles of different charge type [6]. According to the above-mentioned model, the aromatic butylamino group of tetracaine would be located in a highly hydrophobic region. In contrast, a proposal made many years ago [14] placed the aromatic amine at the interfacial region, in close interaction with the phosphate group of lipids. Although the latter model has been abandoned, there are still doubts concerning how deeply tetracaine penetrates in bilayers, i.e. whether a rod-like or a bent structure describes in a better way the configuration assumed by the membrane-bound anaesthetic [ 11. Prom the above-described possible configurations of tetracaine in membranes, it follows that the butylamino group could be an excellent probe for learning more about the topographical aspects of the interaction of tetracaine with bilayers. Surprisingly, very little attention has been paid to that group. One possible reason for the lack of studies focusing on the aromatic amine is its low pK [6]. Ionization of the but&mine is generally considered unimportant from the point of view of the pharmacolcgical effects of the drug, since it takes place far below the physiological pI1. A study of the behaviour of the aromatic amino group at low pH might contribute, however, to a better understanding of processes occurring under physiological conditions.
77
The ability of local anaesthetics to quench the fluorescence of probes located in the membrane can be used to assess the way that the anaesthetics interact with bilayers. Using this kind of methodology, Koblin et al. [15] found that procaine and tetracaine quench, to different extents, the fluorescence of N-octadecyl naphthyl-2-amine 6-sulphonic acid and 12-(9anthroyl) stearic acid in the presence of red cell ghosts. In contrast, lidocaine does not show any quenching capacity. Since both tetracaine and procaine molecules contain an aromatic amine whereas lidocaine does not, Koblin et al. attributed the quenching activity to such a group. The position of the aromatic amino group in the tetracaine molecule is ideal to assess the penetration of the anaesthetic int ; the hydrophobic region of membranes. In addition, the presence of such a group seems to be indispensable for the anaesthetic to behave as quencher through encounter with fluorescent probes [ 151. These properties of tetracaine were utilized in the present study, to determine the location of the anaesthetic in micelles based upon its ability to quench the fluorescence of perylene. The lipophilic dye was solubilized into the hydrophobic core of detergent micelles. The experiments were performed at several pH values to test whether quenching is dependent on the protonation degree of the aromatic amino group of tetracaine. For comparison, a similar study was made using a dioxane-water mixture in order to have both the anaesthetic and the fluorophore freely solubilized in a homogeneous, partially non-aqueous medium.
2. Materials
and methods
The surfactant Brij 35 (polyoxyethylene (23) dodecylether) was purchased from BDH Chemicals Ltd. Crystalline tetracaine hydrochloride ((2dimethylaminoethyl)-4-(n-butylamino) benzoate hydrochloride) was obtained from Sigma. Perylene (peri-dinaphthalene) and spectroscopic grade dioxane (Uvasol) were from Merck and Aldrich respectively. The non-ionic micelles were prepared by dissolving 1% Brij 35 in water. Since the average molecular weight reported for this detergent is 1100 [ 161, such a concentration is equivalent to 9 mM, well above the critical micellar concentration (9.1 X 10m5M [ 171). Perylene is insoluble in water but can be solubilized in micelles [ 21. Incorporation of perylene into micelles was performed by adding a small volume of concentrated solution of the probe in ethanol to the detergent solution, with continuous and vigorous magnetic stirring. The final concentration of perylene was 4 X lo* M. The ethanol added to the aqueous solution was less than 0.4%. Considering the surfactant concentration and the aggregation number of the micelles which has been reported to be 40 [ 171, it can be estimated that the ratio of the number of micelles to the number of solubilized perylene molecules was approximately 56OO:l. Partially non-aqueous solutions of perylene and tetracaine were prepared using mixtures of dioxane and water of known dielectric constant [6-181.
78
For the quenching experiments, the fluorescence of the micellar or partially non-aqueous solution of perylene was measured before and after addition of the appropriate amount of crystalline tetracaine. Perylene fluorescence was excited at 409 nm and read at 469 nm (uncorrected wavelengths). In this wavelength range, no absorption or emission of light by tetracaine was detected. Excitation and emission spectra as well as fluorescence measurements at constant wavelengths were obtained in an Aminco Bowman spectrophotofluorometer equipped with a Hewlett-Packard 7004 B recorder. Slit widths equivalent to 5.5 run bandpass were used throughout. Acid-base titration of tetracaine aromatic amine in dioxane-water mixtures was performed by measuring the pH dependence of the anaesthetic fluorescence (excitation wavelength, 302 run; emission wavelength, 360 run) as previously described [ 61. Titration was performed on 3 PM anaesthetic solutions to avoid the well-known self-quenching effect which appears at millimolar concentrations of fluorophore [ 19 1. All experiments were repeated at least three to five times. The results shown in Figs. 1, 3, 4 and 5 are representative of the corresponding sets of experiments. In the rest of figures, mean values + standard deviation, are plotted. Standard deviation bars smaller than symbols are not shown.
3. Results Figure 1 shows that at pH 5 the local anaesthetic tetracaine quenches the fluorescence of the lipophilic dye perylene incorporated into micelles of the non-ionic detergent Brij 35. The process follows the Stern-Volmer [20] equation, suggesting that quenching occurs through encounter of fluorophore and quencher:
FO F=
1 +&[&I
(1)
where F,, and F are the fluorescence readings in the absence and presence of quencher and Ko is the Stern-Volmer quenching constant. In this case, KQ should be considered as an apparent quenching constant since it is obtained from the bulk quencher concentration [Q] and not from the actual anaesthetic concentration in the micellar phase, which is unknown. Obviously this apparent quenching constant must include implicitly a partition coefficient [ 2 1,22 ] accounting for the distribution of the anaesthetic between the micellar and aqueous phases. It can be observed that the slopes of the straight lines in Fig. 1 decrease as the pH is lowered. A normalized plot of the apparent quenching constant as a function of pH (Fig. 2) shows that KQ decreases sigmoidally as the pH becomes more acidic, the curve showing an inflection point at pH 2.3. Quenching is completely reversed below pH 1. The pH titration of the quenching process seems to follow the ionization of the anaesthetic aromatic amino group for which a pK of 2.14 has been reported [ 61. It appears that the ability of the anaesthetic to quench the fluorescence
79
100
pH=5 PI+=3
c
80
8 =
60
s PI-l= 2
5 a
---__ 40 !
pH=l
1 0 TETRACAINE
I
I
I
0.02
0.04
0.06
HYDROCHLORIDE
(Ml
0
1
2
3
4
5
PH
Fig. 1. Stem-Volmer plot for the quenching of perylene fluorescence by tetracaine in 1% micellar solutions of the non-ionic detergent Bru 35 at different pH values. The ionic strength Z was kept constant at 0.25 with NaCl. Fig. 2. The apparent quenching constant Ko as a function of pH, for quenching of perylene fluorescence by tetracaine in 1% micellar solutions of Brfj 35 (Z=O.25). Quenching constants were obtained from the slopes of straight lines similar to those in Fig. 1. The data were normalized by assigning a value of 100 to the K, corresponding to the plateau above pH 4. The midpoint of the curve is found at pH 2.3.
of perylene incorporated into micelles depends on the absence of charge on its aromatic amine. Quenching of perylene fluorescence by tetracaine takes place without changes in the position of the excitation or emission spectra of the fluorophore, as can be observed in Fig. 3. Control experiments were also performed to determine whether the fluorescence of perylene in micelles is affected by lowering the pH. Figure 4 shows the excitation and fluorescence spectra of perylene micellar solutions of pH 5 and pH 0.5. No effect is detected as the spectra obtained at both pH values are coincident. The effect of pH on the quenching activity of tetracaine on micellar perylene can be observed in the spectra of Fig. 5. This figure shows that, at constant tetracaine concentration (0.05 M), the perylene fluorescence intensity increases as the pH is lowered. The results in Figs. 4 and 5 are better illustrated in Fig. 6 which shows that only in the presence of tetracaine is perylene fluorescence affected by changing the pH from 5 to 0.5. In the absence of anaesthetic, fluorescence remains constant, notwithstanding the pH change. For comparison with the micellar process, quenching was also studied in an homogeneous dioxane-water mixture (30 wt.% dioxane), which makes it possible to solubilize both the local anaesthetic and perylene, the lipophilic fluorophore. In this system, the quenching process also obeys the Stern-Volmer expression, i.e. plots of the quenching parameters according to the Stem-Volmer equation (eqn. (1)) give straight lines with intercepts equal
80
1 300
350
400
450
450
WAVELENGTH
500
550
600
(nm)
3. The effect of tetracaine concentration on the fluorescence spectra of perylene in micelles (1% Brij; 10 mM NaCl; pH 5.0) at different tetracaine concentrations. The molar concentration of tetracaine was varied from 0 to 0.05 as indicated by the arrows. Excitation spectra (left) were scanned with emission tixed at 469 run; emission spectra (right) were obtained setting the excitation at 409 run. Fig.
7-
3 f 2 a
6
b
-
I 300
350
400
I..
I
450
450
WAVELENGTH
pli=0.5
J 500 1 “m
550
600
1
Fig. 4. The effect of pH on the fluorescence spectra of perylene in micelles (1% Brij; 10 mM NaCl). Excitation spectra (left) were scanned with emission 6xed at 469 run; for the emission spectra (right) the excitation was set at 409 run.
81
300
350
500 450 (nm)
450 400 WAVELENGTH
550
600
Fig. 5. Fluorescence spectra of micehar perylene at different pH values in the presence of tetracaine (1% Brij; 10 mM NaCI; 0.05 M tetracaine hydrochloride). The spectra were obtained at the pH values indicated in the figure. Excitation spectra (left) were scanned with the emission fixed at 469 run; for the emission spectra (right) the excitation was at 409 run.
”
n
Fig. 6. Fluorescence intensity (excitation at 409 nm; emission at 469 run) as a function of pH, for micellar perylene in the presence (0) and absence (0) of tetracaine. The perylene was dissolved in 1% Brij and 10 mM NaCl. Fig. 7. The pH dependence of the Stem-Volmer constant I&, for quenching of perylene fluorescence by tetracaine in homogeneous dioxane-water solutions (30 wt.% dioxane). The normalized curve was obtained by assigning a value of 100 to the limiting value of K, reached above pH 2. The inflection point of the curve corresponds to pH 1.4.
82
to 1 (not shown but similar to those in Fig. 1). The pH dependence of the Stem-Volmer constant is presented in Fig. 7. It can be seen that, as in the case of quenching in micellar solutions, the quenching constant decreases with acidification of media. However, there is an important difference: even for pH values as low as 0.5, no complete reversal of quenching is obtained, i.e. the quenching constant does not become zero as in the case of mice&r solutions. What can be observed, instead, is that, below pH 1, it decreases to a constant value, equivalent to approximately 40% of the Ko at neutral pH. Since the partially non-aqueous medium can lower the pK of a cationic acid [ 6,181, it seemed necessary to investigate whether the complete ionization range of the aromatic amine was scanned in the study of the pH dependence of KQ shown in Fig. 7. Thus, acid-base titration of the aromatic amine was performed in different dioxane-water mixtures. Figure 8 shows the pK of the butylamino group in dioxane-water mixtures containing 0 wt.%, 20 wt.%, 45 wt.%, 70 wt.% and 82 wt.% dioxane corresponding to dielectric constants of 78.36, 60.79,38.48, 17.69 and 9.53 respectively [6, 181. It can be observed that, as the proportion of dioxane in the mixtures is increased, the pK diminishes. In solutions containing 30 wt.% dioxane, the interpolated pK value for the aromatic amine is 1.7, slightly above the pH value of 1.4 corresponding to the inflection point of the curve in Fig. 7. Thus it is safe to conclude that the study of the effect of acidification on KQ in the partially non-aqueous medium covers the pH range in which ionization of the aromatic amino group takes place. Figures 2 and 7 make the difference between the pH dependence of KQ in micelles and that in homogeneous, partially non-aqueous solutions apparent. The lack of complete reversal of quenching is not the only characteristic differentiating the behaviour of the tetracaine-perylene pair in dioxane-water with respect to that in micelles. Figure 9 shows that for the micellar solution % WATER 1
100 r’
0
SO
60
I
0
20 %
40 60 DIOXANE
I
80
I
100
SODIUM
Fig. 8. The pK characterizing the acid-base dissociation of tetracaine solutions containing different proportions of dioxane and water.
CHLORIDE aromatic
(M ) amine
in
Fig. 9. The effect of salt concentration on the apparent quenching constant KQ for quenching of perylene fluorescence by tetracaine at pH 5: - --O- -, dioxane(30 wt.%)-water mixture; ,-Z-, 1% micellar solution of Brij 35.
83
a change in NaCl concentration from 0.01 to 0.3 M induces an increase in the apparent quenching constant K,, from 30.6 to 82.3 M-l. In contrast, K, for quenching in a dioxane-water mixture is slightly decreased by addition of salt.
4. Discussion In the present study, fluorescence quenching is used to investigate the adsorption of a local anaesthetic onto micelles. A description is presented of the ability of tetracaine to quench the fluorescence of the lipophilic dye perylene in two different media; a simple model membrane system consisting of micelles of a neutral amphiphile, into the hydrophobic region of which the fluorophore can be incorporated, and a homogeneous dioxane-water mixture where both the fluorophore and the quencher can be freely solubilized. The results presented here are consistent with quenching of perylene by tetracaine taking place by collision, since the Stem-Volmer expression is obeyed both in homogeneous solution and in micelles. Previous studies from other laboratories showed that amines can quench the fluorescence of aromatic hydrocarbons by transfer of an electron to the fluorophore in the excited state [23-251. Although such an explanation could be extended, by way of inference, to the extinction process studied in the present work, a more direct proof that tetracaine quenches perylene fluorescence through electron transfer was needed. Our working hypothesis was that, if quenching occurred by the abovementioned process, it would be reversed by protonation of the aromatic amine, since the positive charge would make that group unable to donate an electron to the fluorophore. Our results show that, in fact, part of the quenching in a partially non-aqueous medium is abolished by protonation of tetracaine aromatic amine. However, the fully protonated anaesthetic molecule is still capable of exhibiting a & equivalent to 40% of the value shown at neutral pH (Fig. 7), indicating that there should be an additional mechanism for the extinction of perylene fluorescence. At the moment, we do not have data supporting any particular mechanism for such remaining quenching activity. What we do know is that it is resistant to extremely acidic media. As for the complete reversal of quenching in the micellar system at low pH (Figs. 1 and 2), it is explained in terms of the two effects that protonation of the aromatic butylamino group can bring about: on the one hand, it decreases the quenching capacity of the anaesthetic molecules colliding with perylene through inhibition of the electron transfer, as already discussed; on the other hand, it makes tetracaine unable to penetrate into the mice&r region where the fluorophore is solubilized. Presumably, the positive charge makes the insertion of the protonated butylamino moiety into the micellar core thermodynamically unfavourable. Partition of tetracaine into the micellar phase appears to be very important for the apparent quenching activity of
84
the anaesthetic, as shown by the results in Fig. 9. Addition of NaCl to neutral micelles markedly increases the apparent quenching constant. In contrast, Ko for quenching in dioxane-water is almost unaffected by salt concentration. These results can be interpreted ln terms of the possible enhancement by the added salt of the ability of tetracaine to establish hydrophobic interactions. The salting-out of the anaesthetic from the aqueous phase would favour its partition into micelles, causing an increase in the apparent quenching constant. Other alternative explanations for the effect of salt on Ko based on a change in the structure of the Brij micelles or on an increased micellization of the detergent can be ruled out. Firstly, non-ionic micelles can hardly be affected by neutral salts [ 16, 171; secondly, even at low salt concentrations, the number of micelles is overwhelmingly large with respect to the number of perylene molecules (5600: 1) [ 171. An important part of this work is based on the reversal of the anaesthetic quenching activity by acidification of media to pH values near or below 1. We have not found any change attributable to perturbations, either of perylene or of the Brij micelle structure, by such an extreme acidic condition. The experiments of Figs. 4 and 6 clearly show that the fluorescence of perylene in the neutral micelles is not modified by acidification of the medium. Only in the presence of tetracaine does perylene fluorescence show a dependence on pH (Figs. 5 and 6). Since such a dependence correlates well with ionization of the anaesthetic aromatic amine [6] (Figs. 2 and S), our results can be explained in terms of the protonation-deprotonation equilibrium of that group. The information obtained on the participation of a functional group of the anaesthetic molecule in a fluorescence quenching process allows us to characterize the mode of insertion of tetracaine in model membrane systems. Our results show that at neutral pH the anaesthetic butyl amine can reach the micelle non-polar core where perylene is located and quench its fluorescence by collision. This is consistent with models suggesting that membranebound tetracaine assumes a rod-like configuration parallel to the surface normal with the aromatic butylamino group located in a highly hydrophobic region [l, 5-101. The nerve-blocking properties of local anaesthetics depend on their capacity to induce perturbations in membranes [ 261. A detailed knowledge of the con&uration of these compounds at the lipid-water interface will favour a better understanding of drug-membrane interactions and, as a consequence, of the molecular basis of their pharmacological effects.
Acknowledgements The partial financial support of Consejo National de Ciencia y Tecnologia de Mexico and Consejo de1 Sistema National de Education Tecnoldglca through research grants to M.S.F. is gratefully acknowledged. Dr. Fern&ndez is a member of the National System of Investigators, Mexico.
85
References 1 Y. Kuroda and Y. Fujiwara, Locations and dynamical perturbations for lipids of cationlc forms of procaine, tetracaine, and dibucaine in small unilameilar phosphatidylchohne vesicles as studied by nuclear Overhauser effects in (l)H nuclear magnetic resonance spectroscopy, Biochim. Biophys. Acta, 903 (1987) 395-410. M. S. Femandez, Formation of miceIles and membrane action of the local anaesthetic tetracaine hydrochloride, Biochim. Biophys. Acta, 597 (1980) 83-91. M. S. Femandez, Disruption of Iiposomes by tetracaine miceIles, B&him. Biophys. Acta, 646 (1981) 27-30. W. A. Frezzatti, Jr., W. R. Toseili and S. Schreier, Spin label study of local anaesthetic--lipid membrane interactions. Phase separation of the uncharged form and bilayer miceilization by the charged form of tetracaine, B&him. Biophys. Acta, 860 (1986) 531-538. 5 M. S. Femandez and J. Cerb6n, The importance of the hydrophobic interactions of local anaesthetics in the displacement of polyvaient cations from artiiicial lipid membranes, Biochim. Biophys. Acta, 298 (1973) 8-14. 6 J. Garcia-Soto and M. S. Fem&ndez, The effect of neutral and charged miceiies on the acid-base dissociation of the local anaesthetic tetracaine, B&him. Biophys. Acta, 731 (1983) 275-281. 7 H. Hauser, S. A. Penkett and D. Chapman, Nuclear magnetic resonance spectroscopic studies of procaine hydrochloride and tetracaine hydrocNoride at lipid-water interfaces, B&him Biophys. Acta, I83 (1969) 466-475. 8 J. Cerb6n, NMR evidence for the hydrophobic interaction of local anaesthetics. Possible relation to their potency, B&him. Biophys. Acta, 290 (1972) 51-57. 9 M. S. Femandez and J. Cerb6n, PMR and viscosity studies of the interaction of local anaesthetics and micelles, Arch. Biochem. Biophys., 172 (1976) 721-725. 10 Y. Boulanger, S. Schreier and I. C. P. Smith, Molecular details of anaesthetic--lipid interaction as seen by deuterium and phosphorus-31 nuclear magnetic resonance, Biochemistry, 20 (1981) 6824-6830. 11 Y. Boulanger, S. Schreier, L. C. Leitch and I. C. P. Smith, Multiple binding sites for local anaesthetics in membranes: characterization of the sites and their equilibria by deuterium NMR of specifically deuterated procaine and tetracaine, Can. J. Biochem., 58 (1980) 986-995. 12 J. Westman, Y. Boulanger, A. Ehrenberg and I. C. P. Smith, Charge and pH dependent drug binding to model membranes. A (2)HNMR and light absorption study, B&him. Biophys. Actu, 685 (1982) 315328. interaction: a (2) H nuclear 13 E. C. Kelusky and I. C. P. Smith, Anaesthetic-membrane magnetic resonance study of the binding of specifically deuterated tetracaine and procaine to phosphatidylchoiine, Can. J. Biochem. Cell Biol., 62 (1984) 178-184. 14 M. B. Feinstein, Reaction of local anaesthetics with phospholipids. A possible chemical basis for anaesthesia, .I. Gen. Physiol., 48 (1964) 357-374. 15 D. D. KobIin, W. D. Pace and H. H. Wang, The penetration of local anaesthetics Into the red blood cell membrane as studied by fluorescence quenching, Arch. Btichem. Biophys., 171 (1975) 176-182. 16 K. Shinoda, T. Nakagawa, B. I. Tamamushi and T. Isemura, Colloidal Su@zctants, Academic Press, New York, 1963. in Micellar and Macromokculur Systems, 17 J. H. Fendler and E. J. FendIer, Cat&&s Academic Press, New York, 1975. 18 M. S. Femandez and P. Fromherz, Lipoid pH indicators as probes of electrical potential and polarity in miceiles, J. Phys. Chem., 81 (1977) 1755-1761. 19 G. G. Guilbault, Practical Fluorescence: Theory, Methods and Techniques, Marcel Dekker, New York, 1973. 20 0. Stem and M. Voimer, The extinction period of fluorescence, Phys. Z., 20 l1919) 183-188.
86 21 G. Papageorgiou and C. Argoudelis, Cation-dependent quenching of the fluorescence of chlorophyll a in viva by nitroaromatic compounds, Arch. Biochem. Biophys., I56 (1973) 134-142. 22 S. G. Bertolotti, M. V. Bohorquez, J. J. Cosa, N. A. Garcia and C. M. Previtali, Mice&r effect on the fluorescence quenching of indolic compounds by aminoacids, Photo&em. Photobid., 46 (1987) 331-335. 23 H. Leonhardt and A. Weller, Elektronentibertragungsreaktionen des angeregten Perylens, Ber. Bunsenges. Phgs. Chem. 67 (1963) 791-795. 24 N. Mataga, K. Ezumi and T. Okada, Temperature effects on charge transfer fluorescence spectra and mechanisms of charge transfer interactions in the excited electronic state, Mol. Phys., 10 (1966) 201-202. 25 N. Mataga, T. Okada and K. Ezumi, Fluorescence of pyrene-iV,Ar-dimethyhmilinecomplex in non-polar solvent, Mol. Phys., 10 (1966) 203-204. 26 J. M. Ritchie and N. M. Greene, Local anaesthetics, in L. S. Goodman and A. Gihnan (eds.), The Phurrna.cological Basis of Th.erapeuks, MacMiian, New York, 6th edn., 1980, pp. 302321.
and Photobiology, B: Biology, 7 (1990)
THE LOCAL ANAESTHETIC QUENCHER OF PERYLENE MICELLES MARTA S. FERNtiDEZ
75-86
75
TETRACAINE AS A FLUORESCENCE IN
and ESTELA CALDER6N
Departamento de Bioquimica, Centro C&T Investigacibn y de Estudias Avanzados de1 Institute Politecnico National, Apartada Postal 14-740, 07000 M&ico, D.F. (Medico) (Received
November
Keywords. micelles.
10, 1989;
accepted
March 21, 1990)
Tetracaine, local anaesthetic, perylene, fluorescence quenching,
At neutral pH the local anaesthetic tetracaine hydrochloride quenches the fluorescence of the lipophik dye perylene incorporated into non-ionic micelles. The process follows the Stern-Volmer equation, suggesting that quenching occurs through encounter of fluorophore and quencher. As the pH is lowered from 5 to 1, the apparent quenching constant decreases sigmoidally, the midpoint of the curve being at pH 2.3, close to the pK value characterizing the ionization of the anaesthetic aromatic butylamino group. Quenching is completely reversed below pH 1. These results show that the ability of tetracaine to quench the fluorescence of perylene incorporated into micelIes depends on the absence of charge on its aromatic amine. Quenching was also studied in homogeneous dioxane-water solution. In this system the quenching constant also decreases sigmoidally as the pH is lowered. The inflection point of the curve is nearly coincident with the pK of tetracaine butylamino group in the same partially non-aqueous medium. Protonation of this group induces 60% reversal of the quenching, suggesting that the main mechanism of fluorescence extinction could be the electron transfer from unprotonated tetracaine aromatic amine to perylene in the excited state. However, an additional process which remains operative even when such an amino group is positively charged must also be involved. It can be concluded that the complete reversal of tetracaine quenching of perylene fluorescence in micelles induced by low pH is due to the inability of the anaesthetic to become partitioned into micelles upon protonation of its aromatic amine. In contrast, at neutral pH the local anaesthetic is able to reach the micelIe non-polar core where perylene is located. This is consistent with the models, suggesting that the membrane-bound tetracaine assumes a rod-like configuration parallel to the surface normal with the aromatic butylamino group located into a highly hydrophobic region.
loll-1344/90/$3.50
0 Elsevier
Sequoia/Printed
in The Netherlands
76
1. Introduction
The local anaesthetic tetracaine hydrochloride interacts both hydrophobically and electrostatically with membranes [ 11. The ability of this drug to establish hydrophobic interactions has been demonstrated by its micelleforming capacity [Z], the consequence of which is a disrupting effect on liposomes occurring when the vesicles are exposed to anaesthetic concentrations above the critical micellar value (approximately 60 mM) [3, 41. As for the electrostatic interaction of tetracaine with membranes, it gives rise to interesting effects concerning the displacement of ions from the membrane-water interface. The cationic form of tetracaine can, for instance, neutralize the phosphate group of phosphatidylcholine and displace P?’ from the surface of liposomes, as judged by the reversal of the shift that the paramagnetic cation induces in the nuclear magnetic resonance signal of the phospholipid trimethylammonium protons [5]. At neutral pH the tetracaine molecule has a clear-cut amphipathic character due to the positively charged aliphatic amine [6] and to a hydrophobic moiety equivalent to a decyl chain [ 21. Such a molecule may become inserted in membranes with its long axis parallel to the hydrocarbon chains and with the cationic dimethylamino group exposed to the lipid-water interface [5, 7,8]. This location, which has been supported by a number of studies [g-13], is also consistent with the extreme sensitivity of the acid-base dissociation of tetracaine dimethylamine to the Gouy-Chapman diffuse double-layer electrostatic potential as well as to the polarity of the interfacial region, which can be detected when the anaesthetic interacts with micelles of different charge type [6]. According to the above-mentioned model, the aromatic butylamino group of tetracaine would be located in a highly hydrophobic region. In contrast, a proposal made many years ago [14] placed the aromatic amine at the interfacial region, in close interaction with the phosphate group of lipids. Although the latter model has been abandoned, there are still doubts concerning how deeply tetracaine penetrates in bilayers, i.e. whether a rod-like or a bent structure describes in a better way the configuration assumed by the membrane-bound anaesthetic [ 11. Prom the above-described possible configurations of tetracaine in membranes, it follows that the butylamino group could be an excellent probe for learning more about the topographical aspects of the interaction of tetracaine with bilayers. Surprisingly, very little attention has been paid to that group. One possible reason for the lack of studies focusing on the aromatic amine is its low pK [6]. Ionization of the but&mine is generally considered unimportant from the point of view of the pharmacolcgical effects of the drug, since it takes place far below the physiological pI1. A study of the behaviour of the aromatic amino group at low pH might contribute, however, to a better understanding of processes occurring under physiological conditions.
77
The ability of local anaesthetics to quench the fluorescence of probes located in the membrane can be used to assess the way that the anaesthetics interact with bilayers. Using this kind of methodology, Koblin et al. [15] found that procaine and tetracaine quench, to different extents, the fluorescence of N-octadecyl naphthyl-2-amine 6-sulphonic acid and 12-(9anthroyl) stearic acid in the presence of red cell ghosts. In contrast, lidocaine does not show any quenching capacity. Since both tetracaine and procaine molecules contain an aromatic amine whereas lidocaine does not, Koblin et al. attributed the quenching activity to such a group. The position of the aromatic amino group in the tetracaine molecule is ideal to assess the penetration of the anaesthetic int ; the hydrophobic region of membranes. In addition, the presence of such a group seems to be indispensable for the anaesthetic to behave as quencher through encounter with fluorescent probes [ 151. These properties of tetracaine were utilized in the present study, to determine the location of the anaesthetic in micelles based upon its ability to quench the fluorescence of perylene. The lipophilic dye was solubilized into the hydrophobic core of detergent micelles. The experiments were performed at several pH values to test whether quenching is dependent on the protonation degree of the aromatic amino group of tetracaine. For comparison, a similar study was made using a dioxane-water mixture in order to have both the anaesthetic and the fluorophore freely solubilized in a homogeneous, partially non-aqueous medium.
2. Materials
and methods
The surfactant Brij 35 (polyoxyethylene (23) dodecylether) was purchased from BDH Chemicals Ltd. Crystalline tetracaine hydrochloride ((2dimethylaminoethyl)-4-(n-butylamino) benzoate hydrochloride) was obtained from Sigma. Perylene (peri-dinaphthalene) and spectroscopic grade dioxane (Uvasol) were from Merck and Aldrich respectively. The non-ionic micelles were prepared by dissolving 1% Brij 35 in water. Since the average molecular weight reported for this detergent is 1100 [ 161, such a concentration is equivalent to 9 mM, well above the critical micellar concentration (9.1 X 10m5M [ 171). Perylene is insoluble in water but can be solubilized in micelles [ 21. Incorporation of perylene into micelles was performed by adding a small volume of concentrated solution of the probe in ethanol to the detergent solution, with continuous and vigorous magnetic stirring. The final concentration of perylene was 4 X lo* M. The ethanol added to the aqueous solution was less than 0.4%. Considering the surfactant concentration and the aggregation number of the micelles which has been reported to be 40 [ 171, it can be estimated that the ratio of the number of micelles to the number of solubilized perylene molecules was approximately 56OO:l. Partially non-aqueous solutions of perylene and tetracaine were prepared using mixtures of dioxane and water of known dielectric constant [6-181.
78
For the quenching experiments, the fluorescence of the micellar or partially non-aqueous solution of perylene was measured before and after addition of the appropriate amount of crystalline tetracaine. Perylene fluorescence was excited at 409 nm and read at 469 nm (uncorrected wavelengths). In this wavelength range, no absorption or emission of light by tetracaine was detected. Excitation and emission spectra as well as fluorescence measurements at constant wavelengths were obtained in an Aminco Bowman spectrophotofluorometer equipped with a Hewlett-Packard 7004 B recorder. Slit widths equivalent to 5.5 run bandpass were used throughout. Acid-base titration of tetracaine aromatic amine in dioxane-water mixtures was performed by measuring the pH dependence of the anaesthetic fluorescence (excitation wavelength, 302 run; emission wavelength, 360 run) as previously described [ 61. Titration was performed on 3 PM anaesthetic solutions to avoid the well-known self-quenching effect which appears at millimolar concentrations of fluorophore [ 19 1. All experiments were repeated at least three to five times. The results shown in Figs. 1, 3, 4 and 5 are representative of the corresponding sets of experiments. In the rest of figures, mean values + standard deviation, are plotted. Standard deviation bars smaller than symbols are not shown.
3. Results Figure 1 shows that at pH 5 the local anaesthetic tetracaine quenches the fluorescence of the lipophilic dye perylene incorporated into micelles of the non-ionic detergent Brij 35. The process follows the Stern-Volmer [20] equation, suggesting that quenching occurs through encounter of fluorophore and quencher:
FO F=
1 +&[&I
(1)
where F,, and F are the fluorescence readings in the absence and presence of quencher and Ko is the Stern-Volmer quenching constant. In this case, KQ should be considered as an apparent quenching constant since it is obtained from the bulk quencher concentration [Q] and not from the actual anaesthetic concentration in the micellar phase, which is unknown. Obviously this apparent quenching constant must include implicitly a partition coefficient [ 2 1,22 ] accounting for the distribution of the anaesthetic between the micellar and aqueous phases. It can be observed that the slopes of the straight lines in Fig. 1 decrease as the pH is lowered. A normalized plot of the apparent quenching constant as a function of pH (Fig. 2) shows that KQ decreases sigmoidally as the pH becomes more acidic, the curve showing an inflection point at pH 2.3. Quenching is completely reversed below pH 1. The pH titration of the quenching process seems to follow the ionization of the anaesthetic aromatic amino group for which a pK of 2.14 has been reported [ 61. It appears that the ability of the anaesthetic to quench the fluorescence
79
100
pH=5 PI+=3
c
80
8 =
60
s PI-l= 2
5 a
---__ 40 !
pH=l
1 0 TETRACAINE
I
I
I
0.02
0.04
0.06
HYDROCHLORIDE
(Ml
0
1
2
3
4
5
PH
Fig. 1. Stem-Volmer plot for the quenching of perylene fluorescence by tetracaine in 1% micellar solutions of the non-ionic detergent Bru 35 at different pH values. The ionic strength Z was kept constant at 0.25 with NaCl. Fig. 2. The apparent quenching constant Ko as a function of pH, for quenching of perylene fluorescence by tetracaine in 1% micellar solutions of Brfj 35 (Z=O.25). Quenching constants were obtained from the slopes of straight lines similar to those in Fig. 1. The data were normalized by assigning a value of 100 to the K, corresponding to the plateau above pH 4. The midpoint of the curve is found at pH 2.3.
of perylene incorporated into micelles depends on the absence of charge on its aromatic amine. Quenching of perylene fluorescence by tetracaine takes place without changes in the position of the excitation or emission spectra of the fluorophore, as can be observed in Fig. 3. Control experiments were also performed to determine whether the fluorescence of perylene in micelles is affected by lowering the pH. Figure 4 shows the excitation and fluorescence spectra of perylene micellar solutions of pH 5 and pH 0.5. No effect is detected as the spectra obtained at both pH values are coincident. The effect of pH on the quenching activity of tetracaine on micellar perylene can be observed in the spectra of Fig. 5. This figure shows that, at constant tetracaine concentration (0.05 M), the perylene fluorescence intensity increases as the pH is lowered. The results in Figs. 4 and 5 are better illustrated in Fig. 6 which shows that only in the presence of tetracaine is perylene fluorescence affected by changing the pH from 5 to 0.5. In the absence of anaesthetic, fluorescence remains constant, notwithstanding the pH change. For comparison with the micellar process, quenching was also studied in an homogeneous dioxane-water mixture (30 wt.% dioxane), which makes it possible to solubilize both the local anaesthetic and perylene, the lipophilic fluorophore. In this system, the quenching process also obeys the Stern-Volmer expression, i.e. plots of the quenching parameters according to the Stem-Volmer equation (eqn. (1)) give straight lines with intercepts equal
80
1 300
350
400
450
450
WAVELENGTH
500
550
600
(nm)
3. The effect of tetracaine concentration on the fluorescence spectra of perylene in micelles (1% Brij; 10 mM NaCl; pH 5.0) at different tetracaine concentrations. The molar concentration of tetracaine was varied from 0 to 0.05 as indicated by the arrows. Excitation spectra (left) were scanned with emission tixed at 469 run; emission spectra (right) were obtained setting the excitation at 409 run. Fig.
7-
3 f 2 a
6
b
-
I 300
350
400
I..
I
450
450
WAVELENGTH
pli=0.5
J 500 1 “m
550
600
1
Fig. 4. The effect of pH on the fluorescence spectra of perylene in micelles (1% Brij; 10 mM NaCl). Excitation spectra (left) were scanned with emission 6xed at 469 run; for the emission spectra (right) the excitation was set at 409 run.
81
300
350
500 450 (nm)
450 400 WAVELENGTH
550
600
Fig. 5. Fluorescence spectra of micehar perylene at different pH values in the presence of tetracaine (1% Brij; 10 mM NaCI; 0.05 M tetracaine hydrochloride). The spectra were obtained at the pH values indicated in the figure. Excitation spectra (left) were scanned with the emission fixed at 469 run; for the emission spectra (right) the excitation was at 409 run.
”
n
Fig. 6. Fluorescence intensity (excitation at 409 nm; emission at 469 run) as a function of pH, for micellar perylene in the presence (0) and absence (0) of tetracaine. The perylene was dissolved in 1% Brij and 10 mM NaCl. Fig. 7. The pH dependence of the Stem-Volmer constant I&, for quenching of perylene fluorescence by tetracaine in homogeneous dioxane-water solutions (30 wt.% dioxane). The normalized curve was obtained by assigning a value of 100 to the limiting value of K, reached above pH 2. The inflection point of the curve corresponds to pH 1.4.
82
to 1 (not shown but similar to those in Fig. 1). The pH dependence of the Stem-Volmer constant is presented in Fig. 7. It can be seen that, as in the case of quenching in micellar solutions, the quenching constant decreases with acidification of media. However, there is an important difference: even for pH values as low as 0.5, no complete reversal of quenching is obtained, i.e. the quenching constant does not become zero as in the case of mice&r solutions. What can be observed, instead, is that, below pH 1, it decreases to a constant value, equivalent to approximately 40% of the Ko at neutral pH. Since the partially non-aqueous medium can lower the pK of a cationic acid [ 6,181, it seemed necessary to investigate whether the complete ionization range of the aromatic amine was scanned in the study of the pH dependence of KQ shown in Fig. 7. Thus, acid-base titration of the aromatic amine was performed in different dioxane-water mixtures. Figure 8 shows the pK of the butylamino group in dioxane-water mixtures containing 0 wt.%, 20 wt.%, 45 wt.%, 70 wt.% and 82 wt.% dioxane corresponding to dielectric constants of 78.36, 60.79,38.48, 17.69 and 9.53 respectively [6, 181. It can be observed that, as the proportion of dioxane in the mixtures is increased, the pK diminishes. In solutions containing 30 wt.% dioxane, the interpolated pK value for the aromatic amine is 1.7, slightly above the pH value of 1.4 corresponding to the inflection point of the curve in Fig. 7. Thus it is safe to conclude that the study of the effect of acidification on KQ in the partially non-aqueous medium covers the pH range in which ionization of the aromatic amino group takes place. Figures 2 and 7 make the difference between the pH dependence of KQ in micelles and that in homogeneous, partially non-aqueous solutions apparent. The lack of complete reversal of quenching is not the only characteristic differentiating the behaviour of the tetracaine-perylene pair in dioxane-water with respect to that in micelles. Figure 9 shows that for the micellar solution % WATER 1
100 r’
0
SO
60
I
0
20 %
40 60 DIOXANE
I
80
I
100
SODIUM
Fig. 8. The pK characterizing the acid-base dissociation of tetracaine solutions containing different proportions of dioxane and water.
CHLORIDE aromatic
(M ) amine
in
Fig. 9. The effect of salt concentration on the apparent quenching constant KQ for quenching of perylene fluorescence by tetracaine at pH 5: - --O- -, dioxane(30 wt.%)-water mixture; ,-Z-, 1% micellar solution of Brij 35.
83
a change in NaCl concentration from 0.01 to 0.3 M induces an increase in the apparent quenching constant K,, from 30.6 to 82.3 M-l. In contrast, K, for quenching in a dioxane-water mixture is slightly decreased by addition of salt.
4. Discussion In the present study, fluorescence quenching is used to investigate the adsorption of a local anaesthetic onto micelles. A description is presented of the ability of tetracaine to quench the fluorescence of the lipophilic dye perylene in two different media; a simple model membrane system consisting of micelles of a neutral amphiphile, into the hydrophobic region of which the fluorophore can be incorporated, and a homogeneous dioxane-water mixture where both the fluorophore and the quencher can be freely solubilized. The results presented here are consistent with quenching of perylene by tetracaine taking place by collision, since the Stem-Volmer expression is obeyed both in homogeneous solution and in micelles. Previous studies from other laboratories showed that amines can quench the fluorescence of aromatic hydrocarbons by transfer of an electron to the fluorophore in the excited state [23-251. Although such an explanation could be extended, by way of inference, to the extinction process studied in the present work, a more direct proof that tetracaine quenches perylene fluorescence through electron transfer was needed. Our working hypothesis was that, if quenching occurred by the abovementioned process, it would be reversed by protonation of the aromatic amine, since the positive charge would make that group unable to donate an electron to the fluorophore. Our results show that, in fact, part of the quenching in a partially non-aqueous medium is abolished by protonation of tetracaine aromatic amine. However, the fully protonated anaesthetic molecule is still capable of exhibiting a & equivalent to 40% of the value shown at neutral pH (Fig. 7), indicating that there should be an additional mechanism for the extinction of perylene fluorescence. At the moment, we do not have data supporting any particular mechanism for such remaining quenching activity. What we do know is that it is resistant to extremely acidic media. As for the complete reversal of quenching in the micellar system at low pH (Figs. 1 and 2), it is explained in terms of the two effects that protonation of the aromatic butylamino group can bring about: on the one hand, it decreases the quenching capacity of the anaesthetic molecules colliding with perylene through inhibition of the electron transfer, as already discussed; on the other hand, it makes tetracaine unable to penetrate into the mice&r region where the fluorophore is solubilized. Presumably, the positive charge makes the insertion of the protonated butylamino moiety into the micellar core thermodynamically unfavourable. Partition of tetracaine into the micellar phase appears to be very important for the apparent quenching activity of
84
the anaesthetic, as shown by the results in Fig. 9. Addition of NaCl to neutral micelles markedly increases the apparent quenching constant. In contrast, Ko for quenching in dioxane-water is almost unaffected by salt concentration. These results can be interpreted ln terms of the possible enhancement by the added salt of the ability of tetracaine to establish hydrophobic interactions. The salting-out of the anaesthetic from the aqueous phase would favour its partition into micelles, causing an increase in the apparent quenching constant. Other alternative explanations for the effect of salt on Ko based on a change in the structure of the Brij micelles or on an increased micellization of the detergent can be ruled out. Firstly, non-ionic micelles can hardly be affected by neutral salts [ 16, 171; secondly, even at low salt concentrations, the number of micelles is overwhelmingly large with respect to the number of perylene molecules (5600: 1) [ 171. An important part of this work is based on the reversal of the anaesthetic quenching activity by acidification of media to pH values near or below 1. We have not found any change attributable to perturbations, either of perylene or of the Brij micelle structure, by such an extreme acidic condition. The experiments of Figs. 4 and 6 clearly show that the fluorescence of perylene in the neutral micelles is not modified by acidification of the medium. Only in the presence of tetracaine does perylene fluorescence show a dependence on pH (Figs. 5 and 6). Since such a dependence correlates well with ionization of the anaesthetic aromatic amine [6] (Figs. 2 and S), our results can be explained in terms of the protonation-deprotonation equilibrium of that group. The information obtained on the participation of a functional group of the anaesthetic molecule in a fluorescence quenching process allows us to characterize the mode of insertion of tetracaine in model membrane systems. Our results show that at neutral pH the anaesthetic butyl amine can reach the micelle non-polar core where perylene is located and quench its fluorescence by collision. This is consistent with models suggesting that membranebound tetracaine assumes a rod-like configuration parallel to the surface normal with the aromatic butylamino group located in a highly hydrophobic region [l, 5-101. The nerve-blocking properties of local anaesthetics depend on their capacity to induce perturbations in membranes [ 261. A detailed knowledge of the con&uration of these compounds at the lipid-water interface will favour a better understanding of drug-membrane interactions and, as a consequence, of the molecular basis of their pharmacological effects.
Acknowledgements The partial financial support of Consejo National de Ciencia y Tecnologia de Mexico and Consejo de1 Sistema National de Education Tecnoldglca through research grants to M.S.F. is gratefully acknowledged. Dr. Fern&ndez is a member of the National System of Investigators, Mexico.
85
References 1 Y. Kuroda and Y. Fujiwara, Locations and dynamical perturbations for lipids of cationlc forms of procaine, tetracaine, and dibucaine in small unilameilar phosphatidylchohne vesicles as studied by nuclear Overhauser effects in (l)H nuclear magnetic resonance spectroscopy, Biochim. Biophys. Acta, 903 (1987) 395-410. M. S. Femandez, Formation of miceIles and membrane action of the local anaesthetic tetracaine hydrochloride, Biochim. Biophys. Acta, 597 (1980) 83-91. M. S. Femandez, Disruption of Iiposomes by tetracaine miceIles, B&him. Biophys. Acta, 646 (1981) 27-30. W. A. Frezzatti, Jr., W. R. Toseili and S. Schreier, Spin label study of local anaesthetic--lipid membrane interactions. Phase separation of the uncharged form and bilayer miceilization by the charged form of tetracaine, B&him. Biophys. Acta, 860 (1986) 531-538. 5 M. S. Femandez and J. Cerb6n, The importance of the hydrophobic interactions of local anaesthetics in the displacement of polyvaient cations from artiiicial lipid membranes, Biochim. Biophys. Acta, 298 (1973) 8-14. 6 J. Garcia-Soto and M. S. Fem&ndez, The effect of neutral and charged miceiies on the acid-base dissociation of the local anaesthetic tetracaine, B&him. Biophys. Acta, 731 (1983) 275-281. 7 H. Hauser, S. A. Penkett and D. Chapman, Nuclear magnetic resonance spectroscopic studies of procaine hydrochloride and tetracaine hydrocNoride at lipid-water interfaces, B&him Biophys. Acta, I83 (1969) 466-475. 8 J. Cerb6n, NMR evidence for the hydrophobic interaction of local anaesthetics. Possible relation to their potency, B&him. Biophys. Acta, 290 (1972) 51-57. 9 M. S. Femandez and J. Cerb6n, PMR and viscosity studies of the interaction of local anaesthetics and micelles, Arch. Biochem. Biophys., 172 (1976) 721-725. 10 Y. Boulanger, S. Schreier and I. C. P. Smith, Molecular details of anaesthetic--lipid interaction as seen by deuterium and phosphorus-31 nuclear magnetic resonance, Biochemistry, 20 (1981) 6824-6830. 11 Y. Boulanger, S. Schreier, L. C. Leitch and I. C. P. Smith, Multiple binding sites for local anaesthetics in membranes: characterization of the sites and their equilibria by deuterium NMR of specifically deuterated procaine and tetracaine, Can. J. Biochem., 58 (1980) 986-995. 12 J. Westman, Y. Boulanger, A. Ehrenberg and I. C. P. Smith, Charge and pH dependent drug binding to model membranes. A (2)HNMR and light absorption study, B&him. Biophys. Actu, 685 (1982) 315328. interaction: a (2) H nuclear 13 E. C. Kelusky and I. C. P. Smith, Anaesthetic-membrane magnetic resonance study of the binding of specifically deuterated tetracaine and procaine to phosphatidylchoiine, Can. J. Biochem. Cell Biol., 62 (1984) 178-184. 14 M. B. Feinstein, Reaction of local anaesthetics with phospholipids. A possible chemical basis for anaesthesia, .I. Gen. Physiol., 48 (1964) 357-374. 15 D. D. KobIin, W. D. Pace and H. H. Wang, The penetration of local anaesthetics Into the red blood cell membrane as studied by fluorescence quenching, Arch. Btichem. Biophys., 171 (1975) 176-182. 16 K. Shinoda, T. Nakagawa, B. I. Tamamushi and T. Isemura, Colloidal Su@zctants, Academic Press, New York, 1963. in Micellar and Macromokculur Systems, 17 J. H. Fendler and E. J. FendIer, Cat&&s Academic Press, New York, 1975. 18 M. S. Femandez and P. Fromherz, Lipoid pH indicators as probes of electrical potential and polarity in miceiles, J. Phys. Chem., 81 (1977) 1755-1761. 19 G. G. Guilbault, Practical Fluorescence: Theory, Methods and Techniques, Marcel Dekker, New York, 1973. 20 0. Stem and M. Voimer, The extinction period of fluorescence, Phys. Z., 20 l1919) 183-188.
86 21 G. Papageorgiou and C. Argoudelis, Cation-dependent quenching of the fluorescence of chlorophyll a in viva by nitroaromatic compounds, Arch. Biochem. Biophys., I56 (1973) 134-142. 22 S. G. Bertolotti, M. V. Bohorquez, J. J. Cosa, N. A. Garcia and C. M. Previtali, Mice&r effect on the fluorescence quenching of indolic compounds by aminoacids, Photo&em. Photobid., 46 (1987) 331-335. 23 H. Leonhardt and A. Weller, Elektronentibertragungsreaktionen des angeregten Perylens, Ber. Bunsenges. Phgs. Chem. 67 (1963) 791-795. 24 N. Mataga, K. Ezumi and T. Okada, Temperature effects on charge transfer fluorescence spectra and mechanisms of charge transfer interactions in the excited electronic state, Mol. Phys., 10 (1966) 201-202. 25 N. Mataga, T. Okada and K. Ezumi, Fluorescence of pyrene-iV,Ar-dimethyhmilinecomplex in non-polar solvent, Mol. Phys., 10 (1966) 203-204. 26 J. M. Ritchie and N. M. Greene, Local anaesthetics, in L. S. Goodman and A. Gihnan (eds.), The Phurrna.cological Basis of Th.erapeuks, MacMiian, New York, 6th edn., 1980, pp. 302321.
