- Email: [email protected]
Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases
Accepted Manuscript Title: Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases Authors: Huaxia Xin, Zhuoshun Dai, Jianfeng Cai, Yanxiong Ke, Hui Shi, Qing Fu, Yu Jin, Xinmiao Liang PII: DOI: Reference:
S0021-9673(17)30869-5 http://dx.doi.org/doi:10.1016/j.chroma.2017.06.020 CHROMA 358590
To appear in:
Journal of Chromatography A
Received date: Revised date: Accepted date:
11-4-2017 5-6-2017 9-6-2017
Please cite this article as: Huaxia Xin, Zhuoshun Dai, Jianfeng Cai, Yanxiong Ke, Hui Shi, Qing Fu, Yu Jin, Xinmiao Liang, Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases, Journal of Chromatography Ahttp://dx.doi.org/10.1016/j.chroma.2017.06.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases Huaxia Xina, Zhuoshun Daia, Jianfeng Caia, Yanxiong Kea, Hui Shia, Qing Fu a,*, Yu Jina,*, Xinmiao Lianga,b a
Engineering Research Center of Pharmaceutical Process Chemistry, Ministry of
Education, School of Pharmacy, East China University of Science and Technology, Shanghai 200237, PR China b
Key Lab of Separation Science for Analytical Chemistry, Key Lab of Natural
Medicine, Liaoning Province, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, PR China *Corresponding author：Fax: +86-021-64250622; Tel: +86-021-64250633 E-mail address: [email protected] (Y. Jin), [email protected] (Q. Fu)
Highlights 1. Chiral SFC in the separation of diastereoisomers in Piper kadsura. 2. Unconventional modifier DCM providing excellent separation. 3. Stacked automated injections bringing high efficiency in the purification.
Abstract Supercritical fluid chromatography (SFC) with chiral stationary phases (CSPs) is an advanced solution for the separation of achiral compounds in Piper kadsura. Analogues and stereoisomers are abundant in natural products, but there are obstacles in separation using conventional method. In this paper, four lignan diastereoisomers, (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin, from Piper kadsura were separated and purified by chiral SFC. Purification strategy was designed, considering of the compound enrichment, sample purity and purification throughput. Two step achiral purification method on chiral preparative columns with stacked automated injections was developed. Unconventional mobile phase modifier dichloromethane (DCM) was applied to improve the sample solubility. Four diastereoisomers was prepared at the respective weight of 103.1 mg, 10.0 mg, 152.3 mg and 178.6 mg from 710 mg extract with the purity of greater than 98%.
Keywords: Chiral supercritical fluid chromatography, Achiral separation, Diastereoisomers, Purification, Piper kadsura
1 Introduction Due to the complex chemical constitution, the separation of natural products remains a challenge [1], moreover, isomers in natural products tend to exist in forms of analogues and stereoisomers, which are hard to be well separated and purified. We even encounter the sticky situation that several compounds could not be separated at the final stage because of their similar structures. Though various separation strategies have been applied, including columns packing with small size core-shell particles tolerating ultra-high pressure and high temperature [2-7] and multi-dimensional
chromatography methods [8-9], there are still obstacles in solving the separation problem. In addition to the superior performance in the separation of chiral compounds [10-12], researchers found the advantage of chiral stationary phase in the separation of closely-related mixtures and diastereoisomers analytes by chiral SFC [13]. Closely-related species, like a variety of complex mixtures of drug metabolites and analogs separated with SFC using CSPs, have been illustrated [14-16]. Stereoisomerism is also common in natural products, while there are seldom reports on separation of diastereoisomers in traditional Chinese medicine by chiral SFC. Therefore, solving achiral separation problem using chiral column would significantly broaden the application of SFC, especially in the field of traditional Chinese medicine. Though preparative supercritical fluid chromatography (prep-SFC) has gradually emerged as a preferred technique, chromatography theory of prep-SFC in fields of injection conditions [17-19], modifier, pressure and temperature [20] still need further research. Among them, the modifier caught our attention. Modifiers improve the mobile phase polarity of CO2, providing the mixed fluid with elution ability. Alcohol is taken as the conventional modifier for SFC. The most commonly used modifier is methanol (MeOH), but sometimes neat alcohol offers weak sample solubility, resulting in a poor peak shape and the low recovery [21], especially in prep-SFC. To solve this problem, unconventional modifiers like dichloromethane (DCM), ethyl acetate (EA) and methyl tert-butyl ether (MTBE) and so on are used to increase the solubility and to avoid the peak from splitting and tailing [21-23]. In this way, the possibility of sample precipitation would be reduced, preventing the column pressure drop from rising. In addition to the promotion of the solubility, the unconventional modifier could even influence the retention time and selectivity. In this paper, diastereoisomers coming from Piper kadsura (Choisy) Ohwi (P. kadsura, known as Haifengteng) were taken as the research targets. P. kadsura is rich in lignans and neolignans [24-26]; these compounds exist in various kinds of isomers and form abundant steric configuration, which result in different pharmacological
activity and also the difficulty in separation and purification. Before getting the fraction containing four lignan diastereoisomers, (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin (Fig. 1), a series of separation had been done, including preparative high performance liquid chromatography (prep-HPLC). However, it was difficult to separate these four diastereoisomers; ultra-high performance chromatography (UHPLC) and SFC with sub-2 μm particle stationary phase had been tried, but no ideal result was received. Thus, we sought solutions from CSPs column under SFC mode. In the process of SFC purification, the effects of modifier on peak shape were optimized. As to the unconventional modifiers, DCM added in MeOH/CO2 mobile phase could improve solubility of the sample and achieved ideal sample loads and peak shape. The stacked injection methods were optimized by accurate calculation so that the purification throughput was greatly improved. The advantages of SFC like rapid analysis, less mobile phase consuming, labor saving and superior separation were prominently presented in the purification procedure. Successfully, the chiral column realized the separation and purification of four lignan diastereoisomers from one fraction of the extract of P. kadsura.
2 Experimental 2.1 Reagents and materials MeOH of analytical and HPLC grade and DCM of HPLC grade were purchased from Anhui Fulltime Specialized Solvents and J&K (Beijing, China), respectively; The water used in this study was purified by Milli-Q water purification system (Millipore, Bedford, MA, USA). Liquid CO2 (food-grade, 99.9% pure) was purchased from ZhenxinGaisi (Shanghai, China). P. kadsura were collected in Anguo Herb Market, Hebei province (China). After rough sample pretreatment, powered crude medicine P.kadsura was extracted with ethanol, the filtrate was following extracted with petroleum ether (PE). The concentrated PE extract was used as the samples for HPLC preparation. In the primary HPLC preparation, samples were separated into 23 fractions. Four target diastereoisomers, of which the structures were confirmed by NMR,
were contained in one of the 23 fractions. The target analyte fraction used in the following experiments was abbreviated as “Fraction”.
2.2 Instrument UHPLC analysis was performed using a Waters ACQUITY UHPLC® H-Class system which included a quaternary solvent manager, a sample manager-FTN, a column manager and a PDA detector. Data acquisition and processing were conducted using Waters Empower 3 software. SFC separation was carried out on the system of Waters ACQUITY Ultra Performance Convergence ChromatographyTM (ACQUITY UPC2), which includes a binary solvent delivery pump, an auto sampler, a column manager, a photodiode array (PDA) detector and an automatic back pressure regulator (ABPR). SFC experiment control and data acquisition were performed by the Waters EmpowerTM Pro 3 Software. Prep-SFC device was an SFC prep 80 system. It includes a high-pressure CO2 pump, high-pressure co-solvent pump, a mass flow meter, a UV Detector, an automated back pressure regular (ABPR), a manual back pressure regular (MBPR), a PDA detector and six high pressure fraction collection cyclones from Waters (Milford, MA, USA). Data acquisition and processing were conducted using SuperChrom software. NMR spectra was recorded at 400 MHz for 1H and 100 MHz for 13C on a BRUKER AVANCE Ⅲ-400 spectrometer.
2.3 Chromatographic conditions UHPLC anlysis was performed as follows: column was an Acquity UHPLC HSS C18 (100 × 2.1 mm i.d., 1.8 µm, Waters, USA), abbreviated as HSS C18. The mobile phase A was H2O and the mobile phase B was MeOH. The linear gradient was from 40% B to 90%B in 40 min at a flow of 0.2 mL/min. The injection volume was 1 µL. UV detection was performed at 235 nm. SFC analysis was carried out in conditions that: columns were Acquity UPC2™ BEH (50 × 2.1 mm i.d., 1.7 µm), Acquity UPC2™ BEH 2-EP (50 × 2.1 mm i.d., 1.7
µm), and Acquity UPC2™ CSH Fluoro-Phenyl (50 × 2.1 mm i.d., 1.7 µm) were purchased from Waters (Milford, MA, USA), abbreviated as BEH, BEH 2-EP and CSH FP, respectively. The XAmide column (150 × 4.6 mm i.d., 5 µm), CSP-1 (immobilized 3,5-dimethylphenylcarbamate amylose derivative) and CSP-2 (immobilized 3,5-dimethylphenylcarbamate cellulose derivative) columns (250 × 4.6 mm i.d., 5 µm) were synthesized by our laboratory. The mobile phase A was CO2 and B was MeOH. The linear gradient for BEH, BEH 2-EP and CSH FP columns was from 0.5% B to 5% B in 3 min at a flow of 0.8 mL/min. The linear gradient for XAmide column was from 0.5% B to 10% B in 5 min at a flow of 3.0 mL/min. Isocratic elution for CSP-1 and CSP-2 columns was applied at 15% B at a flow of 3.0 mL/min. Other conditions were set at as follows: the back pressure was 13.8 MPa; the column temperature was 30°C; the injection volume was 0.5 µL; UV detection was performed at 235 nm. Prep-SFC were realized as follows: The preparative CSP-1and CSP-2 columns (250 × 20 mm i.d., 5 µm) were synthesized by our laboratory. (1) Method for Fraction primary separated into pairs was performed on CSP-1 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM=1/1. Isocratic purification was used at 15% B at a flow of 60 g/min. 1.8 ml of 50 mg/ml (90 mg) MeOH/DCM =1/1 solution injected. (2) Method for Compounds 1 and 2 was performed on CSP-2 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM =1/1. Isocratic purification was used at 20% B at a flow of 60 g/min. 0.8 ml of 15 mg/ml (12 mg) MeOH/DCM =1/1 solution injected. (3) Method for Compounds 3 and 4 was performed on CSP-2 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM =1/1. Isocratic purification was used at 20% B at a flow of 60 g/min. 1.5 ml of 30 mg/ml (45 mg) MeOH/DCM =1/1 solution injected. Other conditions were set at as follows: the back pressure was 15 MPa; the column temperature was 30°C; UV detection was performed at 235 nm.
3 Results and discussion 3.1 Column screening on SFC
UHPLC and SFC methods were employed to examine the purity of Fraction. Results were summarized in Fig. 2. Though single peak was detected by HPLC system, phenomenon of peak splitting was detected, when tested on UHPLC system (Fig. 2A). The sample was eluted at the retention time of nearly 40 min. Two pairs of slight shoulder peaks were noticed; it could be concluded that HPLC had limited separation ability for this sample. Thereafter, SFC was employed. When analyzed on BEH, BEH 2-EP columns with sub-2 μm particle size, two peaks could be found from Fig. 2B1 and 2B2, but three peaks were separated on CSH-FP column with better resolution (Fig. 2B3). Compared to HPLC, SFC presented preferred resolution. Then, amide-bonded stationary phase XAmide column was tested. Though XAmide column is a conventional specification column with the silica gel particle size of 5.0 μm and generally applied in hydrophilic interaction liquid chromatography, it even showed better resolution than the results of BEH and BEH 2-EP. But all the results above could not meet the resolution demand of purification. In this case, two kinds of chiral columns CSP-1 and CSP-2 were employed. These two polysaccharide derivatization bonded chiral stationary phases were synthesized by our laboratory. The CSP-1 column presented outstanding selectivity, separating four peaks from the Fraction (Fig. 2D1). Surprisingly, a low-content component was found, with four target compounds approximately separated into two pairs. The components were numbered from Compound 1 to Compound 4, respectively. On the CSP-2 column, the retention time of the compounds was changed. Though it did not give an ideal result, different selectivity was showed, eluting the compounds in the order of 2, 1 and 3, 4 (Fig. 2D2). After further purification and structure identification, they were found in forms of diastereoisomers. Compared to the results on HPLC and achiral SFC, chiral SFC provided better separation. In addition, four target diastereoisomers showed different separation selectivity on two chiral columns. Through the design of a reasonable purification strategy, namely, the combination of the CSP-1 column and the CSP-2 column, it was possible to realize the purification of these four compounds.
3.2 Purification of two pairs of isomers on CSP-1 Column with Stacked Automated Injections The Fraction was well separated into two pairs on CSP-1 column. While, it would be difficult to separate these four diastereoisomers in one step on either CSP column. The separation resolution of Compound 1 and 2 with great difference in content is not satisfactory for purification on CSP-1 column (Fig. 2D1). Coelution of Compound 1 and 3 was observed on CSP-2 column (Fig. 2D2). In this way, for the sake of both purification throughput and purity, a two-step purification method for the Fraction was developed. In the first step, the Fraction would be separated into Fraction 1 and Fraction 2 on CSP-1 column, respectively containing Compounds 1, 2 and Compounds 3, 4. Considered the long-time gap nearly 7.5 min between Compounds 2 and 3, stacked automated injections method was employed for the first step (Fig.3). The former pair started at the retention time of 3.9 min and ended at 5.5 min, covering 1.6 min for peak width (Fig. 3A). In addition to the interval time at every 1.6 min, the preparative system would consume about 0.4 min for injection. Taking these two parts into calculation, we could insert three injections in the time gap of 7.5 min (Fig. 3B). Thus, samples could be injected for four times in one cycle of 30 min, and the productivity increased by 277% (from one injection completed in 22.5 min to four injections completed in 32.5 min). In this procedure, four diastereoisomers were gathered into two parts on CSP-1 column, named as Fraction 1 and Fraction 2. Meanwhile, low-content Compound 2 was enriched. 710 mg samples were prepared in twice experiment with four injections in each experiment. Fractions weighed 150 mg and 470 mg, respectively.
3.3 Purification of Compounds 1 and 2 on CSP-2 Column The elution of compound 1 and 2 was reversed on these two CSP columns. Compounds 2 was eluted before the Compound 1 on CSP-2 column. In this study, CSP-2 was chosen to purify the Compound 1 and 2 mainly based on the following reasons. The difference in content between Compound 1 and 2 was obvious. When scaled up to prep-SFC, the peak tailing of Compound 1 would become more serious
on the condition of sample overloading. If Compound 2 was eluted after Compound 1, Compound 2 would be eluted in the peak tail of Compound 1. On the contrary, the reversed elution order could avoid the overlap. The Fig. 4A showed the satisfactory separation of Compound 1 and 2 on the CSP-2 column. Since the peak tailing of Compound 1 was obvious, the purification of compound 2 could be achieved on this column. Based on this, a stacked automated injections method was applied in compound purification (Fig. 4B), reducing labor and time cost greatly. Compound 2 started at the retention time of 5.9 min, and Compound 1 ended at 7.7 min, covering 1.8 min for time width. The second injection was made 2 min later and so on. The cycle time was reduced, with the productivity increased by 291% (from one injection completed in 7.7 min to fourteen injections completed in 37 min). In this way, 150 mg sample was prepared. 103.1 mg Compound 1 and 10.0 mg Compound 2 were respectively gathered in 37 min.
3.4 Purification of Compounds 3 and 4 on CSP-2 Column In the purification of Fraction 2, Compound 3 and 4 were analyzed with MeOH as a modifier. The mobile phase remained the same when scaled up. From the result of Fig. 5A, the peak shape was influenced by the poor sample solubility, with the column pressure risen at the same time. In this case, unconventional modifier DCM was added. Compared to neat MeOH, modifier with DCM provided better peak shape and better resolution (Fig. 5B). DCM made the separation of Compounds 3 and 4 could be implemented, presenting marked improvement. Therefore, as an unconventional modifier, DCM was necessary in purification. In the purification process of Compounds 3 and 4, stacked automated injections method was also applied on CSP-2 column. Similarly, Compound 3 started at the retention time of 5.5 min, and Compound 4 ended at 7.9 min, covering 2.4 min for time width. The second injection was made 3 min later and so on. The cycle time was reduced, with the productivity increased by 235% (from one injection completed in 8 min to ten injections completed in 34 min).
Unconventional modifier DCM provided the sample with excellent solubility and resolution. Compounds 3 and 4 remained baseline separated on preparative scale by CSP-2 column. In this way, 470 mg sample was prepared. 152.3 mg Compound 3 and 178.6 mg Compound 4 were respectively gathered in 34 min.
4 Conclusions In the experiment, four lignan diastereoisomers with purity over 98% were separated from the extract of P. kadsura. Screening of different stationary phases was carried out on UHPLC andSFC system. From the trials of HSS C18, BEH, BEH 2-EP, CSH-FP, XAmide, CSP-1 and CSP-2 columns, polysaccharide-based chiral stationary phases CSP-1 column showed the best result on SFC. Two-step method firstly separated the Fraction into two pairs, followed using CSP-2 column to realize the second separation. In the method, we found different selectivity towards four target compounds between CSP-1 and CSP-2 columns, which solved the enrichment of low level content and problem of poor resolution. Process of scaling up focused on the sample solubility. Also, unconventional mobile phase modifier DCM applied, which provided excellent solubility for the sample and greatly improved the resolution. Stacked automated injections was used in the purification to reduce the labor, time and mobile phase consume. Unit interval productivity was increased to 277% and 291% in the purification of Fraction 1 and 2, respectively. Finally, four diastereoisomers are prepared at the respective weight of 103.1 mg, 10.0 mg, 152.3 mg and 178.6 mg from 710 mg extract.
Acknowledgment This work was supported by National Natural Science Funds (81403100) and Fundamental Research Funds for the Central Universities (222201717016)
References [1] M. Ganzera, A. Murauer, Separation of Natural Products, in: C. F. Poole, Supercritical Fluid Chromatography, Elsevier Inc., Netherlands, 2017, pp. 439-460. [2] J. De Vos, C. Stassen, A. Vaast, G. Desmet, S. Eeltink, High-resolution separations of tryptic digest mixtures using core-shell particulate columns operated at 1,200 bar, J. Chromatogr. A 1264 (2012) 57-62. [3] D. Guillarme, E. Grata, G. Glauser, J.L. Wolfender, J.L. Veuthey, S. Rudaz, Some solutions to obtain very efficient separations in isocratic and gradient modes using small particles size and ultra-high pressure, J. Chromatogr. A 1216 (2009) 3232-3243. [4] J. De Vos, M. De Pra, G. Desmet, R. Swart, T. Edge, F. Steiner, S. Eeltink, High-speed isocratic and gradient liquid-chromatography separations at 1500bar, J. Chromatogr. A 1409 (2015) 138-145. [5] M. Biba, E.L. Regalado, N. Wu, C.J. Welch, Effect of particle size on the speed and resolution of chiral separations using supercritical fluid chromatography, J. Chromatogr. A 1363 (2014) 250-256. [6] J.M. Cunliffe, D.P. Dreyer, R.N. Hayes, R.P. Clement, J.X. Shen, Using temperature to optimize resolution and reduce analysis times for bioanalytical diastereomer LC-MS/MS separations, J. Pharm. Biomed. Anal. 54 (2011) 179-185. [7] S. Fekete, J.L. Veuthey, D. Guillarme, Comparison of the most recent chromatographic approaches applied for fast and high resolution separations: Theory and practice, J. Chromatogr. A 1408 (2015) 1-14. [8] P.Q. Tranchida, P. Donato, F. Cacciola, M. Beccaria, P. Dugo, L. Mondello, Potential of comprehensive chromatography in food analysis, Trends Anal. Chem. 52 (2013) 186-205. [9] M. Sarrut, A. Corgier, G. Cretier, A. Le Masle, S. Dubant, S. Heinisch, Potential and limitations of on-line comprehensive reversed phase liquid chromatographyxsupercritical fluid chromatography for the separation of neutral
compounds: An approach to separate an aqueous extract of bio-oil, J. Chromatogr. A 1402 (2015) 124-133. [10] D. Speybrouck, E. Lipka, Preparative supercritical fluid chromatography: A powerful tool for chiral separations, J. Chromatogr. A 1467 (2016) 33-55. [11] L. Miller, Preparative enantioseparations using supercritical fluid chromatography, J. Chromatogr. A 1250 (2012) 250-255. [12] E. Lemasson, S. Bertin, C. West, Use and practice of achiral and chiral supercritical fluid chromatography in pharmaceutical analysis and purification, J. Sep. Sci. 39 (2016) 212-233. [13] E.L. Regalado, C.J. Welch, Separation of achiral analytes using supercritical fluid chromatography with chiral stationary phases, Trends Anal. Chem. 67 (2015) 74-81. [14] E.L. Regalado, R. Helmy, M.D. Green, C.J. Welch, Chromatographic resolution of closely related species: drug metabolites and analogs, J. Sep. Sci. 37 (2014) 1094-1102. [15] E.L. Regalado, W. Schafer, R. McClain, C.J. Welch, Chromatographic resolution of closely related species: separation of warfarin and hydroxylated isomers, J. Chromatogr. A 1314 (2013) 266-275. [16] E.L. Regalado, A.A. Makarov, R. McClain, M. Przybyciel, C.J. Welch, Search for improved fluorinated stationary phases for separation of fluorine-containing pharmaceuticals from their desfluoro analogs, J. Chromatogr. A 1380 (2015) 45-54. [17] L. Miller, I. Sebastian, Evaluation of injection conditions for preparative supercritical fluid chromatography, J. Chromatogr. A 1250 (2012) 256-263. [18] Y. Dai, G. Li, A. Rajendran, Peak distortions arising from large-volume injections in supercritical fluid chromatography, J. Chromatogr. A 1392 (2015) 91-99. [19] M. Enmark, D. Asberg, A. Shalliker, J. Samuelsson, T. Fornstedt, A closer study of peak distortions in supercritical fluid chromatography as generated by the injection, J. Chromatogr. A 1400 (2015) 131-139. [20] D. Asberg, M. Enmark, J. Samuelsson, T. Fornstedt, Evaluation of co-solvent fraction, pressure and temperature effects in analytical and preparative supercritical
fluid chromatography, J. Chromatogr. A 1374 (2014) 254-260. [21] J.O. DaSilva, B. Coes, L. Frey, I. Mergelsberg, R. McClain, L. Nogle, C.J. Welch, Evaluation of non-conventional polar modifiers on immobilized chiral stationary phases for improved resolution of enantiomers by supercritical fluid chromatography, J. Chromatogr. A 1328 (2014) 98-103. [22] L. Miller, Evaluation of non-traditional modifiers for analytical and preparative enantioseparations using supercritical fluid chromatography, J. Chromatogr. A 1256 (2012) 261-266. [23] L. Miller, Use of dichloromethane for preparative supercritical fluid chromatographic enantioseparations, J. Chromatogr. A 1363 (2014) 323-330. [24] K.H. Kim, J.W. Choi, S.K. Ha, S.Y. Kim, K.R. Lee, Neolignans from Piper kadsura and their anti-neuroinflammatory activity, Bioorg. Med. Chem. Lett. 20 (2010) 409-412. [25] L.C. Lin, C.C. Shen, Y.C. Shen, T.H. Tsai, Anti-inflammatory Neolignans from Piper kadsura, J. Nat. Prod. 69 (2006) 842-844. [26] T. Konisihi, T. Konoshima, A. Daikonya, S. Kitanakak, Neolignans from Piper futokadsura and Their Inhibition of Nitric Oxide Production, Chem. Pharm. Bull. 53(1) (2005) 121-124.
Figure captions Fig.1 Four diastereoisomers separated from the extract of P. kadsura. Compounds 1, 2, 3 and 4 correspond to (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin. Fig. 2 Retention behaviors on the different stationary phases. (A) Acquity UHPLC HSS C18 (100 × 2.1 mm i.d., 1.8 µm) (B1) Acquity UPC2™BEH (50 × 2.1 mm i.d., 1.7 µm) (B2) Acquity UPC2™ BEH 2-EP (50 × 2.1 mm i.d., 1.7 µm) (B3) Acquity UPC2™ CSH Fluoro-Phenyl (50 × 2.1 mm i.d., 1.7 µm) (C) XAmide (150 × 4.6 mm i.d., 5 µm)(D1) CSP-1 (250 × 4.6 mm i.d., 5 µm) (D2) CSP-2(250 × 4.6 mm i.d., 5 µm). Fig. 3 Prep-chromatography of separation step one. (A) Separate four isomers into two pairs on CSP-1 column (250 × 20 mm i.d., 5 µm). The flow rate was 60 g/min. The modifier was MeOH/DCM=1/1. Isocratic purification was used at 15% modifier. 1.8 ml of 50 mg/ml (90 mg) MeOH/DCM =1/1 solution injected. (B) Stacked automated injections on CSP-1 column. Fig. 4 Prep-chromatography of separation step two. (A) Separation of Compound 1 and 2 on CSP-2 column (250 × 20 mm i.d., 5 µm). The flow rate was 60 g/min. The modifier was MeOH/DCM=1/1. Isocratic purification was used at 20% modifier. 0.8 ml of 15 mg/ml (12 mg) MeOH/DCM =1/1 solution injected. (B) Stacked automated injections on CSP-2 column. Fig. 5 Prep-chromatography of separation step two. The flow rate was 60 g/min. (A) Separation of Compound 3 and 4 on CSP-2 column with neat MeOH modifier. Isocratic purification was used at 25% modifier. 1.0 ml of 10 mg/ml (10 mg) MeOH solution injected. (B) Separation of Compound 3 and 4 on CSP-2 column with MeOH/DCM=1/1 modifier. Isocratic purification was used at 20% modifier. 1.5 ml of 30 mg/ml (45 mg) MeOH/DCM =1/1 solution injected. (C) Stacked automated injections on CSP-2 column
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
S0021-9673(17)30869-5 http://dx.doi.org/doi:10.1016/j.chroma.2017.06.020 CHROMA 358590
To appear in:
Journal of Chromatography A
Received date: Revised date: Accepted date:
11-4-2017 5-6-2017 9-6-2017
Please cite this article as: Huaxia Xin, Zhuoshun Dai, Jianfeng Cai, Yanxiong Ke, Hui Shi, Qing Fu, Yu Jin, Xinmiao Liang, Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases, Journal of Chromatography Ahttp://dx.doi.org/10.1016/j.chroma.2017.06.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Rapid purification of diastereoisomers from Piper kadsura using supercritical fluid chromatography with chiral stationary phases Huaxia Xina, Zhuoshun Daia, Jianfeng Caia, Yanxiong Kea, Hui Shia, Qing Fu a,*, Yu Jina,*, Xinmiao Lianga,b a
Engineering Research Center of Pharmaceutical Process Chemistry, Ministry of
Education, School of Pharmacy, East China University of Science and Technology, Shanghai 200237, PR China b
Key Lab of Separation Science for Analytical Chemistry, Key Lab of Natural
Medicine, Liaoning Province, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, PR China *Corresponding author：Fax: +86-021-64250622; Tel: +86-021-64250633 E-mail address: [email protected] (Y. Jin), [email protected] (Q. Fu)
Highlights 1. Chiral SFC in the separation of diastereoisomers in Piper kadsura. 2. Unconventional modifier DCM providing excellent separation. 3. Stacked automated injections bringing high efficiency in the purification.
Abstract Supercritical fluid chromatography (SFC) with chiral stationary phases (CSPs) is an advanced solution for the separation of achiral compounds in Piper kadsura. Analogues and stereoisomers are abundant in natural products, but there are obstacles in separation using conventional method. In this paper, four lignan diastereoisomers, (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin, from Piper kadsura were separated and purified by chiral SFC. Purification strategy was designed, considering of the compound enrichment, sample purity and purification throughput. Two step achiral purification method on chiral preparative columns with stacked automated injections was developed. Unconventional mobile phase modifier dichloromethane (DCM) was applied to improve the sample solubility. Four diastereoisomers was prepared at the respective weight of 103.1 mg, 10.0 mg, 152.3 mg and 178.6 mg from 710 mg extract with the purity of greater than 98%.
Keywords: Chiral supercritical fluid chromatography, Achiral separation, Diastereoisomers, Purification, Piper kadsura
1 Introduction Due to the complex chemical constitution, the separation of natural products remains a challenge [1], moreover, isomers in natural products tend to exist in forms of analogues and stereoisomers, which are hard to be well separated and purified. We even encounter the sticky situation that several compounds could not be separated at the final stage because of their similar structures. Though various separation strategies have been applied, including columns packing with small size core-shell particles tolerating ultra-high pressure and high temperature [2-7] and multi-dimensional
chromatography methods [8-9], there are still obstacles in solving the separation problem. In addition to the superior performance in the separation of chiral compounds [10-12], researchers found the advantage of chiral stationary phase in the separation of closely-related mixtures and diastereoisomers analytes by chiral SFC [13]. Closely-related species, like a variety of complex mixtures of drug metabolites and analogs separated with SFC using CSPs, have been illustrated [14-16]. Stereoisomerism is also common in natural products, while there are seldom reports on separation of diastereoisomers in traditional Chinese medicine by chiral SFC. Therefore, solving achiral separation problem using chiral column would significantly broaden the application of SFC, especially in the field of traditional Chinese medicine. Though preparative supercritical fluid chromatography (prep-SFC) has gradually emerged as a preferred technique, chromatography theory of prep-SFC in fields of injection conditions [17-19], modifier, pressure and temperature [20] still need further research. Among them, the modifier caught our attention. Modifiers improve the mobile phase polarity of CO2, providing the mixed fluid with elution ability. Alcohol is taken as the conventional modifier for SFC. The most commonly used modifier is methanol (MeOH), but sometimes neat alcohol offers weak sample solubility, resulting in a poor peak shape and the low recovery [21], especially in prep-SFC. To solve this problem, unconventional modifiers like dichloromethane (DCM), ethyl acetate (EA) and methyl tert-butyl ether (MTBE) and so on are used to increase the solubility and to avoid the peak from splitting and tailing [21-23]. In this way, the possibility of sample precipitation would be reduced, preventing the column pressure drop from rising. In addition to the promotion of the solubility, the unconventional modifier could even influence the retention time and selectivity. In this paper, diastereoisomers coming from Piper kadsura (Choisy) Ohwi (P. kadsura, known as Haifengteng) were taken as the research targets. P. kadsura is rich in lignans and neolignans [24-26]; these compounds exist in various kinds of isomers and form abundant steric configuration, which result in different pharmacological
activity and also the difficulty in separation and purification. Before getting the fraction containing four lignan diastereoisomers, (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin (Fig. 1), a series of separation had been done, including preparative high performance liquid chromatography (prep-HPLC). However, it was difficult to separate these four diastereoisomers; ultra-high performance chromatography (UHPLC) and SFC with sub-2 μm particle stationary phase had been tried, but no ideal result was received. Thus, we sought solutions from CSPs column under SFC mode. In the process of SFC purification, the effects of modifier on peak shape were optimized. As to the unconventional modifiers, DCM added in MeOH/CO2 mobile phase could improve solubility of the sample and achieved ideal sample loads and peak shape. The stacked injection methods were optimized by accurate calculation so that the purification throughput was greatly improved. The advantages of SFC like rapid analysis, less mobile phase consuming, labor saving and superior separation were prominently presented in the purification procedure. Successfully, the chiral column realized the separation and purification of four lignan diastereoisomers from one fraction of the extract of P. kadsura.
2 Experimental 2.1 Reagents and materials MeOH of analytical and HPLC grade and DCM of HPLC grade were purchased from Anhui Fulltime Specialized Solvents and J&K (Beijing, China), respectively; The water used in this study was purified by Milli-Q water purification system (Millipore, Bedford, MA, USA). Liquid CO2 (food-grade, 99.9% pure) was purchased from ZhenxinGaisi (Shanghai, China). P. kadsura were collected in Anguo Herb Market, Hebei province (China). After rough sample pretreatment, powered crude medicine P.kadsura was extracted with ethanol, the filtrate was following extracted with petroleum ether (PE). The concentrated PE extract was used as the samples for HPLC preparation. In the primary HPLC preparation, samples were separated into 23 fractions. Four target diastereoisomers, of which the structures were confirmed by NMR,
were contained in one of the 23 fractions. The target analyte fraction used in the following experiments was abbreviated as “Fraction”.
2.2 Instrument UHPLC analysis was performed using a Waters ACQUITY UHPLC® H-Class system which included a quaternary solvent manager, a sample manager-FTN, a column manager and a PDA detector. Data acquisition and processing were conducted using Waters Empower 3 software. SFC separation was carried out on the system of Waters ACQUITY Ultra Performance Convergence ChromatographyTM (ACQUITY UPC2), which includes a binary solvent delivery pump, an auto sampler, a column manager, a photodiode array (PDA) detector and an automatic back pressure regulator (ABPR). SFC experiment control and data acquisition were performed by the Waters EmpowerTM Pro 3 Software. Prep-SFC device was an SFC prep 80 system. It includes a high-pressure CO2 pump, high-pressure co-solvent pump, a mass flow meter, a UV Detector, an automated back pressure regular (ABPR), a manual back pressure regular (MBPR), a PDA detector and six high pressure fraction collection cyclones from Waters (Milford, MA, USA). Data acquisition and processing were conducted using SuperChrom software. NMR spectra was recorded at 400 MHz for 1H and 100 MHz for 13C on a BRUKER AVANCE Ⅲ-400 spectrometer.
2.3 Chromatographic conditions UHPLC anlysis was performed as follows: column was an Acquity UHPLC HSS C18 (100 × 2.1 mm i.d., 1.8 µm, Waters, USA), abbreviated as HSS C18. The mobile phase A was H2O and the mobile phase B was MeOH. The linear gradient was from 40% B to 90%B in 40 min at a flow of 0.2 mL/min. The injection volume was 1 µL. UV detection was performed at 235 nm. SFC analysis was carried out in conditions that: columns were Acquity UPC2™ BEH (50 × 2.1 mm i.d., 1.7 µm), Acquity UPC2™ BEH 2-EP (50 × 2.1 mm i.d., 1.7
µm), and Acquity UPC2™ CSH Fluoro-Phenyl (50 × 2.1 mm i.d., 1.7 µm) were purchased from Waters (Milford, MA, USA), abbreviated as BEH, BEH 2-EP and CSH FP, respectively. The XAmide column (150 × 4.6 mm i.d., 5 µm), CSP-1 (immobilized 3,5-dimethylphenylcarbamate amylose derivative) and CSP-2 (immobilized 3,5-dimethylphenylcarbamate cellulose derivative) columns (250 × 4.6 mm i.d., 5 µm) were synthesized by our laboratory. The mobile phase A was CO2 and B was MeOH. The linear gradient for BEH, BEH 2-EP and CSH FP columns was from 0.5% B to 5% B in 3 min at a flow of 0.8 mL/min. The linear gradient for XAmide column was from 0.5% B to 10% B in 5 min at a flow of 3.0 mL/min. Isocratic elution for CSP-1 and CSP-2 columns was applied at 15% B at a flow of 3.0 mL/min. Other conditions were set at as follows: the back pressure was 13.8 MPa; the column temperature was 30°C; the injection volume was 0.5 µL; UV detection was performed at 235 nm. Prep-SFC were realized as follows: The preparative CSP-1and CSP-2 columns (250 × 20 mm i.d., 5 µm) were synthesized by our laboratory. (1) Method for Fraction primary separated into pairs was performed on CSP-1 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM=1/1. Isocratic purification was used at 15% B at a flow of 60 g/min. 1.8 ml of 50 mg/ml (90 mg) MeOH/DCM =1/1 solution injected. (2) Method for Compounds 1 and 2 was performed on CSP-2 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM =1/1. Isocratic purification was used at 20% B at a flow of 60 g/min. 0.8 ml of 15 mg/ml (12 mg) MeOH/DCM =1/1 solution injected. (3) Method for Compounds 3 and 4 was performed on CSP-2 column: mobile phase A was CO2 and the mobile phase B was MeOH/DCM =1/1. Isocratic purification was used at 20% B at a flow of 60 g/min. 1.5 ml of 30 mg/ml (45 mg) MeOH/DCM =1/1 solution injected. Other conditions were set at as follows: the back pressure was 15 MPa; the column temperature was 30°C; UV detection was performed at 235 nm.
3 Results and discussion 3.1 Column screening on SFC
UHPLC and SFC methods were employed to examine the purity of Fraction. Results were summarized in Fig. 2. Though single peak was detected by HPLC system, phenomenon of peak splitting was detected, when tested on UHPLC system (Fig. 2A). The sample was eluted at the retention time of nearly 40 min. Two pairs of slight shoulder peaks were noticed; it could be concluded that HPLC had limited separation ability for this sample. Thereafter, SFC was employed. When analyzed on BEH, BEH 2-EP columns with sub-2 μm particle size, two peaks could be found from Fig. 2B1 and 2B2, but three peaks were separated on CSH-FP column with better resolution (Fig. 2B3). Compared to HPLC, SFC presented preferred resolution. Then, amide-bonded stationary phase XAmide column was tested. Though XAmide column is a conventional specification column with the silica gel particle size of 5.0 μm and generally applied in hydrophilic interaction liquid chromatography, it even showed better resolution than the results of BEH and BEH 2-EP. But all the results above could not meet the resolution demand of purification. In this case, two kinds of chiral columns CSP-1 and CSP-2 were employed. These two polysaccharide derivatization bonded chiral stationary phases were synthesized by our laboratory. The CSP-1 column presented outstanding selectivity, separating four peaks from the Fraction (Fig. 2D1). Surprisingly, a low-content component was found, with four target compounds approximately separated into two pairs. The components were numbered from Compound 1 to Compound 4, respectively. On the CSP-2 column, the retention time of the compounds was changed. Though it did not give an ideal result, different selectivity was showed, eluting the compounds in the order of 2, 1 and 3, 4 (Fig. 2D2). After further purification and structure identification, they were found in forms of diastereoisomers. Compared to the results on HPLC and achiral SFC, chiral SFC provided better separation. In addition, four target diastereoisomers showed different separation selectivity on two chiral columns. Through the design of a reasonable purification strategy, namely, the combination of the CSP-1 column and the CSP-2 column, it was possible to realize the purification of these four compounds.
3.2 Purification of two pairs of isomers on CSP-1 Column with Stacked Automated Injections The Fraction was well separated into two pairs on CSP-1 column. While, it would be difficult to separate these four diastereoisomers in one step on either CSP column. The separation resolution of Compound 1 and 2 with great difference in content is not satisfactory for purification on CSP-1 column (Fig. 2D1). Coelution of Compound 1 and 3 was observed on CSP-2 column (Fig. 2D2). In this way, for the sake of both purification throughput and purity, a two-step purification method for the Fraction was developed. In the first step, the Fraction would be separated into Fraction 1 and Fraction 2 on CSP-1 column, respectively containing Compounds 1, 2 and Compounds 3, 4. Considered the long-time gap nearly 7.5 min between Compounds 2 and 3, stacked automated injections method was employed for the first step (Fig.3). The former pair started at the retention time of 3.9 min and ended at 5.5 min, covering 1.6 min for peak width (Fig. 3A). In addition to the interval time at every 1.6 min, the preparative system would consume about 0.4 min for injection. Taking these two parts into calculation, we could insert three injections in the time gap of 7.5 min (Fig. 3B). Thus, samples could be injected for four times in one cycle of 30 min, and the productivity increased by 277% (from one injection completed in 22.5 min to four injections completed in 32.5 min). In this procedure, four diastereoisomers were gathered into two parts on CSP-1 column, named as Fraction 1 and Fraction 2. Meanwhile, low-content Compound 2 was enriched. 710 mg samples were prepared in twice experiment with four injections in each experiment. Fractions weighed 150 mg and 470 mg, respectively.
3.3 Purification of Compounds 1 and 2 on CSP-2 Column The elution of compound 1 and 2 was reversed on these two CSP columns. Compounds 2 was eluted before the Compound 1 on CSP-2 column. In this study, CSP-2 was chosen to purify the Compound 1 and 2 mainly based on the following reasons. The difference in content between Compound 1 and 2 was obvious. When scaled up to prep-SFC, the peak tailing of Compound 1 would become more serious
on the condition of sample overloading. If Compound 2 was eluted after Compound 1, Compound 2 would be eluted in the peak tail of Compound 1. On the contrary, the reversed elution order could avoid the overlap. The Fig. 4A showed the satisfactory separation of Compound 1 and 2 on the CSP-2 column. Since the peak tailing of Compound 1 was obvious, the purification of compound 2 could be achieved on this column. Based on this, a stacked automated injections method was applied in compound purification (Fig. 4B), reducing labor and time cost greatly. Compound 2 started at the retention time of 5.9 min, and Compound 1 ended at 7.7 min, covering 1.8 min for time width. The second injection was made 2 min later and so on. The cycle time was reduced, with the productivity increased by 291% (from one injection completed in 7.7 min to fourteen injections completed in 37 min). In this way, 150 mg sample was prepared. 103.1 mg Compound 1 and 10.0 mg Compound 2 were respectively gathered in 37 min.
3.4 Purification of Compounds 3 and 4 on CSP-2 Column In the purification of Fraction 2, Compound 3 and 4 were analyzed with MeOH as a modifier. The mobile phase remained the same when scaled up. From the result of Fig. 5A, the peak shape was influenced by the poor sample solubility, with the column pressure risen at the same time. In this case, unconventional modifier DCM was added. Compared to neat MeOH, modifier with DCM provided better peak shape and better resolution (Fig. 5B). DCM made the separation of Compounds 3 and 4 could be implemented, presenting marked improvement. Therefore, as an unconventional modifier, DCM was necessary in purification. In the purification process of Compounds 3 and 4, stacked automated injections method was also applied on CSP-2 column. Similarly, Compound 3 started at the retention time of 5.5 min, and Compound 4 ended at 7.9 min, covering 2.4 min for time width. The second injection was made 3 min later and so on. The cycle time was reduced, with the productivity increased by 235% (from one injection completed in 8 min to ten injections completed in 34 min).
Unconventional modifier DCM provided the sample with excellent solubility and resolution. Compounds 3 and 4 remained baseline separated on preparative scale by CSP-2 column. In this way, 470 mg sample was prepared. 152.3 mg Compound 3 and 178.6 mg Compound 4 were respectively gathered in 34 min.
4 Conclusions In the experiment, four lignan diastereoisomers with purity over 98% were separated from the extract of P. kadsura. Screening of different stationary phases was carried out on UHPLC andSFC system. From the trials of HSS C18, BEH, BEH 2-EP, CSH-FP, XAmide, CSP-1 and CSP-2 columns, polysaccharide-based chiral stationary phases CSP-1 column showed the best result on SFC. Two-step method firstly separated the Fraction into two pairs, followed using CSP-2 column to realize the second separation. In the method, we found different selectivity towards four target compounds between CSP-1 and CSP-2 columns, which solved the enrichment of low level content and problem of poor resolution. Process of scaling up focused on the sample solubility. Also, unconventional mobile phase modifier DCM applied, which provided excellent solubility for the sample and greatly improved the resolution. Stacked automated injections was used in the purification to reduce the labor, time and mobile phase consume. Unit interval productivity was increased to 277% and 291% in the purification of Fraction 1 and 2, respectively. Finally, four diastereoisomers are prepared at the respective weight of 103.1 mg, 10.0 mg, 152.3 mg and 178.6 mg from 710 mg extract.
Acknowledgment This work was supported by National Natural Science Funds (81403100) and Fundamental Research Funds for the Central Universities (222201717016)
References [1] M. Ganzera, A. Murauer, Separation of Natural Products, in: C. F. Poole, Supercritical Fluid Chromatography, Elsevier Inc., Netherlands, 2017, pp. 439-460. [2] J. De Vos, C. Stassen, A. Vaast, G. Desmet, S. Eeltink, High-resolution separations of tryptic digest mixtures using core-shell particulate columns operated at 1,200 bar, J. Chromatogr. A 1264 (2012) 57-62. [3] D. Guillarme, E. Grata, G. Glauser, J.L. Wolfender, J.L. Veuthey, S. Rudaz, Some solutions to obtain very efficient separations in isocratic and gradient modes using small particles size and ultra-high pressure, J. Chromatogr. A 1216 (2009) 3232-3243. [4] J. De Vos, M. De Pra, G. Desmet, R. Swart, T. Edge, F. Steiner, S. Eeltink, High-speed isocratic and gradient liquid-chromatography separations at 1500bar, J. Chromatogr. A 1409 (2015) 138-145. [5] M. Biba, E.L. Regalado, N. Wu, C.J. Welch, Effect of particle size on the speed and resolution of chiral separations using supercritical fluid chromatography, J. Chromatogr. A 1363 (2014) 250-256. [6] J.M. Cunliffe, D.P. Dreyer, R.N. Hayes, R.P. Clement, J.X. Shen, Using temperature to optimize resolution and reduce analysis times for bioanalytical diastereomer LC-MS/MS separations, J. Pharm. Biomed. Anal. 54 (2011) 179-185. [7] S. Fekete, J.L. Veuthey, D. Guillarme, Comparison of the most recent chromatographic approaches applied for fast and high resolution separations: Theory and practice, J. Chromatogr. A 1408 (2015) 1-14. [8] P.Q. Tranchida, P. Donato, F. Cacciola, M. Beccaria, P. Dugo, L. Mondello, Potential of comprehensive chromatography in food analysis, Trends Anal. Chem. 52 (2013) 186-205. [9] M. Sarrut, A. Corgier, G. Cretier, A. Le Masle, S. Dubant, S. Heinisch, Potential and limitations of on-line comprehensive reversed phase liquid chromatographyxsupercritical fluid chromatography for the separation of neutral
compounds: An approach to separate an aqueous extract of bio-oil, J. Chromatogr. A 1402 (2015) 124-133. [10] D. Speybrouck, E. Lipka, Preparative supercritical fluid chromatography: A powerful tool for chiral separations, J. Chromatogr. A 1467 (2016) 33-55. [11] L. Miller, Preparative enantioseparations using supercritical fluid chromatography, J. Chromatogr. A 1250 (2012) 250-255. [12] E. Lemasson, S. Bertin, C. West, Use and practice of achiral and chiral supercritical fluid chromatography in pharmaceutical analysis and purification, J. Sep. Sci. 39 (2016) 212-233. [13] E.L. Regalado, C.J. Welch, Separation of achiral analytes using supercritical fluid chromatography with chiral stationary phases, Trends Anal. Chem. 67 (2015) 74-81. [14] E.L. Regalado, R. Helmy, M.D. Green, C.J. Welch, Chromatographic resolution of closely related species: drug metabolites and analogs, J. Sep. Sci. 37 (2014) 1094-1102. [15] E.L. Regalado, W. Schafer, R. McClain, C.J. Welch, Chromatographic resolution of closely related species: separation of warfarin and hydroxylated isomers, J. Chromatogr. A 1314 (2013) 266-275. [16] E.L. Regalado, A.A. Makarov, R. McClain, M. Przybyciel, C.J. Welch, Search for improved fluorinated stationary phases for separation of fluorine-containing pharmaceuticals from their desfluoro analogs, J. Chromatogr. A 1380 (2015) 45-54. [17] L. Miller, I. Sebastian, Evaluation of injection conditions for preparative supercritical fluid chromatography, J. Chromatogr. A 1250 (2012) 256-263. [18] Y. Dai, G. Li, A. Rajendran, Peak distortions arising from large-volume injections in supercritical fluid chromatography, J. Chromatogr. A 1392 (2015) 91-99. [19] M. Enmark, D. Asberg, A. Shalliker, J. Samuelsson, T. Fornstedt, A closer study of peak distortions in supercritical fluid chromatography as generated by the injection, J. Chromatogr. A 1400 (2015) 131-139. [20] D. Asberg, M. Enmark, J. Samuelsson, T. Fornstedt, Evaluation of co-solvent fraction, pressure and temperature effects in analytical and preparative supercritical
fluid chromatography, J. Chromatogr. A 1374 (2014) 254-260. [21] J.O. DaSilva, B. Coes, L. Frey, I. Mergelsberg, R. McClain, L. Nogle, C.J. Welch, Evaluation of non-conventional polar modifiers on immobilized chiral stationary phases for improved resolution of enantiomers by supercritical fluid chromatography, J. Chromatogr. A 1328 (2014) 98-103. [22] L. Miller, Evaluation of non-traditional modifiers for analytical and preparative enantioseparations using supercritical fluid chromatography, J. Chromatogr. A 1256 (2012) 261-266. [23] L. Miller, Use of dichloromethane for preparative supercritical fluid chromatographic enantioseparations, J. Chromatogr. A 1363 (2014) 323-330. [24] K.H. Kim, J.W. Choi, S.K. Ha, S.Y. Kim, K.R. Lee, Neolignans from Piper kadsura and their anti-neuroinflammatory activity, Bioorg. Med. Chem. Lett. 20 (2010) 409-412. [25] L.C. Lin, C.C. Shen, Y.C. Shen, T.H. Tsai, Anti-inflammatory Neolignans from Piper kadsura, J. Nat. Prod. 69 (2006) 842-844. [26] T. Konisihi, T. Konoshima, A. Daikonya, S. Kitanakak, Neolignans from Piper futokadsura and Their Inhibition of Nitric Oxide Production, Chem. Pharm. Bull. 53(1) (2005) 121-124.
Figure captions Fig.1 Four diastereoisomers separated from the extract of P. kadsura. Compounds 1, 2, 3 and 4 correspond to (-)-Galbelgin, (-)-Ganschisandrin, Galgravin and (-)-Veraguensin. Fig. 2 Retention behaviors on the different stationary phases. (A) Acquity UHPLC HSS C18 (100 × 2.1 mm i.d., 1.8 µm) (B1) Acquity UPC2™BEH (50 × 2.1 mm i.d., 1.7 µm) (B2) Acquity UPC2™ BEH 2-EP (50 × 2.1 mm i.d., 1.7 µm) (B3) Acquity UPC2™ CSH Fluoro-Phenyl (50 × 2.1 mm i.d., 1.7 µm) (C) XAmide (150 × 4.6 mm i.d., 5 µm)(D1) CSP-1 (250 × 4.6 mm i.d., 5 µm) (D2) CSP-2(250 × 4.6 mm i.d., 5 µm). Fig. 3 Prep-chromatography of separation step one. (A) Separate four isomers into two pairs on CSP-1 column (250 × 20 mm i.d., 5 µm). The flow rate was 60 g/min. The modifier was MeOH/DCM=1/1. Isocratic purification was used at 15% modifier. 1.8 ml of 50 mg/ml (90 mg) MeOH/DCM =1/1 solution injected. (B) Stacked automated injections on CSP-1 column. Fig. 4 Prep-chromatography of separation step two. (A) Separation of Compound 1 and 2 on CSP-2 column (250 × 20 mm i.d., 5 µm). The flow rate was 60 g/min. The modifier was MeOH/DCM=1/1. Isocratic purification was used at 20% modifier. 0.8 ml of 15 mg/ml (12 mg) MeOH/DCM =1/1 solution injected. (B) Stacked automated injections on CSP-2 column. Fig. 5 Prep-chromatography of separation step two. The flow rate was 60 g/min. (A) Separation of Compound 3 and 4 on CSP-2 column with neat MeOH modifier. Isocratic purification was used at 25% modifier. 1.0 ml of 10 mg/ml (10 mg) MeOH solution injected. (B) Separation of Compound 3 and 4 on CSP-2 column with MeOH/DCM=1/1 modifier. Isocratic purification was used at 20% modifier. 1.5 ml of 30 mg/ml (45 mg) MeOH/DCM =1/1 solution injected. (C) Stacked automated injections on CSP-2 column
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5