CHIRALITY (2014)

Maltodextrins as Chiral Selectors in CE: Molecular Structure Effect of Basic Chiral Compounds on the Enantioseparation HADI TABANI, ALI REZA FAKHARI,* AND SAEED NOJAVAN Department of Pure Chemistry, Faculty of Chemistry, Shahid Beheshti University, G. C., P.O. Box 19396-4716, Evin, Tehran, I.R. Iran

ABSTRACT Prediction of chiral separation for a compound using a chiral selector is an interesting and debatable work. For this purpose, in this study 23 chiral basic drugs with different chemical structures were selected as model solutes and the influence of their chemical structures on the enantioseparation in the presence of maltodextrin (MD) as chiral selector was investigated. For chiral separation, a 100-mM phosphate buffer solution (pH 3.0) containing 10% (w/v) MD with dextrose equivalent (DE) of 4-7 as chiral selector at the temperature of 25°C and voltage of 20 kV was used. Under this condition, baseline separation was achieved for nine chiral compounds and partial separation was obtained for another six chiral compounds while no enantioseparation was obtained for the remaining eight compounds. The results showed that the existence of at least two aromatic rings or cycloalkanes and an oxygen or nitrogen atom or –CN group directly bonded to the chiral center are necessary for baseline separation. With the obtained results in this study, chiral separation of a chiral compound can be estimated with MD-modified capillary electrophoresis before analysis. This prediction will minimize the number of preliminary experiments required to resolve enantiomers and will save time and cost. Chirality 00:000–000, 2014. © 2014 Wiley Periodicals, Inc. KEY WORDS: capillary electrophoresis; chiral separation; maltodextrin; prediction INTRODUCTION

Despite significant developments over the last three decades, separation of enantiomers still remains one of the hot topics of separation science. In order to address the question of possible enantiomeric differences with respect to activity, disposition, or toxicity of racemic drugs or to further elucidate enantioselective mechanisms, the availability of stereoselective analysis methods is essential. A broad range of enantioselective analysis methods has been developed using different methodologies. Chromatographic methods such as gas chromatography (GC),1,2 high-performance liquid chromatography (HPLC),3–5 supercritical fluid chromatography (SFC),6,7 and thin-layer chromatography (TLC)8,9 have been developed using different chiral separation principles. More recently, capillary electrophoresis (CE)10–13 has been shown to be a powerful alternative to chromatographic methods. One of the successful application areas of CE is in pharmaceutical analysis, where the focus is on relatively small synthetic pharmaceutical compounds. Among various applications, separation and quantitation of enantiomers are key applications of CE in drug analysis. CE enantiomer separation has several attractive features such as simple and fast method development in addition to the above-mentioned advantages. Many reviews and books have been published for enantiomer separation by CE.14–20 These advantages make CE a firm candidate for a variety of separations (particularly those involving compound families). A large number of chiral selectors are currently available, especially prominent among which are cyclodextrins (CDs), chiral crown ethers, and linear polysaccharides. The efficient chiral selector for enantioseparation is usually selected by trial and error, which is expensive and time-consuming. However, there are some reports that predicted separation capabilities of chiral selectors and provided criteria for their choice. Their results showed that molecular size, charge, and the presence of specific functional groups or substructures in the analytes are very important. © 2014 Wiley Periodicals, Inc.

Blanco and Valverde predicted the use of various CDs for enantioresolving of some acidic and basic drugs.21 They found that analytes with an aromatic ring bearing a single or no substituent could be resolved with α-CD that has the smallest cavity. The β-CD could resolve analytes with a doubly substituted aromatic ring or two fused aromatic rings, while γ-CD could resolve analytes with three or four fused rings or analytes with a single extensively substituted ring. Substituted aromatic rings exhibit steric hindrance that restricts inclusion into the β-CD cavity and they can be better enantioresolved with γ-CD. Gubitz and Schmid showed that 2-hydroxypropyl-β-CD (HP-β-CD) could enantioresolve analytes possessing some aromatic rings in addition to a hydroxyl group on the chiral carbon.22 Anionic charged modified CDs such as carboxymethyl-β-CD, sulfated-β-CD, and sulphobutylether-β-CD could resolve neutral or cationic analytes (particularly those having some aromatic rings).23–25 Kuhn showed that crown ethers could discriminate enantiomers of analytes bearing a primary amine. Primary amines are held inside the cavity via three hydrogen bonds in a tripod-like arrangement.26 Stalcup and Agyei introduced heparin as a suitable chiral selector for analytes, which have at least two nitrogens with one of them incorporated in a heterocyclic aromatic ring.27 Also, some regression or computational models such as artificial neural networks (ANNs),28 quantitative structure property relationship (QSPR),29,30 quantitative structure binding relationships (QSBR),31 particle swarm optimization (PSO), and support vector machines (SVMs)32 approaches were used for predicting the thermodynamic parameters for the inclusion complexation of chiral guests with CD. *Correspondence to: Ali Reza Fakhari, Department of Pure Chemistry, Faculty of Chemistry, Sahid Beheshti University, G. C., P.O. Box 19396-4716, Evin, Tehran, I.R. Iran. E-mail: [email protected] Received for publication 15 January 2014; Accepted 6 May 2014 DOI: 10.1002/chir.22344 Published online in Wiley Online Library (wileyonlinelibrary.com).

TABANI ET AL.

Although, for most chiral selectors available, the attempts of prediction have failed until the present, recognition mechanisms involved in the chiral separation are complex and not always clearly known. Another group of chiral selectors are maltodextrins (MDs) but there is no report for prediction of chiral separation with this type of chiral selector. MDs are complex malto-oligo and polysaccharide mixtures, which were obtained from starch by partial acid and/or enzymatic hydrolysis.33,34 MDs are characterized by the average degree of polymerization (DP) (defined as the number of saccharide monomers within the oligo- or polysaccharide chain) and the dextrose equivalent (DE) (defined as the percent reducing sugars calculated as glucose on a dry substance basis); a high DE indicates that the oligomeric chains in a mixture are short. MDs in general show a wide enantioselectivity, probably due to their helical structure.35 A full turn in the helix requires at least six glucose units. It is believed that different energies in the interaction between the enantiomers and MD can lead to the enantioseparation. D’Hulst and Verbeke first introduced MDs to CE enantiomer separation in 1992.36 They further investigated the enantiomer separation of three coumarinic anticoagulant drugs and two chlorinated derivatives employing MD with DE 2.37 An extensive study of the chiral selector MD with DE 2 was also performed.38 The effect of various parameters such as pH of background electrolyte (BGE), concentration of the BGE, etc., on the enantioseparation was evaluated by the same group.34 Soini et al. used a range of MDs (Dextrin 10, Dextrin 15 and Dextrin 20) to separate four acidic drugs and three basic drugs.39 High molecular mass dextrins (ca. 3700 to >10 000) were employed by Nishi et al. for the enantiomer separations of more than 40 drugs, including both acidic and basic drugs.40 However, in these reports the effect of chemical structures of drugs on the enantioseparation was not investigated. Prediction of chiral separation for a compound with a chiral selector is an interesting and debatable work. However, the recognition mechanisms involved in chiral recognition are complex and not always clearly known, and therefore the selection of the appropriate selector is usually done by a trial-and-error approach, which is highly costly and too timeconsuming. Consequently, in this work it was decided to explore further and in more general terms the chiral separation probability with MDs. For this purpose, 23 chiral basic drugs with different chemical structures were used as model solutes, and enantioseparation with MD-modified CE was investigated. With the aim of obtaining results in this study, chiral separation of a chiral basic compound can be estimated with MD-modified CE before analysis. This prediction will minimize the number of preliminary experiments required to resolve enantiomers and will cause savings in time and cost.

from Sigma–Aldrich (St. Louis, MO). Trihexyphenidyl (THP), oxyphencyclimine (OXY), bromodiphenhydramine (BDH), atenolol (ATE), oxomemazine (OXO), propranolol (PRO), viloxazine (VIL), verapamil (VP), and norverapamil (NVP) were obtained from Darou Pakhsh Pharmaceutical company (Tehran, Iran) and were used without further purification. MDs (DE: 4–7, 13–17, and 16.5–19.5) were purchased from Fluka (Buchs, Switzerland). H3PO4, NaH2PO4.2H2O, Na2HPO4, NaOH, and HCl were purchased from Merck (Darmstadt, Germany). HPLC grade ® water was obtained through a Milli-Q system (Millipore, Bedford, MA) and was used to prepare all solutions.

Stock and Standard Solutions -1

Stock solutions of each chiral drug were prepared as 1000 mg L in methanol. The stock solutions were protected from light exposure using aluminum foil and stored for a month at 4°C with no evidence of decomposition. Then the required working standard solutions were freshly prepared by appropriate dilution of the stock solutions with HPLC grade water to the required concentrations.

CE Equipment CE was carried out using a Lumex Capel 105 (St. Petersburg, Russia) equipped with a UV detector. The electrophoretic experiments were performed in an uncoated fused-silica capillary (St. Petersburg, Russia) 60 cm × 50 μm I.D. (50 cm effective length). Before use, the capillary was conditioned for 5 min with 0.5 M HCl, 5 min with water, 5 min with 0.5 M NaOH, and 5 min with water. Additionally, the capillary was washed for 1 min with 0.5 M NaOH, 1 min with water, and 2 min with the running buffer with positive pressure applied at the injection end between each run. The samples were injected by hydrodynamic injection at 50 mbar for 5 s at anodic end of capillary and detected at 214 nm. Acquisition of electropherograms was computer-controlled by Chrom & Spec software v. 1.5.

Data Processing The analytical data were acquired and treated with the software defined in the above section. Resolution (Rs) was calculated according to the following formula: Rs ¼ 2

t2  t1 w 1 þ w2

(1)

where t1 and t2 are the migration times (in min) of the last and the first eluting peak, respectively, and W1 and W2 are the baseline widths (in min) of these peaks. The electrophoretic mobility (μ) can be determined experimentally from the migration time and the field strength:    L Lt μ¼ (2) tr V where L is the distance from the inlet to the detection point, tr is the time required for the analyte to reach the detection point (migration time), V is the applied voltage (field strength), and Lt is the total length of the capillary. Separation selectivity (α) in CE is taken from chromatographic techniques and can be expressed as follows41–44   μ1 (3) α¼ μ2

RESULTS AND DISCUSSION EXPERIMENTAL Reagents and Materials All chemicals used in the analyses were analytical grade. To prevent capillary blockage, all buffers and samples were filtered through 0.45-μm filter membranes (Millipore, Bedford, MA). Amlodipine (AM), medetomidine (MED), methamphetamine (MET), rivastigmine (RIV), sitagliptin (SIT), tolterodine (TOL), trimipramine (TRI), cetirizine (CT), hydroxyzine (HCZ), and meclozine (MEC) were obtained from Tofigh Daru (Tehran, Iran) and used without further purification. The pure substance of transtramadol (TRA) was purchased from Grunenthal (Stolberg, Germany). Fluoxetine (FLU), citalopram (CIT), and venlafaxine (VEN) were purchased Chirality DOI 10.1002/chir

In this work, MDs as simple and inexpensive chiral selectors were applied for enantioseparation of 23 chiral basic pharmaceutical compounds. Different interactions of chiral analytes with the helical structure of MD establish the basis of the enantioselectivity. The change in conformation from a flexible coil to a helix in the presence of chiral analytes and buffer salts play an important role in selective interactions leading to enantioresolution.33,34 The helical structure has a hydrophobic inside, much like the cavity of CDs, but MDs represent a considerably more flexible entity than CDs,

ENANTIOSEPARATION WITH MALTODEXTRIN AS CHIRAL SELECTOR

leading to less restriction toward a steric approach of an interacting solute.39 The formation of a helical structure supported by additional interactions, such as hydrogen bonds and dipole–dipole interactions, is assumed to be responsible for chiral recognition.45–48 In this study, model basic chiral compounds were analyzed in a constant preliminary condition. This preliminary condition (based on previous study) was a 100-mM phosphate buffer solution (pH 3.0) containing 10% (w/v) MD with DE 4–7 as the chiral selector at the temperature of 25°C and 18 kV applied CE voltage. Also, all of the chiral compounds under investigation were basic drugs, and considering their pKa values, all of these basic drugs were cationic or dicationic in the preliminary condition (pH 3.0). Thus, there is no suitable link between terms of pKa and the patterns observed. In CE, it is common to observe two baseline resolved peaks for the mixture of enantiomers even in cases when the selectivity of separation does not exceed 1.01. This is difficult to achieve in alternative separation techniques such as GC or SFC and simply unimaginable in HPLC.41 Thus, in this study Rs was used for the classification of model compounds. Classification of Model Compounds

In the preliminary condition, nine chiral compounds (series-I) were separated in baseline (Rs > 1.5) (Fig. 1), and partial separation was obtained for six other compounds (series-II)

(0 < Rs < 1.5) (Fig. 2), while no separation was obtained for last eight chiral compounds (series-III) (Rs = 0) (Fig. 3). The preliminary condition for series-I was optimized to obtain short migration times (with the risk of losing the resolution close to 1.5). Thus, the temperature of capillary column and applied voltage were increased and conditions with the shortest migration times (still having Rs 1.5) were selected as optimum ones. For series-II and III, we tried to improve the Rs by changing important parameters such as pH of BGE, temperature of capillary column, applied voltage, and concentration and type of MD. For a survey of this classification and different manner of these chiral drugs with MD, several structural properties of these drugs such as partition coefficient (log P), hydrogen bond donors (HBD) and acceptors (HBA), polarizability (POL), molar volume (MV), and polar surface area (PSA) were investigated (Table 1). The ChemSpider chemical structure database 49 was used to obtain the values of variables for all chiral drugs listed in Table 1. As can be seen in Table 1, there is no suitable relation between these parameters and enantioseparation. For example, there are same partition coefficients for compounds of different series or HBD and HBA parameters for all of series are randomly. But, it was found that situation of chiral center and type of functional groups on chiral carbon are effective on the enantioseparation with MDs. Thus, in the follow, these parameters were investigated.

Fig. 1. Electropherograms of the chiral drugs with Rs > 1.5 (series-I). Experimental conditions: capillary column: 60 cm (50 cm effective length) × 50 μm I.D.; detection: 214 nm; applied voltage: 18 kV; injection: 60 mbar × 5 s; separation solution: 100 mM phosphate buffer (pH 3.0, 10% (w/v) MD with DE 4-7). Chirality DOI 10.1002/chir

TABANI ET AL.

Fig. 2. Electropherograms of the chiral drugs with 0 < Rs < 1.5 (series-II). CE conditions were the same as in Figure 1.

Fig. 3. Electropherograms of the chiral drugs with Rs = 0 (series-III). CE conditions were the same as in Figure 1. Chirality DOI 10.1002/chir

ENANTIOSEPARATION WITH MALTODEXTRIN AS CHIRAL SELECTOR

TABLE 1. Structural properties of the chiral basic drugs Series I

II

III

a

Compounds

ID

CT HCZ MEC BDH OXY THP AM FLU CIT TRA VEN TOL MED VP NVP ATE PRO TRI MET OXO RIV SIT VIL

2577 3531 58260 2350 4481 5371 2077 3269 2669 5322 5454 391967 61868 2425 94724 2162 4777 5382 8947 18281 70377 3571948 5464

Log P

HBD

HBA

POL (×10 -24 cm3)

MV (cm3)

PSA (Å2)

2.16 2.03 — 4.43 4.69 4.49 4.16 4.09 2.51 2.51 2.91 5.77 3.10 3.90 4.02 0.10 3.10 5.15 1.94 2.85 2.14 1.30 1.10

1 1 — 0 1 1 3 1 0 1 1 1 1 0 1 4 2 0 1 0 0 2 1

5 4 — 2 5 2 7 2 3 3 3 2 2 6 6 5 3 2 1 4 4 6 4

42.0 42.0 — 34.6 38.7 36.4 41.8 31.7 36.5 30.9 32.8 40.8 24.7 52.3 50.4 29.4 31.3 37.1 — 37.2 29.0 33.8 26.0

314.2 317.7 — 265.4 290.0 289.7 333.0 266.7 272.6 251.4 261.7 324.3 190.2 429.4 414.4 236.7 237.0 286.1 — 274.9 241.0 252.4 223.0

53.01 53.94 — 12.47 62.13 23.47 99.88 21.26 36.26 32.7 32.7 23.47 28.68 63.95 72.74 84.85 41.49 6.48 3.24 49.0 32.78 77.04 39.72

a

ChemSpider identification number (ID).

Chiral compounds of series-I (Rs > 1.5). In the preliminary con-

dition, enantiomers of nine compounds (CT, HCZ, MEC, BDH, OXY, THP, AM, FLU, and CIT) were separated in baseline. The results showed that the existence of two aromatic rings or cycloalkanes and oxygen (such as OH) or nitrogen (incorporated in a heterocyclic ring) atom directly bonding to the chiral center is necessary for baseline separation of enantiomers of chiral compounds (Fig. 1). For example, BDH has two aromatic rings and one etheric oxygen at the chiral carbon in its chemical structure; THP has a hydroxyl group (OH), one aromatic ring, and one cycloalkane directly bonded to the chiral carbon. In this condition (BGE: pH 3.0, 10% (w/v) MD with DE 4-7; temperature of capillary column: 25°C; applied voltage: 18 kV), high Rs values for CT (2.66), HCZ (2.68), MEC (1.76), BDH (2.89), OXY (1.63), THP (3.33), AM (1.64), FLU (1.85) and CIT (6.54) were obtained.

In the following step, in order to obtain suitable Rs and short migration times, the temperature of capillary column was studied between 13 to 30°C and applied voltage was investigated in the range of 15 to 22 kV. The optimal conditions (with Rs above 1.5) which gave the shortest migration times were shown in Table 2. Chiral compounds of series-II (0 < Rs < 1.5). In the preliminary

condition, compounds of series-II (TRA, VEN, TOL, MED, VP, and NVP) were partially separated (Fig. 2). For example, TOL in its chemical structure has two aromatic rings bonded to the chiral center but has no O or N atom or –CN group at the chiral center. Therefore, TOL enantiomers resolved partially. For VP and NVP, they have in their chemical structures just one aromatic ring bonded to the chiral center and there is a –CN group at the chiral carbon. Probably the presence of polar –CN group in the molecular structure, which

TABLE 2. Optimal CE conditions for the enantioseparation of series-I Migration time (min) Compounds CT HCZ MEC BDH OXY THP AM FLU CIT

a

Temperature (°C)

Voltage (kV)

t1

t2

Rs

30 30 30 30 25 30 25 30 30

20 20 20 20 18 22 18 20 22

18.93 18.14 22.52 17.95 23.85 17.21 19.41 18.12 10.91

19.72 18.75 23.95 18.75 24.18 17.83 19.79 18.58 11.87

2.46 2.28 1.61 2.30 1.63 2.42 1.64 1.68 4.45

α

a

1.042 1.034 1.063 1.045 1.014 1.036 1.02 1.025 1.087

a

The concentration of MD (DE 4-7) for all compounds was 10% (w/v). Chirality DOI 10.1002/chir

TABANI ET AL.

may exhibit hydrogen-bonding abilities in the enantiospecific complexation with the MD, is responsible for this partial resolution. Also, there is no O or N atom on chiral carbons of VEN and MED, and thus partial resolutions were obtained in a preliminary defined MD-modified CE condition. These results verified that the absence of one of the effective issues (two rings and an O or N atom or a –CN group) could result in a partial resolution. In the case of TRA, we expected its enantiomers to be resolved in baseline. Despite that TRA has two rings and an O atom directly bonded to its chiral center (suitable functions on chiral carbon), partial resolution was obtained with the preliminary condition. The investigation of chemical structure of TRA shows that the chiral carbon was incorporated in a cycloalkane ring and the position of this carbon is not between two rings. This difference in position of chiral carbon might be the reason for partial resolution. To obtain the best Rs, effective parameters on enantioseparation such as pH of BGE, temperature of capillary column, applied voltage, and DE and concentration of MD were investigated. In this study, three MDs (DE 4–7, 13–17, and 16.5–19.5) were used as chiral selectors. The effect of DE on chiral recognition was investigated using a 100-mM phosphate buffer solution (pH 3.0) containing 10% (w/v) chiral selector. The results obtained showed that resolutions increased with decreasing of DE of MD. As shown in Table 3, higher resolutions were obtained with a lower DE value. Also, Soini et al.39 have reported that the higher oligomers are particularly effective as chiral selectors. This may be attributed to the greater amphiphilic character of the longer chain MDs, which provided highly hydrophobic sites in the center of the helical structure. Thus, MD with DE 4–7 was used as the chiral selector. In the following step, the effect of MD (DE 4–7) concentration on the enantioseparation of chiral drugs was investigated using 100-mM phosphate buffer solution (pH 3.0) over a concentration range of 5.0 to 25.0% (w/v). Since the average molecular weight of MD is unknown, concentrations are given in percentages. The concentration of MD was limited to 25% (w/v) considering the high viscosity of the solutions. The migration times of enantiomers increased with increasing concentration of MD. Increases in the drug-chiral selector complex adduct and in viscosity of the BGE were the main causes of the retarded migration for enantiomers. For compounds of series-II, the resolutions increased as MD concentration rose, but then reached the maximum at 20% (w/v) MD addition. Figure 4 shows the resolutions of enantiomers with different MD concentrations. At higher concentration, the resolutions decreased slightly due to saturated

Fig. 4. The effect of MD concentration with DE 4-7 on the resolution of the series-II chiral drugs.

complexation between the enantiomers and chiral selector. These results corresponded well with those derived from the Wren and Rowe model concerning the existence of a maximum resolution at certain concentrations of chiral selector.50,51Considering the successful separation and appropriate analysis time, an MD concentration of 20% (w/v) was chosen. The effect of BGE pH on the migration time and resolution was investigated in the range of 2.0 to 6.0 using phosphate buffer solution (100 mM) containing 20% (w/v) MD (DE 4–7). The Rs of these drugs increased with decreasing pH (Fig. 5). This may easily be explained by a decrease in the electroosmotic flow (EOF). According to the Rs equation given by Jorgenson and Lukacs,52 maximum Rs of cationic species is achieved when the EOF is completely suppressed. As a result, pH 2.0 gave the highest Rs for these drugs. The effect of applied voltage on chiral CE analysis can be discussed from different aspects. Increased amounts of voltage usually improve the CE analysis efficiency, decrease the migration time by affecting both EOF and electrophoretic

TABLE 3. The effect of DE type on the enantioseparation of compounds of series-II RS Compounds TRA VEN TOL MED VP NVP

MD with DE 4-7

MD with DE 13-17

MD with DE 16-19.5

1.22 1.26 1.16 1.28 1.30 1.14

1.05 1.10 0.9 1.03 1.1 0.8

0.85 0.85 0.74 0.89 0.93 0.66

Chirality DOI 10.1002/chir

Fig. 5. The effects of buffer pH on the resolution of the series-II chiral drugs.

ENANTIOSEPARATION WITH MALTODEXTRIN AS CHIRAL SELECTOR

TABLE 4. Optimal CE conditions for the enantioseparation of series-II Migration time (min) Compounds TRA VEN TOL MED VP NVP

t1

t2

Rs

α

24.25 27.65 26.30 23.15 28.50 26.34

24.85 28.34 26.78 23.55 29.32 26.72

1.85 1.93 1.78 1.52 2.21 1.47

1.025 1.025 1.018 1.017 1.029 1.014

flow, and have a significant role in Joule heating through increasing the electric flow. The effect of applied voltage on the enantioseparation of chiral drugs was investigated using 100 mM phosphate solution (pH 2.0, 20% (w/v) MD with DE 4–7) in the range of 15 to 22 kV. Best resolutions for all of these drugs were obtained at 18 kV. The cartridge temperature can affect the EOF and electrophoretic flow through changing the BGE viscosity, which also explains its effect on the resolution. Also, the host–guest complexation mechanism is a kinetically and thermodynamically driven process.53 Thus, the main effect of temperature is observed on migration velocity and the efficiency of the separation. In this study, a circulating coolant containing water was used to maintain a constant temperature inside the capillary cartridge. To determine the optimum temperature for the separation, several electrophoretic runs at different cartridge temperatures (ranging from 13 to 30°C) were performed and the separation of the enantiomers was examined. Based on the results, 20°C was chosen to carry

out the experiments. Under the optimized conditions (BGE: pH 2.0, 20% (w/v) MD with DE 4–7; temperature of capillary column: 20°C; applied voltage: 18 kV), baseline separation was obtained for enantiomers of all compounds in this category (series-II) except for NVP (Table 4). Chiral compounds of series-III (Rs =0). In preliminary condi-

tion, for the last eight chiral compounds (ATE, PRO, TRI, MET, OXO, RIV, SIT, and VIL) no separation was achieved. The investigation of their chemical structures shows that there is no aromatic ring bonded to the chiral carbon (except for RIV). In other words, these results verified that for baseline separation two aromatic rings along with O or N directly bonded to chiral carbon and for partial resolution at least one aromatic ring along with a -CN group at the chiral carbon are required. However, partial resolutions were expected for RIV enantiomers, while no separation was obtained in the preliminary condition. The investigation of RIV chemical structure shows that the N atom bonded to the chiral carbon is not incorporated in a heterocyclic ring. To obtain best Rs, effective parameters on enantioseparation such as pH of BGE, temperature of capillary column, applied voltage and concentration, and DE of MD were optimized. The pH of BGE in the range of 2.0–6.0, applied voltage in the range of 15–22 kV, and concentration of MD in the range of 5–25% were investigated. By changing these conditions, none of the compounds were separated. The lack of resolution between enantiomers of this series again highlight the role of rings and O/N atoms or –CN group directly bonded to the chiral carbon in enantioseparation with MD (Fig. 3). The relation between Rs and chemical structures (effective functions on chiral carbon) of basic chiral drugs is shown in Table 5.

TABLE 5. The effect of chemical structures (effective functions on chiral carbon) on the enantioseparation The number of effective functions on chiral carbon Compounds Series-I

Series-II

Series-III

CT HCZ MEC BDH OXY THP a AM FLU CIT TRA VEN TOL MED VP NVP ATE PRO TRI MET OXO RIV SIT VIL

Rs

Aromatic ring

Cycloalkane ring

O

N

CN

Preliminary

Optimal

2 2 2 2 1 1 1 1 2 1 1 2 1 1 1 0 0 0 0 0 1 0 0

0 0 0 0 1 1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0

0 0 0 1 1 1 0 1 1 1 0 0 0 0 0 1 1 0 0 0 0 0 1

1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0

2.66 2.68 1.76 2.89 1.63 3.33 1.64 1.85 6.54 1.22 1.26 1.16 1.28 1.30 1.14 — — — — — — — —

2.46 2.28 1.61 2.30 1.63 2.42 1.64 1.68 4.45 1.85 1.93 1.78 1.52 2.21 1.47 — — — — — — — —

a

AM has one aromatic ring and another ring has two double-bonds. Chirality DOI 10.1002/chir

TABANI ET AL.

CONCLUSIONS

To obtain an enantioseparation with good resolution, the analytical conditions, especially the type of chiral selector, must be optimized. These operations are laborious, timeconsuming, and expensive. Therefore, the probable prediction of enantioseparation in the presence of one chiral selector can be useful. For this purpose, in this study 23 chiral basic pharmaceutical compounds with different chemical structures were selected as model solutes and the influence of their chemical structures on the enantioseparation in the presence of MD (as chiral selector) was investigated. In the preliminary condition, baseline separation (Rs > 1.5) was obtained for compounds of series-I, and partial separation was obtained for compounds of series-II (0 < Rs < 1.5), while no separation was obtained for compounds of series-III (Rs = 0). The investigation of chemical structures of different series showed that: 1) For baseline separation, at least two aromatic rings or one aromatic and one cycloalkane ring along with O and N (incorporated in a heterocyclic ring) on chiral carbon are required. 2) The absence of one of the mentioned functions on the chiral carbon could result in partial resolution between enantiomers. 3) The absence of both aromatic rings could result in no resolution between enantiomers. These results can help to predict the possibility of enantioseparation for each compound in the presence of MD. This prediction will reduce the number of preliminary tests required for enantioseparation and will save time and costs.

11.

12.

13.

14. 15. 16. 17. 18.

19. 20. 21. 22. 23.

ACKNOWLEDGMENTS

The financial support from the Research Affairs of Shahid Beheshti University is gratefully acknowledged. There are no financial or commercial conflicts of interest. LITERATURE CITED 1. Nolin TD, Frye RF. Stereoselective determination of the CYP2C19 probe drug mephenytoin in human urine by gas chromatography–mass spectrometry. J Chromatogr B 2003;783:265–271. 2. Juvancz Z, Markides KE, Petersson P, Johnson DF, Bradshaw JS, Lee ML. Copolymeric (1R-trans)-N,N-1,2-cyclohexylene-bisbenzamide oligodimethylsiloxane chiral stationary phase for gas chromatography. J Chromatogr A 2002;982:119–126. 3. Matarashvili I, Chankvetadze L, Fanali S, Farkas T, Chankvetadze B. HPLC separation of enantiomers of chiral arylpropionic acid derivatives using polysaccharide-based chiral columns and normal-phase eluents with emphasis on elution order. J Sep Sci 2013;36:140–147. 4. Gubitz G, Schmid MG. Chiral separation by chromatographic and electromigration techniques. Biopharm Drug Dispos 2001;22:291–336. 5. Gasparrini F, Misiti D, Villani C. High-performance liquid chromatography chiral stationary phases based on low-molecular-mass selectors. J Chromatogr A 2001;906:35–50. 6. Garzotti M, Hamdan M. A novel sensor based on electropolymerization of β-cyclodextrin and l-arginine on carbon paste electrode for determination of fluoroquinolones. J Chromatogr B 2002;770:53–61. 7. Toribio L, Bernal JL, del Nozal MJ, Jimenez JJ, Nieto EM. Applications of the Chiralpak AD and Chiralcel OD chiral columns in the enantiomeric separation of several dioxolane compounds by supercritical fluid chromatography. J Chromatogr A 2001;921:305–313. 8. Armstrong DW, Faulkner JR, Han SM. Use of hydroxypropyl- and hydroxyethyl-derivatized β-cyclodextrins for the thin-layer chromatographic separation of enantiomers and diastereomers. J Chromatogr 1988;452:323–330. 9. Aboul-Enein HY, El-Awady MI, Heard CM, Nicholls PJ. Application of thin-layer chromatography in enantiomeric chiral analysis—an overview. Biomed Chromatogr 1999;13:531–537. 10. Fakhari AR, Tabani H, Nojavan S, Abedi H. Electromembrane extraction combined with cyclodextrin-modified capillary electrophoresis for the Chirality DOI 10.1002/chir

24. 25.

26. 27. 28.

29.

30.

31.

32.

33. 34.

35. 36.

quantification of trimipramine enantiomers. Electrophoresis 2012;33: 506–515. Lodén H , Pettersson C, Arvidsson T, Amini A. Quantitative determination of salbutamol in tablets by multiple-injection capillary zone electrophoresis. J Chromatogr A 2008;1207:181–185. Lomsadze K, Vega ED, Salgado A, Crego AL, Scriba GKE, Marina ML, Chankvetadze B. Separation of enantiomers of norephedrine by capillary electrophoresis using cyclodextrins as chiral selectors: Comparative CE and NMR studies. Electrophoresis 2012;33:1637–1647. Tabani H, Fakhari AR, Shahsavani A. Simultaneous determination of acidic and basic drugs using dual hollow fibre electromembrane extraction combined with CE. Electrophoresis 2013;34:269–276. Terabe S, Otsuka K, Nishi H. Separation of enantiomers by capillary electrophoretic techniques. J Chromatogr A 1994;666:295–319. Chankvetadze B. Capillary electrophoresis in chiral analysis. New York: John Wiley & Sons; 1997. Fanali S. Controlling enantioselectivity in chiral capillary electrophoresis with inclusion–complexation. J Chromatogr A 1997;792:227–67. Nishi H, Terabe S, editors. Applications of chiral cCapillary electrophoresis. J Chromatogr A vol. 8752000:3–484. Special volume. Chankvetadze B. Enantioseparations by using capillary electrophoretic techniques: The story of 20 and a few more years. J Chromatogr A 2007;1168:45–70. Rocco A, Maruška A, Fanali S. Enantiomeric separations by means of nano-LC. J Sep Sci 2013;36:421–444. Gubitz G, Schmid MG. Chiral separations: Methods and protocols. Totowa, NJ: Humana Press; 2004. Blanco M, Valverde I. Choice of chiral selector for enantioseparation by capillary electrophoresis. Trends Anal Chem 2003;22:428–439. Gubitz G, Schmid MG. Chiral separation principles in capillary electrophoresis. J Chromatogr A 1997;792:179–225. Fillet M, Hubert P, Crommen J. Enantioseparation of nonsteroidal antiinflammatory drugs by capillary electrophoresis using mixtures of anionic and uncharged β-cyclodextrins as chiral additives. Electrophoresis 1997;18:1013–1018. Stalcup AM, Gahm KH. Application of sulfated cyclodextrins to chiral separations by capillary zone electrophoresis. Anal Chem 1996;68:1360–1368. Tait R, Thompson D, Stella V, Stobaugh J. Sulfobutyl ether betacyclodextrin as a chiral discriminator for use with capillary electrophoresis. Anal Chem 1994;66:4013–4018. Kuhn R. Enantiomeric separation by capillary electrophoresis using a crown ether as chiral selector. Electrophoresis 1999;20:2605–2613. Stalcup AM, Agyei NM. Heparin: A chiral mobile-phase additive for capillary zone electrophoresis. Anal Chem 1994;66:3054–3059. Dohnal V, Farková M, Havel J. Prediction of chiral separations using a combination of experimental design and artificial neural networks. Chirality 1999;11:616–621. Katritzky AR, Fara DC, Yang H, Karelson M, Suzuki T, Solov’ev VP, Varnek A. Quantitative structure  property relationship modeling of βcyclodextrin complexation free energies. J Chem Inf Comput Sci 2004;44:529–541. Asensi-Bernardi L, Escuder-Gilabert L, Martín-Biosca Y, MedinaHernández MJ, Sagrado S. Modeling the chiral resolution ability of highly sulfated β-cyclodextrin for basic compounds in electrokinetic chromatography. J Chromatogr A 2013;1308:152–160. Loukas YL. Quantitative structure-binding relationships (QSBR) and artificial neural networks: improved predictions in drug: Cyclodextrin inclusion complexes. Int J Pharm 226;207:211. Prakasvudhisarn C, Wolschann P, Lawtrakul L. Predicting complexation thermodynamic parameters of β-cyclodextrin with chiral guests by using swarm intelligence and support vector machines. Int J Mol Sci 2009;10:2107–2121. D’Hulst A, Verbeke N. Carbohydrates as chiral selectors for capillary electrophoresis of racemic drugs. Enantiomer 1997;2:69–79. D’Hulst A, Verbeke N. Quantitation in chiral capillary electrophoresis: Theoretical and practical considerations. Electrophoresis 1994;15: 854–863. Rundle RE, Foster JF, Baldwin RR. On the nature of the starch-iodine complex. J Am Chem Soc 1944;66:2116–2120. D’Hulst A, Verbeke N. Chiral separation by capillary electrophoresis with oligosaccharides. J Chromatogr 1992;608:275–287.

ENANTIOSEPARATION WITH MALTODEXTRIN AS CHIRAL SELECTOR 37. D’Hulst A, Verbeke N. Separation of the enantiomers of coumarinic anticoagulant drugs by capillary electrophoresis using maltodextrins as chiral modifiers. Chirality 1994;6:225–229. 38. D’Hulst A, Verbeke N. Chiral analysis of basic drugs by oligosaccharidemediated capillary electrophoresis. J Chromatogr A 1996;735:283–293. 39. Soini H, Stefansson M, Riekkola ML, Novotny M. Maltooligosaccharides as chiral selectors for the separation of pharmaceuticals by capillary electrophoresis. Anal Chem 1994;66:3477–3284. 40. Nishi H, Izumoto S, Nakamura K, Nakai H, Sato T. Dextran and dextrin as chiral selectors in capillary zone electrophoresis. Chromatographia 1996;42:617–630. 41. Chankvetadze B. Separation selectivity in chiral capillary electrophoresis with charged selectors. J Chromatogr A 1997;792:269–295 42. Baumy P, Morin P, Dreux M, Viaud MC, Boye S, Guillaumet G, Determination of β-cyclodextrin inclusion complex constants for 3,4-dihydro-2-H-1benzopyran enantiomers by capillary electrophoresis. J Chromatogr A 1995;707:311–326. 43. Rawjee YY, Staerk DU, Vigh G. Capillary electrophoretic chiral separations with cyclodextrin additives: I. Acids: Chiral selectivity as a function of pH and the concentration of β-cyclodextrin for fenoprofen and ibuprofen. J Chromatogr 1993;635:291–306. 44. Guttman A, Paulus A, Cohen AS, Grinberg N, Karger BL. Use of complexing agents for selective separation in high-performance capillary electrophoresis: Chiral resolution via cyclodextrins incorporated within polyacrylamide gel columns. J Chromatogr 1988;448:41–53.

45. Nishi H. Enantioselectivity in chiral capillary electrophoresis with polysaccharides. J Chromatogr A 1997;792:327–332. 46. Mikus FF, Hixon RM, Rundle RE. The complex of fatty acids with amylase. J Am Chem Soc 1946;68:1115–1123. 47. Aoyama Y, Otsuki J, Nagai Y, Kobayashi K, Toi H. Host-guest complexation of oligosaccharides: Interaction of maltodextrins with hydrophobic fluorescence probes in water. Tetrahedron Lett 1992;33: 3775–3778. 48. Jane JL, Robyt JF, Huang DH. Structure studies of amylose-V complexes and retrograded amylose by action of alpha amylases, and a new method for preparing amylodextrins. Carbohydr Res 1985;140: 21–35. 49. http://www.chemspider.com/ (accessed 31.04.2014). 50. Wren SAC, Rowe RC. Theoretical aspects of chiral separation in capillary electrophoresis: I. Initial evaluation of a model. J Chromatogr 1992;603: 235–241. 51. Wren SAC, Rowe \RC, Payne RS. A theoretical approach to chiral capillary electrophoresis with some practical implications. Electrophoresis 1994;15: 774–778. 52. Jorgenson JW, Lukacs KD. Zone electrophoresis in open-tubular glass capillaries. Anal Chem 1981;53:1298–1302. 53. Peng ZL, Yi F, Guo B, Lin JM. Temperature effects on the enantioselectivity of basic analytes in capillary EKC using sulfated beta-CDs as chiral selectors. Electrophoresis 2007;28:3753–3758.

Chirality DOI 10.1002/chir