Chiral Interactions of the Drug Propranolol and α1 ... - ACS Publications

Article pubs.acs.org/ac

Chiral Interactions of the Drug Propranolol and α1-Acid-Glycoprotein at a Micro Liquid−Liquid Interface Paula Lopes and Ritu Kataky* Durham University, Department of Chemistry, South Road, Durham, DH1 3LE, U.K. ABSTRACT: The investigation of chiral interactions of drugs with plasma proteins is of fundamental importance for drug efficacy and toxicity studies. In this paper, we demonstrate a simple liquid−liquid interface procedure for investigating chiral interactions. Chiral discrimination of the enantiomers of a basic drug, propranolol, was achieved at a micro liquid− liquid interface, using α1-acid-glycoprotein (AGP) as a chiral acute phase plasma protein. When the protein is added to an aqueous phase containing the enantiomers of propranalol hydrochloride, the binding of (S)- and (R)-propranolol hydrochloride to the protein results in a decrease in the cyclic voltammetry (CV) and differential pulse voltammetry (DPV) current responses corresponding to the decrease in transfer of propranolol at an aqueous−1,2-dichloroethane interface. This decrease is a consequence of the complexation of the drug and the protein. The complex drug−protein does not transfer across the interface nor changes the transfer potential of the uncomplexed form of propranolol enantiomers. The bound concentration of propranolol enantiomers in the presence of AGP was found to be greater for (S)-propranolol than (R)-propranolol for solutions containing constant concentrations of AGP (50 μM). Scatchard analysis yielded association constants of 2.7 and 1.3 × 105 M−1 for (S)- and (R)-propranolol, respectively.

C

one of the major acute phase8 proteins of human blood. Thus changes in the level of AGP in plasma during physiological and pathological conditions can have a profound effect on drug disposition and pharmacological activity.12,13 The nature of drug binding to AGP has been characterized mainly, by hydrophobic interactions, due to the hydrophobic residues near the AGP binding site. Hydrogen bonding, van der Waal’s forces, and electrostatic interactions also play a significant role in the binding interactions.1,2,5,9−11 Additionally, it has been demonstrated that AGP has the ability to stereoselectively bind enantiomers of widely different character.14−16 The binding interaction takes place in the chiral sites of the protein, which appear to be located both in the peptide chain and in the carbohydrate units.14 This peptide chain is composed of several different chiral groups, beside the hydrophobic groups, hydrogen bonding amides, and anionic and cationic groups. On the other hand, the carbohydrate units are built up sialic acid, hexosamine and neutral hexoses, and all containing chiral carbons, which are involved in the retention of enantiomers.17−19 Enantioselective analytical techniques20 have become increasingly important in the field of drug analysis in the last decades. A very popular way of performing enantioselective analysis is by the use of chiral stationary phases (CSPs) in highperformance liquid chromatography (HPLC). The CSPs may be divided into categories according to the type of chiral selector. One category comprises a group of phases where an immobilized protein is used as a chiral selector, one of the most successful ones being α1-acid-gycoprotein on silica, which was first introduced by Hermansson.18,19 The binding of drugs to

hirality is an intrinsic property of many molecules, and the investigation of chiral interactions are crucially important for a diverse range of applications such as food flavors, cosmetics, environmental pollution, the clinical, medical, and pharmaceutical industries, and metabolism of living systems, among others. The binding of chiral drugs to serum proteins is of particular interest since it affects the transportation and distribution of pharmaceutical agents in the body.1,2 Albumin, α1-acidgycoprotein, and lipoproteins, are three major proteins to which drugs bind in plasma.1 Changes in drug binding by plasma proteins affect the plasma unbound drug concentrations and, therefore, influence the pharmacologic effects caused by the drug.2,3 Human serum albumin (HSA) in plasma binds mainly acidic drugs whereas α1-acid-gycoprotein (AGP) is a primary carrier of basic drugs,7,8 such as propranolol (β-adrenergic receptor blockers), imipramine (antidepressants), and lidocaine (local anesthetics)2−4 underscoring its importance as a drugbinding transport protein.5 The human plasma concentration of AGP is about 1 g/L.1,6 Serum AGP, also called orosomucoid, is a negatively charged (pI = 2.7−3.2)7 acidic (pKa = 2.6)8 glycoprotein, due to the presence of sialic acids (12% of the carbohydrate moiety). It is highly heterogeneous, extensively glycoslated (Mw ≈ 41 000), with a carbohydrate content of 45% (w/w). It is composed of a single polypeptide chain of 183 amino acids with up to five carbohydrate moieties attached to the protein core via five N-linked glycans.5,7−9 It has been estimated that there are 12−2010 different forms of AGP in serum due to variations in its amino acid sequence and the types and numbers of carbohydrate groups attached to its polypeptide chain. These carbohydrate moieties are thought to be located on the outside of this protein, giving it a hydrophobic core and a hydrophilic exterior.11 As a member of the lipocalin protein family, AGP is © 2012 American Chemical Society

Received: November 6, 2011 Accepted: January 17, 2012 Published: January 17, 2012 2299

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304

Analytical Chemistry

Article

AGP has been examined by a wide range of techniques,1,12,20,26−29 including HPLC,5,21 circular dichroism,21 equilibrium dialysis and liquid chromatography,8,31 and fluorescence spectrophotometry.22 Stereoselective interactions between chiral molecules have tremendous importance both for drug discovery and biochemical processes. Using electrochemistry at a liquid−liquid interface can add valuable information to these studies. However, very few reports exist on interaction of chiral ions and molecules at the liquid−liquid interface. The authors have, previously, reported the detection and discrimination of a chiral drug in the presence of a chiral stationary phase.23 Scholz et al.24,25 have shown the differences of the Gibbs energies of solvation of chiral ions in chiral solvents. The transfer of racemic (±)-propranolol hydrochloride (Figure 1) has already been studied at the liquid−liquid interface.26−28

lithium chloride (LiCl, 10 mM), and the aqueous reference solution was made of LiCl (10 mM) and BTPPACl (1 mM). All the experiments were performed using a Compactstat (Ivium Technologies, Eindhoven, The Netherlands). The micropipets were made from borosilicate glass capillaries (o.d./i.d., 1.5/1.17 mm) obtained from Harvard Apparatus Ltd., using a laser-based pipet puller model P-97 (Sutter Instrument Co. Novato). The proper choice of the pulling parameters was critical to make a pipet with a short shank and flat orifice. An environmental scanning electron microscope (ESEM) was used to examine the quality and measure the diameter of the micropipets before and after the measurements. The experiments were performed with an organic solution inside the pipet; its inner wall was silanized to render it hydrophobic. This was done by filling the pipets with a trimethylchlorosilane solution. The solution was removed after 30 min, and the pipets were allowed to dry in the air overnight.38 The CV and DPV responses presented were obtained with micropipets of similar size, 10 ± 1 μm radius, Figure 2.

Figure 1. Structure of (±)-propranolol hydrochloride (±)-1-isopropylamino-3-(1-naphthyoxy)-2-propranolol hydrochloride (pKa = 9.45− 9.53).47,48

Fantini et al.26 studied the influence of a gel in the water phase in a macro water−1,2-dichloroethane interface, and Collins et al.27,28 reported the detection of (±)-propranolol in artificial saliva and in the presence of HSA at a micro-ITIES array. In this article we report a simple and sensitive method for the determination of chiral interactions of propranolol enantiomers with AGP, at a micro liquid−liquid interface, supported at the tip of a micropipet offering the advantages of microelectrode measurements.29−34 Protonated (S)- and (R)-propranolol enantiomers were examined at physiological pH (pH = 7.4) using cyclic voltammetry (CV) and differential pulse voltammetry (DPV) in the presence of α1-acid-glycoprotein. It is known that propranolol binds to AGP in blood with a high association constant.2,22,35,36

■

EXPERIMENTAL SECTION Chemicals. All chemicals were reagent grade and used as received unless indicated otherwise. All aqueous solutions were prepared with deionized water from a Milli-Q system (Watford, U.K.). 1,2-Dichloroethane was HPLC grade (Aldrich) and is known as a suspect carcinogen, so it was handled with all necessary precautions in order to avoid inhalation and skin contact. The organic-phase supporting electrolyte salt bis(triphenylphosphoranylidine) ammonium tetrakis(4-chlorophenyl)borate (BTPPATPBCl, 10 mM) was prepared by metathesis37 of equimolar quantities of the corresponding salts, bis(triphenylphosphoranylidene) ammonium chloride (BTPPACl, Aldrich), and potassium tetrakis(4-chlorophenyl-borate) (KTPBCl, Fluka) in a minimum amount of a 2:1 methanol/ water mixture. The resulting precipitate was filtered and recrystalized from acetone before use. (S)-(−)-Propranolol hydrochloride, (R)-(+)-propranolol hydrochloride, α1-acidglycoprotein from bovine plasma, and trimethylchlorosilane were purchased from Sigma. The aqueous phase electrolyte was

Figure 2. ESEM pictures of the micropipets (a) before and (b) after performing the ion transfer voltammetry measurements.

The electrochemical cell used was a customized two-electrode cell, where the interfacial potential difference was applied between two reference electrodes. Both of them were 0.25 mm diameter Ag wires coated with AgCl (prepared by potentiostatic oxidation of silver wires in a 3 M solution of KCl). One reference electrode was inserted into the pipet from the back, in contact with the aqueous reference solution, and another one was immersed into the external solution, containing the analyte. A simple two electrode arrangement can be used as nanoamp levels of current are measured at the microelectrode tip 2300

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304

Analytical Chemistry

Article

Scheme 1. Cell Configuration Useda

a

PH+ is the protonated drug cations, (S)- and (R)-propranolol, and χ = 100, 87.5, 75.0, 62.5, 37.5, 20.0, and 10.0 μM.

minimizing significant ohmic drop and charging currents.33,39−41 The waveform parameters used for all the differential pulse voltammetry (DPV) experiments were as follows: pulse amplitude, 0.05 V; sampling width, 0.060 s; step height, 5 mV. According to the interface polarization convention adopted in the experimental setup, the transfer of ionic species from the aqueous to the organic phase is defined as a positive current. A background current response was obtained using a cell without propranolol in the outer phase. All the experiments were carried out inside a Faraday cage at room temperature (21 ± 2 °C). The transfer of propranolol enantiomers across the micro liquid−liquid interface was studied using the cell shown in Scheme 1. In order to investigate the effect of the chiral selector on the propranolol enantiomers, 50 μM of AGP was added to the aqueous phase. The drug−protein mixed solutions were prepared in potassium phosphate buffer (pH = 7.4).

to transfer and ingress of the propranolol to the organic phase and a peak-shaped response on the reverse scan, corresponding to the transfer and egress of the propranolol from the organic phase to the aqueous phase. This type of diffusion regime was pioneered by Shao et al.31 who observed a pseudosteady state wave (spherical diffusion) when the ion transfer process was controlled by the species entering the micropipet, and a peakshaped voltammogram (linear diffusion) occurred when the transfer process was controlled by the species leaving the micropipet. Therefore, the apparent steady-state in the forward scan and the peak observed in the reverse scan are due to the transfer of the protonated (S)- and (R)-propranolol cations, consistent with the different diffusion fields on either side of the microinterface.31,39 The standard potential of transfer of the (S)- and (R)propranolol cations were determined from the CV experiments and referred to an internal reference ion, TEA+, according to eq 1.

■

RESULTS AND DISCUSSION Initial experiments were focused toward establishing whether the effect of AGP on the potential window of the background electrolyte, which was approximately 600 mV, was limited by the background electrolyte transfers at extremes of the potentials (dotted line, Figure 3). We found no marked effect on the

E1/2 − Δow ϕ1o′ = E1/2 − Δow ϕoTEA+′ TEA+

(1)

o where Δo wϕTEA ′ = 52 mV,43,44 is the standard transfer potential of TEA+ across the water−1,2-dichloroethane interface. E1/2 and 1/2 + ETEA are the experimental half-wave potentials of propranolol cations and TEA+, respectively. 0,w→o A value for the free Gibbs energy of transfer, ΔGtr,i , can then be calculated using eq 2,

→o ΔGtr0,w = zlF × Δow ϕio′ ,i

(2)

where zi stands for the charge of the transferring species and F is the Faraday constant. The standard potential of the propranolol enantiomers, Δowϕio′, was found to be the same and equal to 197 ± 4 mV, and the free Gibbs energy of transfer is equal to 19.0 ± 0.4 kJ mol−1. These values are in good agreement with values already report1/2+ 0,w→o ed in the literature (Δowϕio′ = 202 ± 5 mV vs ETEA , ΔGtr,i = −1 26 12.6 ± 0.5 kJ mol ). According to Beattie et al.40, the microinterface supported at a micropipet tip displays a mass transport behavior and current response similar to a microelectrode, according to following equation,

Figure 3. CV of the supporting electrolytes (background) (dotted line) and with AGP (dashed line) at physiological pH.

potential window when the protein is added, meaning that there is no significant contribution to the increasing of the charging current (dashed line, Figure 3). This is corroborated by the findings of Vanysek and Sun,42 where in the study of bovine serum albumin adsorption at the water−nitrobenzene interface they showed that the capacitance increases for pHs below the isoelectric point and that at more basic pHs, the opposite occurs. In order to study the interactions between AGP and (S)- and (R)-propranolol at the liquid−liquid interface, transfer measurements were performed with and without the AGP in the aqueous phase. Initially, the transfer of the two enantiomers of propranolol was characterized in the absence of AGP (Figure 4), using both CV and DPV, at pH 7.4 (propranolol hydrochloride is 90% ionized at this pH). Linear concentration dependence was observed with a limit of detection of 10 μM using DPV. CVs of propranolol enantiomers produced an asymmetric shape with an apparent steady-state on the forward scan corresponding

Iss = A × πziaFDC

(3)

where Iss is the limiting current, D the diffusion coefficient, C the bulk concentration of the transferring species, a the radius of the micropipet tip (10 ± 1 μm), and the empirical factor A is 4, assuming a disk-shaped ITIES.38,39 The diffusion coefficient of (S)- and (R)-propranolol was determined using eq 3, and the limiting currents from the CV in Figure 4c,d) were found to be 1.34 × 10−6 ± 0.1 and 1.26 × 10−6 ± 0.1 cm2 s−1, respectively. The calculated values are in agreement with those reported by Fantini et al.26 (5 ± 5 × 10−6 ± 0.1 cm2 s−1) for a nongellified interface. It was expected that on addition of AGP to the aqueous phase, the enantiomers of propanolol hydrochloride would enantioselectively bind to AGP and influence the transfer to the nonaqueous phase. Accordingly, a constant concentration (50 μM) 2301

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304

Analytical Chemistry

Article

Figure 4. Transfer of propranolol enantiomers in PBS buffer (pH7.4) at the micro ITIES: (a and b) DPV (background subtracted) of increasing concentrations of (R)- and (S)-propranolol, respectively. Response for 0.010 (light gray), 0.025, 0.0375, 0.050, 0.0625, 0.0750, 0.0875, and 0.10 mM (black). Inset: calibration curve of peak current vs concentration. (c and d) CV (scan rate, 10 mV s−1) of increasing concentrations of (R)- and (S)propranolol. Response for 0.0375 (light gray), 0.050, 0.0625, 0.0750, 0.0875, 0.10 mM (black). Inset: calibration curve of steady-state current vs concentration. The steady-state currents in the calibration curve were obtained by subtraction of the currents at the potentials 0.2 and 0.6 V.

Figure 5. (a and b) DPV (background subtracted) and (c and d) CV of (R)- and (S)-propranolol, respectively, with 0.050 mM of AGP in solution. Response for 0.0375 (light gray), 0.050, 0.0625, 0.0750, 0.0875, and 0.10 mM (black).

Scheme 2. Subscripts (w) and (o) Refer to the Aqueous and Organic, Respectively, and “Uncomp” Refers to the Uncomplexed Propranolol

DPV responses was observed upon addition of AGP to the aqueous phase (Figure 5), due to the binding of AGP, which suppresses the transfer of the propranolol enantiomers across

of AGP was added to the aqueous solution and its effect studied under the same conditions (PBS buffer, pH 7.4), with CV and DPV. A significant decrease in current in the CV and 2302

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304

Analytical Chemistry

Article

the microinterface. As represented in Scheme 2, only the uncomplexed propranolol is able to transfer across the interface. Similar findings were obtained by Horrocks et al.33 in the binding study of a cation, N-methylphenanthroline, to DNA. In one of the approaches developed, where the cation is initially present in the aqueous phase, they also found upon the addition of high molecular weight DNA to the aqueous phase that the concentration of free cation decreases, which resulted in a decrease in the ion transfer current. In this study, the binding appears to be stronger in the case of the (S)-propranolol enantiomer (Figure 5b,d), where the peak in DPV and the apparent steady-state in CV are clearly of lower intensity, compared to the ones for (R)-propranolol in Figure 5b,d. The DPV response provided a more sensitive detection signal for lower concentrations of propranolol enantiomers. It was observed that that the peak current intensity for the DPV and the apparent steady-state for the CV, in the presence of AGP, was on average about 33% and 52% times smaller for (S)and (R)-propranolol, respectively, when compared to the measurements without AGP. At lower concentrations of propranolol, i.e., 10 and 25 μM, 100% complexation of propranolol to AGP is possible resulting in the current being suppressed. These values were therefore not included in the graphs in Figure 5. It is known that the interaction of most drugs with the plasma proteins is a dynamic, reversible process with dissociation of bound drug molecules from the drug−protein complex occurring very rapidly, probably within milliseconds or less.45 Quantitatively, the binding of drugs to plasma proteins may be described by the law of mass action and analyzed in terms of the number of binding sites (n) and the apparent association constant (Ka), a measure of affinity for the interaction of the drug and protein. These two binding constants combine to regulate the fraction of unbound drug available. For an equilibrium system, at the binding sites, bound and free drug can be calculated according to the Scatchard model when the drug binds to a single receptor population. If it binds to more than one type of site or if there is co-operativity in binding, the result will be a curved Scatchard plot. The Scatchard plot is defined by eqs 4 and 5,

r = nK a − K ar Cf r=

Cb Cp

Figure 6. (a) Scatchard plot of the interaction between (S)- and (R)propranolol with AGP. [AGP] = 50 μM, [S- and [R-propranolol] = 37.5, 50.0, 62.5, 75.0, 87.5, and 100.0 μM. (b) Concentration of (S)and (R)-propranolol bound for different [propranolol]/[AGP] ratios.

due to the chiral interactions of AGP with the propranolol enantiomers. The number of binding sites, n, on AGP molecules is less than unity for both enantiomers (Table 1). A value of n < 1 probably represents the heterogeneity of the binding sites. Table 1. Binding Parameters of S- and R-Propranolol with α1-Acid Glycoprotein at pH = 7.4

(S)-propranolol (R)-propranolol

Ka (× 105 M−1)

n

nKa (× 105 M−1)

Δ(ΔG) (kJ mol−1)

2.7 ± 0.1 1.3 ± 0.1

0.73 0.50

1.97 ± 0.1 0.65 ± 0.1

−2.9

The chiral free energy of binding between (S)- and (R)propranolol with AGP can also be related using following equations:

(4)

(5)

where r represents the number of moles of bound propranolol (Cb) per mole of AGP, n can represent the number of binding sites on the AGP molecule, Ka is the association constant, Cf is the concentration of free propranolol, and Cp is the total concentration of AGP. The Scatchard plots for (S)- and (R)-propranolol which relate the value of r/Cf to the value of r are presented in Figure 6a. A plot of the amount of drug bound to the protein at different ratios is presented in Figure 6b. From these plots, the binding parameters of Ka and n, for the interactions of (S)- and (R)propranolol with AGP, can be easily obtained using eq 4. The difference in the binding parameters, 2.7 × 105 M−1, n = 0.73, and 1.3 × 105 M−1, n = 0.50, for (S)- and (R)-propranolol clearly shows a chiral selective binding for the two enantiomers with AGP, with the (S)- enantiomer having a higher affinity compared to (R)-propranolol using the same [propranolol]/[AGP] ratios (Figure 6a,b). The difference obtained in the affinity parameters is

ΔG(S·AGP) = − RT ln K a(S·AGP)

(6)

ΔG(R·AGP) = − RT ln K a(R·AGP)

(7)

where Ka(R·AGP) and Ka(R·AGP) are the stability constants of the complexes formed (calculated from the Scatchard plot), R the gas constant (8.314 J mol−1 K−1) and T, the temperature (K). ΔK(S·AGP) and ΔK(R·AGP) are the free energies of the reactions 1′ and 2′ (Scheme 2), respectively. The enantioselectivity of the chiral selector is given by the difference between the free energies of reactions 1′ and 2′:

Δ(ΔG) = ΔG(S·AGP) − ΔG(R·AGP)

(8)

The estimated binding parameters, the total binding affinity (nKa), and stereoselectivity, Δ(ΔG)chiral, are presented in Table 1. The association constant of (S)-propranolol to AGP, Ka(S·AGP), is about 2 times bigger than that of Ka(R·AGP). The calculated values are in good agreement with the literature, where it was found that (S)-propranolol is significantly more strongly bound to AGP than (R)-propranolol.2,22,36 The values obtained for the 2303

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304

Analytical Chemistry

Article

(15) Hermansson, J.; Hermansson, I. J. Chromatogr., A 1994, 666, 181−191. (16) de Vries, J. X.; Schmitz-Kummer, E. J. Chromatogr., A 1993, 644, 315−320. (17) Li, S.; Lloyd, D. K. Chirality 1993, 65, 3684−3690. (18) Hermansson, J.; Ström, K.; Sandberg, R. Chromatographia 1987, 24, 520−526. (19) Hermansson, J.; Grahn, A. J. Chromatogr., A 1995, 694, 57−69. (20) Izake, E. L. J. Pharm. Sci. 2007, 96, 1659−1676. (21) Oravcová, J.; Bystricky, S.; Trnovec, T. Biochem. Pharmacol. 1989, 38, 2575−2579. (22) Zhang, F.; Du, Y.; Ye, B.; Li, P. J. Photochem. Photobiol. B 2007, 86, 246−251. (23) Kataky, R.; Lopes, P. Chem. Commun. 2009, 1490−2. (24) Scholz, F.; Gulaboski, R.; Mirčeski, V.; Langer, P. Electrochem. Commun. 2002, 4, 659−662. (25) Scholz, F.; Gulaboski, R. Faraday Discuss. 2005, 129, 169. (26) Fantini, S.; Clohessy, J.; Gorgy, K.; Fusalba, F.; Johans, C.; Kontturi, K.; Cunnane, V. J. Eur. J. Pharm. Sci. 2003, 18, 251−257. (27) Collins, C. J.; Arrigan, D. W. M. Anal. Chem. 2009, 81, 2344−9. (28) Collins, C. J.; Lyons, C.; Strutwolf, J.; Arrigan, D. W. M. Talanta 2010, 80, 1993−1998. (29) Wilke, S.; Osborne, M. D.; Girault, H. H. J. Electroanal. Chem. 1997, 436, 53−64. (30) Lee, H. J.; Beriet, C.; Girault, H. H. J. Electroanal. Chem. 1998, 453, 211−219. (31) Shao, Y.; Osborne, M. D.; Girault, H. H. J. Electroanal. Chem. 1991, 318, 101−109. (32) Liu, S.; Li, Q.; Shao, Y. Chem. Soc. Rev. 2011, 40, 2236−2253. (33) Horrocks, B. R.; Mirkin, M. V. Anal. Chem. 1998, 70, 4653− 4660. (34) Wickens, J.; Dryfe, R. A. W.; Mair, F. S.; Pritchard, R. G.; Hayes, R.; Arrigan, D. W. M. New J. Chem. 2000, 24, 149−154. (35) Mfiller, W. E.; Stillbauer, A. E. J. Pharmacol. Exp. Ther. 1983, 322, 170−173. (36) Albani, F.; Riva, R.; Contin, M.; Baruzzi, A. Br. J. Clin. Pharmacol. 1984, 18, 244−246. (37) Amemiya, S.; Bard, A. J. Anal. Chem. 2000, 72, 4940−4948. (38) Shao, Y.; Mirkin, M. V. Anal. Chem. 1998, 70, 3155−3161. (39) Liu, B.; Mirkin, M. V. Electroanalysis 2000, 12, 1433−1446. (40) Beattie, P. D.; Delay, A.; Girault, H. H. J. Electroanal. Chem. 1995, 380, 167−175. (41) Beattie, P. Electrochim. Acta 1995, 40, 2961−2969. (42) Vanýsek, P.; Sun, Z. Bioelectrochem. Bioenerg. 1990, 298, 177− 194. (43) Zhang, M.; Sun, P.; Chen, Y.; Li, F.; Gao, Z.; Shao, Y. Chin. Sci. Bull. 2003, 48, 1234−1239. (44) Reymond, F.; Chopineaux-Courtois, V.; Steyaert, G.; Bouchard, G.; Carrupt, P.-A.; Testa, B.; Girault, H. H. J. Electroanal. Chem. 1999, 462, 235−250. (45) Lindup, W. E.; Orme, M. C. L. Br. Med. J 1981, 282, 212−214. (46) Becker, B. a; Larive, C. K. J. Phys. Chem. B 2008, 112, 13581− 13587. (47) Hinderling, P. H.; Hartmann, D. Ther. Drug Monit. 2005, 27. (48) Caron, G.; Steyaert, G.; Pagliara, A.; Reymond, F.; Crivori, P.; Gaillard, P.; Carrupt, P.-A.; Avdeef, A.; Comer, J.; Box, K. J.; Girault, H. H.; Testa, B. Helv. Chim. Acta 1999, 82, 1211−1222.

number of binding sites, n, of 0.50 and 0.73 for (R)- and (S)propranolol, respectively, are further supported by saturation transfer difference (STD) experiments, reported by Becker and Larive.46 They found that the STD maps provided information about the different orientations of the enantiomers of propranolol with respect to the AGP binding pocket. Their experiments showed although that both enantiomers interact with the protein on proton H4, with stronger hydrophobic interactions for the (R)-propranolol methyl protons, a higher level of saturation was found for the (S)-enantiomer, with stronger interactions on the proton at the chiral center and the two adjacent alkyl protons (H1, H2, H3), Figure 1.

■

CONCLUSIONS This study shows a simple and effective method for studying chiral interactions of drugs and proteins. We have shown that using a micro liquid−liquid interface, the stereoselective binding of propranolol enantiomers to AGP can be quantitatively evaluated. Both enantiomers bind to AGP with different affinities, which were reflected in the values of association constants and number of bindings. The (S)-propranolol has a higher association constant (2.7 × 105 ± 0.1 M−1) and a total binding affinity about 1.3 times stronger than that of (R)propranolol. We illustrate the chiral transfer of the drug using an acute phase protein, AGP, as a chiral selector. This work has significance not only for the chiral detection and separation but in a wider context for evaluation of drug−protein interactions in a simple manner.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected].

■

ACKNOWLEDGMENTS This work was funded by EPSRC/RSC/AD and Analytical Chemistry Trust Funds (ACTF).

■

REFERENCES

(1) Tomlin, M. E. Pharmacology and Pharmacokinetics; SpringerVerlag London Limited: London, 2010. (2) Hanada, K.; Ohta, T.; Hirai, M.; Arai, M.; Ogata, H. J. Pharm. Sci. 2000, 89, 751−757. (3) Kishino, S.; Itoh, S.; Nakagawa, S.; Miyazaki, S. Eur. J. Clin. Pharmacol. 2001, 57, 583−587. (4) Verbeeck, R.; Cardinal, J.; Wallace, S. Eur. J. Clin. Pharmacol. 1984, 27, 91−97. (5) Israili, Z. H.; Dayton, P. G. Drug Metab. Rev. 2001, 33, 161−235. (6) Zsila, F.; Molnár, P.; Deli, J.; Lockwood, S. F. Bioorg. Chem. 2005, 33, 298−309. (7) Hochepied, T.; Berger, F. G.; Baumann, H.; Libert, C. Cytokine Growth Factor Rev. 2003, 14, 25−34. (8) Fournier, T.; Medjoubi-N, N.; Porquet, D. Biochim. Biophys. Acta 2000, 1482, 157−171. (9) Yoshima, H.; Matsumoto, A.; Mizuochi, T.; Kawasaki, T.; Kobata, A. J. Biol. Chem. 1981, 256, 8476−8484. (10) Xuan, H.; Hage, D. S. Anal. Biochem. 2005, 346, 300−310. (11) Aubert, J.-P.; Biserte, G.; Loucheux-Lefebvre, M.-H. Arch. Biochem. Biophys. 1976, 175, 410−418. (12) Dijk, W. V.; Havenaar, E. C.; Linden, E. C. M. B.-V. D. Glycoconjugate J. 1995, 233, 227−233. (13) Kuroda, Y.; Shibukawa, A.; Nakagawa, T. Anal. Biochem. 1999, 268, 9−14. (14) Enquist, M.; Hermansson, J. J. Chromatogr., A 1990, 519, 271− 283. 2304

dx.doi.org/10.1021/ac2029425 | Anal. Chem. 2012, 84, 2299−2304