Spectrum-effect relationship between serum HPLC fingerprints and activation of blood circulation and removal of blood stasis by Chuanxiong

Purpose: To study the active ingredients of Chuanxiong (CX) in promoting blood circulation and removing blood stasis. Methods: Blood-activating effects and serum HPLC fingerprints of CX extracts from different parts of China were studied and the spectrum-effect relationship between them was established by grey correlation analysis. Results: After treatment with CX extracted using different solvents, hemorheology indices were lower than those in the model group, and the effect of dichloromethane was better than that of other solvents (p < 0.05 or p < 0.01). There were 6 common peaks by fingerprint analysis. Peaks 1 5 were identified as senkyunolide I, senkyunolide H, senkyunolide A, coniferyl ferulate and ligustilide, respectively. Conclusion: Analysis of the spectrum–effect relationship indicates the contribution of the five components to the blood-activating effect of CX. The findings lend some scientific justification for using CX to remove blood stasis, and selection of quality evaluation indices for CX.


INTRODUCTION
Chuanxiong (CX) is the dried rhizome of Ligusticum chuanxiong Hort. CX is a well-known plant used in traditional Chinese medicine (TCM) formulations which activates the blood circulation and relieves pain [1]. Usually, it is employed with Angelica sinensis radix and Salvia miltiorrhiza for the treatment of irregular menstruation, migraine, coronary heart disease, and angina [2][3][4]. The main chemical components of CX are volatile oils, alkaloids, organic acids and phenols.
According to TCM theory, if the blood is clear, then pain will be absent. Studies on the pharmacodynamic basis of the effects of CX have focused primarily on chuanxiongzine, ferulic acid, ligustilide and senkyunolide with little or no chuanxiongzine in the original CX or CX extracts. Studies [5][6][7] have shown that the main components in the plasma of rats after CX administration are ferulic acid, ligustilide and senkyunolide, but chuanxiongzine is absent. Li and colleagues [8] showed that endoesters were the main active ingredients of CX, and that CX had a good effect on ischemic cardiac disease and cerebrovascular disease. Hence, the identity of the main components of CX that have pharmacodynamic effects is not known.
We investigated the active ingredients of CX that promote blood circulation and remove blood stasis. The aim of the study was to determine the "spectrum-effect relationship" between serum high-performance liquid chromatography (HPLC), as well as the effect of CX on blood circulation and removal of blood stasis by grey correlation analyses.

EXPERIMENTAL Ethical approval of the study protocol
Animal experiments were undertaken with the approval of the Ethics Committee of Chengdu University of Traditional Chinese Medicine (Chengdu, China). Animal treatments were conducted in strict compliance with the Guide for the Care and Use of Laboratory Animals (US National Institutes of Health, Bethesda, MD, USA) [11].

Materials
Twenty-two batches of CX were collected from major producing areas in Sichuan Province, one batch of CX was collected from Fujian Province, and one batch of CX was collected from Shanxi Province, in China, and identified by Professor Yu-Ying Ma and Professor Gui-Hua Jiang (College of Pharmacy, Chengdu University of TCM, Chengdu, China). Information on these 24 batches of CX is shown in Table 1.
Blood-activating effect of CX extracted using different solvents Sample preparation CX samples (250 g/batch) were ground into powder, soaked with 70% ethanol for 10 h, and extracted eight times with 70% ethanol by seepage. The extracts were combined, and concentrated to a flow extract (30 mL) under reduced pressure, then dispersed with the appropriate amount of pure water. Then, they were extracted thrice with petroleum ether, dichloromethane, ethyl acetate or N-butanol, respectively. Extracts of each part were combined and concentrated under reduced pressure, and diluted to 250 mL with 0.5% carboxymethylcellulose sodium (CMC-Na) to obtain samples.

Grouping and treatment of rats
Rats were divided randomly into seven groups: blank; model; compound danshen tablets (CDT); test (four groups). Rats in the blank group did not receive any treatment; rats in the model group only replicated the blood stasis model without gavage drugs; rats in the CDT group and test groups replicated the blood stasis model and were gavaged the corresponding drugs. The specific treatment method was as follows: rats in the blank group and model group were administered pure water at 1 mL/kg/day for 7 days. Rats in the CDT group were given CDTs at 1.8 g/kg/day for 7 days. Rats in different test groups were given the corresponding extract at 1 g/kg/day for 7 days, respectively. All treatments were given by gavage. Eight hours after gavage on day-7, rats in the model group, CDT group, and test groups were administered (s.c.) 0.1% epinephrine hydrochloride injection (0.8 mL/kg) twice at an interval of 4 h, whereas rats in the control group were administered subcutaneously (sc) 0.9 % sodium chloride injection at the same dose as the test groups. Two hours after the first injection of 0.1 % epinephrine hydrochloride, rats in all groups except for the control group were placed in an iced water bath for 5 min. Thirty minutes after gavage on day-8, rats in all groups were anesthetized with 10% chloral hydrate, and blood was collected from the abdominal aorta.

Treatment and assessment of blood samples
First, parameters from whole-blood samples were measured by an automatic hemorrheometer (SA-6000; Succeeder). Then, samples were centrifuged (3000 rpm for 30 min at room temperature), the supernatant separated, and analyzed.

Sample preparation
The method of sample preparation was identical to that described as sample preparation in bloodactivating effect of CX extracted using different solvents.

Grouping and treatment of rats
Rats were divided randomly into 27 groups: blank; model; CDT; CX1-CX24. Rats in the CX1-CX24 groups were given the corresponding extracts at 1 g/kg/day (bodyweight) for 7 days, respectively. Other treatments were identical to those described as grouping and treatment of rats in blood-activating effect of CX extracted using different solvents.

Preparation of serum samples for HPLC
CX powder (100 g) was soaked with 70% ethanol for 10 h, and extracted eight times with 70% ethanol by seepage. The extracts were combined, and concentrated to a flow extract under reduced pressure, and diluted to 20 mL with the appropriate amount of 0.5% CMC-Na to obtain samples for pharmacodynamic studies.
Rats were divided randomly into 25 groups: blank; CX1-CX24. Rats in the blank group were administered 0.5 % CMC-Na (1 mL/kg/day bodyweight) for 3 days. CX1-CX24 groups were administered the corresponding pharmacodynamic sample at 1 mL/kg bodyweight once a day for 3 days, respectively. Thirty minutes after gavage on day-3, rats in all groups were anesthetized with 10% chloral hydrate, and blood from the abdominal aorta was collected. After blood had coagulated, the supernatant was separated after centrifugation (3000 rpm for 10 min at room temperature).
Methanol was added thrice to precipitate proteins. Then, the supernatant was aspirated and filtered through a 0.22-μm membrane to yield serum samples for HPLC.

Preparation of mixed standard solutions
Appropriate amounts of ligustilide, senkyunolide A, senkyunolide I, coniferyl ferulate, and senkyunolide H were accurately weighed, and dissolved in methanol to prepare a series of stock solutions of different concentrations. Appropriate volumes of 5 stock solutions were transferred into the same volumetric flask; methanol was added to obtain mixed reference substance solutions of different concentrations.

HPLC conditions
HPLC was done on a LC-2012 system (Shimadzu) and chromatographic separation was carried on a SP-120-5 C18 column (4.6 mm × 250 mm, 5 μm) operated at 25 ℃. The mobile phase comprised methanol (A) and water (B), with a linear gradient of A: 0-40 min (30-85%) [12]. The injection volume was 20 μL, and the flow rate was 1.0 mL/min. The detection wavelength of the HPLC fingerprint was 280 nm.

Validation of HPLC fingerprint method
Precision, stability within 16 h, and repeatability were used to assess the performance of instrument, stability of the target, and operational consistency, respectively [12].

Similarity evaluation of fingerprints
The HPLC chromatograms of 24 samples of serum from rats administered CX were imported into Similarity Evaluation System for Chromatographic Fingerprint of TCM (SESCF, Version 2004A) (Beijing, China), and generated reference standard fingerprint (R), including 6 characteristic peaks (common peaks) by multipoint correction and peak matching, and the similarity values between the chromatogram of each serum from rats administered CX and the reference fingerprint were evaluated by this software.

Statistical analysis
Data from pharmacodynamic test are presented as mean ± standard deviation (SD). Differences among different groups were analyzed by oneway ANOVA on SPSS 21.0 (IBM, Armonk, NY, USA). Bivariate correlation analyses between different hemorheology indices of CX and common peak areas were done using the grey correlation coefficient.

Selected extraction solvent for bloodactivating effect of CX
Compared with the model group, almost all hemorheology indices in the test groups were reduced. After treatment with CX extracted using different solvents (petroleum ether, dichloromethane, ethyl acetate, N-butanol), the hemorheology indices decreased compared with those in the model group. The effect of dichloromethane solvent was better than that of other solvents (Table 2).

Blood-circulation effects of CX extracts
Compared with the blank group, the hemorheology indices we assessed in the test groups were increased, which suggested that the iced water-induced coagulation model had been created successfully. After treatment with different CX extracts, the hemorheology indices changed compared with those in the model group (Table 3).

Validation of the HPLC-fingerprint method
Testing of precision, stability within 16 h, and repeatability indicated that the relative standard deviation (RSD, n = 6) of retention time and peak area value of 6 common peaks were <5% (Table  4, Table 5 and Table 6, Table 7, Table 8, Table  9) [12]. Hence, our method was feasible for analyzing the HPLC fingerprints of CX extracts.

HPLC fingerprints
The HPLC fingerprints and the reference fingerprint of 24 batches rat serum were shown in Figure 1. And similarities among the HPLC chromatograms of these 24 samples of rat serum and reference standard fingerprints were analyzed. The RSD of the six common peak areas in the 24 samples of rat serum were >35%, indicating that the areas of each common peak in different serum samples varied widely.
The similarities in the HPLC chromatograms of 24 samples of rat serum were in the range 0.012-1.000. Similarities between the HPLC chromatograms of 24 samples and reference standard fingerprints were in range 0.174-0.98.

Identification of common peaks
The common peaks were identified as senkyunolide I, senkyunolide H, senkyunolide A,      coniferyl ferulate, and ligustilide, respectively, by comparing the retention time and ultraviolet-absorption curves of target peaks with those of standards ( Figure 2).

DISCUSSION
Clinical studies have shown that cardiovascular diseases and cerebrovascular diseases are related mostly to blood stasis, so improving blood stasis would be a rational approach. Reports [13,14] have shown that ligustilide, senkyunolide I, senkyunolide H, senkyunolide A, ferulic acid and ligustrazine can improve blood stasis, respectively. However, not all of these compounds have been reported to be in CX or in the body fluids of animals. Therefore, there was insufficient evidence in the previous studies to demonstrate that these components are the active ingredients of CX.
Subcutaneous injection of epinephrine and an iced water bath is the most commonly used method to replicate acute blood stasis [15][16][17]. Water temperature, the injection site, and the time blood is collected have a considerable influence on experimental results. Epinephrine is a vasoactive drug, which causes the death of rats by increasing the blood pressure suddenly when injected close to the head or injected into a blood vessel by mistake. Therefore, injecting epinephrine accurately and mastering the time to start blood collection are the key to successful modeling.
Thirty minutes after the last gavage was selected as the time to start blood collection because there were no more peaks after 30 min. In this work, methanol was selected as the organic mobile phase by comparing the shape and separation of the peak.
The spectrum -effect relationship can be used to determine the main medicinal ingredients of a target drug based on the contribution of its different components. In this study, we studied the active part of CX for blood -activating, and compared the blood -activating effects and serum HPLC fingerprints of CX from different origins, then evaluated the spectrum -effect relationship of them by grey correlation coefficient. The greater the correlation coefficient, the greater was the contribution of the corresponding peak to the blood-circulation activity of CX extracts [18]. We revealed that the contributions of six common peaks to the bloodcirculation activity of CX extracts were in the order: peak 1 (senkyunolide I) > peak 5 (ligustilide) > peak 3 (senkyunolide A) > peak 4 (coniferyl ferulate) > peak 2 (senkyunolide H) > peak 6. Compared with the HPLC fingerprints of medicinal materials, serum HPLC fingerprints can better explain the effect of different ingredients on drug efficacy. This study revealed the blood promoting effect of 5 active ingredients through systematic research. Combined with the previous reports, senkyunolide I, ligustilide, senkyunolide A, coniferyl ferulate and senkyunolide H may be the medicinal material basis of CX for blood -activating.

CONCLUSION
In this study, we investigated the spectrum-effect relationship between the blood-circulation activity and serum HPLC fingerprints of CX extracts for the first time. Our study suggested that ligustilide, senkyunolide A, coniferyl ferulate, senkyunolide H and senkyunolide I helped to improve blood circulation in rats. Our study provides scientific justification for using CX to remove blood stasis.

Data availability
The data used to support the findings of this study are available from the corresponding author upon request. 4.0) and the Budapest Open Access Initiative (http://www.budapestopenaccessinitiative.org/rea d), which permit unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.