1083
J. Sep. Sci. 2014, 37, 1083–1088
Maia Gegenava1 Lali Chankvetadze1 Tivadar Farkas2 Bezhan Chankvetadze1 1 Institute
of Physical and Analytical Chemistry, School of Exact and Natural Sciences, Tbilisi State University, Tbilisi, Georgia 2 Phenomenex Inc., Torrance, CA, USA Received December 8, 2013 Revised February 21, 2014 Accepted February 22, 2014
Research Article
Enantioseparation of selected chiral sulfoxides in high-performance liquid chromatography with polysaccharide-based chiral selectors in polar organic mobile phases with emphasis on enantiomer elution order The separation of the enantiomers of 17 chiral sulfoxides was studied on polysaccharidebased chiral columns in polar organic mobile phases. Enantiomer elution order (EEO) was the primary objective in this study. Two of the six chiral columns, especially those based on amylose tris(3,5-dimethylphenylcarbamate) and cellulose tris(4-chloro-3methylphenylcarbamate) (Lux Cellulose-4) proved to be most successful in the separation of the enantiomers of the studied sulfoxides. Interesting examples of EEO reversal were observed depending on the chiral selector or the composition of the mobile phase. For instance, the R-(+) enantiomer of lansoprazole eluted before the S-(−) enantiomer on Lux Cellulose-1 in both methanol or ethanol as the mobile phase, while the elution order was opposite in the same eluents on amylose tris(3,5-dimethylphenylcarbamate) with the S-(−) enantiomer eluting before the R-(+) enantiomer. The R-(+) enantiomer of omeprazole eluted first on Lux Amylose-2 in methanol but it was second when acetonitrile was used as the mobile phase with the same chiral selector. Several other examples of reversal in EEO were observed in this study. An interesting example of the separation of four stereoisomers of phenaminophos sulfoxide containing chiral sulfur and phosphor atoms is also reported here. Keywords: Chiral sulfoxides / Enantiomer elution order / Enantiomeric separation DOI 10.1002/jssc.201301318
1 Introduction Chiral sulfoxides are widely used as valuable drug compounds (such as omeprazole, lansoprazole, pantoprazole, rabeprazole) [1, 2], pesticides (fipronil, propargit, methiocarb sulfoxide, fensulfothion, etc.) [2], chiral auxiliaries, and chiral ligands to metals [3, 4], in addition they have been incorporated into a variety of natural products and their synthetic derivatives. The synthesis of enantiomerically enriched or pure sulfoxides is of primary interest [5–7]. The enantioselective biological activity of chiral sulfoxides is well documented and became the reason for developing some of them as enantiomerically pure chiral drugs [1]. For instance, together with racemic omeprazole, its enantiomerically pure analogue, (S)-omeprazole is used in clinical practice and Correspondence: Professor Bezhan Chankvetadze, Institute of Physical and Analytical Chemistry, School of Exact and Natural Sciences, Tbilisi State University, Chavchavadze Avenue 3, Tbilisi 0179, Georgia E-mail: [email protected]
Abbreviations: ADMPC, amylose tris(3,5-dimethylphenylcarbamate); EEO, enantiomer elution order C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
some other analogues are also under development [8]. Not enough attention is yet paid to the enantioselective action, toxicity, and biodegradation of chiral sulfoxides used as agrochemicals. However, a few studies indicate the significant difference between the enantiomers of chiral sulfoxides from these points of view [9]. Chromatographic methods have been used for the separation of enantiomers of chiral sulfoxides for more than 20 years [10–25] and together with analytical [10–12, 17–20], preparative, and product-scale potential of these methods are illustrated [13, 16, 21, 22]. Polysaccharidebased chiral columns are well suitable for the separation of enantiomers of chiral sulfoxides [10, 11, 13, 14, 18–25] and very high selectivity of enantioseparation has been observed for some of them [20]. Although most of the separations of chiral sulfoxide enantiomers with polysaccharide-based chiral columns are reported with normal-phase eluents, the potential of polar organic mobile phases for this purpose has been also demonstrated in earlier studies [18, 20–22]. The enantiomer elution order (EEO) of some chiral sulfoxides has been reported in earlier studies. However, not enough attention has been paid to the ways EEO can be reversed by the use of various chiral columns and mobile phases. This is an interesting topic from the viewpoint of both analytical and
www.jss-journal.com
1084
M. Gegenava et al.
J. Sep. Sci. 2014, 37, 1083–1088
Figure 1. Structure of chiral sulfoxides included in this study.
preparative scale separations, as well as from a mechanistic point of view. The goal of the present study was the systematic screening of six polysaccharide-based chiral columns for the separation of 17 selected chiral sulfoxides under polar organic mobile phase conditions with emphasis on EEO and the effects of the nature of chiral selector and composition of the mobile phase on it.
2 Materials and methods 2.1 Chemicals Commercially available chiral sulfoxides, adrafinil, chlorbenzid sulfoxide, fenaminophos sulfoxide, fensulfothion, fenthionsulfoxide, lansoprazole, methyl-p-tolyl sulfoxide, methiocarb sulfoxide, methylphenyl sulfoxide, methyl-2phenylsulfinyl acetate, modafinil, omeprazole, pantoprazole, phenylvinyl sulfoxide, propargit, rabeprazole, and ricobendazole were purchased from Sigma–Aldrich (Taufkirchen, Germany). The structures of these compounds are shown in Fig. 1. HPLC quality solvents such as methanol, ethanol, 2-propanol, and acetonitrile were acquired from Carl Roth (Karlsruhe, Germany).
2.2 Chiral columns The amylose tris(3,5-dimethylphenylcarbamate) (ADMPC) based column was an experimental column provided by Enantiosep (M¨unster, Germany), while the other chiral C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 2. Schematic representation of the structures of the chiral selectors in the chiral columns used in this study.
columns, Lux Amylose-2, Lux Cellulose-1, Lux Cellulose-2, Lux Cellulose-3, and Lux Cellulose-4, were kindly provided by Phenomenex (Torrance, CA, USA). The schematic structures of the chiral selectors of these chiral columns are shown in Fig. 2. All columns had the dimensions 250 × 4.6 mm and were packed with 5 m particles. www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 1083–1088 Table 1. Number of enantioselectively recognized chiral compounds with single center of chirality on a given column in a given mobile phase
Column
MeOH
EtOH
2-PrOH
ACN
Total number of recognized compounds
Lux Cellulose-1 Lux Cellulose-2 Lux Cellulose-3 Lux Cellulose-4 Lux Amylose-2 ADMPC
5 8 6 11 5 12
3 6 6 7 5 10
6 2 6 5 5 5
2 5 1 7 8 11
10 12 10 14 11 15
2.3 HPLC All HPLC experiments were performed with an Agilent 1200 HPLC instrument (Agilent Technologies, Waldbronn, Germany) equipped with a G1367C HiP ALS-SL autosampler, G1316B TCC-SL temperature controller, G1311A quaternary pump, and G1314D variable wavelength detector. Chemstation software (version B.03.02-SR2) was used for instrument control, data acquisition, and data handling. If not stated otherwise, the samples were dissolved in the mobile phase used for the respective separation at a concentration of 0.2 mg/mL. HPLC separations were performed at 20⬚C with 1.00 mL/min mobile phase flow rate and detection was performed at 220 nm. The elution order of enantiomers was determined in each case by spiking racemates with enantiomerically pure isomers (in the case of omeprazole) or by correlation with previous studies that reported the EEO with the same chiral selectors under similar conditions. This latter strategy was used to establish the EEO for lansoprazol based on the results of previous study by Andersson et al. [22] and adrafinil and modafinil based on previous study by Rao et al. [26].
3 Results and Discussion 3.1 General overview of separation results The results of the separation of 15 chiral sulfoxides having a single center of chirality are summarized in Table 1. As seen from this table, the most universal column under polar organic mobile phase conditions was the one based on the ADMPC chiral selector. This column alone was able to separate the enantiomers of 14 chiral compounds studied here in either methanol or acetonitrile mobile phase. The remaining 15th chiral compound could be also resolved, in ethanol, making ADMPC a useful chiral selector for the group of chiral sulfoxides included in this study. The second best chiral column was Lux Cellulose-4 resolving the enantiomers of 14 chiral analytes studied here when used in one of the three following mobile phases: methanol, 2-propanol, or acetonitrile. The number of successfully resolved pairs of enantiomers in a single mobile phase was also highest on the ADMPC C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1085
chiral selector (12 pairs in methanol) while the second best was again Lux Cellulose-4 (11 pairs, also in methanol). Cellulose-based columns containing only methyl substituents on the phenyl moiety (Lux Cellulose-1 and Lux Cellulose-3) were less successful compared with the ones possessing both methyl and chloro substituents in almost every mobile phase and especially in acetonitrile. Both amylosebased columns performed better than the cellulose-based columns when acetonitrile was used as mobile phase (Table 1). With the use of mobile phases made of an alcohol of higher molecular weight, the success rate decreased. This result is in agreement with previous observations by Andersson et al. for a smaller group of chiral sulfoxides [22]. Only few exceptions were observed to this rule. This finding indicates that hydrophobic interactions do not positively contribute to chiral recognition for most of studied analytes. The most useful mobile phases proved to be methanol and acetonitrile. At the same time, these two insured the highest degree of complementarity for most of studied columns (i.e. the highest number of unique separations on each column).
3.2 Effect of chiral selector on EEO Among the chiral columns used in this study, only two, namely Lux Cellulose-1 and amylose ADMPC contain the same pendant groups on polysaccharide chains while the chains themselves are different (cellulose and amylose chains, respectively). Still, there is a certain degree of similarity even between these two selectors: both cellulose and amylose consist of the same monomeric D-glucopyranose building blocks linked at positions 1 and 4 of each two neighboring unit; still, the linkage type is  in the case of cellulose and ␣ in the case of amylose. This difference in linkage type is responsible for the difference in the secondary structure of these polysaccharides as well as of their derivatives. Based on the experimentally determined X-ray structure of cellulose trisphenylcarbamate and molecular modeling/molecular mechanics calculations, Okamoto and co-workers proposed cellulose tris(3,5-dimethylphenylcarbamate) to form a lefthanded threefold (3/2) helix with the glucose units regularly arranged along the helical axis [27–30]. In contrast, ADMPC chiral selector is believed to assume a tight, left-handed 4/1 or 5/1 helix, as well as left-handed helical structure (4/3) [27–30]. Wainer and Booth proposed a different model for ADMPC based on computer simulations according to which the helix seems to be larger than 4/1 or 5/1 [31]. These structural differences between the tris(3,5-dimethylphenylcarbamates) of cellulose and amylose may be reflected in their chiral recognition patterns toward some analytes as it has been documented in several previous studies [22, 32]. In the present study, several examples of EEO reversal were observed between Lux Cellulose-1 and the ADMPC column. In particular, the R-(+) enantiomer of lansoprazole eluted before the S-(−) enantiomer on Lux Cellulose-1 in ethanol as mobile phase, while the elution order on ADMPC column was the S-(−) enantiomer before the R-(+) enantiomer in the same www.jss-journal.com
1086
M. Gegenava et al.
Figure 3. Separation of a mixture of enantiomers of lansoprazole (S enantiomer in excess) on Lux Cellulose-1 (A) and ADMPC (B) columns in ethanol as mobile phase. For other experimental conditions see Section 2.
Figure 4. Separation of mixture of enantiomers of modafinil (S enantiomer in excess) on Lux Cellulose-3 (A) and Lux Cellulose-4 (B) columns in methanol as mobile phase. For other experimental conditions see Section 2.
eluent (Fig. 3). The same applied to rabeprazole enantiomers in ethanol as mobile phase, while the S-(−) enantiomer of omeprazole eluted first in 2-propanol as the mobile phase on Lux Cellulose-1 and as the second peak on ADMPC. A reversal in EEO for modafinil was observed between Lux Cellulose-3 and Lux Cellulose-4 using methanol as the mobile phase (Fig. 4). The major difference between these two columns is that the chiral selector in the former has benzoate pendant groups while the latter has carbamate pendant groups. Since there is also a difference in the substituents on the phenyl moiety, it is difficult to assign the reason for reversal in EEO to either the type of pendant group or to the substituent on the phenyl moiety. The fact that the type of substituent on the phenyl moiety may affect the EEO is confirmed by results shown in Fig. 5. The two columns involved are both prepared with amylose derivatives containing carbamate-based pendant groups but with different substituents, still they demonstrate opposite EEO of omeprazole enantiomers in ethanol mobile phase. Similar results have been reported earlier for few chiral sulfoxides by Andersson and co-workers [22]. Most interestingly, the modafinil enantiomers elute in reversed order off two structurally closely related chiral columns (Lux Cellulose-2 and Lux Cellulose-4; see Fig. 6). This example illustrates once again that a rather minor change in the chemistry of the chiral selector can cause a dramatic change in chiral recognition pattern. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2014, 37, 1083–1088
Figure 5. Separation of a mixture of enantiomers of omeprazole (S enantiomer in excess) on Lux Amylose-2 (A) and ADMPC (B) columns in ethanol mobile phase. For other experimental conditions see Section 2.
Figure 6. Separation of a mixture of enantiomers of modafinil (S enantiomer in excess) on Lux Cellulose-2 (A) and Lux Cellulose-4 (B) columns in methanol mobile phase. For other experimental conditions see Section 2.
Figure 7. Separation of a mixture of enantiomers of omeprazole (S enantiomer in excess) on Lux Amylose-2 in acetonitrile (A) and methanol (B) mobile phases. For other experimental conditions see Section 2.
3.3 Effect of mobile phase composition on EEO The effect of mobile phase composition on EEO has been investigated in our previous studies [32–36] as well as by several other groups [22, 33, 37]. Due to the significant differences between alcohols and acetonitrile in terms of polarity and also in their ability to form hydrogen bonds, the reversal of EEO observed in these mobile phases seems quite logical. Thus, the R-(+) enantiomer of omeprazole elutes first off Lux Amylose-2 with methanol as mobile phase while last when acetonitrile is used as mobile phase with the same column (Fig. 7). The EEO of methylphenyl sulfoxide was also opposite on Lux Cellulose-2 with 2-propanol and acetonitrile www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 1083–1088
1087
Figure 9. Separation of a mixture of stereoisomers of phenaminophos sulfoxide on ADMPC column in acetonitrile mobile phase. For other experimental conditions see Section 2.
4 Concluding remarks Figure 8. Separation of a mixture of enantiomers of omeprazole (A) and rabeprazole (B) on ADMPC column in methanol and 2-propanol (A) and methanol and ethanol (B) mobile phases. For other experimental conditions see Section 2.
as mobile phases (data not shown). More striking seems the EEO reversal caused by changing one alcohol-based mobile phase with another (results summarized in Fig. 8). Thus on the ADMPC column, S-(−)-omeprazole elutes before R-(+)omeprazole when methanol is used as mobile phase while the EEO is opposite when 2-propanol is used as the eluent (Fig. 8a). On the same ADMPC column, the EEO of rabeprazole reverts when even the more similar alcohols, methanol or ethanol, are used as alternative mobile phases (Fig. 8b). These results clearly indicate that the intermolecular forces involved in chiral recognition are complex and their understanding is a challenging task.
This study illustrates that the enantiomer affinity pattern of chiral sulfoxides changes depending on the composition of chiral selector and nature of mobile phase. Even very fine differences in the structure of chiral selectors may cause opposite affinity pattern of enantiomers. The same applies to the composition of the mobile phase. Thus, the enantiomer affinity pattern to the chiral selector may be opposite in closely related mobile phases such as methanol and ethanol. This study emphasizes one more time the importance of determining the EEO in chiral HPLC from both, theoretical and practical points of view. This study was financially supported in part by the Rustaveli Georgian National Science Foundation grant No 31/90 for fundamental research.
5
References
[1] Bentley, R., Chem. Soc. Rev. 2005, 34, 609–624.
3.4 Separation of chiral compounds with multiple stereogenic centers including a sulfur atom as chiral center
´ [2] Fernandez, I., Khiar, N., Chem. Rev. 2003, 103, 3651– 3705. ´ G., [3] Hanquet, G., Colobert, F., Lanners, S., Solladie, Arkivoc 2003, 2003, 328–401. [4] Pellissier, H., Tetrahedron 2006, 62, 5559–5601.
Several compounds included in this project possess diverse stereogenic centers including a chiral sulfur atom. Such a complex mixture of stereoisomers proves difficult to resolve under polar organic mobile phase conditions due to the limited analyte retention typically observed under such conditions. Two chiral compounds in the present study contained a chiral sulfur atom along with other stereogenic centers. For example, the chiral agrochemical propargite contains two chiral carbon atoms present in its cyclohexane ring and a sulfoxide group in the side arm resulting in eight stereoisomers. No more than six peaks could be observed with any of the studied columns and mobile phase combinations. In the case of another chiral agrochemical phenaminophos sulfoxide containing together with a chiral sulfur atom, also a chiral phosphorus atom, all four stereoisomers were almost baseline resolved on the ADMPC column with acetonitrile as the mobile phase (Fig. 9). C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[5] O’Mahony, G. E., Kelly, P., Lawrence, S. E., Arkivoc 2011, 2011, 1–110. ´ ´ [6] Wojaczynska, E., Wojaczynski, J., Chem. Rev. 2010, 110, 4303–4356. [7] Pinna, G., Bellucci, M. C., Malpezzi, L., Pisani, L., Superchi, S., Volonterio, A., Zanda, M., Tetrahedron 2011, 67, 5268–5281. [8] Pai, V., Pai, N., J. Indian Med. Assoc. 2007, 105, 469–470, 472, 474. [9] Liu, D. H., Wang, P., Zhu, W. D., Gu, X., Zhou, W. F., Zhou, Z. Q., Food Chem. 2008, 110, 399–405. ¨ [10] Kusters, E., Loux, V., Schmid, E., J. Chromatogr. A 1994, 666, 421–432. [11] Montanari, C. A., Cass, Q. B., Tiritan, M. E., Soares de Souza, A. L., Anal. Chim. Acta 2000, 419, 93–100. [12] Berthod, A., Xiao, T. L., Liu, Y., Jenks, W. S., Armstrong, D. W., J. Chromatogr. A 2002, 955, 53–69.
www.jss-journal.com
1088
M. Gegenava et al.
[13] Tiritan, M. E., Cass, Q. B., Del Alamo, A., Matlin, S. A., Grieb, S. J., Chirality 1998, 10, 573–577. [14] Cass, Q. B., Kohn, C. K., Calafatti, S. A., Aboul-Enein, H. Y., J. Pharm. Biomed. Anal. 2001, 26, 123–130. ˇ ´ [15] Mericko, D., Lehotay, J., Skacˇ ani, I., Armstrong D. W., J. Liq. Chromatogr. Related Technol. 2008, 32, 331–347. [16] Belaz, K. R. A., Denadai, M., Almeida, A. P., Lima, R. T., Vasconcelos, M. H., Pinto, M. M., Cass, Q. B., Oliveira, R. V., J. Pharm. Biomed. Anal. 2012, 66, 100–108.
J. Sep. Sci. 2014, 37, 1083–1088
[25] Cirilli, R., Ferretti, R., Gallinella, B., Turchetto, L., Zanitti, L., La Torre, F., J. Pharm. Biomed. Anal. 2009, 50, 9–14. [26] Rao, R. N., Shinde, D. D., Biomed. Chromatogr. 2009, 23, 811–816. [27] Yamamoto, C., Yashima, E., Okamoto, Y., Bull. Chem. Soc. Jpn. 1999, 72, 1815–1825. [28] Okamoto, Y., Yashima, E., Angew. Chem. Int. Ed. 1998, 37, 1020–1043.
[17] Lourenc¸o, T. C., Armstrong, D. W., Cass, Q. B., Chromatographia 2010, 71, 361–372.
[29] Chen, X., Yamamoto, C., Okamoto Y., Pure Appl. Chem. 2007, 79, 1561–1573.
[18] Chankvetadze, B., Yamamoto, C., Okamoto, Y., Comb. Chem. High Throughput Screening 2000, 3, 497–508.
[30] Ikai, T., Okamoto, Y., Chem. Rev. 2009, 109, 6077–6101.
[19] Chankvetadze, B., Yamamoto, C., Okamoto, Y., J. Chromatogr. A 2001, 922, 127–137. [20] Chankvetadze, B., Yamamoto, C., Okamoto, Y., Chem. Lett. 2000, 29, 1176–1177. [21] Lourenc¸o, T. C., Batista, J. M. Jr., Furlan, M., He, Y., Nafie, L. A., Santana, C. C., Cass, Q. B., J. Chromatogr. A 2012, 1230, 61–65. ¨ [22] Andersson, S., Nelander, H., Ohlen, K., Chirality 2007, 19, 706–715.
[31] Booth, T. D., Wainer, I. W., J. Chromatogr. A 1996, 737, 157–169. [32] Matarashvili, I., Chankvetadze, L., Fanali, S., Farkas, T., Chankvetadze, B., J. Sep. Sci. 2013, 36, 140–147. [33] Chankvetadze, B., J. Chromatogr. A 2012, 1269, 26–51. [34] Mskhiladze, A., Karchkhadze, M., Dadianidze, A., Fanali, S., Farkas, T., Chankvetadze, B., Chromatographia 2013, 76, 1449–1458. [35] Dossou, K. S. S., Edorh, P. A., Chiap, P., Chankvetadze, B., Servais, A.-C., Fillet, M., Crommen, J., J. Sep. Sci., 2011, 34, 1772–1780.
[23] Zanitti, L., Ferretti, R., Gallinella, B., La Torre, F., Sanna, M. L., Mosca, A., Cirilli, R., J. Pharm. Biomed. Anal. 2010, 52, 665–671.
[36] Mosiashvili, L., Chankvetadze, L., Farkas, T., Chankvetadze, B., J. Chromatogr. A 2013, 1317, 167–174.
[24] Cirilli, R., Ferretti, R., Gallinella, B., De Santis, E., Zanitti, L., La Torre, F., J. Chromatogr. A 2008, 1177, 105–113.
[37] Ma, S., Shen, S., Lee, H., Eriksson, M., Zeng, X., Xu, J., Fandrick, K., Yee, N., Senanayake, C., Grinberg, N., J. Chromatogr. A 2009, 1216, 3784–3793.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
J. Sep. Sci. 2014, 37, 1083–1088
Maia Gegenava1 Lali Chankvetadze1 Tivadar Farkas2 Bezhan Chankvetadze1 1 Institute
of Physical and Analytical Chemistry, School of Exact and Natural Sciences, Tbilisi State University, Tbilisi, Georgia 2 Phenomenex Inc., Torrance, CA, USA Received December 8, 2013 Revised February 21, 2014 Accepted February 22, 2014
Research Article
Enantioseparation of selected chiral sulfoxides in high-performance liquid chromatography with polysaccharide-based chiral selectors in polar organic mobile phases with emphasis on enantiomer elution order The separation of the enantiomers of 17 chiral sulfoxides was studied on polysaccharidebased chiral columns in polar organic mobile phases. Enantiomer elution order (EEO) was the primary objective in this study. Two of the six chiral columns, especially those based on amylose tris(3,5-dimethylphenylcarbamate) and cellulose tris(4-chloro-3methylphenylcarbamate) (Lux Cellulose-4) proved to be most successful in the separation of the enantiomers of the studied sulfoxides. Interesting examples of EEO reversal were observed depending on the chiral selector or the composition of the mobile phase. For instance, the R-(+) enantiomer of lansoprazole eluted before the S-(−) enantiomer on Lux Cellulose-1 in both methanol or ethanol as the mobile phase, while the elution order was opposite in the same eluents on amylose tris(3,5-dimethylphenylcarbamate) with the S-(−) enantiomer eluting before the R-(+) enantiomer. The R-(+) enantiomer of omeprazole eluted first on Lux Amylose-2 in methanol but it was second when acetonitrile was used as the mobile phase with the same chiral selector. Several other examples of reversal in EEO were observed in this study. An interesting example of the separation of four stereoisomers of phenaminophos sulfoxide containing chiral sulfur and phosphor atoms is also reported here. Keywords: Chiral sulfoxides / Enantiomer elution order / Enantiomeric separation DOI 10.1002/jssc.201301318
1 Introduction Chiral sulfoxides are widely used as valuable drug compounds (such as omeprazole, lansoprazole, pantoprazole, rabeprazole) [1, 2], pesticides (fipronil, propargit, methiocarb sulfoxide, fensulfothion, etc.) [2], chiral auxiliaries, and chiral ligands to metals [3, 4], in addition they have been incorporated into a variety of natural products and their synthetic derivatives. The synthesis of enantiomerically enriched or pure sulfoxides is of primary interest [5–7]. The enantioselective biological activity of chiral sulfoxides is well documented and became the reason for developing some of them as enantiomerically pure chiral drugs [1]. For instance, together with racemic omeprazole, its enantiomerically pure analogue, (S)-omeprazole is used in clinical practice and Correspondence: Professor Bezhan Chankvetadze, Institute of Physical and Analytical Chemistry, School of Exact and Natural Sciences, Tbilisi State University, Chavchavadze Avenue 3, Tbilisi 0179, Georgia E-mail: [email protected]
Abbreviations: ADMPC, amylose tris(3,5-dimethylphenylcarbamate); EEO, enantiomer elution order C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
some other analogues are also under development [8]. Not enough attention is yet paid to the enantioselective action, toxicity, and biodegradation of chiral sulfoxides used as agrochemicals. However, a few studies indicate the significant difference between the enantiomers of chiral sulfoxides from these points of view [9]. Chromatographic methods have been used for the separation of enantiomers of chiral sulfoxides for more than 20 years [10–25] and together with analytical [10–12, 17–20], preparative, and product-scale potential of these methods are illustrated [13, 16, 21, 22]. Polysaccharidebased chiral columns are well suitable for the separation of enantiomers of chiral sulfoxides [10, 11, 13, 14, 18–25] and very high selectivity of enantioseparation has been observed for some of them [20]. Although most of the separations of chiral sulfoxide enantiomers with polysaccharide-based chiral columns are reported with normal-phase eluents, the potential of polar organic mobile phases for this purpose has been also demonstrated in earlier studies [18, 20–22]. The enantiomer elution order (EEO) of some chiral sulfoxides has been reported in earlier studies. However, not enough attention has been paid to the ways EEO can be reversed by the use of various chiral columns and mobile phases. This is an interesting topic from the viewpoint of both analytical and
www.jss-journal.com
1084
M. Gegenava et al.
J. Sep. Sci. 2014, 37, 1083–1088
Figure 1. Structure of chiral sulfoxides included in this study.
preparative scale separations, as well as from a mechanistic point of view. The goal of the present study was the systematic screening of six polysaccharide-based chiral columns for the separation of 17 selected chiral sulfoxides under polar organic mobile phase conditions with emphasis on EEO and the effects of the nature of chiral selector and composition of the mobile phase on it.
2 Materials and methods 2.1 Chemicals Commercially available chiral sulfoxides, adrafinil, chlorbenzid sulfoxide, fenaminophos sulfoxide, fensulfothion, fenthionsulfoxide, lansoprazole, methyl-p-tolyl sulfoxide, methiocarb sulfoxide, methylphenyl sulfoxide, methyl-2phenylsulfinyl acetate, modafinil, omeprazole, pantoprazole, phenylvinyl sulfoxide, propargit, rabeprazole, and ricobendazole were purchased from Sigma–Aldrich (Taufkirchen, Germany). The structures of these compounds are shown in Fig. 1. HPLC quality solvents such as methanol, ethanol, 2-propanol, and acetonitrile were acquired from Carl Roth (Karlsruhe, Germany).
2.2 Chiral columns The amylose tris(3,5-dimethylphenylcarbamate) (ADMPC) based column was an experimental column provided by Enantiosep (M¨unster, Germany), while the other chiral C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 2. Schematic representation of the structures of the chiral selectors in the chiral columns used in this study.
columns, Lux Amylose-2, Lux Cellulose-1, Lux Cellulose-2, Lux Cellulose-3, and Lux Cellulose-4, were kindly provided by Phenomenex (Torrance, CA, USA). The schematic structures of the chiral selectors of these chiral columns are shown in Fig. 2. All columns had the dimensions 250 × 4.6 mm and were packed with 5 m particles. www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 1083–1088 Table 1. Number of enantioselectively recognized chiral compounds with single center of chirality on a given column in a given mobile phase
Column
MeOH
EtOH
2-PrOH
ACN
Total number of recognized compounds
Lux Cellulose-1 Lux Cellulose-2 Lux Cellulose-3 Lux Cellulose-4 Lux Amylose-2 ADMPC
5 8 6 11 5 12
3 6 6 7 5 10
6 2 6 5 5 5
2 5 1 7 8 11
10 12 10 14 11 15
2.3 HPLC All HPLC experiments were performed with an Agilent 1200 HPLC instrument (Agilent Technologies, Waldbronn, Germany) equipped with a G1367C HiP ALS-SL autosampler, G1316B TCC-SL temperature controller, G1311A quaternary pump, and G1314D variable wavelength detector. Chemstation software (version B.03.02-SR2) was used for instrument control, data acquisition, and data handling. If not stated otherwise, the samples were dissolved in the mobile phase used for the respective separation at a concentration of 0.2 mg/mL. HPLC separations were performed at 20⬚C with 1.00 mL/min mobile phase flow rate and detection was performed at 220 nm. The elution order of enantiomers was determined in each case by spiking racemates with enantiomerically pure isomers (in the case of omeprazole) or by correlation with previous studies that reported the EEO with the same chiral selectors under similar conditions. This latter strategy was used to establish the EEO for lansoprazol based on the results of previous study by Andersson et al. [22] and adrafinil and modafinil based on previous study by Rao et al. [26].
3 Results and Discussion 3.1 General overview of separation results The results of the separation of 15 chiral sulfoxides having a single center of chirality are summarized in Table 1. As seen from this table, the most universal column under polar organic mobile phase conditions was the one based on the ADMPC chiral selector. This column alone was able to separate the enantiomers of 14 chiral compounds studied here in either methanol or acetonitrile mobile phase. The remaining 15th chiral compound could be also resolved, in ethanol, making ADMPC a useful chiral selector for the group of chiral sulfoxides included in this study. The second best chiral column was Lux Cellulose-4 resolving the enantiomers of 14 chiral analytes studied here when used in one of the three following mobile phases: methanol, 2-propanol, or acetonitrile. The number of successfully resolved pairs of enantiomers in a single mobile phase was also highest on the ADMPC C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1085
chiral selector (12 pairs in methanol) while the second best was again Lux Cellulose-4 (11 pairs, also in methanol). Cellulose-based columns containing only methyl substituents on the phenyl moiety (Lux Cellulose-1 and Lux Cellulose-3) were less successful compared with the ones possessing both methyl and chloro substituents in almost every mobile phase and especially in acetonitrile. Both amylosebased columns performed better than the cellulose-based columns when acetonitrile was used as mobile phase (Table 1). With the use of mobile phases made of an alcohol of higher molecular weight, the success rate decreased. This result is in agreement with previous observations by Andersson et al. for a smaller group of chiral sulfoxides [22]. Only few exceptions were observed to this rule. This finding indicates that hydrophobic interactions do not positively contribute to chiral recognition for most of studied analytes. The most useful mobile phases proved to be methanol and acetonitrile. At the same time, these two insured the highest degree of complementarity for most of studied columns (i.e. the highest number of unique separations on each column).
3.2 Effect of chiral selector on EEO Among the chiral columns used in this study, only two, namely Lux Cellulose-1 and amylose ADMPC contain the same pendant groups on polysaccharide chains while the chains themselves are different (cellulose and amylose chains, respectively). Still, there is a certain degree of similarity even between these two selectors: both cellulose and amylose consist of the same monomeric D-glucopyranose building blocks linked at positions 1 and 4 of each two neighboring unit; still, the linkage type is  in the case of cellulose and ␣ in the case of amylose. This difference in linkage type is responsible for the difference in the secondary structure of these polysaccharides as well as of their derivatives. Based on the experimentally determined X-ray structure of cellulose trisphenylcarbamate and molecular modeling/molecular mechanics calculations, Okamoto and co-workers proposed cellulose tris(3,5-dimethylphenylcarbamate) to form a lefthanded threefold (3/2) helix with the glucose units regularly arranged along the helical axis [27–30]. In contrast, ADMPC chiral selector is believed to assume a tight, left-handed 4/1 or 5/1 helix, as well as left-handed helical structure (4/3) [27–30]. Wainer and Booth proposed a different model for ADMPC based on computer simulations according to which the helix seems to be larger than 4/1 or 5/1 [31]. These structural differences between the tris(3,5-dimethylphenylcarbamates) of cellulose and amylose may be reflected in their chiral recognition patterns toward some analytes as it has been documented in several previous studies [22, 32]. In the present study, several examples of EEO reversal were observed between Lux Cellulose-1 and the ADMPC column. In particular, the R-(+) enantiomer of lansoprazole eluted before the S-(−) enantiomer on Lux Cellulose-1 in ethanol as mobile phase, while the elution order on ADMPC column was the S-(−) enantiomer before the R-(+) enantiomer in the same www.jss-journal.com
1086
M. Gegenava et al.
Figure 3. Separation of a mixture of enantiomers of lansoprazole (S enantiomer in excess) on Lux Cellulose-1 (A) and ADMPC (B) columns in ethanol as mobile phase. For other experimental conditions see Section 2.
Figure 4. Separation of mixture of enantiomers of modafinil (S enantiomer in excess) on Lux Cellulose-3 (A) and Lux Cellulose-4 (B) columns in methanol as mobile phase. For other experimental conditions see Section 2.
eluent (Fig. 3). The same applied to rabeprazole enantiomers in ethanol as mobile phase, while the S-(−) enantiomer of omeprazole eluted first in 2-propanol as the mobile phase on Lux Cellulose-1 and as the second peak on ADMPC. A reversal in EEO for modafinil was observed between Lux Cellulose-3 and Lux Cellulose-4 using methanol as the mobile phase (Fig. 4). The major difference between these two columns is that the chiral selector in the former has benzoate pendant groups while the latter has carbamate pendant groups. Since there is also a difference in the substituents on the phenyl moiety, it is difficult to assign the reason for reversal in EEO to either the type of pendant group or to the substituent on the phenyl moiety. The fact that the type of substituent on the phenyl moiety may affect the EEO is confirmed by results shown in Fig. 5. The two columns involved are both prepared with amylose derivatives containing carbamate-based pendant groups but with different substituents, still they demonstrate opposite EEO of omeprazole enantiomers in ethanol mobile phase. Similar results have been reported earlier for few chiral sulfoxides by Andersson and co-workers [22]. Most interestingly, the modafinil enantiomers elute in reversed order off two structurally closely related chiral columns (Lux Cellulose-2 and Lux Cellulose-4; see Fig. 6). This example illustrates once again that a rather minor change in the chemistry of the chiral selector can cause a dramatic change in chiral recognition pattern. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2014, 37, 1083–1088
Figure 5. Separation of a mixture of enantiomers of omeprazole (S enantiomer in excess) on Lux Amylose-2 (A) and ADMPC (B) columns in ethanol mobile phase. For other experimental conditions see Section 2.
Figure 6. Separation of a mixture of enantiomers of modafinil (S enantiomer in excess) on Lux Cellulose-2 (A) and Lux Cellulose-4 (B) columns in methanol mobile phase. For other experimental conditions see Section 2.
Figure 7. Separation of a mixture of enantiomers of omeprazole (S enantiomer in excess) on Lux Amylose-2 in acetonitrile (A) and methanol (B) mobile phases. For other experimental conditions see Section 2.
3.3 Effect of mobile phase composition on EEO The effect of mobile phase composition on EEO has been investigated in our previous studies [32–36] as well as by several other groups [22, 33, 37]. Due to the significant differences between alcohols and acetonitrile in terms of polarity and also in their ability to form hydrogen bonds, the reversal of EEO observed in these mobile phases seems quite logical. Thus, the R-(+) enantiomer of omeprazole elutes first off Lux Amylose-2 with methanol as mobile phase while last when acetonitrile is used as mobile phase with the same column (Fig. 7). The EEO of methylphenyl sulfoxide was also opposite on Lux Cellulose-2 with 2-propanol and acetonitrile www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 1083–1088
1087
Figure 9. Separation of a mixture of stereoisomers of phenaminophos sulfoxide on ADMPC column in acetonitrile mobile phase. For other experimental conditions see Section 2.
4 Concluding remarks Figure 8. Separation of a mixture of enantiomers of omeprazole (A) and rabeprazole (B) on ADMPC column in methanol and 2-propanol (A) and methanol and ethanol (B) mobile phases. For other experimental conditions see Section 2.
as mobile phases (data not shown). More striking seems the EEO reversal caused by changing one alcohol-based mobile phase with another (results summarized in Fig. 8). Thus on the ADMPC column, S-(−)-omeprazole elutes before R-(+)omeprazole when methanol is used as mobile phase while the EEO is opposite when 2-propanol is used as the eluent (Fig. 8a). On the same ADMPC column, the EEO of rabeprazole reverts when even the more similar alcohols, methanol or ethanol, are used as alternative mobile phases (Fig. 8b). These results clearly indicate that the intermolecular forces involved in chiral recognition are complex and their understanding is a challenging task.
This study illustrates that the enantiomer affinity pattern of chiral sulfoxides changes depending on the composition of chiral selector and nature of mobile phase. Even very fine differences in the structure of chiral selectors may cause opposite affinity pattern of enantiomers. The same applies to the composition of the mobile phase. Thus, the enantiomer affinity pattern to the chiral selector may be opposite in closely related mobile phases such as methanol and ethanol. This study emphasizes one more time the importance of determining the EEO in chiral HPLC from both, theoretical and practical points of view. This study was financially supported in part by the Rustaveli Georgian National Science Foundation grant No 31/90 for fundamental research.
5
References
[1] Bentley, R., Chem. Soc. Rev. 2005, 34, 609–624.
3.4 Separation of chiral compounds with multiple stereogenic centers including a sulfur atom as chiral center
´ [2] Fernandez, I., Khiar, N., Chem. Rev. 2003, 103, 3651– 3705. ´ G., [3] Hanquet, G., Colobert, F., Lanners, S., Solladie, Arkivoc 2003, 2003, 328–401. [4] Pellissier, H., Tetrahedron 2006, 62, 5559–5601.
Several compounds included in this project possess diverse stereogenic centers including a chiral sulfur atom. Such a complex mixture of stereoisomers proves difficult to resolve under polar organic mobile phase conditions due to the limited analyte retention typically observed under such conditions. Two chiral compounds in the present study contained a chiral sulfur atom along with other stereogenic centers. For example, the chiral agrochemical propargite contains two chiral carbon atoms present in its cyclohexane ring and a sulfoxide group in the side arm resulting in eight stereoisomers. No more than six peaks could be observed with any of the studied columns and mobile phase combinations. In the case of another chiral agrochemical phenaminophos sulfoxide containing together with a chiral sulfur atom, also a chiral phosphorus atom, all four stereoisomers were almost baseline resolved on the ADMPC column with acetonitrile as the mobile phase (Fig. 9). C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[5] O’Mahony, G. E., Kelly, P., Lawrence, S. E., Arkivoc 2011, 2011, 1–110. ´ ´ [6] Wojaczynska, E., Wojaczynski, J., Chem. Rev. 2010, 110, 4303–4356. [7] Pinna, G., Bellucci, M. C., Malpezzi, L., Pisani, L., Superchi, S., Volonterio, A., Zanda, M., Tetrahedron 2011, 67, 5268–5281. [8] Pai, V., Pai, N., J. Indian Med. Assoc. 2007, 105, 469–470, 472, 474. [9] Liu, D. H., Wang, P., Zhu, W. D., Gu, X., Zhou, W. F., Zhou, Z. Q., Food Chem. 2008, 110, 399–405. ¨ [10] Kusters, E., Loux, V., Schmid, E., J. Chromatogr. A 1994, 666, 421–432. [11] Montanari, C. A., Cass, Q. B., Tiritan, M. E., Soares de Souza, A. L., Anal. Chim. Acta 2000, 419, 93–100. [12] Berthod, A., Xiao, T. L., Liu, Y., Jenks, W. S., Armstrong, D. W., J. Chromatogr. A 2002, 955, 53–69.
www.jss-journal.com
1088
M. Gegenava et al.
[13] Tiritan, M. E., Cass, Q. B., Del Alamo, A., Matlin, S. A., Grieb, S. J., Chirality 1998, 10, 573–577. [14] Cass, Q. B., Kohn, C. K., Calafatti, S. A., Aboul-Enein, H. Y., J. Pharm. Biomed. Anal. 2001, 26, 123–130. ˇ ´ [15] Mericko, D., Lehotay, J., Skacˇ ani, I., Armstrong D. W., J. Liq. Chromatogr. Related Technol. 2008, 32, 331–347. [16] Belaz, K. R. A., Denadai, M., Almeida, A. P., Lima, R. T., Vasconcelos, M. H., Pinto, M. M., Cass, Q. B., Oliveira, R. V., J. Pharm. Biomed. Anal. 2012, 66, 100–108.
J. Sep. Sci. 2014, 37, 1083–1088
[25] Cirilli, R., Ferretti, R., Gallinella, B., Turchetto, L., Zanitti, L., La Torre, F., J. Pharm. Biomed. Anal. 2009, 50, 9–14. [26] Rao, R. N., Shinde, D. D., Biomed. Chromatogr. 2009, 23, 811–816. [27] Yamamoto, C., Yashima, E., Okamoto, Y., Bull. Chem. Soc. Jpn. 1999, 72, 1815–1825. [28] Okamoto, Y., Yashima, E., Angew. Chem. Int. Ed. 1998, 37, 1020–1043.
[17] Lourenc¸o, T. C., Armstrong, D. W., Cass, Q. B., Chromatographia 2010, 71, 361–372.
[29] Chen, X., Yamamoto, C., Okamoto Y., Pure Appl. Chem. 2007, 79, 1561–1573.
[18] Chankvetadze, B., Yamamoto, C., Okamoto, Y., Comb. Chem. High Throughput Screening 2000, 3, 497–508.
[30] Ikai, T., Okamoto, Y., Chem. Rev. 2009, 109, 6077–6101.
[19] Chankvetadze, B., Yamamoto, C., Okamoto, Y., J. Chromatogr. A 2001, 922, 127–137. [20] Chankvetadze, B., Yamamoto, C., Okamoto, Y., Chem. Lett. 2000, 29, 1176–1177. [21] Lourenc¸o, T. C., Batista, J. M. Jr., Furlan, M., He, Y., Nafie, L. A., Santana, C. C., Cass, Q. B., J. Chromatogr. A 2012, 1230, 61–65. ¨ [22] Andersson, S., Nelander, H., Ohlen, K., Chirality 2007, 19, 706–715.
[31] Booth, T. D., Wainer, I. W., J. Chromatogr. A 1996, 737, 157–169. [32] Matarashvili, I., Chankvetadze, L., Fanali, S., Farkas, T., Chankvetadze, B., J. Sep. Sci. 2013, 36, 140–147. [33] Chankvetadze, B., J. Chromatogr. A 2012, 1269, 26–51. [34] Mskhiladze, A., Karchkhadze, M., Dadianidze, A., Fanali, S., Farkas, T., Chankvetadze, B., Chromatographia 2013, 76, 1449–1458. [35] Dossou, K. S. S., Edorh, P. A., Chiap, P., Chankvetadze, B., Servais, A.-C., Fillet, M., Crommen, J., J. Sep. Sci., 2011, 34, 1772–1780.
[23] Zanitti, L., Ferretti, R., Gallinella, B., La Torre, F., Sanna, M. L., Mosca, A., Cirilli, R., J. Pharm. Biomed. Anal. 2010, 52, 665–671.
[36] Mosiashvili, L., Chankvetadze, L., Farkas, T., Chankvetadze, B., J. Chromatogr. A 2013, 1317, 167–174.
[24] Cirilli, R., Ferretti, R., Gallinella, B., De Santis, E., Zanitti, L., La Torre, F., J. Chromatogr. A 2008, 1177, 105–113.
[37] Ma, S., Shen, S., Lee, H., Eriksson, M., Zeng, X., Xu, J., Fandrick, K., Yee, N., Senanayake, C., Grinberg, N., J. Chromatogr. A 2009, 1216, 3784–3793.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jss-journal.com
