- Email: [email protected]
Protolytic Equilibria of Sartans in Micellar Solutions of Differently Charged Surfactants
Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Contents lists available at ScienceDirect
Journal of Pharmaceutical Sciences journal homepage: www.jpharmsci.org
Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Protolytic Equilibria of Sartans in Micellar Solutions of Differently Charged Surfactants Maja Gruji c 1, Marija Popovi c 1, Gordana Popovi c 2, *, Katarina Nikolic 1, Danica Agbaba 1 1 2
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Belgrade, Belgrade, Serbia Department of General and Inorganic Chemistry, Faculty of Pharmacy, University of Belgrade, Belgrade, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 March 2016 Revised 20 May 2016 Accepted 7 June 2016 Available online 13 July 2016
Protolytic equilibria of irbesartan, losartan, and valsartan have been investigated in the presence and absence of differently charged anionic (sodium dodecyl sulfate), cationic (cetyltrimethylammonium bromide), and nonionic (4-octylphenol polyethoxylate and polyoxyethylene (23) lauryl ether) surfactants. Ionization constants were determined by potentiometric titration at a constant ionic strength (0.1 M NaCl) and temperature 25 C. The effect of surfactants was estimated, based on a shift in apparent w ionization constants (pKapp a ) determined in micellar solutions against the pKa values in water. The values of sartans (up to 1.72 pK units), while the anionic surfactant caused an increase in the pKapp a values (up to 1.44 pK units). cationic surfactant had an opposite effect and caused a reduction in pKapp a These results point out to the fact that the ionizable groups of sartans are involved in electrostatic invalues in the presence of teractions with the charged surface of the ionic micelles. Shift in the pKapp a nonionic surfactants (from 0.86 to þ1.30) is a consequence of the interactions of drugs with the hydrophilic palisade layer. Significant changes in the distribution profiles of the equilibrium forms (from 44% to þ80%) are observed at the biopharmaceutically important pH 4.5 value and can be considered in terms of the potential influence on intestinal absorption and bioavailability. © 2016 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Keywords: physicochemical properties acid-base equilibria surfactants micelle biomimetics drug interaction
Introduction Angiotensin II receptor blockers also known as sartans represent an important class of drugs that act as competitive and selective antagonists of angiotensin II at the angiotensin AT1 receptors.1 Angiotensin II is a potent vasoconstrictor and the primary vasoactive peptide of the renin-angiotensin system, which plays an important role in the pathology of many cardiovascular diseases. Sartans reduce pressor effect of angiotensin II and cause the pharmacological effect of lowering blood pressure. Therefore these drugs are used in the treatment of hypertension as well as in the treatment of cardiac insufficiency, myocardial infarction, and diabetic nephropathy.2 Accurate mechanism of interaction of sartans with AT1 receptor is still not completely resolved. It is considered that the interaction takes place in a 2-step process, which includes spontaneously inserting into the membrane, and then lateral
Abbreviations used: Brij 35, polyoxyethylene (23) lauryl ether; CTAB, cetyltrimethylammonium bromide; SDS, sodium dodecyl sulfate; TX-100, 4-octylphenol polyethoxylate. * Correspondence to: Gordana Popovi c (Telephone: þ381-11-3951215; Fax: þ38111-3972840). E-mail address: [email protected] (G. Popovi c).
diffusion to the relevant transmembrane domain.3-5 These data suggest that not only the conformation of active form of the sartans is important for the activity, but also the ionization state in physiological conditions that can affect partitioning between plasma and biomembranes. Most of the pharmacologically active substances in their chemical structures contain weak acidic and basic groups that partly and gradually ionize in aqueous solution until they establish equilibria between molecular and ionized forms (cationic, anionic, zwitterionic) which may express different physicochemical and pharmacokinetic properties. A degree of ionization of each compound can be predicted for any pH value of the solution based on its pKa value.6 Even at very early stages of the drug discovery, research and development in various fields of pharmacy are unfeasible without knowing the pKa values of the drugs.7 The ionization profile directly affects water solubility of the drug which is important for the analysis of a drug and its biopharmaceutical characterization. Determination of the pKa values of drugs is particularly significant for the prediction of their behavior under physiological conditions and an evaluation of drug bioavailability and distribution through biological membranes.8,9 After application, the drug has to pass through many biological membranes in order to reach the target site of action. In that way, it
http://dx.doi.org/10.1016/j.xphs.2016.06.007 0022-3549/© 2016 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
can be involved in various physicochemical and biological interactions with biomolecules present under physiological conditions. Consequently, bioenvironment can cause changes in protolytic equilibria and ionization might be different from that in an aqueous solution. For this reason, the physicochemical parameter values defined exclusively for the aqueous solution are not sufficient for an accurate estimation of drug behavior in physiological conditions. A better insight could be provided by the determination of the pKa value in an environment with the properties more similar to biological ones. Surfactant micelles have been used as simplified systems of biomembranes, although cell membranes characterize with a very complex structure responsible for the control of biological processes.10-14 Biomimetic nature of micellar solutions is based on structural and functional properties which are considered to mimic the most elementary membrane functions. Due to their amphiphilic properties, the molecules of surfactants are able to self-associate in a manner analogous to membrane phospholipids contributing to the compartmentalization of the molecules and reactions like in biological cells. Surfactant micelles are considered as confined systems which may influence reaction rates, products, and stereochemistry that may be different from those observed in the surfactant-free solutions.10,15 Micelles express a solubilizing effect on the compounds sparingly soluble in water. Reversible interactions between hydrophobic drugs and the micelle lead to the formation of a stable solution with the reduced thermodynamic activity of the solubilized compound.16 It has been shown that specific microenvironment formed in a micellar solution may affect spectral characteristics,17 the acidbase properties,18,19 and isomerization20 of solubilized drugs as well as membrane permeability thereof.21 Depending on their hydrophilic or lipophilic properties, the molecules of drugs can be solubilized in the hydrophobic interior or on the hydrophilic surface of the micelles. The nature of the drug-micelles interactions which involve their hydrophilic and lipophilic parts is important for the prediction of complex biological processes such as the transport of drugs through the cell membrane.22,23 Drugs with complex chemical structures can be involved in various interactions with the micelles and consequently it is generally impossible to predict shifts in protolytic equilibria without experimental investigations. From the chemical point of view, sartans represent acids or ampholytes (Fig. 1). Irbesartan and losartan are ampholytes with one acidic center (tetrazole ring) and one basic center (nitrogen of the imidazole ring). Valsartan is a diprotic acid with the tetrazole ring and the carboxyl group. Ionizable groups of sartans are directly involved in the interaction with the AT1-receptor and they are an
2445
integral part of the chemical structure which is required for their pharmacological activity.1,3 The literature survey revealed that there are not much data on determination of the pKa values for sartans,24-27 where in most cases only one pKa value is experimentally determined for the molecules with 2 ionizable groups.25-28 Also, information about an effect of micelles on the pKa values of sartans is lacking in the literature. Only an effect of a cationic surfactant CTAB (cetyltrimethylammonium bromide) on the acid-base equilibria that includes a tetrazole ring of losartan25 and valsartan27 has been investigated so far. The aim of this study was to investigate an effect of differently charged micelles, as a membrane mimicking systems on protolytic equilibria which include all ionizable groups of irbesartan, losartan, and valsartan in the pH range from 0 to 14. Investigations were performed potentiometrically in the presence and in the absence of the anionic (sodium dodecyl sulfate, SDS), cationic (CTAB), and nonionic (4-octylphenol polyethoxylate [TX-100] and polyoxyethylene (23) lauryl ether [Brij 35]) surfactants (Fig. 2). The surfactant concentration can affect the micellar phase concentrations and the micellar phase volumes, but in this study all the surfactants were applied in the same concentration. The primary objective was to investigate the behavior of sartans in the environments with a different charge or polarity in relation to water, as well as to compare their ionization in every specific surfactant solution with ionization in water.
Materials and Methods Chemicals and Reagents The examined compounds (losartan (2-n-butyl-4-chloro5-hydroxymethyl-1-[(20 -(1H-tetrazol-5-yl)biphenyl-4-yl)methyl] imidazole, potassium salt); irbesartan (2-butyl-3-[(20 -(1H-tetrazol-5-yl)biphenyl-4-yl)methyl]-1,3-diazaspiro[4,4]non-1-en-4-one); and valsartan ((S)-N-valeryl-N-{[20 -(1H-tetrazol-5-yl)biphenyl-4-yl] methyl}-valine) were kindly donated from the Medicines and Medical Devices Agency of Serbia (Belgrade, Serbia). Sodium chloride and methanol for analysis were purchased from Merck (Darmstadt, Germany). Sodium dodecyl sulfate (J.T. Baker, 95% purity), CTAB (Acros Organic, 99% purity), Triton TX-100 (Acros Organic, 98% purity), and Brij 35 solution (30% wt/wt in water) for biochemistry (Merck) were used to prepare micellar solutions. All solutions were prepared with double distilled water. Standard solutions of HCl and CO2-free NaOH were standardized by potentiometry.
Figure 1. Chemical structures of sartans.
2446
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 2. Chemical structures of surfactants.
Potentiometric titrations were performed with a titration system 798 MPT Titrino with a combined electrode (LL unitrode Pt1000; Metrohm). Constant temperature of the titrated solutions was maintained using a Huber Polystat CC2 thermostat.
Potentiometric Titration Ionization constants of irbesartan, losartan, and valsartan were determined in the absence and in the presence of the 102 M surfactant (SDS, CTAB, TX-100, and Brij 35). For the pH measurements, the electrode was regularly calibrated with the standard buffer solutions (pH 4.01 and 7.00). An addition of the surfactants in the above concentration had no significant effect on pH of the buffers (under ±0.02 pH units). All surfactants were used in the concentrations significantly above their critical micellar concentration, thus the influence of the other molecules present in the solution at critical micellar concentration can be neglected. All measurements were carried out at 25 C with a continuous magnetic stirring. Constant ionic strength was adjusted to 0.1 M with NaCl. In the surfactant-free media, ionization constants pKa* were obtained in the different methanol-water mixtures (30%, 40%, 45%, 50%, and 55% wt/wt). To 40 mL solutions of sartans (5 104 to 103 M) in the methanol-water mixtures, 1.0 mL HCl solution (0.1041 M) was added and titrated with the 0.02 mL aliquots of standard NaOH solution (0.0996 M). Determination of the pKapp a values of sartans in the presence of surfactants was performed by applying the same procedure, but instead of the water-methanol mixture, the 0.01 M aqueous solutions of the surfactants were used.
The measured pH values are expressed as pcH values (pcH ¼ log [Hþ]), according to the relation pcH ¼ pH A. The correction factor A was determined experimentally by titrating the standard hydrochloric acid solution at the ionic strength of 0.1 M (NaCl) with the standard NaOH solution for all methanol-water mixtures and the surfactant supplied solutions.28,29 Based on the data obtained by potentiometric titrations, the ionization constants were calculated using a computer program Hyperquad, which enables determination of equilibrium constants in complex systems containing the overlapped acid-base equilibria.30 Results and Discussion Due to poor water solubility of sartans in their molecular forms, the aqueous pKw a values were obtained indirectly from the pKa* values potentiometrically determined in the different methanolwater mixtures (30%-55% methanol, wt/wt).31,32 Methanol was chosen as a solvent because its general effect on the ionization constants is close to that of pure water.31 The pKw a values were obtained by extrapolation of the pKa* values to 0% of methanol (Fig. 3). As a cosolvent in the mixed solvent solutions, methanol contributes to the better solubility of sartans but it relatively little affects their pKw a values, particularly in the case of losartan and irbesartan. This fact is demonstrated by the plots of pKa* versus % methanol (Fig. 3) with the slope values lower than 0.01 for both losartan and irbesartan, and 0.02 for the both constants of valsartan. Numerical values obtained in this study and the literature pKw a values are listed in Table 1. Because of values of the ionization constants for the molecules of the examined sartans, it can be
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
2447
Table 1 Potentiometrically Determined pKw a Values of Irbesartan, Losartan, and Valsartan, and the Literature Data Sartan
pKw a Potentiometrically Determined
Literature Data
Valsartan
pKw a1 pKw a2 pKw a1 pKw a2 pKw a1 pKw a2
3.6024 4.7024, 4.9026, 5.0625 2.9524 4.2524, 3.1526, 4.7027 3.6924 4.4224, 4.7026
Losartan Irbesartan
Figure 3. Plots of pKa* versus %MeOH (wt/wt) for (a) irbesartan, (b) losartan, and (c) valsartan. C, pKa1*; -, pKa2*.
concluded that the ionization processes are overlapped. Moreover, little data are available in the literature regarding complete definition of the protolytic equilibria of sartans in aqueous solutions.24 In most cases, only one constant has been experimentally determined and attributed to the ionization of the tetrazole ring.25-27 However, ionization of the carboxyl group in valsartan and ionization of the imidazole nitrogen in irbesartan and losartan must be taken into account in order to get a better insight in their real ionization profiles. It is necessary to assess which equilibrium forms are present in the solution at a particular pH value. Small differences between the pKw a values determined for the examined sartans in this study and the earlier obtained values24 could be a
3.79 4.55 3.27 4.60 3.88 4.55
consequence of the different working conditions (ionic strength and electrolyte). The obtained pKw a values are attributed to the corresponding ionizable functional groups, based on the analysis of the chemical structure of sartans. The common part of the chemical structures of the examined sartans is the tetrazole ring linked to the biphenyl group; therefore in each studied compound the ionizable groups are sufficiently well separated not to affect ionization of each other by the electron effects. Accordingly, it can be expected that the deprotonation of nitrogen in the tetrazole ring occurs in approximately the same way with all investigated compounds resulting in similar pKw a values of the tetrazole. The carboxyl group is more acidic compared to the tetrazol ring, indicating that in the molecule of valsartan with 2 acidic groups, a lower pKw a1 value (3.79) can be assigned to the carboxylic group, while pKw a2 4.55 corresponds to the tetrazole ionization. These values are in agreement with the literature data for the pKw a values of the carboxylic acids and the 5-substituted tetrazole derivatives.33 In the molecules of irbesartan and losartan, the pKw a2 values (4.55 and 4.60, respectively) are related to the tetrazole deprotonation. In these 2 molecules, the pKw a1 values correspond to the weakly basic centers, the nitrogen in the imidazole structure of losartan (3.27), and 4,5-dihydro-5-oxo1H-imidazole of irbesartan (3.88). The chlorine substituent in the losartan imidazole reduces the electron density on the nitrogen atom by the electron withdrawing effect which hinders proton binding and causes a decrease in the pKw a value of the losartan imidazole compared to the same ionizable group in irbesartan. The described ionization and the existing equilibrium species of sartans are shown in Figure 4. Protolytic equilibria of valsartan and losartan were examined in the presence of the 4 surfactants, SDS, CTAB, TX-100, and Brij 35, without the use of cosolvent because the surfactants contributed to increasing solubility thereof. Because the sartans are poorly soluble in water, an increase in their solubility in the micellar solutions points out to the interaction between the sartans and the micelles, and indicates that the equilibrium species are present in a micellar pseudophase. This means that the apparent ionization constant, pKapp a , represents a kind of a hybrid between drug ionization in the aqueous phase and ionization within the micellar pseudophase.34 Because of poor solubility in micellar solutions of nonionic surfactants (TX-100 and Brij 35), the pKapp values of irbesartan are a determined only in the presence of the ionic surfactants SDS and CTAB. Poor solubility can indicate that interactions of irbesartan with the nonionic micelles are lacking. Potentiometric curves obtained by titration of sartan solutions in the presence of surfactants are shown in Figure 5. For the purpose of comparison, the titration curves of the methanol-water mixtures without the surfactants, with the lowest percentage of methanol are also shown in Figure 5. The difference between the pKapp values determined in the sura factant solutions and the pKw a values that define the ionization in 35 the surfactant-free solutions (DpKapp ¼ pKapp - pKw demona a a) strates the shift in protolytic equilibria of sartans caused by micelles
2448
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
þ Figure 4. Ionization profiles of (a) irbesartan (IrbH, molecular form; IrbHþ 2 cationic form; Irb , anionic form), (b) losartan (LosH, molecular form; LosH2 , cationic form; Los , anionic form), and (c) valsartan (ValH2, molecular form; ValH, monoanionic form; Val2, dianionic form).
(Table 2). The shifts in distribution of the equilibrium forms of the examined sartans in the presence of micelles can be clearly seen from the distribution diagrams (Fig. 6). Changes in the pKapp values a are the consequence of interactions that could be based on the
hydrophobic effects between the hydrophobic groups of the examined molecules and the lipophilic interior of the micelle as well as on electrostatic effects, which depend on the charge of the ionized groups of sartans and the charged or polar micelle
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 5. Potentiometric curves for (a) irbesartan, (b) losartan, and (c) valsartan solutions in the absence and presence of 102 M surfactant (SDS, CTAB, TX-100, and Brij 35) titrated with standard NaOH solution. I ¼ 0.1 M (NaCl) and t ¼ 25 C.
surface.36 Also, in the case of the nonionic micelles, formation of the hydrogen bonds and dipole interactions with solubilized drugs are possible. The direction of the protolytic equilibrium shift can
2449
indicate the type of interaction that predominates. The obtained DpKapp values ranged from 1.44 to þ1.72. Generally, from Table 2 a it can be seen that the anionic surfactant SDS causes an increase in the pKapp values of the acidic groups of valsartan, and of the acidic a and basic groups of irbesartan and losartan which represent the ampholytes. This effect points out to a decrease in ionization of carboxyl group (DpKapp ¼ þ1.21) and tetrazol ring a (DpKapp ¼ þ1.49), and the increase in ionization of the imidazole a nitrogen in these compounds (DpKapp ¼ þ1.72). Due to the negaa tively charged surface of the anionic surfactant micelles, they can participate in electrostatic interactions with the ionized forms of the examined compounds. In the interaction with the negatively charged groups (the carboxylate anion and deprotonated nitrogen of the tetrazole), the repulsion forces are predominant. This effect stabilizes neutral forms of these functional groups which are protonated and thus hinders ionization, leading to an increase in the pKapp values. Accordingly, in the case of the anionic surfactant the a electrostatic forces of attraction can stabilize the cationic forms of the imidazole ring in losartan and irbesartan, which facilitates their protonation. Contrary to the negatively charged SDS micelles, the CTAB micelles with the positively charged surface show the shift in protolytic equilibria of sartans in the opposite direction. The pKapp values a determined in the presence of cationic micelles are lower compared to those in the aqueous solution. A positively charged surface of the CTAB micelles causes an increase in the ionization of the carboxyl group (DpKapp ¼ 0.50) and the tetrazole ring a (DpKapp ¼ 1.44), but reduces ionization of the imidazole nitrogen a (DpKapp ¼ 1.31). The observed shifts also point out to the fact that a electrostatic forces are prevailing in the interactions of the micelles with the ionized groups. It seems that the CTAB micelles stabilize negatively charged carboxyl anion and the deprotonated tetrazole ring. On the other hand, the forces of repulsion hinder protonation of the imidazole nitrogen which causes a decrease of its pKapp value. a Micelles of the nonionic surfactants (TX-100, Brij 35) do not have a charge and counter ions in a surface layer which is formed by the coils of the hydrated polyethylene oxide chains.16 Although they are not ionized, the surfaces of these micelles are stabilized by the hydrogen bonds and dipole interactions dominate in the palisade hydrated hydrophilic layer.37 The polar moieties of the drugs, their proton donor, and proton acceptor groups can predominantly be retained in the hydrophilic layer of the nonionic micelles.38 The 2 applied nonionic surfactants form neutral micelles with a spherical shape. However, due to the differences in their chemical structures and the number of the hydrophilic oxyethylene units, their micellar properties are not necessarily identical so that the differences may occur in the interactions with the same drug.39 The shift in the pKapp values from 0.86 (the imidazole moiety of losa artan) in the presence of Brij 35 to þ1.30 (the carboxyl group of valsartan) in the presence of TX-100 could be explained in terms of
Table 2 The pKapp Values of Valsartan, Losartan, and Irbesartan Potentiometrically Determined in the Presence of 102 M Surfactants a Sartan
Valsartan pKapp a1 pKapp a2 Losartan pKapp a1 pKapp a2 Irbesartan pKapp a1 pKapp a2
SDS
CTAB
TX-100
Brij 35
pKapp a
DpKapp a
pKapp a
DpKapp a
pKapp a
DpKapp a
pKapp a
DpKapp a
5.00 ± 0.06 5.14 ± 0.15
þ1.21 þ0.59
3.29 ± 0.04 3.83 ± 0.06
0.50 0.72
5.09 ± 0.12 5.55 ± 0.17
þ1.30 þ1.00
4.35 ± 0.03 4.94 ± 0.04
þ0.56 þ0.39
4.75 ± 0.05 6.01 ± 0.06
þ1.48 þ1.41
1.96 ± 0.20 3.16 ± 0.13
1.31 1.44
3.66 ± 0.08 5.29 ± 0.09
þ0.39 þ0.69
2.41 ± 0.04 4.72 ± 0.04
0.86 þ0.12
5.60 ± 0.20 6.04 ± 0.12
þ1.72 þ1.49
3.08 ± 0.04 3.14 ± 0.02
0.80 1.41
¼ pKapp pKw I ¼ 0.1 M NaCl, t ¼ 25 C, DpKapp a a a.
2450
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 6. Distribution of the equilibrium forms of sartans in the presence and absence of surfactants, as a function of pH: (a) irbesartan, (b) and (c) losartan, (d) and (e) valsartan. Abbreviations of the sartan equilibrium forms are shown in the legend of Figure 4.
the hydrogen bonding and the dipole interactions in the palisade layer of the nonionic micelles. Ionization of the acidic groups of valsartan (carboxyl group and tetrazole ring) and losartan (tetrazole ring) is decreased in the presence of both nonionic surfactant micelles. Hydrogen bonds formed with the polar oxygen atoms in the hydrated layer are probably a dominant effect in the process of solubilization which hinders deprotonation of the acidic groups and shifts the equilibria in the direction of neutral forms. TX-100 and Brij 35 exert opposite effects on ionization of the basic center of losartan. Namely, ionization of imidazole nitrogen of losartan is increased in the presence of TX-100 (DpKapp ¼ þ0.39) and a decreased in the presence of the Brij 35 micelles (DpKapp ¼ 0.86). a It is possible that different effects of TX-100 and Brij 35 on
ionization of imidazole are a consequence of different polarity of the palisade layers which is higher in the Brij 35 micelles, where the solvation process occurs stronger pronounced compared to the TX100 micelles. These differences of the palisade polarity can also affect orientation of the molecules in this layer. All observed shifts of the pKapp values indicate that the ina teractions between sartans and the micelles take place mostly in the surface layers of the micelles, where ionizable groups of these drugs participate in electrostatic effects and formation of hydrogen bond. Surfactants significantly affect distribution of the molecular and ionized forms in the pH range from 2 to 7. This interval includes the pH values of a biopharmaceutical importance which can indicate a
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
significant distribution change at physiological conditions, due to the interaction of drugs with biomolecules of different charge and polarity. Table 3 shows an effect of the surfactants on distribution of the equilibrium forms of sartans at the pH values of biopharmaceutical significance (pH 1.2, 4.5, 6.8, 7.4). The effect of micelles is most evident at pH 4.5, which matches the value in the proximal part of the small intestine where the largest numbers of orally administered drugs are absorbed but where many charged and polar biomolecules are present. Mostly the nonionized fraction of a drug is able to cross the cell membrane because of the lipid nature of the membrane. The behavior of sartans observed with the differently charged surfactants at pH 4.5 can be discussed in terms of human intestinal absorption and oral bioavailability. The presence of the anionic SDS micelles shifts the equilibrium toward the formation of the cationic form of irbesartan, whose content increases by 81% while the content of molecular and anionic forms decreases by 40% compared to surfactant-free solution. The cationic CTAB micelles shift protolytic equilibria of irbesartan in the opposite direction in relation to SDS by increasing the content of anionic form by 46%. These results point out to an increase in the ionized form of irbesartan in the presence of the positively or negatively charged biomolecules present in small intestine. Because for the ionized species it is more difficult to pass through biomembranes, interactions of irbesartan with charged biomolecules in small intestine can potentially decrease intestinal absorption and oral bioavailability. A similar influence of the ionic micelles can be noticed in the case of losartan, where SDS also favors protonation of the imidazole nitrogen (þ57%) and CTAB contributes to an almost complete deprotonation of tetrazole (þ43%). Higher displacement of the equilibria in the direction of the cationic form of irbesartan compared to losartan in the presence of the anionic micelles at pH 4.5 is probably due to the lower electron density on the imidazole nitrogen which is reduced by the electron withdrawing effect of the chlorine substituent. It can be observed that SDS reduces ionization of valsartan, moving the equilibria toward the molecular form (þ63%), while CTAB increases the tetrazole ionization by increasing the dianionic form (39%). Therefore, the presence of the positively charged biomolecules can decrease the content of the nonionized valsartan species which pass easily through the membranes, while the negatively charged ones increase this content. At pH 4.5, the
2451
contents of molecular forms of losartan and valsartan are increased under the influence of the nonionic surfactants. This finding indicates that an interaction of valsartan and losartan with the noncharged polar molecules can potentially affect an increase in the absorption and bioavailability of these drugs. Conclusion In this study, it was shown that the micelles of the differently charged surfactants shift the pKapp values of irbesartan, losartan, a and valsartan. In line with this finding, it can be assumed that microenvironment changes can significantly affect the protolytic equilibria of these compounds also in physiological conditions. Ionization of compounds in a biological system depends on polarity and charge of the environment and directly affects pharmacokinetic parameters and pharmacological behavior of drugs. Decreased ionization of the acidic groups and increased ionization of basic groups in the presence of the anionic SDS micelles, and an opposite effect on ionization in the presence of the cationic CTAB micelles point out to predominant electrostatic effects which involve ionizable groups of sartans and the charged surface of the ionic micelles. Based on a shift in protolytic equilibria in the presence of the nonionic micelles, it can be assumed that ionizable groups exert dipole interactions and form hydrogen bond, while the drug is predominantly retained in the palisade layer of the micelles. The aggregates of surfactants exhibit structural and functional properties of the cell membranes, so that the observed changes in ionization of sartans can point out to specific interactions with biological membranes and to localization in them. Sartans are the orally administered drugs which pass through many membranes, until they reach the target site of action. For this reason, this study can help defining the form of the drug which interacts with the transmembrane domain of the AT1 receptors. Moreover, the pKa* values valid for the methanol-water mixtures have an analytical significance and can be used to optimize chromatographic analysis in which the investigated compounds are analyzed by applying a mobile phase consisting of methanol as an organic modifier. Acknowledgments
Table 3 Percentage of the Equilibrium Forms of Sartans at the pH Values of Biopharmaceutical Importance in the Surfactant-Free and the Surfactant Containing Media Variable
Irbesartan H2O SDS CTAB Losartan H2O SDS CTAB TX-100 Brij 35
Valsartan H2O SDS CTAB TX-100 Brij 35
pH 1.2
pH 4.5
pH 6.8
pH 7.4
Cat
Mol An
Cat
Mol An
Cat
Mol An
Cat
100 100 99
0 0 1
0 0 0
11 92 0
47 7 4
42 0 96
0 1 0
1 15 0
99 0 84 0 100 0
0 4 0
100 96 100
100 100 85 100 94
0 0 15 0 6
0 0 0 0 0
6 63 0 11 1
48 36 4 77 62
45 1 96 12 37
0 0 0 0 0
1 14 0 3 1
99 86 100 97 99
0 0 0 0 0
0 4 0 1 0
100 96 100 99 100
Mol An
Mol MA
DA Mol MA
DA Mol MA
DA
Mol MA
DA
100 100 99 100 100
0 0 0 0 0
43 5 82 1 18
99 98 100 90 99
0 0 0 0 0
100 99 100 97 100
0 0 1 0 0
9 72 1 86 34
48 23 17 14 48
0 0 0 0 0
1 2 0 10 1
0 1 0 3 0
Equilibrium forms: Cat, cationic; An, anionic; Mol, molecular; MA, monoanionic; DA, dianionic.
This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, contract #172033. References 1. Lemke TL, Williams DA, Roche VF, Zito SW. Foye’s Principles of Medicinal Chemistry. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2013. 2. Doulton TW, He FJ, MacGregor GA. Systematic review of combined angiotensin-converting enzyme inhibition and angiotensin receptor blockade in hypertension. Hypertension. 2005;45(5):880-886. 3. Zoumpoulakis P, Daliani I, Zervou M, et al. Losartan’s molecular basis of interaction with membranes and AT1 receptor. Chem Phys Lipids. 2003;125(1): 13-25. 4. Fotakis C, Christodouleas D, Zoumpoulakis P, et al. Comparative biophysical studies of sartan class drug molecules losartan and candesartan (CV-11974) with membrane bilayers. J Phys Chem B. 2011;115(19):6180-6192. 5. Zervou M, Cournia Z, Potamitis C, et al. Insights into the molecular basis of action of the AT1 antagonist losartan using a combined NMR spectroscopy and computational approach. Biochimica et Biophysica Acta Biomembranes. 2014;1838(3):1031-1046. 6. Cairns D. Essentials of Pharmaceutical Chemistry. 4th ed. Great Britain: Pharmaceutical Press; 2012. 7. Manallack DT, Prankerd RJ, Nassta GC, Ursu O, Oprea TI, Chalmers DK. A chemogenomic analysis of ionization constantsdimplications for drug discovery. Chem Med Chem. 2013;8(2):242-255. 8. Manallack DT, Prankerd RJ, Yuriev E, Oprea TI, Chalmers DK. The significance of acid/base properties in drug discovery. Chem Soc Rev. 2013;42(2):485-496.
2452
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
9. Peters JU, Hert J, Bissantz C, et al. Can we discover pharmacological promiscuity early in the drug discovery process? Drug Discov Today. 2012;17(7):325-335. 10. Fendler JH. Atomic and molecular clusters in membrane mimetic chemistry. Chem Rev. 1987;87(5):877-899. 11. Garavito RM, Ferguson-Miller S. Detergents as tools in membrane biochemistry. J Biol Chem. 2001;276(35):32403-32406. 12. Mahiuddin S, Zech O, Raith S, Touraud D, Kunz W. Catanionic micelles as a model to mimic biological membranes in the presence of anesthetic alcohols. Langmuir. 2009;25(21):12516-12521. 13. Ghosh D, Chattopadhyay N. Equilibrium and dynamic effects on ligand binding to biomacromolecules and biomimetic model systems. Int Rev Phys Chem. 2013;32(3):435-466. 14. Pereira-Leite C, Carneiro C, Soares JX, et al. Biophysical characterization of the drugemembrane interactions: the case of propranolol and acebutolol. Eur J Pharm Biopharm. 2013;84(1):183-191. 15. Mirgorodskaya AB, Yackevich EI, Kudryashova YR, et al. Design of supramolecular biomimetic catalysts of high substrate specificity by noncovalent selfassembly of calix [4] arenes with amphiphilic and polymeric amines. Colloids Surf B Biointerfaces. 2014;1(117):497-504. 16. Rangel-Yagui CO, Pessoa Jr A, Tavares LC. Micellar solubilization of drugs. J Pharm Pharm Sci. 2005;8(2):147-163. 17. Varghese B, Al-Busafi SN, Suliman FO, Al-Kindy SM. Synthesis, spectroscopic characterization and photophysics of a novel environmentally sensitive dye 3naphthyl-1-phenyl-5-(4-carboxyphenyl)-2-pyrazoline. J Lumin. 2015;159:9-16. 18. Jaiswal PV, Ijeri VS, Srivastava AK. Effect of surfactants on the dissociation constants of ascorbic and maleic acids. Colloids Surf B Biointerfaces. 2005;46(1):45-51. 19. Popovi c MR, Popovi c GV, Agbaba DD. The effects of anionic, cationic, and nonionic surfactants on acidebase equilibria of ACE inhibitors. J Chem Eng Data. 2013;58(9):2567-2573. 20. Popovi c MR, Popovi c GV, Filipi c SV, Nikoli c KM, Agbaba DD. The effects of micelles of differently charged surfactants on the equilibrium between (Z)-and (E)-diastereomers of five ACE inhibitors in aqueous media. Monatsh Chem Chem Mon. 2015;146(6):913-921. 21. Miller JM, Beig A, Krieg BJ, et al. The solubilityepermeability interplay: mechanistic modeling and predictive application of the impact of micellar solubilization on intestinal permeation. Mol Pharm. 2011;8(5):1848-1856. 22. Mahajan S, Mahajan RK. Interactions of phenothiazine drugs with surfactants: a detailed physicochemical overview. Adv Colloid Interface Sci. 2013;199:1-4. 23. Morrow BH, Wang Y, Wallace JA, Koenig PH, Shen JK. Simulating pH titration of a single surfactant in ionic and nonionic surfactant micelles. J Phys Chem B. 2011;115(50):14980-14990. 24. Tosco P, Rolando B, Fruttero R, et al. Physicochemical profiling of sartans: a detailed study of ionization constants and distribution coefficients. Helv Chim Acta. 2008;91(3):468-482.
25. Cudina O, Brbori c J, Jankovi c I, Karljikovi c-Raji c K, Vladimirov S. Study of valsartan interaction with micelles as a model system for biomembranes. Colloids Surf B Biointerfaces. 2008;65(1):80-84. 26. Cagigal E, Gonzalez L, Alonso RM, Jimenez RM. pKa determination of angiotensin II receptor antagonists (ARA II) by spectrofluorimetry. J Pharm Biomed Anal. 2001;26(3):477-486. 27. Abdel-Fattah L, Abdel-Aziz L, Gaied M. Enhanced spectrophotometric determination of Losartan potassium based on its physicochemical interaction with cationic surfactant. Spectrochim Acta A Mol Biomol Spectrosc. 2015;136: 178-184. 28. Irving HM, Miles MG, Pettit LD. A study of some problems in determining the stoicheiometric proton dissociation constants of complexes by potentiometric titrations using a glass electrode. Anal Chim Acta. 1967;38: 475-488. 29. Albert A, Serjeant EP. The Determination of Ionization Constants. 3rd ed. London: Chapman and Hall; 1984. 30. Gans P, Sabatini A, Vacca A. Investigation of equilibria in solution. Determination of equilibrium constants with the HYPERQUAD suite of programs. Talanta. 1996;43(10):1739-1753. 31. Avdeef A. Absorption and Drug Development: Solubility, Permeability, and Charge State. Hoboken, NJ: John Wiley & Sons; 2012. bel R, Chmurzyn ski L. Potentiometric pKa determination of standard sub32. Wro stances in binary solvent systems. Anal Chim Acta. 2000;405(1):303-308. 33. Charton M. Electrical effects of ortho substituents in imidazoles and benzimidazoles. J Org Chem. 1965;30(10):3346-3350. 34. Mchedlov-Petrossyan NO, Kamneva NN, Kharchenko AY, Shekhovtsov SV, Marinin AI, Kryshtal AP. The influence of the micellar pseudophase of the double-chained cationic surfactant di-n-tetradecyldimethylammonium bromide on the absorption spectra and protolytic equilibrium of indicator dyes. Colloids Surf A Physicochem Eng Asp. 2015;476:57-67. 35. Mchedlov-Petrossyan NO, Vodolazkaya NA, Kamneva NN. Acid-base equilibrium in aqueous micellar solutions of surfactants. Micelles: Structural biochemistry, formation and functions and usage. New York: Nova Science Pub Inc; 2013:352. 36. Khossravi D. Drug-surfactant interactions: effect on transport properties. Int J Pharm. 1997;155(2):179-190. 37. Narang AS, Delmarre D, Gao D. Stable drug encapsulation in micelles and microemulsions. Int J Pharm. 2007;345(1):9-25. 38. Dar AA, Chat OA. Cosolubilization of Coumarin30 and Warfarin in cationic, anionic, and nonionic micelles: a micelleewater interfacial charge dependent FRET. J Phys Chem B. 2015;119(35):11632-11642. 39. Kumbhakar M, Goel T, Mukherjee T, Pal H. Role of micellar size and hydration on solvation dynamics: a temperature dependent study in Triton-X-100 and Brij-35 micelles. J Phys Chem B. 2004;108(50):19246-19254.
Contents lists available at ScienceDirect
Journal of Pharmaceutical Sciences journal homepage: www.jpharmsci.org
Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Protolytic Equilibria of Sartans in Micellar Solutions of Differently Charged Surfactants Maja Gruji c 1, Marija Popovi c 1, Gordana Popovi c 2, *, Katarina Nikolic 1, Danica Agbaba 1 1 2
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Belgrade, Belgrade, Serbia Department of General and Inorganic Chemistry, Faculty of Pharmacy, University of Belgrade, Belgrade, Serbia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 March 2016 Revised 20 May 2016 Accepted 7 June 2016 Available online 13 July 2016
Protolytic equilibria of irbesartan, losartan, and valsartan have been investigated in the presence and absence of differently charged anionic (sodium dodecyl sulfate), cationic (cetyltrimethylammonium bromide), and nonionic (4-octylphenol polyethoxylate and polyoxyethylene (23) lauryl ether) surfactants. Ionization constants were determined by potentiometric titration at a constant ionic strength (0.1 M NaCl) and temperature 25 C. The effect of surfactants was estimated, based on a shift in apparent w ionization constants (pKapp a ) determined in micellar solutions against the pKa values in water. The values of sartans (up to 1.72 pK units), while the anionic surfactant caused an increase in the pKapp a values (up to 1.44 pK units). cationic surfactant had an opposite effect and caused a reduction in pKapp a These results point out to the fact that the ionizable groups of sartans are involved in electrostatic invalues in the presence of teractions with the charged surface of the ionic micelles. Shift in the pKapp a nonionic surfactants (from 0.86 to þ1.30) is a consequence of the interactions of drugs with the hydrophilic palisade layer. Significant changes in the distribution profiles of the equilibrium forms (from 44% to þ80%) are observed at the biopharmaceutically important pH 4.5 value and can be considered in terms of the potential influence on intestinal absorption and bioavailability. © 2016 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Keywords: physicochemical properties acid-base equilibria surfactants micelle biomimetics drug interaction
Introduction Angiotensin II receptor blockers also known as sartans represent an important class of drugs that act as competitive and selective antagonists of angiotensin II at the angiotensin AT1 receptors.1 Angiotensin II is a potent vasoconstrictor and the primary vasoactive peptide of the renin-angiotensin system, which plays an important role in the pathology of many cardiovascular diseases. Sartans reduce pressor effect of angiotensin II and cause the pharmacological effect of lowering blood pressure. Therefore these drugs are used in the treatment of hypertension as well as in the treatment of cardiac insufficiency, myocardial infarction, and diabetic nephropathy.2 Accurate mechanism of interaction of sartans with AT1 receptor is still not completely resolved. It is considered that the interaction takes place in a 2-step process, which includes spontaneously inserting into the membrane, and then lateral
Abbreviations used: Brij 35, polyoxyethylene (23) lauryl ether; CTAB, cetyltrimethylammonium bromide; SDS, sodium dodecyl sulfate; TX-100, 4-octylphenol polyethoxylate. * Correspondence to: Gordana Popovi c (Telephone: þ381-11-3951215; Fax: þ38111-3972840). E-mail address: [email protected] (G. Popovi c).
diffusion to the relevant transmembrane domain.3-5 These data suggest that not only the conformation of active form of the sartans is important for the activity, but also the ionization state in physiological conditions that can affect partitioning between plasma and biomembranes. Most of the pharmacologically active substances in their chemical structures contain weak acidic and basic groups that partly and gradually ionize in aqueous solution until they establish equilibria between molecular and ionized forms (cationic, anionic, zwitterionic) which may express different physicochemical and pharmacokinetic properties. A degree of ionization of each compound can be predicted for any pH value of the solution based on its pKa value.6 Even at very early stages of the drug discovery, research and development in various fields of pharmacy are unfeasible without knowing the pKa values of the drugs.7 The ionization profile directly affects water solubility of the drug which is important for the analysis of a drug and its biopharmaceutical characterization. Determination of the pKa values of drugs is particularly significant for the prediction of their behavior under physiological conditions and an evaluation of drug bioavailability and distribution through biological membranes.8,9 After application, the drug has to pass through many biological membranes in order to reach the target site of action. In that way, it
http://dx.doi.org/10.1016/j.xphs.2016.06.007 0022-3549/© 2016 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
can be involved in various physicochemical and biological interactions with biomolecules present under physiological conditions. Consequently, bioenvironment can cause changes in protolytic equilibria and ionization might be different from that in an aqueous solution. For this reason, the physicochemical parameter values defined exclusively for the aqueous solution are not sufficient for an accurate estimation of drug behavior in physiological conditions. A better insight could be provided by the determination of the pKa value in an environment with the properties more similar to biological ones. Surfactant micelles have been used as simplified systems of biomembranes, although cell membranes characterize with a very complex structure responsible for the control of biological processes.10-14 Biomimetic nature of micellar solutions is based on structural and functional properties which are considered to mimic the most elementary membrane functions. Due to their amphiphilic properties, the molecules of surfactants are able to self-associate in a manner analogous to membrane phospholipids contributing to the compartmentalization of the molecules and reactions like in biological cells. Surfactant micelles are considered as confined systems which may influence reaction rates, products, and stereochemistry that may be different from those observed in the surfactant-free solutions.10,15 Micelles express a solubilizing effect on the compounds sparingly soluble in water. Reversible interactions between hydrophobic drugs and the micelle lead to the formation of a stable solution with the reduced thermodynamic activity of the solubilized compound.16 It has been shown that specific microenvironment formed in a micellar solution may affect spectral characteristics,17 the acidbase properties,18,19 and isomerization20 of solubilized drugs as well as membrane permeability thereof.21 Depending on their hydrophilic or lipophilic properties, the molecules of drugs can be solubilized in the hydrophobic interior or on the hydrophilic surface of the micelles. The nature of the drug-micelles interactions which involve their hydrophilic and lipophilic parts is important for the prediction of complex biological processes such as the transport of drugs through the cell membrane.22,23 Drugs with complex chemical structures can be involved in various interactions with the micelles and consequently it is generally impossible to predict shifts in protolytic equilibria without experimental investigations. From the chemical point of view, sartans represent acids or ampholytes (Fig. 1). Irbesartan and losartan are ampholytes with one acidic center (tetrazole ring) and one basic center (nitrogen of the imidazole ring). Valsartan is a diprotic acid with the tetrazole ring and the carboxyl group. Ionizable groups of sartans are directly involved in the interaction with the AT1-receptor and they are an
2445
integral part of the chemical structure which is required for their pharmacological activity.1,3 The literature survey revealed that there are not much data on determination of the pKa values for sartans,24-27 where in most cases only one pKa value is experimentally determined for the molecules with 2 ionizable groups.25-28 Also, information about an effect of micelles on the pKa values of sartans is lacking in the literature. Only an effect of a cationic surfactant CTAB (cetyltrimethylammonium bromide) on the acid-base equilibria that includes a tetrazole ring of losartan25 and valsartan27 has been investigated so far. The aim of this study was to investigate an effect of differently charged micelles, as a membrane mimicking systems on protolytic equilibria which include all ionizable groups of irbesartan, losartan, and valsartan in the pH range from 0 to 14. Investigations were performed potentiometrically in the presence and in the absence of the anionic (sodium dodecyl sulfate, SDS), cationic (CTAB), and nonionic (4-octylphenol polyethoxylate [TX-100] and polyoxyethylene (23) lauryl ether [Brij 35]) surfactants (Fig. 2). The surfactant concentration can affect the micellar phase concentrations and the micellar phase volumes, but in this study all the surfactants were applied in the same concentration. The primary objective was to investigate the behavior of sartans in the environments with a different charge or polarity in relation to water, as well as to compare their ionization in every specific surfactant solution with ionization in water.
Materials and Methods Chemicals and Reagents The examined compounds (losartan (2-n-butyl-4-chloro5-hydroxymethyl-1-[(20 -(1H-tetrazol-5-yl)biphenyl-4-yl)methyl] imidazole, potassium salt); irbesartan (2-butyl-3-[(20 -(1H-tetrazol-5-yl)biphenyl-4-yl)methyl]-1,3-diazaspiro[4,4]non-1-en-4-one); and valsartan ((S)-N-valeryl-N-{[20 -(1H-tetrazol-5-yl)biphenyl-4-yl] methyl}-valine) were kindly donated from the Medicines and Medical Devices Agency of Serbia (Belgrade, Serbia). Sodium chloride and methanol for analysis were purchased from Merck (Darmstadt, Germany). Sodium dodecyl sulfate (J.T. Baker, 95% purity), CTAB (Acros Organic, 99% purity), Triton TX-100 (Acros Organic, 98% purity), and Brij 35 solution (30% wt/wt in water) for biochemistry (Merck) were used to prepare micellar solutions. All solutions were prepared with double distilled water. Standard solutions of HCl and CO2-free NaOH were standardized by potentiometry.
Figure 1. Chemical structures of sartans.
2446
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 2. Chemical structures of surfactants.
Potentiometric titrations were performed with a titration system 798 MPT Titrino with a combined electrode (LL unitrode Pt1000; Metrohm). Constant temperature of the titrated solutions was maintained using a Huber Polystat CC2 thermostat.
Potentiometric Titration Ionization constants of irbesartan, losartan, and valsartan were determined in the absence and in the presence of the 102 M surfactant (SDS, CTAB, TX-100, and Brij 35). For the pH measurements, the electrode was regularly calibrated with the standard buffer solutions (pH 4.01 and 7.00). An addition of the surfactants in the above concentration had no significant effect on pH of the buffers (under ±0.02 pH units). All surfactants were used in the concentrations significantly above their critical micellar concentration, thus the influence of the other molecules present in the solution at critical micellar concentration can be neglected. All measurements were carried out at 25 C with a continuous magnetic stirring. Constant ionic strength was adjusted to 0.1 M with NaCl. In the surfactant-free media, ionization constants pKa* were obtained in the different methanol-water mixtures (30%, 40%, 45%, 50%, and 55% wt/wt). To 40 mL solutions of sartans (5 104 to 103 M) in the methanol-water mixtures, 1.0 mL HCl solution (0.1041 M) was added and titrated with the 0.02 mL aliquots of standard NaOH solution (0.0996 M). Determination of the pKapp a values of sartans in the presence of surfactants was performed by applying the same procedure, but instead of the water-methanol mixture, the 0.01 M aqueous solutions of the surfactants were used.
The measured pH values are expressed as pcH values (pcH ¼ log [Hþ]), according to the relation pcH ¼ pH A. The correction factor A was determined experimentally by titrating the standard hydrochloric acid solution at the ionic strength of 0.1 M (NaCl) with the standard NaOH solution for all methanol-water mixtures and the surfactant supplied solutions.28,29 Based on the data obtained by potentiometric titrations, the ionization constants were calculated using a computer program Hyperquad, which enables determination of equilibrium constants in complex systems containing the overlapped acid-base equilibria.30 Results and Discussion Due to poor water solubility of sartans in their molecular forms, the aqueous pKw a values were obtained indirectly from the pKa* values potentiometrically determined in the different methanolwater mixtures (30%-55% methanol, wt/wt).31,32 Methanol was chosen as a solvent because its general effect on the ionization constants is close to that of pure water.31 The pKw a values were obtained by extrapolation of the pKa* values to 0% of methanol (Fig. 3). As a cosolvent in the mixed solvent solutions, methanol contributes to the better solubility of sartans but it relatively little affects their pKw a values, particularly in the case of losartan and irbesartan. This fact is demonstrated by the plots of pKa* versus % methanol (Fig. 3) with the slope values lower than 0.01 for both losartan and irbesartan, and 0.02 for the both constants of valsartan. Numerical values obtained in this study and the literature pKw a values are listed in Table 1. Because of values of the ionization constants for the molecules of the examined sartans, it can be
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
2447
Table 1 Potentiometrically Determined pKw a Values of Irbesartan, Losartan, and Valsartan, and the Literature Data Sartan
pKw a Potentiometrically Determined
Literature Data
Valsartan
pKw a1 pKw a2 pKw a1 pKw a2 pKw a1 pKw a2
3.6024 4.7024, 4.9026, 5.0625 2.9524 4.2524, 3.1526, 4.7027 3.6924 4.4224, 4.7026
Losartan Irbesartan
Figure 3. Plots of pKa* versus %MeOH (wt/wt) for (a) irbesartan, (b) losartan, and (c) valsartan. C, pKa1*; -, pKa2*.
concluded that the ionization processes are overlapped. Moreover, little data are available in the literature regarding complete definition of the protolytic equilibria of sartans in aqueous solutions.24 In most cases, only one constant has been experimentally determined and attributed to the ionization of the tetrazole ring.25-27 However, ionization of the carboxyl group in valsartan and ionization of the imidazole nitrogen in irbesartan and losartan must be taken into account in order to get a better insight in their real ionization profiles. It is necessary to assess which equilibrium forms are present in the solution at a particular pH value. Small differences between the pKw a values determined for the examined sartans in this study and the earlier obtained values24 could be a
3.79 4.55 3.27 4.60 3.88 4.55
consequence of the different working conditions (ionic strength and electrolyte). The obtained pKw a values are attributed to the corresponding ionizable functional groups, based on the analysis of the chemical structure of sartans. The common part of the chemical structures of the examined sartans is the tetrazole ring linked to the biphenyl group; therefore in each studied compound the ionizable groups are sufficiently well separated not to affect ionization of each other by the electron effects. Accordingly, it can be expected that the deprotonation of nitrogen in the tetrazole ring occurs in approximately the same way with all investigated compounds resulting in similar pKw a values of the tetrazole. The carboxyl group is more acidic compared to the tetrazol ring, indicating that in the molecule of valsartan with 2 acidic groups, a lower pKw a1 value (3.79) can be assigned to the carboxylic group, while pKw a2 4.55 corresponds to the tetrazole ionization. These values are in agreement with the literature data for the pKw a values of the carboxylic acids and the 5-substituted tetrazole derivatives.33 In the molecules of irbesartan and losartan, the pKw a2 values (4.55 and 4.60, respectively) are related to the tetrazole deprotonation. In these 2 molecules, the pKw a1 values correspond to the weakly basic centers, the nitrogen in the imidazole structure of losartan (3.27), and 4,5-dihydro-5-oxo1H-imidazole of irbesartan (3.88). The chlorine substituent in the losartan imidazole reduces the electron density on the nitrogen atom by the electron withdrawing effect which hinders proton binding and causes a decrease in the pKw a value of the losartan imidazole compared to the same ionizable group in irbesartan. The described ionization and the existing equilibrium species of sartans are shown in Figure 4. Protolytic equilibria of valsartan and losartan were examined in the presence of the 4 surfactants, SDS, CTAB, TX-100, and Brij 35, without the use of cosolvent because the surfactants contributed to increasing solubility thereof. Because the sartans are poorly soluble in water, an increase in their solubility in the micellar solutions points out to the interaction between the sartans and the micelles, and indicates that the equilibrium species are present in a micellar pseudophase. This means that the apparent ionization constant, pKapp a , represents a kind of a hybrid between drug ionization in the aqueous phase and ionization within the micellar pseudophase.34 Because of poor solubility in micellar solutions of nonionic surfactants (TX-100 and Brij 35), the pKapp values of irbesartan are a determined only in the presence of the ionic surfactants SDS and CTAB. Poor solubility can indicate that interactions of irbesartan with the nonionic micelles are lacking. Potentiometric curves obtained by titration of sartan solutions in the presence of surfactants are shown in Figure 5. For the purpose of comparison, the titration curves of the methanol-water mixtures without the surfactants, with the lowest percentage of methanol are also shown in Figure 5. The difference between the pKapp values determined in the sura factant solutions and the pKw a values that define the ionization in 35 the surfactant-free solutions (DpKapp ¼ pKapp - pKw demona a a) strates the shift in protolytic equilibria of sartans caused by micelles
2448
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
þ Figure 4. Ionization profiles of (a) irbesartan (IrbH, molecular form; IrbHþ 2 cationic form; Irb , anionic form), (b) losartan (LosH, molecular form; LosH2 , cationic form; Los , anionic form), and (c) valsartan (ValH2, molecular form; ValH, monoanionic form; Val2, dianionic form).
(Table 2). The shifts in distribution of the equilibrium forms of the examined sartans in the presence of micelles can be clearly seen from the distribution diagrams (Fig. 6). Changes in the pKapp values a are the consequence of interactions that could be based on the
hydrophobic effects between the hydrophobic groups of the examined molecules and the lipophilic interior of the micelle as well as on electrostatic effects, which depend on the charge of the ionized groups of sartans and the charged or polar micelle
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 5. Potentiometric curves for (a) irbesartan, (b) losartan, and (c) valsartan solutions in the absence and presence of 102 M surfactant (SDS, CTAB, TX-100, and Brij 35) titrated with standard NaOH solution. I ¼ 0.1 M (NaCl) and t ¼ 25 C.
surface.36 Also, in the case of the nonionic micelles, formation of the hydrogen bonds and dipole interactions with solubilized drugs are possible. The direction of the protolytic equilibrium shift can
2449
indicate the type of interaction that predominates. The obtained DpKapp values ranged from 1.44 to þ1.72. Generally, from Table 2 a it can be seen that the anionic surfactant SDS causes an increase in the pKapp values of the acidic groups of valsartan, and of the acidic a and basic groups of irbesartan and losartan which represent the ampholytes. This effect points out to a decrease in ionization of carboxyl group (DpKapp ¼ þ1.21) and tetrazol ring a (DpKapp ¼ þ1.49), and the increase in ionization of the imidazole a nitrogen in these compounds (DpKapp ¼ þ1.72). Due to the negaa tively charged surface of the anionic surfactant micelles, they can participate in electrostatic interactions with the ionized forms of the examined compounds. In the interaction with the negatively charged groups (the carboxylate anion and deprotonated nitrogen of the tetrazole), the repulsion forces are predominant. This effect stabilizes neutral forms of these functional groups which are protonated and thus hinders ionization, leading to an increase in the pKapp values. Accordingly, in the case of the anionic surfactant the a electrostatic forces of attraction can stabilize the cationic forms of the imidazole ring in losartan and irbesartan, which facilitates their protonation. Contrary to the negatively charged SDS micelles, the CTAB micelles with the positively charged surface show the shift in protolytic equilibria of sartans in the opposite direction. The pKapp values a determined in the presence of cationic micelles are lower compared to those in the aqueous solution. A positively charged surface of the CTAB micelles causes an increase in the ionization of the carboxyl group (DpKapp ¼ 0.50) and the tetrazole ring a (DpKapp ¼ 1.44), but reduces ionization of the imidazole nitrogen a (DpKapp ¼ 1.31). The observed shifts also point out to the fact that a electrostatic forces are prevailing in the interactions of the micelles with the ionized groups. It seems that the CTAB micelles stabilize negatively charged carboxyl anion and the deprotonated tetrazole ring. On the other hand, the forces of repulsion hinder protonation of the imidazole nitrogen which causes a decrease of its pKapp value. a Micelles of the nonionic surfactants (TX-100, Brij 35) do not have a charge and counter ions in a surface layer which is formed by the coils of the hydrated polyethylene oxide chains.16 Although they are not ionized, the surfaces of these micelles are stabilized by the hydrogen bonds and dipole interactions dominate in the palisade hydrated hydrophilic layer.37 The polar moieties of the drugs, their proton donor, and proton acceptor groups can predominantly be retained in the hydrophilic layer of the nonionic micelles.38 The 2 applied nonionic surfactants form neutral micelles with a spherical shape. However, due to the differences in their chemical structures and the number of the hydrophilic oxyethylene units, their micellar properties are not necessarily identical so that the differences may occur in the interactions with the same drug.39 The shift in the pKapp values from 0.86 (the imidazole moiety of losa artan) in the presence of Brij 35 to þ1.30 (the carboxyl group of valsartan) in the presence of TX-100 could be explained in terms of
Table 2 The pKapp Values of Valsartan, Losartan, and Irbesartan Potentiometrically Determined in the Presence of 102 M Surfactants a Sartan
Valsartan pKapp a1 pKapp a2 Losartan pKapp a1 pKapp a2 Irbesartan pKapp a1 pKapp a2
SDS
CTAB
TX-100
Brij 35
pKapp a
DpKapp a
pKapp a
DpKapp a
pKapp a
DpKapp a
pKapp a
DpKapp a
5.00 ± 0.06 5.14 ± 0.15
þ1.21 þ0.59
3.29 ± 0.04 3.83 ± 0.06
0.50 0.72
5.09 ± 0.12 5.55 ± 0.17
þ1.30 þ1.00
4.35 ± 0.03 4.94 ± 0.04
þ0.56 þ0.39
4.75 ± 0.05 6.01 ± 0.06
þ1.48 þ1.41
1.96 ± 0.20 3.16 ± 0.13
1.31 1.44
3.66 ± 0.08 5.29 ± 0.09
þ0.39 þ0.69
2.41 ± 0.04 4.72 ± 0.04
0.86 þ0.12
5.60 ± 0.20 6.04 ± 0.12
þ1.72 þ1.49
3.08 ± 0.04 3.14 ± 0.02
0.80 1.41
¼ pKapp pKw I ¼ 0.1 M NaCl, t ¼ 25 C, DpKapp a a a.
2450
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
Figure 6. Distribution of the equilibrium forms of sartans in the presence and absence of surfactants, as a function of pH: (a) irbesartan, (b) and (c) losartan, (d) and (e) valsartan. Abbreviations of the sartan equilibrium forms are shown in the legend of Figure 4.
the hydrogen bonding and the dipole interactions in the palisade layer of the nonionic micelles. Ionization of the acidic groups of valsartan (carboxyl group and tetrazole ring) and losartan (tetrazole ring) is decreased in the presence of both nonionic surfactant micelles. Hydrogen bonds formed with the polar oxygen atoms in the hydrated layer are probably a dominant effect in the process of solubilization which hinders deprotonation of the acidic groups and shifts the equilibria in the direction of neutral forms. TX-100 and Brij 35 exert opposite effects on ionization of the basic center of losartan. Namely, ionization of imidazole nitrogen of losartan is increased in the presence of TX-100 (DpKapp ¼ þ0.39) and a decreased in the presence of the Brij 35 micelles (DpKapp ¼ 0.86). a It is possible that different effects of TX-100 and Brij 35 on
ionization of imidazole are a consequence of different polarity of the palisade layers which is higher in the Brij 35 micelles, where the solvation process occurs stronger pronounced compared to the TX100 micelles. These differences of the palisade polarity can also affect orientation of the molecules in this layer. All observed shifts of the pKapp values indicate that the ina teractions between sartans and the micelles take place mostly in the surface layers of the micelles, where ionizable groups of these drugs participate in electrostatic effects and formation of hydrogen bond. Surfactants significantly affect distribution of the molecular and ionized forms in the pH range from 2 to 7. This interval includes the pH values of a biopharmaceutical importance which can indicate a
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
significant distribution change at physiological conditions, due to the interaction of drugs with biomolecules of different charge and polarity. Table 3 shows an effect of the surfactants on distribution of the equilibrium forms of sartans at the pH values of biopharmaceutical significance (pH 1.2, 4.5, 6.8, 7.4). The effect of micelles is most evident at pH 4.5, which matches the value in the proximal part of the small intestine where the largest numbers of orally administered drugs are absorbed but where many charged and polar biomolecules are present. Mostly the nonionized fraction of a drug is able to cross the cell membrane because of the lipid nature of the membrane. The behavior of sartans observed with the differently charged surfactants at pH 4.5 can be discussed in terms of human intestinal absorption and oral bioavailability. The presence of the anionic SDS micelles shifts the equilibrium toward the formation of the cationic form of irbesartan, whose content increases by 81% while the content of molecular and anionic forms decreases by 40% compared to surfactant-free solution. The cationic CTAB micelles shift protolytic equilibria of irbesartan in the opposite direction in relation to SDS by increasing the content of anionic form by 46%. These results point out to an increase in the ionized form of irbesartan in the presence of the positively or negatively charged biomolecules present in small intestine. Because for the ionized species it is more difficult to pass through biomembranes, interactions of irbesartan with charged biomolecules in small intestine can potentially decrease intestinal absorption and oral bioavailability. A similar influence of the ionic micelles can be noticed in the case of losartan, where SDS also favors protonation of the imidazole nitrogen (þ57%) and CTAB contributes to an almost complete deprotonation of tetrazole (þ43%). Higher displacement of the equilibria in the direction of the cationic form of irbesartan compared to losartan in the presence of the anionic micelles at pH 4.5 is probably due to the lower electron density on the imidazole nitrogen which is reduced by the electron withdrawing effect of the chlorine substituent. It can be observed that SDS reduces ionization of valsartan, moving the equilibria toward the molecular form (þ63%), while CTAB increases the tetrazole ionization by increasing the dianionic form (39%). Therefore, the presence of the positively charged biomolecules can decrease the content of the nonionized valsartan species which pass easily through the membranes, while the negatively charged ones increase this content. At pH 4.5, the
2451
contents of molecular forms of losartan and valsartan are increased under the influence of the nonionic surfactants. This finding indicates that an interaction of valsartan and losartan with the noncharged polar molecules can potentially affect an increase in the absorption and bioavailability of these drugs. Conclusion In this study, it was shown that the micelles of the differently charged surfactants shift the pKapp values of irbesartan, losartan, a and valsartan. In line with this finding, it can be assumed that microenvironment changes can significantly affect the protolytic equilibria of these compounds also in physiological conditions. Ionization of compounds in a biological system depends on polarity and charge of the environment and directly affects pharmacokinetic parameters and pharmacological behavior of drugs. Decreased ionization of the acidic groups and increased ionization of basic groups in the presence of the anionic SDS micelles, and an opposite effect on ionization in the presence of the cationic CTAB micelles point out to predominant electrostatic effects which involve ionizable groups of sartans and the charged surface of the ionic micelles. Based on a shift in protolytic equilibria in the presence of the nonionic micelles, it can be assumed that ionizable groups exert dipole interactions and form hydrogen bond, while the drug is predominantly retained in the palisade layer of the micelles. The aggregates of surfactants exhibit structural and functional properties of the cell membranes, so that the observed changes in ionization of sartans can point out to specific interactions with biological membranes and to localization in them. Sartans are the orally administered drugs which pass through many membranes, until they reach the target site of action. For this reason, this study can help defining the form of the drug which interacts with the transmembrane domain of the AT1 receptors. Moreover, the pKa* values valid for the methanol-water mixtures have an analytical significance and can be used to optimize chromatographic analysis in which the investigated compounds are analyzed by applying a mobile phase consisting of methanol as an organic modifier. Acknowledgments
Table 3 Percentage of the Equilibrium Forms of Sartans at the pH Values of Biopharmaceutical Importance in the Surfactant-Free and the Surfactant Containing Media Variable
Irbesartan H2O SDS CTAB Losartan H2O SDS CTAB TX-100 Brij 35
Valsartan H2O SDS CTAB TX-100 Brij 35
pH 1.2
pH 4.5
pH 6.8
pH 7.4
Cat
Mol An
Cat
Mol An
Cat
Mol An
Cat
100 100 99
0 0 1
0 0 0
11 92 0
47 7 4
42 0 96
0 1 0
1 15 0
99 0 84 0 100 0
0 4 0
100 96 100
100 100 85 100 94
0 0 15 0 6
0 0 0 0 0
6 63 0 11 1
48 36 4 77 62
45 1 96 12 37
0 0 0 0 0
1 14 0 3 1
99 86 100 97 99
0 0 0 0 0
0 4 0 1 0
100 96 100 99 100
Mol An
Mol MA
DA Mol MA
DA Mol MA
DA
Mol MA
DA
100 100 99 100 100
0 0 0 0 0
43 5 82 1 18
99 98 100 90 99
0 0 0 0 0
100 99 100 97 100
0 0 1 0 0
9 72 1 86 34
48 23 17 14 48
0 0 0 0 0
1 2 0 10 1
0 1 0 3 0
Equilibrium forms: Cat, cationic; An, anionic; Mol, molecular; MA, monoanionic; DA, dianionic.
This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, contract #172033. References 1. Lemke TL, Williams DA, Roche VF, Zito SW. Foye’s Principles of Medicinal Chemistry. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2013. 2. Doulton TW, He FJ, MacGregor GA. Systematic review of combined angiotensin-converting enzyme inhibition and angiotensin receptor blockade in hypertension. Hypertension. 2005;45(5):880-886. 3. Zoumpoulakis P, Daliani I, Zervou M, et al. Losartan’s molecular basis of interaction with membranes and AT1 receptor. Chem Phys Lipids. 2003;125(1): 13-25. 4. Fotakis C, Christodouleas D, Zoumpoulakis P, et al. Comparative biophysical studies of sartan class drug molecules losartan and candesartan (CV-11974) with membrane bilayers. J Phys Chem B. 2011;115(19):6180-6192. 5. Zervou M, Cournia Z, Potamitis C, et al. Insights into the molecular basis of action of the AT1 antagonist losartan using a combined NMR spectroscopy and computational approach. Biochimica et Biophysica Acta Biomembranes. 2014;1838(3):1031-1046. 6. Cairns D. Essentials of Pharmaceutical Chemistry. 4th ed. Great Britain: Pharmaceutical Press; 2012. 7. Manallack DT, Prankerd RJ, Nassta GC, Ursu O, Oprea TI, Chalmers DK. A chemogenomic analysis of ionization constantsdimplications for drug discovery. Chem Med Chem. 2013;8(2):242-255. 8. Manallack DT, Prankerd RJ, Yuriev E, Oprea TI, Chalmers DK. The significance of acid/base properties in drug discovery. Chem Soc Rev. 2013;42(2):485-496.
2452
M. Grujic et al. / Journal of Pharmaceutical Sciences 105 (2016) 2444-2452
9. Peters JU, Hert J, Bissantz C, et al. Can we discover pharmacological promiscuity early in the drug discovery process? Drug Discov Today. 2012;17(7):325-335. 10. Fendler JH. Atomic and molecular clusters in membrane mimetic chemistry. Chem Rev. 1987;87(5):877-899. 11. Garavito RM, Ferguson-Miller S. Detergents as tools in membrane biochemistry. J Biol Chem. 2001;276(35):32403-32406. 12. Mahiuddin S, Zech O, Raith S, Touraud D, Kunz W. Catanionic micelles as a model to mimic biological membranes in the presence of anesthetic alcohols. Langmuir. 2009;25(21):12516-12521. 13. Ghosh D, Chattopadhyay N. Equilibrium and dynamic effects on ligand binding to biomacromolecules and biomimetic model systems. Int Rev Phys Chem. 2013;32(3):435-466. 14. Pereira-Leite C, Carneiro C, Soares JX, et al. Biophysical characterization of the drugemembrane interactions: the case of propranolol and acebutolol. Eur J Pharm Biopharm. 2013;84(1):183-191. 15. Mirgorodskaya AB, Yackevich EI, Kudryashova YR, et al. Design of supramolecular biomimetic catalysts of high substrate specificity by noncovalent selfassembly of calix [4] arenes with amphiphilic and polymeric amines. Colloids Surf B Biointerfaces. 2014;1(117):497-504. 16. Rangel-Yagui CO, Pessoa Jr A, Tavares LC. Micellar solubilization of drugs. J Pharm Pharm Sci. 2005;8(2):147-163. 17. Varghese B, Al-Busafi SN, Suliman FO, Al-Kindy SM. Synthesis, spectroscopic characterization and photophysics of a novel environmentally sensitive dye 3naphthyl-1-phenyl-5-(4-carboxyphenyl)-2-pyrazoline. J Lumin. 2015;159:9-16. 18. Jaiswal PV, Ijeri VS, Srivastava AK. Effect of surfactants on the dissociation constants of ascorbic and maleic acids. Colloids Surf B Biointerfaces. 2005;46(1):45-51. 19. Popovi c MR, Popovi c GV, Agbaba DD. The effects of anionic, cationic, and nonionic surfactants on acidebase equilibria of ACE inhibitors. J Chem Eng Data. 2013;58(9):2567-2573. 20. Popovi c MR, Popovi c GV, Filipi c SV, Nikoli c KM, Agbaba DD. The effects of micelles of differently charged surfactants on the equilibrium between (Z)-and (E)-diastereomers of five ACE inhibitors in aqueous media. Monatsh Chem Chem Mon. 2015;146(6):913-921. 21. Miller JM, Beig A, Krieg BJ, et al. The solubilityepermeability interplay: mechanistic modeling and predictive application of the impact of micellar solubilization on intestinal permeation. Mol Pharm. 2011;8(5):1848-1856. 22. Mahajan S, Mahajan RK. Interactions of phenothiazine drugs with surfactants: a detailed physicochemical overview. Adv Colloid Interface Sci. 2013;199:1-4. 23. Morrow BH, Wang Y, Wallace JA, Koenig PH, Shen JK. Simulating pH titration of a single surfactant in ionic and nonionic surfactant micelles. J Phys Chem B. 2011;115(50):14980-14990. 24. Tosco P, Rolando B, Fruttero R, et al. Physicochemical profiling of sartans: a detailed study of ionization constants and distribution coefficients. Helv Chim Acta. 2008;91(3):468-482.
25. Cudina O, Brbori c J, Jankovi c I, Karljikovi c-Raji c K, Vladimirov S. Study of valsartan interaction with micelles as a model system for biomembranes. Colloids Surf B Biointerfaces. 2008;65(1):80-84. 26. Cagigal E, Gonzalez L, Alonso RM, Jimenez RM. pKa determination of angiotensin II receptor antagonists (ARA II) by spectrofluorimetry. J Pharm Biomed Anal. 2001;26(3):477-486. 27. Abdel-Fattah L, Abdel-Aziz L, Gaied M. Enhanced spectrophotometric determination of Losartan potassium based on its physicochemical interaction with cationic surfactant. Spectrochim Acta A Mol Biomol Spectrosc. 2015;136: 178-184. 28. Irving HM, Miles MG, Pettit LD. A study of some problems in determining the stoicheiometric proton dissociation constants of complexes by potentiometric titrations using a glass electrode. Anal Chim Acta. 1967;38: 475-488. 29. Albert A, Serjeant EP. The Determination of Ionization Constants. 3rd ed. London: Chapman and Hall; 1984. 30. Gans P, Sabatini A, Vacca A. Investigation of equilibria in solution. Determination of equilibrium constants with the HYPERQUAD suite of programs. Talanta. 1996;43(10):1739-1753. 31. Avdeef A. Absorption and Drug Development: Solubility, Permeability, and Charge State. Hoboken, NJ: John Wiley & Sons; 2012. bel R, Chmurzyn ski L. Potentiometric pKa determination of standard sub32. Wro stances in binary solvent systems. Anal Chim Acta. 2000;405(1):303-308. 33. Charton M. Electrical effects of ortho substituents in imidazoles and benzimidazoles. J Org Chem. 1965;30(10):3346-3350. 34. Mchedlov-Petrossyan NO, Kamneva NN, Kharchenko AY, Shekhovtsov SV, Marinin AI, Kryshtal AP. The influence of the micellar pseudophase of the double-chained cationic surfactant di-n-tetradecyldimethylammonium bromide on the absorption spectra and protolytic equilibrium of indicator dyes. Colloids Surf A Physicochem Eng Asp. 2015;476:57-67. 35. Mchedlov-Petrossyan NO, Vodolazkaya NA, Kamneva NN. Acid-base equilibrium in aqueous micellar solutions of surfactants. Micelles: Structural biochemistry, formation and functions and usage. New York: Nova Science Pub Inc; 2013:352. 36. Khossravi D. Drug-surfactant interactions: effect on transport properties. Int J Pharm. 1997;155(2):179-190. 37. Narang AS, Delmarre D, Gao D. Stable drug encapsulation in micelles and microemulsions. Int J Pharm. 2007;345(1):9-25. 38. Dar AA, Chat OA. Cosolubilization of Coumarin30 and Warfarin in cationic, anionic, and nonionic micelles: a micelleewater interfacial charge dependent FRET. J Phys Chem B. 2015;119(35):11632-11642. 39. Kumbhakar M, Goel T, Mukherjee T, Pal H. Role of micellar size and hydration on solvation dynamics: a temperature dependent study in Triton-X-100 and Brij-35 micelles. J Phys Chem B. 2004;108(50):19246-19254.