Journal of Natural Medicines

, Volume 63, Issue 1, pp 9–14

Triterpenes from Cynomorium songaricium—analysis of HCV protease inhibitory activity, quantification, and content change under the influence of heating

  • Chao-Mei Ma
  • Ying Wei
  • Zhi-Gang Wang
  • Masao Hattori
Original Paper

DOI: 10.1007/s11418-008-0267-7

Cite this article as:
Ma, CM., Wei, Y., Wang, ZG. et al. J Nat Med (2009) 63: 9. doi:10.1007/s11418-008-0267-7

Abstract

Inhibitory activity of the three major triterpenes from the stems of Cynomorium songaricum—ursolic acid, acetyl ursolic acid, and malonyl ursolic acid hemiester—and their related compounds were tested for their inhibitory activity on HCV protease; malonyl ursolic acid hemiester was the most potent. A HPLC–PAD (photo diode array detector)–MS method was established to quantify the contents of each triterpene in C. songaricum. Using this method, the effect heating had on the contents was also investigated. It was found that among the three triterpenes, the content of malonyl ursolic acid hemiester decreased most quickly during the heating process.

Keywords

Ursane Ursolic acid Cynomoriaceae Cynomorium songaricum HCV protease 

Introduction

“SuoYang” (Cynomorium songaricum Rupr.; Cynomoriaceae) is a parasitic plant, which grows mainly in the northern part of China, such as Inner Mongolia Autonomous Region and Gan Su Province. The stems of this plant are reputed in Traditional Chinese Medicine to be a tonic, with few side effects [1]. In addition to medicinal applications, extracts of this plant are frequently added to wines and teas. Previous studies have reported the occurrence of triterpenes, tannins, and other compounds in this plant [2, 3, 4, 5, 6, 7]. The triterpenes and tannins were reported to have inhibitory activity against HIV protease, with malonyl ursolic acid hemiester showing higher activity than ursolic acid [3].

Hepatitis C virus (HCV) infection is a serious problem globally. The NS3/4A protease of HCV is essential for the maturation of the virus and has become one of the important targets for developing anti-HCV drugs. In this study, we tested the HCV protease inhibitory activity of both the triterpenes from C. songaricum and other related compounds.

The quantification of ursolic acid in C. songaricum has been reported previously [8, 9]; however, no report of that of its malonyl hemiester has been available. In the present study, we also developed a HPLC–UV–MS method to quantify the bio-active triterpenes in the stems of C. songaricum. Thus, the influence of heating on the contents of these triterpenes was investigated.

Materials and methods

Compounds and reagents

Compounds 13 were isolated and 47 were synthesized in our laboratory, as described in the literature [3]. HPLC-grade methanol (MeOH), acetonitrile (CH3CN), and trifluoroacetic acid (TFA) were purchased from WAKO Pure Chemical Industries, Ltd. (Osaka, Japan). HPLC-grade water was prepared using an Advantec GS-200 AQUARIUS automatic water distillation apparatus (Advantec Co., Tokyo, Japan).

Plant materials

Cynomorium songaricum Rupr. (Cynomoriaceae) stems were collected in Wulateqianqi, Innermongolia, China, in 1995 and kept in the dark at room temperature (20–28°C). Herbarium voucher specimens were deposited at the Division of Metabolic Engineering, Institute of Natural Medicine, University of Toyama, Japan.

HCV protease assay

SensoLyte 520 HCV Protease Assay Kit *Fluorimetric* (lot# AK71145-1011) and HCV NS3/4A protease (lot# 046-079) were purchased from AnaSpec, San Jose, CA, USA.

A 2-μl sample solution (in DMSO) and 8 μl of freshly diluted enzyme (0.5 μg/ml) were added to each well of a 384-well black Assay plates (BD Falcon). The reaction was started by adding 10 μl of freshly diluted substrate (100 × dilution of a DMSO stocking solution). After being incubated at room temperature (28°C) for 30 min, the fluorescence intensities were measured at Ex/Em = 485 nm/535 nm by a TECAN GENios plate reader. A known HCV protease inhibitor, embelin, was used as a positive control [10].

Inhibition percentages were calculated as 100 × (Fvehicle − Fsample)/Fvehicle = % inhibition, where F is the fluorescence value of vehicle control or of sample minus the fluorescence of the substrate control. All samples were assayed in triplicate and the average inhibitory percentages were used to find out the IC50 by plotting with the concentrations.

Sample preparation

To a 100-ml bottle containing precisely 4.8 mg internal standard 4, EtOAc (96 ml) was added to form a 0.05-mg/ml solution. To 50-ml triangular bottles, exactly 3 g of ground powder from the stems of C. songaricum and 17.0 ml of the EtOAc-containing internal standard solution (0.05 mg/ml) at a ratio of 5.67 ml per g stems were added. The bottles were sealed with parafilm and the mixtures were extracted by sonication (on 10 min, off 10 min, and on again 10 min). Part of each solution was transferred to microglass tubes (5 × 50 mm) and centrifuged at 10,000 rpm for 3 min using an Eppendorf MiniSpin before the supernatants were used for analysis.

To test heating effects on the contents of the triterpenes, test tubes (15 × 160 mm), each containing 0.3 g of the plant powder, were heated in an oven at 100 or 50°C. After a certain time (listed in Table 3), the tubes were taken out and cooled to room temperature, and EtOAc solution containing internal standard in the same ratio as described above was added. The mixtures were sonicated and centrifuged before HPLC analysis in the same manner as described.

To test the heating effects on the dry form of compound 3, we added 200 μl of a methanol solution (0.150 mg/ml) of compound 3 to microglass tubes (5 × 50 mm). After the solutions were completely dried under streams of N2 at room temperature, the tubes were heated at 100°C for 6, 12, or 24 h. Once cooled to room temperature, 200 μl of a MeOH solution containing internal standard was added. The tubes were sonicated for 1 min and the solutions were used for HPLC analysis to quantify compounds 13.

To test the heating effect on compound 3 in the presence of water, a flask containing 20 ml of 0.1 mg/ml 3 in H2O was heated at 100°C with a condenser on the top of the bottle. To each 2-ml microcentrifuge tube, 1 ml of the solution was transferred after 6, 12, or 24 h of heating. After the solution was cooled to room temperature, 0.5 ml of EtOAc solution containing internal standard was added and mixed well. The EtOAc layer was used for HPLC analysis.

Chromatography

HPLC–PAD–MS was carried out on an Agilent 1100 system (Agilent Technologies, Waldbronn, Germany) equipped with degasser, binary pump, photo diode array detector, and atmospheric pressure chemical ionization spectrometry (APCI, Esquire 3000plus, Bruker Daltonik GmbH, Bremen, Germany). Data were acquired, integrated, and analyzed using a ChemStation with the software provided by the manufacturer. HPLC separation was performed on COSMOSIL 5C18-MS-II Waters columns (4.6 × 150 mm, Nacalai Tesque, Inc., Kyoto, Japan).

HPLC–PAD–MS conditions for analysis of the constituents

The mobile phase contained solvents A and B, where A was water:formic acid (99.9:0.1) and B was acetonitrile:formic acid (99.9:0.1). The linear gradient profile was from 80 to 100% B in 15 min and kept at 100% B for 5 min. The wavelength of PAD detection was 205 nm. The flow rate was 1 ml/min for HPLC and PAD detection with the column temperature kept at 40°C. A splitter was connected between PAD and MS detectors, which reduced the flow rate to 0.2 ml/min for MS detection.

Pure compounds 13 in methanol at six concentrations (0.017, 0.033, 0.067, 0.100, 0.133, and 0.167 mg/ml) were used for the standard curves. Concentration of the internal standard (4) was kept at 0.05 mg/ml. Standard curves were plotted using the area ratio of compound to internal standard as Y-axis and the concentration of compound as X-axis. Injection volumes were 10 μl for all compounds and extract.

APCI–MS was operated in negative mode with fragmentation of 1.5 Ampl, scanning range of m/z 50–1000, Volt and trap drive of 49.3. High-purity nitrogen (99.99%) was used as dry gas at a flow rate of 9 l/min and a temperature of 350°C.

Results and discussion

HCV protease inhibitory activity

As shown in Table 1, all three triterpenes (13) from C. songaricium showed inhibitory activity on HCV protease, with malonyl ursolic acid hemiester (3) being the most potent. Other related triterpenes, 47, were then tested and the results were compared with the activity on HIV protease (Table 1). Similar to their activity on HIV protease, all the dicarboxylic acid hemiesters (36) showed good inhibitory activity against HCV protease, while ursolic acid methyl ester (7) was less active. However, in the case of HIV protease, most potent inhibitory activity was observed on the glutaryl hemiester (4); in the case of HCV protease, the malonyl hemiester was the most potent. It will be interesting to investigate the detailed inhibitory mechanism of these compounds. As these triterpenes, especially malonyl ursolic acid hemiester, in this plant showed good inhibitory activity against both HIV-1 and HCV proteases, we then tried to develop a method to quantify the contents of these triterpenes and to investigate their stability.
Table 1

HIV and HCV protease inhibitory activity of triterpenes from C. songaricium and related compounds Open image in new window

Compounda

R1

R2

IC50 (μg/ml)

HIV PRb

HCV PR

1

H

COOH

8

16

2

COCH3

COOH

13

11

5

COCOOH

COOH

7

8

3

COCH2COOH

COOH

6

3

6

CO(CH2)2COOH

COOH

6

5

4

CO(CH2)3COOH

COOH

4

10

7

H

COOCH3

14

94

Embelin

   

4

a1 ursolic acid; 2 acetyl ursolic acid; 3 malonyl ursolic acid hemiester; 4 glutaryl ursolic acid hemiester; 5 oxalyl ursolic acid hemiester; 6 succinyl ursolic acid hemiester; 7 ursolic acid methyl ester

bData cited from Ref. [3]

Quantification of the triterpenes from C. songaricium using HPLC–PAD–MS

Methods using LC–UV (205 nm) or LC–MS to quantify these triterpenes were compared. It was found that the intra- and inter-day precision and accuracy of the LC–UV (205 nm) method were better than those of LC–MS, even for these compounds of weak UV absorptivity. The LC–UV (205 nm) method was thus selected for the quantification while LC–UV–MS was used to confirm the identities of the peaks by comparing with that of standard compounds analyzed under the same conditions. Peak areas were integrated manually when the base lines were not a straight line. Figure 1 shows the HPLC–APCI–MS and HPLC–UV (205 nm) chromatograms of a standard mixture (14) as well as the MS spectra of these compounds. All the compounds displayed [M+Cl] as the pseudo-molecular ions. The [Cl] might come from the glassware used in the experiment. As shown in Table 2, good linearity was found for the calibration curves with the area ratios of standard compounds to internal standard as Y-axis and the concentrations of standard compounds as X-axis.
Fig. 1

HPLC–APCI–MS and HPLC–UV chromatograms and MS spectra of compounds 13 (0.067 mg/ml) and internal standard 4 (0.05 mg/ml)

Table 2

Correlation values for compounds 13 by HPLC–UV (205 nm)

Compound

r2

tRa

LOQ (μg/ml)

Linear regression equation

4

 

7.6–7.9

0.017

Internal standard

1

0.98

8.8–9.2

0.008

Y = 41.24x − 0.475

2

0.99

13.0–13.6

0.008

Y = 33.34x − 0.323

3

0.99

6.2–6.5

0.017

Y = 19.39x − 0.228

r2 regression value for the calibration curve; tR retention time; LOQ limit of quantification (signal to noise ratio 10)

aRetention times were the data from two HPLC columns of the same type

Analysis of C. songaricum samples

Figure 2 shows the HPLC–UV (205 nm) chromatogram, HPLC–APCI–MS (base peak chromatogram, BPC) and extracted ion chromatograms of an EtOAc extract of the stems of C. songaricium with the internal standard (4). The peaks of the compounds were identified by comparing their retention times, UV and mass spectra with those of the standards. Under the HPLC condition, peaks of compounds 14 were well separated from other peaks.
Fig. 2

HPLC chromatogram of a C. songaricium extract containing internal standard (4) detected by means of UV (205 nm) and HPLC–APCI–MS chromatogram together with the extracted ion chromatograms of compounds 14

The contents of compounds 13 in untreated and heat-treated stems of C. songaricium are listed in Table 3. The contents of all three triterpenes decreased during heating. In particular, the content of malonyl ursolic acid hemiester (3) decreased most quickly, from the highest to the lowest of the three triterpenes after 12 h heating at 100°C. HPLC–PAD–MS analysis of the heating effect on malonyl ursolic acid hemiester (3) with and without water showed that this compound decomposed to ursolic acid and acetyl ursolic acid in the presence of water, while the dry form of compound 3 predominately decomposed to acetyl ursolic acid (Table 4). The malonyl group tends to release CO2 to become an acetyl group. The conversion of malonyl ursolic acid hemiester to acetyl ursolic acid (2) has been observed and reported previously [6]. The production of ursolic acid after heating 3 in the presence of water might come from the spontaneous hydrolysis of 3. As shown in Table 4, quantitative analysis revealed that the contents of 3 decreased steadily as the heating time was prolonged. Detailed investigation of the stability of compound 3 using an advanced stability test chamber with precisely controlled temperature and humidity may provide more in-depth and accurate stability data for this compound.
Table 3

Quantitative results for compounds 13 in the stems of C. songaricium treated at different temperatures and times

Compound

Stems concentrations (mg/g)

Untreated

100°C, 6 h

100°C, 12 h

100°C, 24 h

50°C, 53 h

1

0.207

0.190

0.177

0.169

0.182

2

0.264

0.238

0.236

0.218

0.197

3

0.369

0.254

0.155

0.123

0.199

Table 4

Yields (%) of compounds 13 after heating of 3

 

0 h

Heating of dry powder

Heating of 3 in H2O (0.1 mg/ml)

6 h

12 h

24 h

6 h

12 h

24 h

3

99.9

90.7

85.8

80.7

97.8

91.4

81.0

1

0.0

0.0

0.0

0.1

2.1

4.1

7.5

2

0.0

3.3

6.4

10.7

0.0

2.6

6.2

These days, drying oven and drying rooms are employed more and more frequently to process herbal drugs in modern factories, because the method is not affected by weather and has the bonus effect of killing any bacterials and fungi in the herbal drugs during the heating process. Because the fresh stems are very juicy, this heating process has also been adopted to dry the stems of C. songaricum, and the resulting taste, color, and smell of the dried stems are the same as when using the traditional method [11]. Ursolic acid is known to have a range of bioactivity. Malonyl ursolic acid hemiester demonstrated a more potent inhibitory activity than ursolic acid on both HIV protease [3] and HCV protease. As the contents of these constituents, especially the content of malonyl ursolic acid hemiester, decreased significantly during heating, extreme caution should be taken while processing this herbal drug at high temperature.

The presently established method to quantify ursolic acid, acetyl ursolic acid, and malonyl ursolic acid hemiester is very convenient because the sample preparation procedure is simple and the HPLC running time is short. This method may be used with a little modification if needed for the detection of these triterpenes in different samples of C. songaricum and its wine or tea products.

Copyright information

© The Japanese Society of Pharmacognosy and Springer 2008

Authors and Affiliations

  • Chao-Mei Ma
    • 1
  • Ying Wei
    • 1
  • Zhi-Gang Wang
    • 1
  • Masao Hattori
    • 1
  1. 1.Institute of Natural Medicine, University of ToyamaToyamaJapan

Personalised recommendations