Comparative studies on the toxicological , anti-inflammatory and analgesic properties of three sources of Xuedan in mice and their rapid identification by electronic tongue

Purpose: To compare the toxicological, anti-inflammatory and analgesic properties of three sources of Xuedan, viz, Hemsleya omeiensis (HO), Hemsleya giganth (HG) and Hemsleya dolichocarpa (HD) in mice, and to study their rapid identification based on electronic tongue (E-tongue). Methods: After 7 days of administration, the median lethal doses (LD50) of the three xuedan decoctions in mice were determined. In addition, the anti-inflammatory and analgesic effects of the three xuedans were evaluated in mice using xylene-induced ear edema and acetic acid-induced pain. Furthermore, Etongue technology was used to identify HO, HG and HD. Principal component analysis (PCA) and discriminant factor analysis (DFA) were used to analyze the data acquired by E-tongue. Results: The median lethal dose (LD50) values of H. omeiensis, H. gigantha and H. dolichocarpa were 32.3, 17.4 and 13.7g/kg, respectively. Compared with normal control group, the anti-inflammatory effects of Xuedan were obvious in xylene-induced ear edema (p < 0.05), and pain sensation was significantly inhibited in acetic acid-induced writhing test (p < 0.05). Furthermore, E-tongue technology effectively identified HO, HG and HD. Conclusion: H. omeiensis exhibits the highest LD50 value and best analgesic effect among the three sources of xuedan. E-tongue technology is effective and rapid in identifying HO, HG and HD.


INTRODUCTION
Xuedan is derived from Hemsleya (Cucurbitaceae) which is distributed in subtropical and temperate regions of Asia, especially in China [1].In Chinese Traditional medicinal theory system, xuedan is used for clearing heat and removing toxins.Xuedan exerts antibacterial and anti-inflammatory effects.Consequently, xuedan is used for treating various diseases such as hepatitis, enteritis, bronchitis and coronary heart disease [2].In China, xuedan comes from three varieties of Hemsleya viz Hemsleya omeiensis, Hemsleya gigantha and Hemsleya dolichocarpa.Studies have revealed that there are wide differences in compositions and in bioactivities among the three sources of xuedan, especially with respect to their bitter taste.However, no studies have been carried out on effective and rapid strategy for discriminating between the three sources of xuedan.The present study was carried out to compare the toxicology and pharmacology of the three sources of xuedan, and to develop a rapid method based on electronic tongue for their identification.

EXPERIMENTAL Plant materials
The three varieties of xuedan used in this study were provided by Sichuan Fuzheng Pharmaceutical Co., Ltd., and identified by Professor Xian-Ming Lu (College of Pharmacy, Chengdu University of Traditional Chinese Medicine, Chengdu, China).

Animal studies
A total of 200 mice (18 -22 g) were obtained from the Chengdu DaShuo Experimental Animal Co., Ltd.(Chengdu, China).All animals received humane care according to the Declaration of Helsinki [3], and fed in accordance with standard laboratory protocols.

Acute toxicity studies
Acute toxicity studies were carried out in accordance with the procedure reported in previous studies, but with minor modifications [4][5][6].First, the complete lethal dose (LD 100 ) and zero (0), i.e., lethal dose-0 (LD 0 ) were determined through oral administration of the test samples.The results showed that the LD 100 and LD 0 values for H. omeiensis were 48 and 24 g/kg, respectively; and LD 100 and LD 0 values for H. gigantha and H. dolichocarpa were 24 and 6 g/kg, respectively.Based on these results, mice were randomly divided into five groups (n = 8) for the testing of Xuedan samples.
The mice were fasted overnight but were permitted free access to water, and were administered the designed dosages (Table 1) orally (intragastrically).The mice were observed for signs of poisoning or death for 14 days.Mortality was calculated from the number of mouse deaths in each group, and LD 50 was calculated using the weighted regression probability unit method (Bliss method) with LD 50 calculation version 2.0 software.The 95 % confidence interval was calculated using the improved method of Käber [6].

Anti-inflammatory studies
Mice were randomly divided into four groups (n = 10): control group, H. omeiensis group, H. gigantha group and H. dolichocarpa group.The test samples were administered orally at a dose of 0.6 g/kg (equivalent to daily human adult dose) for 7 continuous days.Then, 30 min after the last administration, 20 µL of xylene was uniformly applied to both sides of the left auricle of each mouse, while the right ear was used as a control.After 30 min, the mice were euthanized.The ears were cut along the baseline of the auricle.The circular ear was punched in the same part of the left and right ears with an 8 mm diameter puncher [7].The ears were weighed with an electronic balance, and the weight difference of the two ears (W l -W r ) was used as swelling index (Es), calculated as in Eq 1.

Analgesic studies
Mice were randomly divided into four groups (n = 10): control group, H. omeiensis group, H. gigantha group and H. dolichocarpa group.The mice in each group were orally administered corresponding test samples at a dose of 0.6 g/kg/day (equivalent to the daily adult human dose) continuously for 7 days.Then, one hour after the last administration, each mouse was treated with 0.6 % acetic acid (0.1 mL/kg) by intraperitoneal injection.The number of writhes (Nw) (abdominal contraction and extension of hind limbs) within 15 min after acetic acid injection in each group of mice was recorded, and the percentage inhibition (Pi) was calculated as in Eq 3 [8].
where Nwc and Nwt are the number of writhes of control and test groups, respectively.

Electronic tongue test
The αAstree E-tongue, used to investigate the taste of the three varieties of Xuedan, had six cross-sensitive potential sensors (ZZ, AB, GA, CA, DA and JE), Alphasoft data acquisition system, signal processing system and pattern recognition software (Alpha M.O.S, Version 2012.45).The method [8] measures E-tongue under conditions that ensure that the acquired data are reliable and stable.Each test sample powder (1.0 g) was extracted with 200 mL boiled water in a conical flask for 3 h.Thereafter, it was filtered through a 0.22 µL microporous filter membrane, and 25 mL of the filtrate was placed in the auto-sampler tray of the E-tongue apparatus for analysis.Each sample was analyzed for 120 sec in triplicate.The data were recorded using Astree II software.

Statistical analysis
Data are expressed as mean ± standard deviation (SD), and were analyzed using SPSS 21.0 software with single factor analysis of variance between groups.Values of p < 0.05 were considered statistically significant.All Etongue data were analyzed using Alpha M.O.S. statistical software.Principal component analysis (PCA) and discriminant factor analysis (DFA) were used for discrimination analysis of the samples.

Acute toxicity
As can be seen from the Table 1, LD 50 values of the H. omeiensis, H. gigantha, H. dolichocarpa were 32.3, 17.4 and 13.7 g/kg, respectively.All samples showed moderate toxicity, but the H. omeiensis showed the lowest toxicity in mice, when compared to the other two sources of xuedan (32.3 g/kg vs. 17.4 and 13.7 g/kg).

Anti-inflammatory effects
The results of anti-inflammatory studies of the three xuedan sources are shown in Table 2. Compared with the normal group, the three aqueous extracts of samples significantly inhibited xylene-induced ear swelling in mice (p < 0.01).There was no significant difference in percent inhibition of swelling between the three samples.

Analgesic activity
Compared with the normal group, the three varieties of xuedan significantly reduced the number of acetic acid-induced writhes in mice (p < 0.05), and the effect of HO was the most significant (p < 0.01).There were significant differences in the percentage inhibition of writhes among HO, HG and HD (p < 0.05 (Table 3).

E-tongue results
Typical sensor responses recorded by E-tongue apparatus are shown in Figure 1.An average value was acquired based on the stable sensor responses of E-tongue analyses from 100 to 120 sec, and was used as the output data.The repeatability of the E-tongue results using this method of study is shown in Table 4.In this study, 9 batches of samples were collected for each source of xuedan.Each sample was measured and analyzed three times.From the signals acquired through E-tongue measurements, 81 data points were used for further PCA and DFA.The raw data was projected onto the PC axis, and each sample will a set on the PCA three-dimensional map.The distance between sample sets is usually used to indicate similarities in taste.The threedimensional map of PCA is shown in Figure 2, and the three principal components were PC1 (82.83 %), PC2 (16.21%), and PC3 (0.66 %).The total degree of contribution of the three principal components was 99 %.Similar to PCA, a total 81 data points were projected onto the DFA score plots.As can be seen in Figure 3, the two principal components for E-tongue were DF1 = 92.36% and DF2 = 7.64 %, and their total contribution was 100 %.The DFA model effectively classified the three sources xuedan.In addition, DFA is often used to establish a sample database for qualitative discrimination between unknown samples and determination of grouping information of unknown samples.This model can be used to identify samples from unknown base sources of xuedan.

DISCUSSION
It is well known that traditional Chinese medicines (TCMs) have been applied in the treatment of various diseases for thousands years, and studies have demonstrated their pharmacological activities [10].It has been shown that the main active phytochemical components of the genus Hemsleya are triterpenoid saponins which possess significant antitumor, antibacterial and anti-inflammatory properties [1,11].Some scholars have suggested that hemslecin A is a major antibacterial agent in the genus Hemsleya genus [12].It has been directly used as an effective agent for treating inflammation and bacterial infection-related disorders.Subsequent studies revealed that hemslecins A and B exhibit potent anti-HIV-1 activities [13][14][15][16].Therefore, Hemsleya genus is a promising plant resource for discovery of drugs for treating infection and inflammation.However, some TCMs are not derived from one plant but lots of plants, and the use of original sources of the plants would be beneficial for guaranteeing the therapeutic effects of the TCMs.

CONCLUSION
Hemsleya omeiensis possesses the highest LD 50 value and best analgesic effect among the three varieties of Hemsleya.In addition, E-tongue, when combined with PCA and DFA, can effectively and rapidly discriminate between the three sources of xuedan.

Figure 1 :
Figure 1: Typical sensor responses of E-tongue during measurement

Figure 2 :
Figure 2: PCA score plots of E-tongue measurements of three varieties of Hemsleya chinensis (n = 9).Each sample was measured and analyzed three times.HO, HG and HG refer to H. omeiensis, H. Gigantha, H. dolichocarpa, respectively

Figure 3 :
Figure 3: DFA score plots of E-tongue measurements of three varieties of Hemsleya chinensis (n = 9).Each sample was measured and analyzed three times.HO, HG and HG represent H. omeiensis, H. Gigantha, H. dolichocarpa, respectively

Table 1 :
LD50 values of the three sources of xuedan (n = 8)

Table 2 :
Effect of the three sources of xuedan on ear edema induced by xylene in mice

Table 3 :
Effects of the three sources of xuedan on number of writhes in mice

Table 4 :
Repeatability of E-tongue determination method (n = 6) moderate bitter taste instead of extreme bitter taste.In addition, it is produces better bioactive effects, relative to the other two sources.In view of these desirable qualities, a proposal to distinguish H. omeiensis from the other two sources of xuedan was designed in the present study.The bitter taste is the key point for quick and effective discrimination of the 3 sources of xuedan.Electronic tongue technology is an objective quantitative evaluation method for sensory indicators.By collecting objective and accurate data, the results of the taste characteristics of samples are obtained, and the differences in the taste of different samples are scientifically reflected.It is increasingly used in the Chinese medicine industry.These results suggest that Etongue combined with PCA can effectively discriminate between the three varieties of Xuedan, especially H. omeiensis which differs in taste from the other two sources of Xuedan.DFA is a classification technique that optimizes differentiation by recombining sensor data and is another classical analysis model in addition to PCA.It can be used to establish a sample database for qualitative discrimination between samples from unknown base sources of xuedan.This study has demonstrated that electronic tongue technology, when combined with PCA and DFA can effectively distinguish among Xuedans from the three varieties of Hemsleya.
However, based on the feedback from the doctors in clinics and the users, H. omeiensis differs from the other two sources of xuedan which have