Investigation on the origin of 5-HMF in Shengmaiyin decoction by RP-HPLC method

Abstract

The origin of 5-HMF (5-hydroxymethyl-2-furaldehyde) in a Shengmaiyin decoction was investigated by the RP-HPLC method below. A C₁₈ column (250 mm×4.6 mm, i.d. 5 μm) with a column temperature of 25 °C was used. The mobile phase was a mixture of ultra-pure water-acetonitrile (95:5, V/V) and the flow rate was 1.0 ml/min. The detection wavelength was 280 nm. The injection volume was 1 μl and the running time was about 20 min. The addition of Schisandra was regulated to assess the contribution of an acid environment to the production of 5-HMF. In order to confirm the role of saccharides in the production of 5-HMF, different amount of fructose was used. The 5-HMF level in decoctions of processed and unprocessed Schisandra was investigated in order to determine the origin of 5-HMF. The results showed that 5-HMF was derived mainly from the decoction of Schisandra only and not the mixed decoction of Ophionpogon and Schisandra. The appearance of 5-HMF is not simply the result of the decomposition of saccharides under the acid environment created by Schisandra, but the processing procedure plays an important role in the production of 5-HMF.

INTRODUCTION

Shengmaiyin or Shengmaisan, one of the famous herbal recipes in traditional Chinese medicine, is made from Radix ginseng, Ophiopogon and Schisandra. It is used to treat hypotension (Chen et al., 2000; Gui, 2001), cardiovascular system diseases (Qian et al., 2004; Zeng and Zeng, 2004), to increase immunity (Wu et al., 1997; Guo et al., 2004) and cure other diseases (Li et al., 1996; Liu et al., 1998; Fang et al., 2000; Yi et al., 2003; Xu et al., 2003; Wang et al., 2004a; Wang et al., 2004b). Zhu et al.(1998a) found a new component, 5-hydroxymethyl-2-furaldehyde (5-HMF) in the decoction of Shengmaiyin and believed that 5-HMF derives from the decoction of Ophiopogon and Schisandra in the decomposition of saccharide under the acid environment created by Schisandra (Xia et al., 1998). The contribution of 5-HMF to the cure of cardiovascular disease (Zhu et al., 1998b) was also found in many other herbs (Li et al., 1999; Zhang et al., 2001; Meng et al., 2002; Wang et al., 2003; Dong and Wang, 2004; Yuan et al., 2004; Xu et al., 2004).

Normally, 5-HMF is poisonous to the nervous system (Chi et al., 1998), by combining to protein and thus causing the accumulation of poisons in the body; and has potent reproductive toxicity (Zdzienicka et al., 1978; Wang et al., 1994; Qasim et al., 1995), and also damages striated muscles and viscera (Reinald et al., 1995). Many countries strictly limit the content of 5-HMF in honey, beer and glucose injections (Cohen et al., 1994; Lo et al., 1995; Zeng et al., 2002; Li et al., 2004).

It is very important to control the level of 5-HMF for bio-activity and side effects of the herbs. The origin of 5-HMF in the Shengmaiyin was investigated in order to optimize Shengmaiyin production to acheieve quality control and high curative effect with minimal side-effects, and to contribute to better understanding of traditional Chinese medicine.

MATERIALS AND INSTRUMENTS

Materials and reagents

Unprocessed Schisandra is the dry, ripe fruit of Schisandra chinensis (Turcz.) Baill, but medicinal Schisandra is made from both Schisandra chinensis (Turcz.) Baill and Schisandra sphenanthera Rehd. et Wils. by different processes. Ophiopogon is the dry root of Ophiopogon japonicus (Thunb.) Ker-Gawl. All crude herbal and medicinal medicines were authenticated by Professor Chen Kong-rong, Zhejiang College of Traditional Chinese Medicine and met the standards of the Pharmacopoeia of PRC 2005 Edition. Chemicals used included HPLC grade acetonitrile (Merck, Germany), ultra-pure water (milli-Q) and 5-HMF (Sigma, USA).

Instrumentation

PHSJ-4A pH meter and Agilent HP 1100 HPLC systems equipped with a G1322A solvent degasser, a G1354A quaternary gradient solvent pump, a G1313A multiple auto-sampler, a G1316A thermostat column compartment, a G1314A UV-Vis detector and Agilent HP 1100 chromatography workstation were used for analyses.

METHODS

Analytical methods

The standards and samples of 5-HMF were all analyzed by RP-HPLC. Chromatographic conditions were as follow: Chromatographic separation was performed on a Diamonsil C₁₈ column (250 mm×4.6 mm i.d., 5 μm; Dikma) with a guard column (4 mm×3 mm i.d., 5 μm; Phenomenex) packed with the same material at 25 °C. The mobile phase was a mixture of ultra-pure water-acetonitrile (95:5, V/V) filtered through a 0.45 μm filter and delivered at a flow rate of 1.0 ml/min and detection wavelength of 280 nm, injection volume was 1 μl and running time was about 20 min.

Preparation of calibration curve

The stock solution of 5-HMF was prepared by dissolving 13.62 mg 5-HMF in a 50 ml volumetric flask with ultra-pure water and stored at 4 °C. Seven concentrations of 5-HMF standard solutions (2.724, 5.448, 8.172, 10.896, 13.620, 16.344, 27.240 mg/100 ml) were obtained by appropriately diluting the stock solution with mobile phase and prepared for evaluating the linearity by HPLC. The injection volume was 1 μl. The calibration curve was obtained by plotting the peak areas against the concentrations of 5-HMF standard solutions.

Samples preparation

Schisandra and fructose samples: 12.5 g medicinal Schisandra and a series of different weights of fructose (0 g, 0.25 g, 0.5 g, 1.0 g, 1.5 g, 2.0 g) were soaked in 250 ml deionized water for 2 h and weighed, then decocted for 1 h at boiling temperature and reweighed. Deionized water was added to the decocted samples to make them reach the weight before decoction. The decoction was filtered and the filtrate was passed through a 0.45 μm filter for HPLC analysis.

Ophiopogon and acid samples: 25 g medicinal Ophiopogon were soaked in a series of different pH value acid solutions regulated by hydrochloric acid, sulfuric acid and acetic acid (HCl: 2.972, 3.142, 3.454; H₂SO₄: 2.987, 3.058, 3.401; HAc: 2.894, 3.012, 3.331) for 2 h and weighed, then decocted for 1 h at boiling temperature, reweighed, and then deionized water was added to the decocted samples to make them reach the weight of the sample before decoction. The decoction was filtered and the filtrate was passed through a 0.45 μm filter for HPLC analysis. Ophiopogon samples were prepared by soaking 25 g medicinal Ophiopogon in 250 ml deionized water for 2 h and then treated the same way as that for the Ophiopogon and acid samples.

Schisandra and Ophiopogon samples preparation: 25 g medicinal Ophiopogon and a series of different weights of medicinal or unprocessed Schisandra (6.25 g, 12.5 g, 18.75 g, 25 g, 50 g) were soaked in 250 ml deionized water for 2 h and weighed, then decocted for 1 h at boiling temperature, reweighed, and then deionized water was added to the decocted samples to make them reach the weight of the sample before decoction. The decoction was filtered and the filtrate was passed through a 0.45 μm filter for HPLC analysis.

Schisandra samples: different weights of medicinal or unprocessed Schisandra (6.25 g, 12.5 g, 18.75 g, 25 g, 50 g) were soaked in 250 ml deionized water for 2 h and weighed, then decocted for 1 h at boiling temperature and reweighed, and then deionized water was added to the decocted samples to make them reach the weight before decoction. The decoction was filtered and the filtrate was passed through a 0.45 μm filter for HPLC analysis.

Precision

Both standard solutions and samples were injected at a volume of 1 μl six times repeatedly according to the analytical method mentioned above, and the relative standard deviations were calculated.

Stability

Both standard solutions and sample decoctions at 2 h, 4 h, 6 h, 8 h, 10 h, 12 h and 24 h were analyzed according to the analytical method mentioned above, and the relative standard deviations were calculated.

Accuracy

In 250 ml deionized water 12.5 g processed medicinal Schisandra were soaked for 2 h and weighed, then decocted for 1 h at boiling temperature and reweighed, and then deionized water was added to the decocted samples to make them reach the weight before decoction. The decoction was filtered and the filtrate was passed through a 0.45 μm filter for HPLC analysis. Six Schisandra samples were prepared under the same conditions above, and the relative standard deviation of the six samples was calculated.

Recoveries

Different weight 5-HMF was added to the Schisandra decoction samples and the content of 5-HMF in the decoctions was analyzed by the method mentioned above. The recoveries were calculated by using the ratio of the detected content to added content of 5-HMF.

RESULTS AND DISCUSSION

Performance of HPLC system

The mobile phase containing different proportions of ultra-pure water and acetonitrile was tested to find the optimal combination. After several trials, a mobile phase consisting of ultra-pure water and acetonitrile (95:5, V/V) was used which could achieve effective separation and acceptable sensitivity, as well as symmetric peak shape and short analytical period. Representative chromatograms are shown in Fig.1. The retention time of 5-HMF was about 13 min, no interferences were observed.

Fig. 1.

Chromatograms of standard and samples

(a) Chromatogram of 5-HMF standard solution; (b) Chromatogram of unprocessed Schisandra decoction from Schisandra chinensis (Turcz.) Baill; (c) Chromatogram of medicinal Schisandra decoction from Schisandra chinensis (Turcz.) Baill processed by factory A; (d) Chromatogram of medicinal Schisandra decoction from Schisandra sphenanthera Rehd. et Wils. processed by factory B; (e) Chromatogram of medicinal Schisandra decoction from Schisandra sphenanthera Rehd. et Wils. processed by factory C

System suitability

The number of theoretical plates of column (N) and the tailing factor (T) to analyze 5-HMF were about 14 300 and 0.96, respectively. Both conformed to the Pharmacopoeia of PRC.

Linearity

The peak area (Y) and concentration (X) of 5-HMF standard solutions were subjected to regression analysis to calculate the calibration equation and correlation coefficient. The calibration equation was Y=7376.8X+6.8016, R ²=0.9999. The results showed excellent correlation between the peak area and the concentration of 5-HMF within the range of concentrations from 0.02724 to 0.2724 mg/ml.

Precision

The relative standard deviations of six injections for standard and samples were 0.14% and 0.71%, respectively. The system yielded excellent performance for analysis of 5-HMF content.

Stability

5-HMF in standard solution and in samples showed excellent stability according to the results. The relative standard deviations of standard solution and sample were 0.15% and 1.23%, respectively.

Repeatability

The content of 5-HMF in 6 Schisandra samples prepared under the same conditions showed no marked difference. The relative standard deviation of the six data was 1.58%.

Recoveries

The average extraction recovery of 5-HMF in decoction was 98.89% and the relative standard deviation was 2.78%. The average recoveries were sufficiently high to reach the normal analytical criterion. The extraction recovery data of 5-HMF are presented in Table 1.

Table 1.

The extraction recovery data on 5-HMF

Sample	Found (mg)	Added (mg)	Extraction recovery (%)	Average (%)	RSD (%)
1	13.46	13.62	98.83	98.89	2.78
2	13.04	13.62	95.77
3	13.64	13.62	100.12
4	13.30	13.62	97.66
5	13.25	13.62	97.31
6	14.12	13.62	103.65

The influence of fructose addition on the 5-HMF level

The effect of adding fructose is presented in Table 2. The results showed that additional fructose had no evident effects on the 5-HMF level of the decoctions (Table 2).

Table 2.

The influence of adding fructose on the 5-HMF level

Fructose (g)	5-HMF concentration (mg/100 ml)	Average (mg/100 ml)	RSD (%)
0	17.6	17.5	2.18
0.2500	17.7
0.7253	17.3
1.0249	17.9
1.2650	17.8
1.6482	16.9
2.3189	17.3

The influence of pH value variation on 5-HMF level

Although variation in the quantity of Schisandra had very little effect on the pH value, there was nonetheless evident effect on the 5-HMF level in the decoctions. The small change of pH value caused by different acids had no evident effect on the production of 5-HMF when compared to the effects produced by Schisandra. Table 3 shows the influence of the quantity of Schisandra on pH value and the pH value variation on 5-HMF level. The results showed that it was not simply the decomposition of saccharide under the acid environment created by Schisandra but that Ophiopogon alone can also produce 5-HMF at a very low level when decocted without Schisandra or acid. Maybe 5-HMF originated mainly from Schisandra, not in combination with Schisandra/Ophiopogon decoction and there are other factors affecting the producing of 5-HMF besides the acid environment created by Schisandra. The result conflicts with Zhu et al.(1998b)’s report. Straightforward detection of 5-HMF in the sample without any treatment in this experiment makes the result more scientific. Some changes in Schisandra components, properties inevitably occur in the course of sample treatment.

Table 3.

The influence of Schisandra quantity on pH value and 5-HMF level

Acid environment creating material		Ophiopogon (g)	pH	5-HMF content (mg/100 ml)
Schisandra (g)	Acid
6.25	-	25	3.094	8.70
12.50	-	25	3.046	16.40
18.75	-	25	3.024	24.20
25.00	-	25	2.993	31.60
50.00	-	25	2.930	59.90
-	HCl	25	2.972	0.39
-	HCl	25	3.142	0.46
-	HCl	25	3.454	0.46
-	H2SO4	25	2.987	0.44
-	H2SO4	25	3.058	0.50
-	H2SO4	25	3.401	0.40
-	HAC	25	2.894	1.15
-	HAC	25	3.012	0.67
-	HAC	25	3.331	0.52
-	-	25	6.576	0.34

The influence of different Schisandra on the 5-HMF level of Ophiopogon existence or inexistence

There were great differences in 5-HMF level among the three decoctions of medicinal or unprocessed Schisandra and Ophiopogon. The 5-HMF level between the decoctions of unprocessed Schisandra only and Schisandra/Ophiopogon group was different but there was no evident difference between the decoctions of medicinal Schisandra only and Schisandra/Ophiopogon group. 5-HMF content differed with the processing methods of Schisandra. The results showed that the presence of Ophiopogon had no evident effect on the production of 5-HMF when Schisandra was processed but that the processing method had great effect on the 5-HMF level in the decoctions (Table 4, Fig.2). The changing 5-HMF level in the decoction was mainly affected by Schisandra processing methods. The results was confirmed in the report that 5-HMF appeared in the course of Ephedra sinica Stapf processing (Xu et al., 2004). The processing time, processing temperature and herb species probably all affect 5-HMF level of medicinal Schisandra during the course of the processing.

Table 4.

The influence of the Ophiopogon and non-Ophiopogon existence on the level of 5-HMF in decoctions

No.	Schisandra content (g)	5-HMF content (mg/100 ml)
		Ophiopogon+Schisandra group			Schisandra group
		a1	b1	c1	a2	b2	c2
1	6.25	8.67	3.98	1.06	8.32	3.28	0.52
2	12.50	16.35	6.72	1.62	16.08	6.22	0.81
3	18.75	24.23	9.95	2.15	24.44	9.53	1.22
4	25.00	31.55	13.26	3.01	29.95	13.37	1.07
5	50.00	59.91	24.25	5.18	62.11	23.67	2.48

Fig. 2.

The influence of Ophiopogon addition on the production of 5-HMF

CONCLUSION

The composition of traditional Chinese medicine is very complex and every procedure of processing may change the results. The method of analysis established can reduce the operational error by a straight analysis of the components of herbal decoctions without extra treatment and provide more accurate results.

It was concluded that 5-HMF of Shengmaiyin was mainly derived from the decoction of Schisandra and not from the combined Ophiopogon/Schisandra decoction, because there was no evident difference in the 5-HMF level between the medicinal Schisandra/Ophiopogon group and Schisandra alone group and Schisandra content or Schisandra processed methods had great effect on the 5-HMF level. The factors affecting 5-HMF level in Schisandra will be reported in succession.

The addition of fructose has no effect on the production of 5-HMF in the decoction, showing that the appearance of 5-HMF is not simply the decomposition of saccharide under the acid environment created by Schisandra. The difference in 5-HMF content between the decoctions of unprocessed and processed Schisandra showed that the processing method played an important role in the production of 5-HMF. Because 5-HMF is found in many herbs, the physiological and biochemical roles of 5-HMF in the herbs deserved further investigation in the future.