J. Cent. South Univ. Technol. (2009) 16: 0201-0205

DOI: 10.1007/s11771-009-0034-2                                                                                                                

Enantiomeric separation of phenylsuccinic acid by cyclodextrin-modified reversed phase high-performance liquid chromatography

MAN Rui-lin(满瑞林)¹, WANG Zhong-hui(王钟辉)¹, TANG Ke-wen(唐课文)^{1, 2}

 (1. School of Chemistry and Chemical Engineering, Central South University, Changsha 410083, China;

2. Department of Chemistry and Chemical Engineering, Hunan Institute of Science and Technology,

Yueyang 414000, China)

Abstract: The chiral separation of phenylsuccinic acid (PSA) was studied by reversed phase high-performance liquid chromatography (RP-HPLC) with cyclodextrins (CDs) as chiral mobile phase additives. The effects of types of CDs, concentration of hydroxypropyl-β-cyclodextrin (HP-β-CD), percentage of organic modifier, pH value and column temperature on enantioselective separation were investigated. The quantification property of the developed RP-HPLC method was examined. The chiral recognition mechanism of PSA was also discussed. The results show that a baseline separation of PSA enantiomers is achieved on a Lichrospher C18 column (4.6 mm (inner diameter)×250 mm, 5 ?m) with HP-β-CD as chiral mobile phase additive. The capacity factors of R-PSA and S-PSA are 3.94 and 4.80, respectively. The separation factor and resolution are respectively 1.22 and 8.03. The mobile phase is a mixture of acetonitrile and deionized water (20?80, volume ratio) containing 10 mmol/L HP-β-CD and 0.05% trifluoroacetic acid (pH 2.5, adjusted with triethylamine) with a flow rate of 1.0 mL/min. The ultraviolet (UV) detector is set at   254 nm. The likely roles are inclusion interaction, induction and hydrogen bonding between HP-β-CD and PSA enantiomers.

Key words: high performance liquid chromatography; hydroxypropyl-β-cyclodextrin; enantiomer separation; phenylsuccinic acid

1 Introduction

Most often the enantiomers of chiral drugs have different pharmacological and toxicological properties. Therefore, there is an increasing demand for optically pure enantiomers in the chemical industry [1-4], and the quantitative enantiomeric compositions of these drugs should also be determined.

Phenylsuccinic acid (PSA) is an important intermediate of N-methyl-α-phensuximide, which is used as raw drug material. S-PSA is a good enantiomeric separation additive for chiral drug chlorpheniramine, and it can also be used to selectively prepare S-PSA-bonded stationary phases [5]. Therefore, enantiomeric separation and purity assessment are vital in assuring good quality in the pharmaceutical production of PSA and in other PSA-related scientific research work as well.

Cyclodextrins (CDs; cyclic oligosaccharides composed of six, seven, or eight α-D-glucopyranose units (α-, β-, γ-CD, respectively)) form a family of excellent chiral selectors that are widely used as chiral mobile phase additives in high-performance liquid chromatography (HPLC) [6-12]. They are inherently chiral and undergo chiral interactions with analytes. CDs separate enantiomers utilizing the phenomenon of host-guest complexation, where a transient diastereomeric complex is formed between the CD and the analyte. Derivatization of the hydroxyl groups increases the solubility and selectivity compared with the native β-CD and the derivatized hydroxyl groups also undergo additional interactions with the analytes, thereby enhancing chiral recognition [13].

Concerning the enantiomeric separation of PSA, it was reported that PSA enantiomers were separable in reversed phase liquid chromatography using Exsil ODS columns with 2, 6-butyl-β-cyclodextrin chemically bonded as a stationary phase [14]. Although the separation results are good, little work has been done on the various factors affecting on the separation and the chiral recognition mechanism.

Since chiral stationary phases are usually quite expensive and difficult to synthesize, the use of chiral mobile phase additives can be a cheap, convenient, yet effective alternative. However, there is nearly no study having been made on HPLC separation of PSA enantiomers on achiral columns with chiral mobile phase additives. In this work, the chiral separation of PSA enantiomers using reversed phase high-performance liquid chromatography (RP-HPLC) was studied withβ-CD and three of its derivatives as chiral mobile phase additives. The effects of types of CDs, concentration of HP-β-CD, percentage of organic modifier, pH value and column temperature on enantioselective separation were investigated, and the method for the determination of PSA enantiomers was validated. In addition, the chiral recognition mechanism of PSA was discussed.2 Experimental

2.1 Chemicals and reagents

Phenylsuccinic acid (racemate, purity≥99.5%, batch number 071015, the chemical structure shown in Fig.1) was provided by Xiangfan Nuoer Chemical Co. Ltd (Hubei, China). Native β-CD, Me-β-CD (degree of substitution 5.1), HE-β-CD (degree of substitution 4.6), HP-β-CD (degree of substitution 5.7, batch number 080102) were all purchased from Shandong Xinda Fine Chemical Co. Ltd (Shandong, China). Acetonitrile (analytical grade, batch number 20070713) was bought from Shanghai Ludu Chemical Reagent Factory. Trifluoroacetic acid (analytical grade, batch number F20070912) was purchased from Sinopharm Chemical Reagent Co. Ltd. And all other chemicals or solvents were analytical reagent and purchased from commercial sources.

Fig.1 Chemical structure of PSA

2.2 Apparatus and chromatographic conditions

Chromatographic studies were performed on Agilent 1200 HPLCs (Agilent, Hewlett-Packard Co., USA) equipped with a thermostated-column device and a variable-wavelength UV detector. An Agilent 1200 series pump and a manual injector with 20 ?L sample loop (Hewlett-Packard Co., USA) were also employed. Data acquisition was performed using the Agilent ChemStation. The pH measurement was performed on a pH meter (Orion, model 818, Shanghai, China).

The separation of the analytes was achieved on a Lichrospher C18 (4.6 mm (inner diameter)×250 mm,  5 ?m) column. The mobile phase (pH 2.5, adjusted with triethylamine) was a mixture of acetonitrile and deionized water containing 10 mmol/L HP-β-CD and 0.05% trifluoroacetic acid (20?80, volume ratio) with a flow rate of 1.0 mL/min. The mobile phase was filtered through a 0.45 ?m filter (Jinteng Associates, Tianjin, China) and sonicated prior to use. The wavelength of UV detector was set at 254 nm and the column was operated at 25 ℃. Injection volume was 20 ?L.

2.3 Preparation of standard solution and mobile phases

For the preparation of standard PSA stock solution (50 mmol/L), 0.971 0 g racemate PSA was accurately weighed, transferred to a 100 mL volumetric flask and dissolved in deionized water to make the standard stock solution of 50 mmol/L. Standard solutions in the range of 0.2-5.0 mg/mL were prepared by appropriate dilutions of the stock solution with deionized water for calibration study. All the solutions were stored at 5 ℃ and kept stable for at least 2 months. The mobile phases with certain concentration of CDs, certain pH value and certain volume ratio of acetonitrile to deionized water were respectively prepared before use according to analytical requirement.

3.1 Optimization of separation

The optimization of the chiral separation was carried out using UV detection. A number of parameters, such as types of CDs, concentration of HP-β-CD, percentage of organic modifier, pH value and column temperature were investigated to determine their exact effect on the resultant chromatography.

3.1.1 Influence of types of CDs

In order to achieve the separation of PSA enantiomers, the influence of the types of CDs on the resolution of PSA enantiomers was firstly examined during optimization. Native β-CD and three of its derivatives (Me-β-CD, HE-β-CD, HP-β-CD) were tested as chiral selectors for the enantiomeric separation of PSA, injected as racemate.

Table 1 lists the influence of the nature of the four cyclodextrins on the enantioseparation of PSA enantiomers. It clearly shows that large differences in chiral resolution are generally obtained with different cyclodextrins at a given concentration. The PAS enantiomers can be completely resolved with all β-CD derivatives. However, with native β-CD, only partial resolution is obtained. Moreover, the stronger the polarity of CD, the better the resolution. This can be explained that molecules interact more easily with CD having stronger polarity. And chemical modification of cyclodextrins has been shown to ‘stretch’ the cavity mouth and therefore change the hydrophobicity of the molecule and the stereoselectivity of the inclusion process. The mouth of the cyclodextrin hydrophobic cavity is surrounded by secondary hydroxyl groups that are considered to be important in chiral recognition [15]. In a derivatized HP-β-CD, some hydroxyl groups are substituted with hydroxypropyl functional groups. This modification allows for a more stereospecific and stronger interaction between the hydroxyl groups and hydrogen-bonding moiety present in the PSA structure. In addition to the aromatic ring which forms inclusion complexes with the CD, PSA (The chemical structure of PSA is shown in Fig.1) has two carboxylic acid groups at the chiral center that could participate in additional interactions with the rim hydroxypropyl groups of the HP-β-CD. Such interactions include induction and hydrogen bonding. So, for further study of the enantiomeric separation of PSA, HP-β-CD was chosen as chiral mobile phase additive.

Table 1 Effect of types of CD on resolution of PSA enantiomers

3.1.2 Influence of concentration of HP-β-CD

With HP-β-CD as chiral mobile phase additive, the influence of concentration of HP-β-CD was examined during optimization. The addition of HP-β-CD (1-20 mmol/L) to the mobile phase composed of 84?16 (volume ratio) aqueous 0.05% trifluoroacetic acid, pH 3.0 (adjusted with triethylamine)/acetonitrile gave a good enantiomeric resolution of PSA enantiomers. The results in Table 2 show that an increase in the concentration of HP-β-CD leads to both a concomitant decrease in retention time and an increase in resolutions, and the shortest retention time and best resolution is achieved with 20 mmol/L HP-β-CD. However, the concentration of HP-β-CD above 10 mmol/L increases pressure obviously and has the trend to decrease column efficiency, so 10 mmol/L HP-β-CD was chosen for further study, with which the baseline separation of PSA enantiomers is achieved. HP-β-CD concentration affects both enantiomer resolution and retention time. However, the interrelationships among the CD concentration, guest-CD complexes and efficiency of chiral resolution still lack theoretical bases.

Table 2 Effect of concentration of HP-β-CD on resolution of PSA enantiomers

3.1.3 Influence of pH value

Since PSA has two carboxylic acid functionalities (pKa₁=3.78, pKa₂=5.55) [14], the molecule dissociates in aqueous solution by releasing two protons. Ionisation suppression by pH control results in longer retention, as expected, and improves the likelihood of chiral discrimination. The results are shown in Fig.2. It clearly shows that better resolutions of PSA enantiomers are obtained at lower pH values. And the resolution consequently decreases as the pH value increases from 2.5 to 5.0. When the pH value reaches up to 3.5, for further increase, the resolution decreases sharply from 6.94 to 2.74. This can be explained by the molecule showing a greater preference for the hydrophobic cavity of the cyclodextrin in the presence of a more ionic mobile phase. So, it is preferable to choose lower pH value. However, lower pH value is not good for the column’s life-span. Here the pH value of 2.5 was chosen for further study.

Fig.2 Effect of pH value on resolution of PSA enantiomers (Mobile phase: V(aqueous 0.05% trifluoroacetic acid)/ V(acetonitrile)=84?16, adjusted with triethylamine to control pH value; c(HP-β-CD)=10 mmol/L)

3.1.4 Influence of percentage of organic modifier

Small amount of organic modifier is especially important in chiral mobile phase additive separation because the addition of organic modifier to the mobile phase greatly decreases the solubility of CDs [16]. And also an increase in the percentage of organic modifier may result in an increase in eluting power and consequently a decrease in retention time and resolution. The results are shown in Fig.3. It can be seen that the resolution decreases sharply as the percentage of acetonitrile increases from 10% to 30%. For further increase of the percentage of acetonitrile, the resolution decreases much more slowly, and finally, nearly no resolution can be achieved. Considering all the reasons above, column ratio of acetonitrile to aqueous is set at 20?80 for further study.

Fig. 3 Influence of percentage of acetonitrile (Mobile phase: mixture of acetonitrile and deionized water containing 0.05% trifluoroacetic acid; pH=2.5; c(HP-β-CD)=10 mmol/L)

3.1.5 Influence of column temperature

The influence of column temperature was also taken into consideration. As shown in Table 3, it can be seen that all the resolutions are good in the temperature range from 10 to 30 ℃. And both of the enantiomers’ retention times constantly decrease as the column temperature increases from 10 to 30 ℃. As we all know that too high column temperature is not good for the column, and too low column temperature influences the column efficiency greatly (as seen in Table 3, the enantiomers’ retention times are too long, although the resolution is a little bigger). So, considering all the reasons above, the column temperature is set at 25 ℃.

Table 3 Effect of column temperature on resolution of PSA enantiomers

Thus, the optimized mobile phase composition is 20% (volume fraction) acetonitrile and aqueous containing  10 mmol/L HP-β-CD and 0.05% trifluoroacetic acid at pH 2.5 (adjusted with triethylamine). The flow rate and detection wavelength are set at 1.0 mL/min and 254 nm, respectively. Column temperature is 25 ℃.

With the optimized mobile phase composition and flow-rate, a baseline separation of PSA enantiomers is achieved with UV detection. The result is shown in Fig.4. Although the peak height of the longer retained solute is somewhat lower than the first peak height, there is no difference in peak area between R-PSA and S-PSA peaks. The chiral selectivity and resolution are found to be 1.22 and 8.03, respectively.

Fig.4 Separation of PSA enantiomers (Mobile phase: mixture of acetonitrile and deconized water (20?80, volume ratio) containing 10 mmol/L HP-β-CD and 0.05% trifluoroacetic acid; pH=2.5)

3.2 Linear relationship between peak area and con- centration

A series of standard solutions containing different amounts of PSA enantiomers were prepared and separated by the above chromatographic procedure using UV detection. The calibration curves of R-PSA and S-PSA were linear over the concentration ranges from 0.2 to 5.0 mg/mL by performing a regression linear analysis of the peak area (y) of each enantiomer versus the concentrations of PSA enantiomers (x) in the standard solutions. The regression equations of the calibration curves are y=1.36+48.68x (r=0.999 05) for R-PSA and y=-1.30+50.40x (r=0.999 31) for S-PSA, respectively.

3.3 Precision

Repeatability of the HPLC method was studied. The racemate PSA at a concentration of 10 mmol/L was used for analyzing. The relative standard deviations (RSDs) of repeatability are 1.5% for R-PSA, and 1.4% for S-PSA, respectively. The results in Table 4 show that the method is precise and accurate.

Table 4 Reproducibility of retention time and peak area

4 Conclusions

(1) The types of CDs, concentration of HP-β-CD, percentage of organic modifier, pH value and column temperature, have a great influence on the resolution and retention time of PSA enantiomers.

(2) With the optimized conditions, a baseline separation of PSA enantiomers is achieved. The separation factor and resolution are respectively 1.22 and 8.03.

(3) The likely roles for chiral recognition mechanism of PSA enantiomers are inclusion interaction, induction and hydrogen bonding between HP-β-CD and PSA enantiomers.

(4) The linear relationships are maintained over the range of 0.2 to 5.0 mg/mL with regression coefficients of 0.999 05 and 0.999 31 for R-PSA and S-PSA respectively.

(5) The HPLC method for resolution of the PSA enantiomers is established. With HP-β-CD as chiral mobile phase additive, the quantification method of the developed RP-HPLC is easy to perform, precise and accurate. The whole procedure may also be extended to the applications on quality control of commercial products.

[1] MIYAKO E, MARUYAMA T, KAMIYA N, GOTO M. Highly enantioselective separation using a supported liquid membrane encapsulating surfactant-enzyme complex [J]. J Am Chem Soc, 2004, 126(28): 8622-8623.

[2] MIRIAM L. Membranes make chiral separations simpler [J]. Manufacturing Chemist, 1995, 10(7): 25-32.

[3] TANG Ke-wen, ZHANG Guo-li, HUANG Ke-long, LI Yuan-jian, YI Jian-min. Resolution of α-cyclohexyl-mandelic acid enantiomers by two-phase (O/W) recognition chiral extraction [J]. Science in China Series B: Chemistry, 2007, 50(6): 764-769.

[4] TANG Ke-wen, HUNG Ke-long. Enantioselective extraction of mandelic enantiomers based on chiral ligand exchange [J]. Journal of Central South University of Technology, 2005, 12(2): 123-128.

[5] STEPHANI R, CESARE V. Determination of the enantiomers of chlorpheniramine and its main monodesmethyl metabolite in urine using achiral-chiral liquid chromatography [J]. Journal of Chromatography B, 1998, 707(2): 235-240.

[6] JANDERA P, BUNC?KOV? S, PLANETA J. Separation of isomeric naphthalenesulphonic acids by micro high-performance liquid chromatography with mobile phases containing cyclodextrin [J]. Journal of Chromatography A, 2000, 871(2): 139-152.

[7] AMEYIBOR E, STEWART J T. Resolution and quantitation of pentazocine enantiomers in human serum by reversed-phase high-performance liquid chromatography using sulfated β- cyclodextrin as chiral mobile phase additive and solid-phase extraction [J]. Journal of Chromatography B, 1997, 703(2): 273-278.

[8] AMEYIBOR E, STEWART J T. HPLC determination of ketoprofen enantiomers in human serum using a nonporous octadecylsilane 1.5 mm column with hydroxypropyl β-cyclodextrin as mobile phase additive [J]. Journal of Pharmaceutical and Biomedical Analysis, 1998, 17(1): 83-88.

[9] ZHONG Q Q, HE L F, BEESLEY T E, TRAHANOVSKY W S, SUN P, WANG C L, ARMSTRONG D W. Development of dinitrophenylated cyclodextrin derivatives for enhanced enantiomeric separations by high-performance liquid chromatography [J]. Journal of Chromatography A, 2006, 1115(1): 19-45.

[10] HEALY L O, MURRIHY J P, TAN A M, COCKER D, MCENERY M, GLENNON J D. Enantiomeric separation of R, S-naproxen by conventional and nano-liquid chromatography with methyl-b- cyclodextrin as a mobile phase additive [J]. Journal of Chromatography A, 2001, 924(2): 459-464.

[11] CHEN De-ying, JIANG Shu-min, CHEN Yu-ying, HU Yu-zhu. HPLC determination of sertraline in bulk drug, tablets and capsules using hydroxypropyl-β-cyclodextrin as mobile phase additive [J]. Journal of Pharmaceutical and Biomedical Analysis, 2004, 34(1): 239-245.

[12] YE Jin-cui, CHEN Guo-sheng, ZENG Su. Enantiomeric separation of norgestrel by reversed phase high-performance liquid chromatography using eluents containing hydroxypropyl-beta- cyclodextrin in stereoselective skin permeation study [J]. Journal of Chromatography B, 2006, 843(2): 289-294.

[13] SPENCER B J, PURDY W C. Effect of the degree of substitution of cyclodextrin derivatives on chiral separations by high-performance liquid chromatography [J]. Liq Chromatogr, 1995, 18(16): 4063-4080.

[14] RUAN Yuan-ping, AO Xiao-ping, ZHANG Xue-man, ZHANG Pei-qiang. Liquid chromatographic resolution of enantiomeric 1, 1′-bi-2-naphthol and phenylsuccinic acid on a coated reversed-phase column with 2, 6-O-butyled-β-cyclodextrins [J]. Chinese Journal of Analytical Chemistry, 2004, 32(7): 949-952. (in Chinese)

[15] BIELEJEWSKA A, NOWAKOWSKI R, DUSZCZYK K, SYBILSKA D. Joint use of cyclodextrin additives in chiral discrimination by reversed-phase high-performance liquid chromatography: Temperature effects [J]. Journal of Chromatography A, 1999, 840(2): 159-170.

[16] CHEN J Z, OHNMACHT C M, HAGE D S. Characterization of drug interactions with soluble β-cyclodextrin by high-performance affinity chromatography [J]. Journal of Chromatography A, 2004, 1033(1): 115-126.

Foundation item: Project(20776038) supported by the National Natural Science Foundation of China

Received date: 2008-07-19; Accepted date: 2008-10-09

Corresponding author: TANG Ke-wen, Professor, PhD; Tel: +86-730-8648500; E-mail: tangkewen@sina.com

(Edited by YANG You-ping)