Ccf

LLa

b

c

a

ARR1AA

KSUcUcCN

1

poasea

(

h0

Journal of Pharmaceutical and Biomedical Analysis 120 (2016) 72–78

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

journa l homepage: www.e lsev ier .com/ locate / jpba

omparison of ultra-high performance supercritical fluidhromatography and ultra-high performance liquid chromatographyor the separation of spirostanol saponins

ing-ling Zhu a, Yang Zhao a, Yong-wei Xu b, Qing-long Sun b, Xin-guang Sun a,i-ping Kang a, Ren-yi Yan c, Jie Zhang c, Chao Liu a,∗, Bai-ping Ma a,∗∗

Beijing Institute of Radiation Medicine, No. 27, Taiping Road, Beijing 100850, PR ChinaWaters Corporation (Shanghai), Shanghai 201206, PR ChinaOvation Health Science and Technology Co., Ltd., ENN Group, Langfang 065001, PR China,

r t i c l e i n f o

rticle history:eceived 12 September 2015eceived in revised form3 November 2015ccepted 1 December 2015vailable online 3 December 2015

eywords:pirostanol saponinsltra-high performance supercritical fluidhromatographyltra-high performance liquidhromatographyhromatographic behavior

a b s t r a c t

Spirostanol saponins are important active components of some herb medicines, and their isolation andpurification are crucial for the research and development of traditional Chinese medicines. We aimed tocompare the separation of spirostanol saponins by ultra-high performance supercritical fluid chromatog-raphy (UHPSFC) and ultra-high performance liquid chromatography (UHPLC). Four groups of spirostanolsaponins were separated respectively by UHPSFC and UHPLC. After optimization, UHPSFC was performedwith a HSS C18 SB column or a Diol column and with methanol as the co-solvent. A BEH C18 column andmobile phase containing water (with 0.1% formic acid) and acetonitrile were used in UHPLC. We foundthat UHPSFC could be performed automatically and quickly. It is effective in separating the spirostanolsaponins which share the same aglycone and vary in sugar chains, and is very sensitive to the numberand the position of hydroxyl groups in aglycones. However, the resolution of spirostanol saponins withdifferent aglycones and the same sugar moiety by UHPSFC was not ideal and could be resolved by UHPLCinstead. UHPLC is good at differentiating the variation in aglycones, and is influenced by double bonds in

atural productsaglycones. Therefore, UHPLC and UHPSFC are complementary in separating spirostanol saponins. Con-sidering the naturally produced spirostanol saponins in herb medicines are different both in aglyconesand in sugar chains, a better separation can be achieved by combination of UHPLC and UHPSFC. UHPSFC isa powerful technique for improving the resolution when UHPLC cannot resolve a mixture of spirostanolsaponins and vice versa.

© 2015 Elsevier B.V. All rights reserved.

. Introduction

Steroidal saponin is a kind of natural products found in thelants of family Dioscoreaceae, Agavaceae, Alliaceae, Liliaceae and son [1]. Steroidal saponins chemically consist of a steroidal aglyconend the linked oligosaccharide moieties. Among them, furostanol

aponin contains two sugar chains at positions C-3 and C-26 gen-rally, and spirostanol saponin has one sugar chain at position C-3nd a closed F ring. Spirostanol saponins are known to be formed

∗ Corresponding author.∗∗ Corresponding author. Fax: +86 10 68214653.

E-mail addresses: liuchao9588@sina.com (C. Liu), mabaiping@sina.comB.-p. Ma).

ttp://dx.doi.org/10.1016/j.jpba.2015.12.002731-7085/© 2015 Elsevier B.V. All rights reserved.

from furostanol saponins by hydrolysis of glucosyl at C-26 and asubsequent dehydration condensation reaction between the newlyformed hydroxyl group at position C-26 and the hydroxyl group atposition C-22 [2]. Spirostanol saponins have a lot of pharmacologi-cal activities including anti-platelet aggregation and cytotoxicity[3–5]. These activities are strictly dependent on the molecularcomposition and configuration of saponins [6–8]. Therefore it isessential to separate these various spirostanol saponins and theirstereoisomers.

Reversed phase high performance liquid chromatography hasbeen utilized for separations of spirostanol saponins [9–11]. How-ever, it is difficult to separate some of spirostanol saponins,

such as 25R/S-isomers and epimers of sugar parts [12–14]. Ultra-high performance liquid chromatography (UHPLC) using columnspacked with porous sub-2 �m particles possesses the advantages

al and

owrs

rHphcsmoaawassfls

(HseTsiac

2

2

tfShLZR

2

t(gm(g3(r3s[At(a

L.-l. Zhu et al. / Journal of Pharmaceutic

f high throughput and improved resolution [15]. UHPLC has beenidely employed in resolving steroidal saponins and can sepa-

ate furostanol saponins well [16,17]. However, the separation ofpirostanol saponins by UHPLC has not been full characterized.

Thin layer chromatography and silica-gel column chromatog-aphy have been used in the separation of spirostanol saponins.owever, these normal phase chromatography methods are noterformed automatically. In addition, they are environmentallyarmful because of the poisonous organic reagents (such ashloroform) used during separation [18–20]. On the contrary,upercritical fluid chromatography (SFC) with CO2 as the majorobile-phase component is often considered as a “green” technol-

gy [21,22]. SFC uses the same columns as standard HPLC systemsnd is suitable for the analysis of non-polar compounds, and canlso be used to separate polar compounds when modifying CO2ith polar solvents [23,24]. It features high speed, high throughput

nd improved chromatographic performance [25], and has beenuccessfully used in separating chiral chemicals or enantiomers,uch as 25R/S-ergostane triterpenoids [26], spirocyclic terpenoidavor compounds [27], diketopiperazines [28], 25 R/S-spirostanolaponin diastereomers [29].

Ultra-high performance supercritical fluid chromatographyUHPSFC) technology has been developed for several years [30].owever, whether UHPSFC can be well applied in separating

pirostanol saponins with structural diversity is still unknownxcept the report on 25 R/S-spirostanol saponin diastereomers [29].herefore, in this study, we compared the separation of spirostanolaponins by UHPSFC and by UHPLC. The UHPSFC conditions,ncluding co-solvents, additives and columns were optimized. Thedvantages and limitations of both UHPSFC and UHPLC are dis-ussed.

. Materials and methods

.1. Reagents

HPLC grade methanol (MeOH), ethanol (EtOH) and acetoni-rile (ACN) were from Fisher Scientific (USA). Reagent gradeormic acid (FA) and ammonium formate (NH4FA) were fromigma–Aldrich Fluka (Germany). Analytical reagent ammoniumydroxide (NH4OH) was from Sinopharm Chemical Reagent Co.,td. (China). High-purity CO2 (≥99.9%) was purchased fromhenxin Gaisi (China). Ultrapure water was obtained from a Milli-QG Purification unit from Millipore (USA).

.2. Spirostanol saponins

A total of 20 spirostanol saponins prepared and iden-ified in our lab were used in this study and listed inFig. 1). They were saponin Pa (1) [31,32], dioscin (2) [33,34],racillin (3) [33,34], deltonin (4) [35], saponin Pb (5) [36],ixture of deglucolanatigonin (6a) and desgalactotigonin

6b) [13], gitogenin 3-O -�-l-rhamnopyranosyl-(1→2)-�-d-lucopyranoside (7) [37], 5�-spirostane-25(27)-en-2�,3�-diol-O -�-l-rhamnopyranosyl-(1→2)-�-d-glucopyranoside8) [38], 25R-spirostane-5-en-2�,3�-diol 3-O -�-l-hamnopyranosyl-(1→2)-�-d-glucopyranoside (9), ophiogenin-O-�-l-rhamnopyranosyl-(1→2)-� -d-glucopyranoside (10) [39],aponin Tb (11) [40], 25R-dracaenoside F (12) [41], progenin III (13)33,34], timosaponin AIII (14) [42], trillin (15) [43], timosaponin

I (16) [44,45], diosgenin (17) [46], sarsasapogenin (18) [47] and

igogenin (19) [48]. Each compound was dissolved in methanol0.5 mg/mL), and then filtered with 0.22 �m membrane beforenalysis.

Biomedical Analysis 120 (2016) 72–78 73

2.3. UHPLC separation

Liquid chromatography was performed on an ACQUITY UPLCTM

system (Waters Corporation, Milford, MA, USA), which consistedof an ACQUITY UPLC BEH C18 column (1.7 �m, 100 mm × 2.1 mm),a binary solvent manager, a sample manager with a fixed loopof 10 �L, an external column oven, and an ELSD detector. Theelution was performed with H2O (with 0.1% FA) and ACN. The gra-dient of ACN was optimized for each separation. The flow rate was0.5 mL/min with a column temperature of 40 ◦C. Data acquisitionand processing were performed using the software Masslynx 4.1(Waters Corporation, Milford, MA, USA).

2.4. UHPSFC separation

SFC system was the ACQUITY Ultra Performance ConvergenceChromatographyTM (UPC2) system (Waters Corporation, Milford,MA, USA), equipped with a binary solvent delivery pump, a sam-ple manager including partial loop volume injection system, abackpressure regulator, and column manager, as well as a evap-orative light scattering detector. The solvent delivery pump wascompatible with mobile phase flow rates up to 4 mL/min andpressures up to 41.34 MPa. The whole system was controlledwith the EmpowerTM Pro3 Software (Waters Corporation, Mil-ford, MA, USA). Chromatographic analyses were performed onACQUITY UPC2 Torus Diol (1.7 �m, 50 mm × 3.0 mm), ACQUITYUPC2 Torus DEA (1.7 �m, 50 mm × 3.0 mm), ACQUITY UPC2 BEH(1.7 �m, 50 mm × 3.0 mm) and ACQUITY UPC2 HSS C18 SB (1.8 �m,50 mm × 2.1 mm) columns. The mobile phase was CO2 with co-solvents. The gradients of the co-solvent were optimized for eachgroup of spirostanol saponins for a better resolution. The back-pressure was 13.78 MPa, and the flow rate was 1.2 mL/min witha column temperature of 40 ◦C.

3. Results and discussion

3.1. Optimization of UHPSFC conditions

According to the structures of steroidal saponins (Fig. 1), theywere grouped into 4 groups: Compounds 1–5 (group 1) had thesame aglycone and different sugar chains; Compounds 6a and 6b(group 2) had the same skeleton of tigogenin, and were only dif-ferent in the configuration of one hydroxyl group in the terminalglycosyl; Compounds 7–13 (group 3) had the same C-3 sugar chainand varied in hydroxyl group at position C-2, C-14 and C-17 orhydrogen bond (double bounds) at position C-5 and C-6; Com-pounds 13–19 (group 4) had various aglycones. The saponins ineach group were mixed respectively and then separated by UHPLCor by UHPSFC to evaluate the separation efficiency.

In order to find an optimal column for each group of spirostanolsaponins, 4 achiral columns, HSS C18 SB, BEH, DEA and Diol wereevaluated. In all cases methanol was used as co-solvent. Com-pounds 1–5 could be separated into 2, 3, 3 and 5 peaks with HSSC18 SB, BEH, DEA and Diol respectively (Fig. 2A). The Diol col-umn provided a better separation. Similarly, compound 7–13 couldalso be separated with column Diol though the resolution wasnot ideal (Fig. 3A). Therefore, the Diol column was applied in theseparation of compounds 1–13 by UHPSFC in the following tests.On the contrary, column HSS C18 SB was suitable for separatingcompounds 13–19 (Fig. 2B). With column HSS C18 SB, all 7 com-pounds could be clearly separated. Among compounds 13–19, three

compounds (17–19) are non-polar sapogenins and the others havedifferent sugar chains. The HSS C18 SB column contains octadecylcarbon chains bonded silica and residual silanol groups on non-end capped stationary phase [49].The octadecyl carbon chains and

74 L.-l. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 120 (2016) 72–78

spiros

srp

Fig. 1. Chemical structures of the

ilanol groups might help to retain aglycones and the sugar chainsespectively, which may contribute to the better separation of com-ounds 13–19 .

tanol saponins used in this study.

Compounds 7–9 were co-eluted by UHPSFC with the Diol col-umn, while compounds 10–13 were well separated (Fig. 3A: Diol).Therefore we tried to use ethanol as the co-solvent instead of

L.-l. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 120 (2016) 72–78 75

Fig. 2. Separations of compounds 1–5 and 13–19 with different columns; (A): separations of compounds 1–5, the concentrations of methanol: 1–26% (0–1.0 min), 26–29%(1.0–10.0 min). (B): separations of compounds 13–19, the concentrations of methanol: 4–5% (0–5.0 min), 5–13% (5.0–5.5 min), 13–38% (5.5–10 min).

F olumn1 methac

msiC(MTfas

rtacaD

ig. 3. Separations of compounds 7–13 by UPHSFC: (A) separations with different c–20% (1–3 min), 20–40% (3–10 min). (B) separations with the Diol column and

oncentrations of methanol: 1–18% (0–1.0 min), 18% (1.0–10.0 min).

ethanol. But the separation could not be improved (data nothown). Then, we optimized the gradient profile and found somemprovement by changing the gradient profile of methanol inO2 from 1% (0–1 min), 1–20% (1–3 min) and 20–40% (3–10 min)Fig. 3A: Diol) to 1–18% (0–1 min) and 18% (1–10 min) (Fig. 3B:

eOH). However, compounds 7 and 9 could not be separated either.hereafter we included additives in the co-solvent (methanol) andound that 0.2% FA, 0.2% NH4OH or 10 mM NH4FA could not provideny improvement (Fig. 3B). Actually the peak shape deterioratedignificantly when NH4OH and NH4FA were added.

The Diol column contains the propanediol bonded silica andetains compounds due to interactions with the solutes throughheir hydroxyl groups. The polar stationary phase had little inter-ction with double bonds. However, compound 9 is different from

ompound 7 only by its double bond between C-5 and C-6 (Fig. 1,glycone E and F). This difference cannot be distinguished by theiol column, no matter whether these additives were added or not.

s and methanol without additives, the concentrations of methanol: 1% (0–1.0 min),nol with the presence of additives, 0.2% FA, 0.2% NH4OH or 10 mM NH4FA, the

3.2. Comparison of UHPSFC and UHPLC

To compare the separation efficiency of UHPSFC and UHPLC, thespirostanol saponins in each group were also separated by UHPLC,in which the analysis conditions, mobile phase solutions and gra-dient profiles were optimized for each group. The representativeresults obtained under the optimized conditions were comparedand shown in Fig. 4. The separation time was shorter by UHPSFC(around 9 min) than by UHPLC (around 14 min).

Compounds 1–5 had the same aglycone and various sugar chains(Fig. 1). Compounds 1, 2 and 3 had different terminal glycosyls intheir sugar chains, corresponding to arabinosyl, rhamnosyl and glu-cosyl, which were baseline separated in Fig. 4A1. When the type andnumber of glycosyls in sugar chains was different, compounds 4 and

5 were well isolated by UHPSFC. UHPSFC was effective in separatingthe spirostanol saponins which share the same aglycone and varyin sugar chains. On the contrary, there was little influence of sugarchains on retention times of them analyzed with UHPLC. Actually

76 L.-l. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 120 (2016) 72–78

Fig. 4. Separations of four groups of spirostanol saponins by UHPSFC and UHPLC: UPHSFC separations with the Diol column (A1, B1 and C1) and the HSS C18 SB column(D1) and with optimized gradients of methanol, A1: 1–26% (0–1.0 min), 26–29% (1.0–10.0 min); B1: 1–35% (0–2.0 min), 35% (2.0–10.0 min); C1: 1–18% (0–1.0 min), 18%( HPLC

A , 48–37

tdsiscTdht

aoc1rbTo1chWpbcahrct

1.0–10.0 min); D1: 4–5% (0–5.0 min), 5–13% (5.0–5.5 min), 13–38% (5.5–10 min). U2: 41% (0–15 min); B2: 42% (0–15 min); C2: 35–44% (0–9.0 min), 48% (9.1–13.0 min)3–43% (13–13.2 min), 43% (13.2–15 min).

hese times were very close and compounds 4 and 5 could not beistinguished in spite of the different type and number of glyco-yls (Fig. 4A2). Similarly, compounds 6a and 6b were only differentn the configuration of one hydroxyl group in the terminal glyco-yls (galactosyl in 6a and glucosyl in 6b) in the sugar chains. Theyould be separated by UHPSFC and not by UHPLC (Fig. 4B1 and B2).hese results suggest that the spirostanol saponins with only oneifferent glycosyl, like glucosyl or galactosyl, the actual epimers of aydroxyl could be separated by UHPSFC. Furthermore, it is possibleo purify the mixture by preparative-scale SFC [50].

Compounds 7–13 had the same sugar chain at C-3 and vari-nt aglycones with the different number and position of hydroxylsr double bonds (Fig. 1). When they were analyzed by UHPSFC,ompounds 7 and 9 could not be differentiated, and compounds0–13 could be well separated (Fig. 4C1). When they were sepa-ated by UHPLC, baseline resolved peaks of compounds 7–9 coulde obtained, but compounds 11 and 12 were co-eluted (Fig. 4C2).o be noted, compounds 10–13 varied in the number or positionf hydroxyl groups in their aglycones. With UHPSFC compounds0 (two hydroxyl groups) and 13 (no hydroxyl group in the agly-one) were eluted after and before compounds 11 and 12 (oneydroxyl group at different position in the aglycones) respectively.ith UHPLC, it seems difficult to distinguish isomers such as com-

ounds 11 and 12 . These results demonstrated that UHPSFC wasetter in separating the spirostanol saponins with the same sugarhain and varied hydroxyl groups in aglycones. In addition, only theglycone of compound 7 had no double bond; compound 8 and 9

ad a double bond at the different positions in aglycone. The sepa-ation of compounds 7–9 by UHPSFC was not ideal. On the contrary,ompounds 7–9 had the better resolution by UHPLC. The retentionimes of compounds 7–9 in Fig. 4C2 had shown a classical reversed-

separations with the BEH C18 column and with optimized gradients of acetonitrile,5% (13.0–13.2 min), 35% (13.2–15 min); D2: 43–73% (0–9.0 min), 73% (9.0–13.0 min),

phase behavior that a double bond reduced the retention for UHPLCelution. The number and position of double bonds had differentinfluence in the retention of spirostanol saponins in UHPLC andUHPSFC.

Compounds 13–19 were baseline separated by UHPLC (Fig. 4D2),and completely resolved by UHPSFC except compounds 17 and 19(Fig. 4D1). Compounds 13–16 had 2, 2, 1 and 1 glycosyls in sugarchains respectively, and compounds 17–19 had no sugar chains.This led to a big polarity difference between compounds 13–16and compounds 17–19, and the former group was eluted afterand before the latter group by UHPSFC and UHPLC respectively.Compound 15 was eluted before compound 16 by UHPSFC, suggest-ing compound 15 had a lower polarity and should be eluted aftercompound 16 by UHPLC. Actually it was washed out also beforecompound 16 . This might be related to the double bound in com-pound 15, since double bond might reduce the retention time inUHPLC. Compound 17 had also one double bond between C5 andC6, which might contribute to the lower retention time than thoseof compounds 18 and 19 in UHPLC.

4. Conclusion

Spirostanol saponins, as important active components of someherb medicines, have complicated function and structure relation-ships. Their isolation and purification by UHPSFC with the achiralcolumns are crucial for the research and development of the tradi-

tional Chinese medicines. In UHPSFC, a HSS C18 SB column wasappropriate for the differentiation of sapogenins, while a Diolcolumn was implemented in resolving the spirostanol saponinswith sugar chains. UHPSFC is time-saving and quite effective in

al and

scnadosigmisdab

sicfo

A

tMF

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

L.-l. Zhu et al. / Journal of Pharmaceutic

eparating the spirostanol saponins which share the same agly-one and vary in sugar chains. This method is also sensitive to theumber and the position of hydroxyl groups in the aglycones. It hasn obvious advantage in distinguishing the saponins with only oneifferent glycosyl, like glucosyl or galactosyl, the actual epimersf a hydroxyl. However, when the spirostanol saponins have theame sugar chain and differ in aglycones, the separation by UHPSFCs not ideal and can be performed by UHPLC instead. UHPLC isood at separating the spirostanol saponins with the same sugaroiety and various aglycones, especially different double bonds

n aglycones. Therefore, UHPLC and UHPSFC are complementary ineparating spirostanol saponins. Considering that the naturally pro-uced spirostanol saponins in herb medicines are different both inglycones and in sugar chains, a better separation can be achievedy combination of UHPLC and UHPSFC.

In conclusion, we compared the separations of some spirostanolaponins by UHPSFC and UHPLC. UHPSFC and UHPLC are effectiven separating spirostanol saponins based on the differences in sugarhains and aglycones respectively. UHPSFC is a powerful techniqueor improving the resolution when UHPLC cannot resolve a mixturef spirostanol saponins and vice versa.

cknowledgements

This work was supported by National Natural Science Founda-ion of China (No. 81373938), National Science and Technology

ajor Project (No. 2013YQ170525) and Beijing Natural Scienceoundation (No. 7152114).

eferences

[1] S.G. Sparg, M.E. Light, J.V. Staden, Biological activities and distribution of plantsaponins, J. Ethnopharmacol. 94 (2004) 219–243.

[2] G.C. Kite, E.A. Porter, M.S.J. Simmonds, Chromatographic behaviour ofsteroidal saponins studied by high-performance liquidchromatography–mass spectrometry, J. Chromatogr. A 1148 (2007) 177–183.

[3] Z. Liu, W.Y. Gao, S.S. Jing, Y. Zhang, S.L. Man, Y. Wang, Correlation amongcytotoxicity, hemolytic activity and the composition of steroidal saponinsfrom Paris L, J. Ethnopharmacol. 149 (2013) 422–430.

[4] E.A. Barbic, M. Rothenhöfer, J. Heilmann, R. Fürst, G. Jürgenliemk, Spirostanolsaponins and esculin from Rusci rhizome reduce the thrombin-inducedhyperpermeability of endothelial cells, Phytochemistry 90 (2013) 106–113.

[5] Y.S. Wen, W. Ni, X.J. Qin, H. Yan, C.X. Chen, Y. Hua, Steroidal saponins withcytotoxic activity from the rhizomes of Paris polyphylla var. yunnanensis,Phytochem. Lett. 12 (2015) 31–34.

[6] Z.Y. Yu, L. Guo, B. Wang, L.P. Kang, Z.H. Zhao, Y.J. Shan, H. Xiao, J.P. Chen, B.P.Ma, Y.W. Cong, Structural requirement of spirostanol glycosides for ratuterine contractility and mode of their synergism, J. Pharm. Pharmacol. 62(2010) 521–529.

[7] Y.L. Fu, Z.Y. Yu, X.M. Tang, Y. Zhao, X.L. Yuan, S. Wang, B.P. Ma, Y.W. Cong,Pennogenin glycosides with a spirostanol structure are strong plateletagonists: structural requirement for activity and mode of platelet agonistsynergism, J. Thromb. Haemost. 6 (2008) 524–533.

[8] B.P. Ma, B. Feng, H.Z. Huang, Y.W. Cong, Biotransformation of Chinese herbsand their ingredients, Mode Tradit. Chin. Med. Mater. Med. 12 (2010)150–154.

[9] L. Dinan, J. Harmatha, R. Lafont, Chromatographic procedures for the isolationof plant steroids, J. Chromatogr. A 935 (2001) 105–123.

10] H.S. Yu, J. Zhang, L.P. Kang, L.F. Han, P. Zhou, Y. Zhao, C.Q. Xiong, D.W. Tan, X.B.Song, K. Yu, B.P. Ma, Three new saponins from the fresh rhizomes ofPolygonatum kingianum, Chem. Pharm. Bull. 57 (2009) 1–4.

11] Y. Zhao, L.P. Kang, Y.X. Liu, Y.G. Liang, D.W. Tan, Z.Y. Yu, Y.W. Cong, B.P. Ma,Steroidal Saponins from the Rhizome of Paris polyphylla and their cytotoxicactivities, Planta Med. 75 (2009) 356–363.

12] Y.Y. Lu, J.G. Luo, D.R. Xu, X.F. Huang, L.Y. Kong, Characterization of spirostanolsaponins in Solanum torvum by high-performance liquidchromatography/evaporative light scattering detector/electrospray ionizationwith multi-stage tandem mass spectrometry, Rapid Commun. MassSpectrom. 22 (2008) 2447–2452.

13] L.P. Kang, K.L. Wu, H.S. Yu, X. Pang, J. Liu, L.F. Han, J. Zhang, Y. Zhao, C.Q. Xiong,X.B. Song, C. Liu, Y.W. Cong, B.P. Ma, Steroidal saponins from Tribulus

terrestris, Phytochemistry 107 (2014) 182–189.

14] Q.A. Zheng, Y.J. Zhang, H.Z. Li, C.R. Yang, Steroidal saponins from fresh stem ofDracaena cochinchinensis, Steroids 69 (2004) 111–119.

15] G. Haghi, A. Hatami, M. Mehran, UPLC and HPLC of caffeoyl esters in wild andcultivated Arctium lappa L, Food Chem. 138 (2013) 321–326.

[

[

Biomedical Analysis 120 (2016) 72–78 77

16] B. Avula, Y.H. Wang, Z. Ali, T.J. Smillie, I.A. Khan, Chemical fingerprint analysisand quantitative determination of steroidal compounds from Dioscoreavillosa, Dioscorea species, and dietary supplements using UHPLC-ELSD,Biomed. Chromatogr. 28 (2014) 281–294.

17] W.H. Wang, Y. Zhao, W.G. Jing, J. Zhang, H. Xiao, Q. Zha, A. Liu,Ultrahigh-performance liquid chromatography-ion trap mass spectrometrycharacterization of the steroidal saponins of Dioscorea panthaica Prain etBurkill and its application for accelerating the isolation and structuralelucidation of steroidal saponins, Steroids 95 (2015) 51–65.

18] Y.H. Xiao, H.L. Yin, L. Chen, Y. Tian, S.J. Liu, G.J. Zhang, H.W. Chen, H. Jin, B. Li,J.X. Dong, Three spirostanol saponins and a flavane-O-glucoside from thefresh rhizomes of Tupistra chinensis, Fitoterapia 102 (2015) 102–108.

19] H.S. Yu, J. Zhang, L.P. Kang, L.F. Han, B.P. Ma, P. Zhou, Y. Zhao, C.Q. Xiong, D.W.Tan, X.B. Song, K. Yu, Saponins from the processed rhizomes of Polygonatumkingianum, Chem. Pharm. Bull. 57 (2009) 1011–1014.

20] L.M. Wei, Y.C. Wu, C.C. Chen, P.W. Hsieh, W.B. Pan, Tupichinins B–D, threenew spirostanol saponins from Tupistra chinensis rhizomes, Nat. Prod. Res. 28(2014) 74–80.

21] D. Spaggiari, F. Mehl, V. Desfontaine, A.G.G. Perrenoud, S. Fekete, S. Rudaz, D.Guillarme, Comparison of liquid chromatography and supercritical fluidchromatography coupled to compact single quadrupole mass spectrometerfor targeted in vitro metabolism assay, J. Chromatogr. A 1371 (2014) 244–256.

22] T.A. Berger, Demonstration of high speeds with low pressure drops using1.8 �m particles in SFC, Chromatographia 72 (2010) 597–602.

23] E. Lesellier, C. West, The many faces of packed column supercritical fluidchromatography—a critical review, J. Chromatogr. A 1382 (2015) 2–46.

24] M. Ganzera, Supercritical fluid chromatography for the separation ofisoflavones, J. Pharm. Biomed. Anal. 107 (2015) 364–369.

25] C. Desmortreux, M. Rothaupt, C. West, E. Lesellier, Improved separation offurocoumarins of essential oils by supercritical fluid chromatography, J.Chromatogr. A 1216 (2009) 7088–7095.

26] X. Qiao, R. An, Y. Huang, S. Ji, L. Li, Y.M. Tzeng, D.A. Guo, M. Ye, Separation of25R/S-ergostane triterpenoids in the medicinal mushroom Antrodiacamphorata using analytical supercritical-fluid chromatography, J.Chromatogr. A 1358 (2014) 252–260.

27] M. Schaffrath, V. Weidmann, W. Maison, Enantioselective high performanceliquid chromatography and supercritical fluid chromatography separation ofspirocyclic terpenoid flavor compounds, J. Chromatogr. A 1363 (2014)270–277.

28] J. Frenkel, C. Wess, W. Vyverman, G. Pohnert, Chiral separation of adiketopiperazine pheromone from marine diatoms using supercritical fluidchromatography, J. Chromatogr. B 951–952 (2014) 58–61.

29] Y. Zhao, J. McCauley, X. Pang, L.P. Kang, H.S. Yu, J. Zhang, C.Q. Xiong, R. Chen,B.P. Ma, Analytical and semipreparative separation of 25 R/S-spirostanolsaponin diastereomers using supercritical fluid chromatography, J. Sep. Sci.36 (2013) 3270–3276.

30] A.G.G. Perrenoud, J.L. Veuthey, D. Guillarme, Comparison of ultra-highperformance supercritical fluid chromatography and ultra-high performanceliquid chromatography for the analysis of pharmaceutical compounds, J.Chromatogr. A 1266 (2012) 158–167.

31] T. Namba, X.L. Huang, Y.Z. Shu, S.L. Huang, M. Hattori, N. Kakiuchi, Q. Wang,G.J. Xu, Chronotropic effect of the methanolic extracts of the plants of theParis species and steroidal glycosides isolated from P. vietnamensis onspontaneous beating of myocardial cells, Planta Med. 55 (1989) 501–505.

32] N. Toshihiro, Y. Hiroko, S. Michihiro, H. Reiko, M. Kazumoto, K. Toshio, Steroidglycosides in Paris polyphylla, Chem. Pharm. Bull. 21 (1973) 1240–1247.

33] S.B. Singh, R.S. Takur, Costusoside-I and Costusoside-J, two new furostanolsaponins from the seeds of Costus speciosus, Phytochemistry 21 (1982)911–915.

34] P.K. Agrawal, D.C. Jain, P.K. Gupta, R.S. Thakur, Carbon-13 NMR spectroscopyof steroidal sapogenins and steroidal saponins, Phytochemistry 24 (1985)2479–2496.

35] D.C. Jain, Antifeedant active saponin from Balanites roxburghii stem bark,Phytochemistry 26 (1987) 2223–2225.

36] Y. Hirai, S. Sanada, Y. Ida, J. Shoji, Studies on the constituents of Palmae plants.III. The constituents of Chamaerops humilis L. and Trachycarpus wagnerianusBECC, Chem. Pharm. Bull. 34 (1986) 82–87.

37] S.H. Shim, E.J. Lee, J.S. Kim, S.S. Kang, H. Ha, H.Y. Lee, C. Kim, J.H. Lee, K.H. Son,Rat growth-hormone release stimulators from fenugreek seeds, Chem.Biodivers. 5 (2008) 1753–1761.

38] X. Pang, H.S. Yu, L.P. Kang, B.X. Han, Y. Zhao, C.Q. Xiong, D.W. Tan, W. Song, B.Liu, Y.W. Cong, B.P. Ma, Spirostanol saponins derivated from the seeds ofTrigonella foenum-graecum by �-glucosidase hydrolysis and their inhibitoryeffects on rat platelet aggregation, Planta Med. 78 (2012) 276–285.

39] H.F. Dai, J. Zhou, N.H. Tan, Z.T. Ding, New steroidal glycosides fromOphiopogon japonicu, Acta Bot. Sin. 43 (2001) 97–100.

40] N. Toshihiro, M. Kazumoto, K. Toshio, Steroid saponins and sapogenins ofunderground parts of Trillium Kamtschaticum Pall II pennogenin- andkryptogenin 3-O-glycosides and related compounds, Chem. Pharm. Bull. 23(1975) 872–885.

41] Q.A. Zheng, Y.J. Zhang, H.Z. Li, C.R. Yang, Steroidal saponins from fresh stem of

Dracaena cochinensis, Steroids 69 (2004) 111–119.

42] V.L. Grishkovets, P.A. Karpov, S.V. Iksanova, V.Y. Chirva, A steroid glycosidefrom the seeds of Yucca macrocarpa, Chem. Nat. Compd. 34 (1998) 738–740.

43] T.R. Seshadri, S. Vydeeswaran, P.R. Rao, R.S. Thakur, Saponins from Parispolyphylla, Indian J. Chem. 10 (1972) 377–378.

7 al and

[

[

[

[

[

[

supercritical fluid chromatography and liquid chromatography, J.

8 L.-l. Zhu et al. / Journal of Pharmaceutic

44] H. Kurt, H.K. Maryse, N. Koji, Molluscicidal saponins from Cornus florida L,Helv. Chim. Acta. 61 (1978) 1990–1995.

45] W. Di, L. Wang, T. Peng, S.Q. Wang, Study on synthesis of spirostanol saponin3-glycosylation derivatives of Anemarrhena asphodeloides Bge, Chin. J. Chem.13 (2003) 324–328.

46] A. Akira, Y. Fumio, T. Masatoshi, K. Kiyomi, O. Tameto, Steroidal componentsof domestic plants. LVIII. Variation in isodiotigenin and diosgenin content inthe aerial parts of Dioscorea tokoro, Phytochemistry 8 (1969) 2213–2217.

47] M. Saleh, Steroidal constituents of Solanum arundo, Planta Med. 23 (1973)377–378.

[

Biomedical Analysis 120 (2016) 72–78

48] H. Pilgrim, Sapogenin formation in Digitalis purpurea suspension cultures,Phytochemistry 16 (1977) 1311–1312.

49] K. Taguchi, E. Fukusaki, T. Bamba, Simultaneous analysis for water- andfat-soluble vitamins by a novel single chromatography technique unifying

Chromatogr. A 1362 (2014) 270–277.50] A. Tarafder, C. Hudalla, P. Iraneta, K.J. Fountain, A scaling rule in supercritical

fluid chromatography. I. Theory for isocratic systems, J. Chromatogr. A 1362(2014) 278–293.