Interaction of papaverine with micelles of surfactants with different charge studied by 1H-NMR

74

Biochimica et Biophysica Acta, 1189 (1994) 74-80

© 1994 Elsevier Science B.V. All rights reserved 0005-2736/94/$07.00

BBAMEM 76156

Interaction of papaverine with micelles of surfactants with different charge studied by H-NMR Victor E. Y u s h m a n o v b, Janice R. Perussi a, H i d e t a k e I m a s a t o a and Marcel T a b a k a.. " Institute of Physics and Chemistry of S~o Carlos, University ofS~o Paulo, Cx.P. 369, 13560-970 S~o Carlos, SP (Brazil) and b Institute of Chemical Physics, Russian Academy of Sciences, 117977 Moscow (Russia) (Received 23 February 1993) (Revised manuscript received 2 September 1993)

Key words: Papaverine; Vasodilator drug; Micelle; Interaction; Surfactant; NMR

The interaction of the vasodilator drug papaverine (PAV) with micelles of surfactants with different charge of headgroups as well as the properties of PAV in D20 solution were studied by ~H-NMR. At pD values above 6.4 deprotonated PAV molecules tend to precipitate, the signals of the heterocycle protons of solubilized PAV molecules being shifted to high field. At PAV concentration above 1 mM its protons experience upfield shifts which increase with pD value and are due to the stacking of aromatic rings. Incorporation into micelles caused shifts of all resonances. This effect is due to changes in the local chemical environment of PAV rather than to stacking, and, possibly, involves the deprotonation of the N atom of PAV heterocycle. Line broadening of PAV protons at the molar ratio surfactant/PAV > 16 indicated their restricted mobility. Different complexes were formed due to interaction between the heterocycle of PAV and polar headgroups of cationic cetyltrimethylammonium chloride (CTAC) or anionic sodium dodecylsulfate (SDS). The binding of PAV to zwitterionic N-hexadecyl-N,N-dimethyl-3-ammonio-l-propanesulfonate (HPS) is similar to that of PAV to CTAC. Association constants were estimated from NMR data as 20, 60 and 350 M -1 at pD = 4.9 + 0.1 for HPS, CTAC and SDS, respectively. Thus, the mode of binding of PAV to HPS is defined by the cationic dimethylammonium headgroup fragment, whereas the negative sulfate fragment attenuates the effective charge of HPS headgroup.

Introduction Papaverine, 6,7-dimethoxy- 1-veratryl-isoquinoline (PAV, Fig. 1), possesses a wide range of biological effects, being commonly known as a cerebral vasodilator and smooth muscle relaxant [1]. It is also used in clinics as a cardiac vasodilator, and its direct action upon coronary blood vessels with a reduction in muscular tonus was associated with the response to stimulation of/3-adrenergic receptors [2,3]. It has been classified more recently as a nonspecific vasodilator [4]. Biological effects imply interactions of PAV with cellular components (membranes, biopolymers). In fact, it has been recently reported that PAV affects the hemoglobin oxygen affinity both in isolated protein and intact red blood cells [5,6]. Some of the PAV effects immediately involve cellular membranes. In particular, PAV alters a transmembrane potential and its temper-

ature dependence, compensatory flow and m e m b r a n e conductivity in epithelial and nerve cells in a pH-dependent manner [7,8]. PAV affects ion transport stimulating Na,K-ATPase of nerve m e m b r a n e [8], and enhances intracellular Ca 2+ accumulation in different tissues, thus modifying the pharmacological action of other drugs [9]. By changing ion transport, PAV influences Ca2+-dependent action potentials in myocardium depolarized by potassium [10]. In addition to changes in transmembrane ion transport, PAV inhibits the transport of adenine [11], cyclic A M P [12,13] and insulin [12] across the cellular membranes in different tissues.

,

H3CO~,~ * Corresponding author. Fax: +55 162 749163 or+55 162 722218. Abbreviations: CTAC, cetyltrimethylammonium chloride; HPS, NhexadecyI-N,N-dimethyl-3-ammonio-l-propanesulfonate; PAV, papaverine; SDS, sodium dodecyl sulfate. SSDI 0005-2736(93)E0264-B

H3CO~

5'

2'

,

OCH3 OCH3

3

Fig. 1. Chemical structure of papaverine.

75 Interaction of drugs with membrane lipids defines and modulates other interactions and drug metabolism in organism. However, underlying physico-chemical mechanisms of such interactions quite rarely can be derived immediately from experiments on natural membranes and whole cells. Therefore, various membrane models are in widespread use. Micellar systems, while being the most simple models, provide useful structural and functional models for more complex bioaggregates [14-16]. Understanding of molecular mechanisms of drug-membrane interaction, besides its theoretical importance, has a practical significance for the design of novel drugs and for the selection of the most relevant drugs for various treatment schemes. This study was initiated in order to gain more insight into both the properties of PAV in aqueous solution and the characteristics of its binding to micellar systems formed by surfactants with different charge of headgroups. Micelles of zwitterionic N-hexadecylN,N-dimethyl-3-ammonio- 1-propanesulfonate (HPS) were chosen as a model of lipid aggregates, while cationic cetyltrimethylammonium chloride (CTAC) and anionic sodium dodecyl sulfate (SDS) may be considered as HPS fragments, suitable for studying the interaction of PAV with different parts of the polar HPS headgroup separately. The HPS, CTAC and SDS micelles are commonly accepted model systems for studying various aspects of membrane interactions [17-20]. Proton NMR allowed us to discriminate between different atoms in the PAV molecule during a dynamic equilibrium process in solution. The results could help to elucidate the molecular details of PAV behavior in vivo. Materials and Methods

Reagents and equipment. PAV hydrochloride (Sigma), HPS (Sigma), SDS (Bio-Rad), sodium acetate (Quimis-Mallinckrodt), D 20 (AnalytiCals), CDC13 (Aldrich) and Hexane (Chimie-Test) were used as purchased. CTAC (Herga) was purified by acetonemethanol extraction. H 2 0 whenever used was distilled and deionized. Aqueous solutions were generally prepared in a 0.02 M acetate buffer. The pD values were measured and adjusted using a Corning Model 130 pH-meter by adding 0.4 to the pH-meter readings [21]. Proton NMR spectra were run on a Bruker AC-200 (resonant frequency 200.13 MHz) at ambient probe temperature (23°C). Residual water suppression in proton spectra was achieved using a presaturation by homonuclear gated decoupling during the delay of 2 s between scans. The recycle time was 5.7 s, the band width was 2 kHz, the pulse length was 2 /xs. The free induction decay signals were digitized using 16 K data points and processed by an exponential filter with a

line broadening factor of 0.2 Hz prior to Fourier transformation. The IR spectra were measured using KBr pellets on an FTIR spectrometer Bomem MB-102, Calculations of binding parameters from N M R data. NMR measurements were performed keeping the concentration of PAV constant and varying the concentration of surfactant in the range of 1 to 220 surfactant molecules per PAV molecule. Assuming the complex formation between PAV and surfactant to be a second-order process and that the equilibrium obeys the law of mass action, and taking into account that chemical exchange is fast in the NMR timescale (only single resonance was observed for each PAV proton in all cases), we can write an equation (6 - 6 f ) / ( 6 b - 6e) = K A S f / ( 1 + KASf) where 3, 6f and 6 b are the observed chemical shifts of a particular PAV resonance and its values for free and bound PAV, respectively; K A is an association constant for the complex, and Sf is the concentration of the non-bound surfactant. In conditions of a large excess of surfactant Sf = So (S o is the total concentration of the surfactant), and we can determine K A and 6 b in any linearizing frame (e.g., Scatchard or HughesKlotz representations) [22,23]. Results and Discussion

PA V in aqueous solution The literature data on PAV spectra in aqueous solution are limited and refer to concentrated solution [24]. The spectra of 1 mM and 10 mM PAV solutions in D20 are shown in Fig. 2A, B. Spectral assignments were made based on 6 values and their pD dependences, and confirmed by double resonance experiments. The acidic N+H proton is not detectable due to the exchange with water. The methylene resonance was observed only if PAV concentration and pD value were relatively high, otherwise it was overlapped by the residual water signal. The same spectra were obtained in H 2 0 as solvent. The pD dependence of the chemical shift of resonances for PAV in 1 mM solution is presented in Fig. 3. The only appreciable chemical shift alterations at high pD values were observed for the protons of the PAV heterocycle which contains the acidic N+H group. The pK a value of this group was estimated as 6.4 by optical spectroscopy (Tabak, M. and Perussi, J.R., unpublished data). At pD values near the pK a of PAV, a partial precipitation of PAV occurs. The precipitate was characterized by the disappearance of N+H resonance and the shifts of other signals in CDC13 (Fig. 4). IR spectra of precipitated PAV also reveal some peculiarities, the disappearance of an intense line at 2503 cm-~ (gener-

76

METH0X 1

2i

3

4

p

5

q

--~--~' ,. 6'

8

3

,

FI

FI

/

5116

,

I

,

8,4

I

,

0.2

I

,

8.0

I

,

7.8

I 7.6

A

I

PPR

7.4

,

I

~

7.2

I 7.0

METHYL ME~ Z METHOX ?~Y ]/

2_'3 6'

8.

~

/ i

/

I

I 6.8

,

/

,l 8.6

~

/

,

1

4.2

,

I

L

4.0

PPM

I 3.1

,

I

a

3.8

Fig. 2. 1H-NMR spectra of papaverine in D 2 0 containing 20 mM acetate buffer: (A) 1 mM, pD 4.31; (B) 10 mM, pD 5.60; (C) 0.5 mM in the presence of 64 mM HPS, pD 4.82. The peak numbers for ring protons are given according to Fig. 1. The methylene resonance in charts A and C is overlapped by the residual water signal. The high field part of chart C is distorted by one of the HPS signals,

ally identified as due to amine salts [25]) being the most remarkable. Hence, deprotonated PAV molecules tend to precipitate at high pD values. The data of Fig. 3 at pD = 7.12 are due to the PAV molecules which

~. 8.0

remained solubilized, and can give an idea on the spectrum of the deprotonated PAV in D20. In Fig. 5 the concentration dependence of PAV chemical shifts is presented. At high concentration of PAV in D 2 0 its protons experience upfield shifts, mainly the protons of the heterocycle. This behavior is known to be characteristic for the stacking of aromatic rings [26,27]. No concentration-dependent changes in peak positions were found using CDC! 3 as solvent. Stacking ability of nitrogen heterocycles decreases if nitrogen atom is protonated, i.e., at low pH [27,28]. Accordingly, the higher the pD value, the lower the 6 values of PAV protons in 10 mM solution (Fig. 6).

~- 7.5 ¢-

Interaction of PA V with miceUes of surfactants v

7.0

0

E ,o

~ 4.0

~.-_o__-.o._.~ i

~

iiiiil|lllllllllJlllllllll

5

Jl i

6

7 pD

Fig. 3. Proton chemical shifts of 1 mM papaverine in D20 containing 20 mM acetate buffer at different pD values.

As follows from Fig. 5, the aggregation of PAV was negligible at 1 mM concentration, which still ensured good signal-to-noise ratio in NMR measurements. At concentration above 1 mM stacking introduced a marked effect to the chemical shift measurements. Therefore, the PAV concentration range of 0.5 to 1 mM was chosen for assessing its interaction with micelles. Incorporation into HPS micelles at pD = 4.9 +0.1 caused shifts of all PAV resonances (Figs. 2C and 7). However, the shift patterns were different from those presented in Figs. 3, 5 and 6. Thus, this effect may be ascribed to changes in the local chemical environment

77

CHCI3, 8

2' METHOXY

5

METHYL

q

2

CHCI3

METHOXY

2' ,6' METHYL l

;

3

I

. . . .

l

17.5 PPH

,

,

+.. . . . .

,.,"

/

/

L/b

5

I .... 8.0

I .... 7.5

V

I .... 7.0

l , , , L I , , , , I $.5

6.0 PPM

.

.

.

.

5.5

I .... 5.0

I .... 4.5

B

I .... 4.0

Fig. 4. I H - N M R s p e c t r a of p a p a v e r i n e in CDCI3: (A) hydrochloride; (B) a f t e r p r e c i p i t a t i o n from D 2 O s o l u t i o n at p D = 12.7. T h e low field p a r t s in e x p a n d e d vertical scale are given in the insets.

of PAV rather than to stacking, and, possibly, involves the deprotonation of nitrogen atom of PAV heterocycle.

8.0

This result naturally put a question as to which part of the HPS headgroup is responsible for the PAV binding. The sulfate fragment of HPS is closer to the micelle-water interface, therefore, the interaction be-

E ¢~ 8.0 n

7.5

~

.~

~:

"-

7.5

1-

o u

7.0 o

E

7.0

~

r"

4.0

e tO

O.

1

10

PAV c o n c e n t r a t i o n ,

100 mM

Fig. 5. P r o t o n c h e m i c a l shifts of p a p a v e r i n e at d i f f e r e n t c o n c e n t r a tions in D 2 0 c o n t a i n i n g 20 m M a c e t a t e buffer at p D 4.80 to 4.97.

4.0

o

,,,.I

3

I

I

4

5

pD

Fig. 6. P r o t o n c h e m i c a l shifts of 10 m M p a p a v e r i n e in D 2 0 contain. ing 20 m M a c e t a t e buffer at d i f f e r e n t p D values.

78

E

~. 8.0

o. 8 . 0 Q.

7.5

7.5 ¢-

t,0}

o

D

7.0

o 7.0 ~ ~ : ~ . ~ E

E tO

c-

c~ 4.0

4.0

1

10

1

100

HPS/PAV, mol/mol Fig. 7. Proton chemical shifts of 0.5 mM papaverine in D20 containing 20 mM acetate buffer and variable amount of HPS at pD = 4.9_+0.1.

t w e e n P A V a n d S D S was studied. A t e q u i m o l a r conc e n t r a t i o n s P A V a n d S D S f o r m e d a n o n s o l u b l e complex. T h e a d d i t i o n o f S D S to S D S / P A V = 2 / 1 led to r e s o l u b i l i z a t i o n o f P A V in t h e form of n o n m i c e l l a r structure, since t h e c o n c e n t r a t i o n o f S D S was b e l o w its critical m i c e l l a r c o n c e n t r a t i o n . This s t r u c t u r e h a d p r o ton c h e m i c a l shifts a p p a r e n t l y d i f f e r e n t from those o f free P A V in s o l u t i o n (Fig. 8). F o l l o w i n g f u r t h e r inc r e a s e in S D S c o n c e n t r a t i o n a new type of c o m p l e x was f o r m e d a n d d o m i n a t e d at high S D S / P A V r a t i o (Fig. 8). T h e c h e m i c a l shift p a t t e r n o f this c o m p l e x has striking d i f f e r e n c e s from t h a t o f HPS. T h e C T A C t i t r a t i o n profiles of c h e m i c a l shifts of P A V r e s o n a n c e s a r e d i f f e r e n t from t h o s e for SDS titration a n d qualitatively similar to that for H P S titration (Fig. 9). T h e i n t e r a c t i o n o f P A V with C T A C a n d H P S is not m e r e l y a h y d r o p h o b i c i n t e r a c t i o n with acyl h y d r o c a r b o n chains, since P A V is not soluble in satur a t e d h y d r o c a r b o n s , e.g., hexane. A n a p p r e c i a b l e line b r o a d e n i n g o f P A V signals occ u r r e d w h e n t h e s u r f a c t a n t / P A V m o l a r r a t i o was equal o r e x c e e d e d 16 (Fig. 2C). This effect was m o r e p r o n o u n c e d for H P S a n d C T A C micelles a n d i n d i c a t e d a r e s t r i c t e d mobility o f P A V in micelles. T h e effects o f s u f f a c t a n t s on the c h e m i c a l shifts o f P A V p r o t o n s s e e m e d to b e less significant in the case o f H P S micelles, o t h e r c o n d i t i o n s b e i n g equal. T h e o b t a i n e d d a t a o n c o n c e n t r a t i o n t i t r a t i o n e n a b l e d the e s t i m a t i o n o f b o t h the association c o n s t a n t K A o f the

10

100

SDS/PAV, m o l / m o l Fig. 8. Proton chemical shifts of 1 mM papaverine in D20 containing 20 mM acetate buffer and variable amount of SDS at pD = 4.9_+0.1. One of methoxy resonances overlapped with the methylene triplet of SDS at 4.04 ppm and was not distinguished at high concentration of SDS.

E o. Q.

8.0

7.5 AZ

oo 7.0

E .gE

o

4.0

~,ll,I

i

1

, i,,111|

10

i

i ,i,,111

i

100

CTAC/PAV, mol/mol Fig. 9. Proton chemical shifts of 1 mM papaverine in D20 containing 20 mM acetate buffer and variable amount of CTAC at pD = 4.9_+0.1.

79 TABLE I

Proton chemical shifts (ppm) of papaverine bound to micelles of surfactants in D20 at pD 4.9 + 0.1 HPS, N-hexadecyl-N,N-dimethyl-3-ammonio-l-propanesulfonate; CTAC, cetyltrimethylammonium chloride; SDS, sodium dodecylsulfate. See Fig. 1 for numbering of ring protons. Surfactant

Proton chemical shift (ppm) Ring protons

HPS CTAC SDS

Methoxyprotons

3

4

5

8

2'

5'

6'

8.35 8.31 8.34

7.79 7.77 8.22

7.58 7.54 7.73

7.43 7.41 7.69

7.13 7.12 7.18

6.81 6.82 6.88

6.81 6.82 6.85

complex and 6 b values for each of the PAV signals. The values of 6b are indeed characteristic of PAV in the micelle-bound state. Titration curves for different resonances gave approximately the same result for KA: about 20, 60 and 350 M-1 for HPS, CTAC and SDS, respectively, at pD = 4.9 _+0.1. The values of 6 b given in Table I provide evidence that PAV forms different complexes with SDS and CTAC. The main differences occurred in the t~b values of the protons of the PAV heterocycle, which apparently interacts in different ways with the cationic and anionic polar headgroups of surfactants. On the contrary, the complexes of PAV with HPS and trimethylammonium fragment of CTAC are similar, indicating that the positively charged dimethylammonium headgroup fragment of HPS defines the mode of binding of PAV to HPS. The sulfate anionic fragment attenuates the effective charge of HPS headgroup, leading to the less significant effects for HPS. It was shown recently that this charge attenuation may result in more weak solvent structure perturbation by HPS as compared with CTAC, due to weaker overall hydration and electrostatic field effects [20]. This was explained by the significant intra- and intermolecular electrostatic interaction between the surfactant head-ions [19]. In conclusion, we have demonstrated that PAV does interact with micelles formed by amphiphilic molecules with different headgroup charge. The aromatic heterocycle of PAV participates in this interaction in different ways dependent on the headgroup charge. The mode of binding of PAV to HPS is defined by the cationic dimethylammonium headgroup fragment, whereas the negative sulfate fragment attenuates the effective charge of HPS headgroup.

Acknowledgements The authors appreciate the financial support of CNPq and FINEP and a visiting grant to V.E.Y. from FAPESP. The authors are indebted to Dr. P. Berci Jr. for the purification of CTAC and to M.R. Fernandes for FTIR measurements.

4.01 4.02 4.15

3.95 3.96 4.06

3.81 3.81 3.88

3.78 3.79 3.83

References 1 Budavari, S. (ed.) (1989) The Merck Index. An Encyclopedia of Chemicals, Drugs, and Biologicals, l l t h Edn., p. 1111, Merck, Rahway, NJ. 2 Pap, J.G. and Fischer, E. (1967) Eur. J. Pharmacol. 2, 1. 3 Charlier, R. (1971) Antianginal Drugs. Handbook of Experimental Pharmacology, Vol. 31, Springer, Berlin. 4 Nikolov, R. and Leniger-Follert, E. (1978) Naunyn-Schmiedeberg's Arch. Pharmacol. 305, 149-154. 5 Meirelles, N.C., Nascimento, O.R. and Tabak, M. (1985) Hemoglobin 9, 399-403. 6 De Paula, E. and Meirelles, N.C. (1992) Braz. J. Med. Biol. Res. 25, 557-565. 7 Simon, L., Porszasz, J., Cibiszerkatalin, P. and Talpas, S.G. (1977) Pharmazie 32, 235-239. 8 Woods, W.T. and Lieberman, E.M. (1976) Neuroscience 1, 383390. 9 Godfraind, T, (1981) Fed. Proc. 40, 2866-2871. 10 Saguineni, M.C. and West, T.C. (1981) J. Pharmacol. Exp. Ther. 219, 715-722. 11 Templetom, B.A. and Chilson, O.P. (1981) J. Biol. Chem. 256, 285-290. 12 Campbell, I.L. and Taylor, K.W. (1981) Biochim. Biophys. Acta 677, 357-364, 13 Rindler, M.J., Bachor, M.M., Spitzer, N, and Saier, M.H., Jr. (1978) J. Biol. Chem. 253, 5431-5436. 14 Cordes, E.H. (1973) Reaction Kinetics in Micelles, Plenum Press, New York. 15 Tanford, C. (1973) The Hydrophobic Effect: Formation of Micelles and Biological Membranes, Wiley-Interscience, New York. 16 Minal, K.L. and Lindman, B. (eds.) (1984) Surfactants in Solution, Plenum Press, New York. 17 Herrmann, K.W. (1966) J. Colloid Interface Sci. 22, 352-359. 18 Reynolds, J.A. (1982) in Lipid-Protein Interactions (Jost, P.C. and Griffith, O.H., eds.), Vol. 2, pp. 193-224, Wiley and Sons, New York. 19 Miiller, S.C. and Ponel, R. (1982) in Solution Behavior of Surfactants. Theoretical and Applied Aspects (Mittal, K.L. and Fendler, E.J., eds.), Vol. 1, pp. 485-504, Plenum Press, New York. 20 El Seoud, M.I., Farah, J.P.S. and El Seoud, O.A. (1989) Ber. Bunsenges. Phys. Chem. 93, 180-183. 21 Glasoe, P.K. and Long, F.A. (1960) J. Phys. Chem. 64, 188-190. 22 Wurmser, R. (1973) in Experimental Methods in Biophysical Chemistry (Nicolau, C., ed.), pp. 613-647, Wiley and Sons, London. 23 Dwek, R.A. (1973) Nuclear Magnetic Resonance (NMR) in Biochemistry. Applications to Enzyme Systems, Clarendon, Oxford. 24 Pouchert, C.J. (1983) The Aldrich Library of NMR Spectra, 2nd Edn., Vol. 2, p. 755, Aldrich Chemical, Milwaukee.

80 25 Silverstein, R.M., Bassler, G.C. and Morrill, T.C. (1991) Spectrometric Identification of Organic Compounds, 5th Edn., p. 160, Wiley and Sons, New York. 26 Cohen, J.S. (1969) Nature 223, 43-46.

27 Granot, J. and Fiat, D. (1977) J. Am. Chem. Soc. 99, 4963-4968. 28 Antonovsky, V.L., Gukovskaya, A.S., Nekrasova, G.V,, Sukhorukov, B.I. and Tchervin, I.I. (1973) Biochim. Biophys. Acta 331, 9-20.