Mechanism and Pharmacodynamic Substance Basis of Raw and Wine-Processed Evodia rutaecarpa on Smooth Muscle Cells of Dysmenorrhea Mice

Objectives Evodia rutaecarpa (ER) is a well-known herbal Chinese medicine traditionally used for analgesia in dysmenorrhea, headaches, abdominal pain, etc. Notably, the analgesic effect of wine-processed Evodia rutaecarpa (PER) was more potent than that of raw ER. This research aimed to investigate the mechanism and pharmacodynamic substance basis of raw ER and PER on smooth muscle cells of dysmenorrhea mice. Methods Metabolomics methods based on UPLC-Q-TOF-MS were utilized to analyse the differential components of ER before and after wine processing. Afterwards, the uterine smooth muscle cells were isolated from the uterine tissue of dysmenorrhea and normal mice. The isolated dysmenorrhea uterine smooth muscle cells were randomly divided into four groups: model group, 7-hydroxycoumarin group (1 mmol/L), chlorogenic acid (1 mmol/L), and limonin (50 μmol/L). The normal group consisted of the isolated normal mouse uterine smooth muscle cells, which were repeated 3 times in each group. The cell contraction and the expression of P2X3 and Ca2+ in vitro were determined using immunofluorescence staining and laser confocal; ELISA was used for detection of PGE2, ET-1, and NO content after 7-hydroxycoumarin, chlorogenic acid, and limonin administered for 24 h. Results The metabolomics results suggested that seven differential compounds were identified in the extracts of raw ER and PER, including chlorogenic acid, 7-hydroxycoumarin, hydroxy evodiamine, laudanosine, evollionines A, limonin, and 1-methyl-2-[(z)-4-nonenyl]-4 (1H)-quinolone. The in vitro results showed that 7-hydroxycoumarin, chlorogenic acid, and limonin were able to inhibit cell contraction and PGE2, ET-1, P2X3, and Ca2+ in dysmenorrhea mouse uterine smooth muscle cells and increase the content of NO. Conclusion Our finding suggested that the compounds of the PER were different from those of the raw ER, and 7-hydroxycoumarin, chlorogenic acid, and limonin could improve dysmenorrhea in mice whose uterine smooth muscle cell contraction was closed with endocrine factors and P2X3-Ca2+ pathway.

Euodia rutaecarpa (Juss.) Benth. Var. Bodinieri (node) Huang (ER) is commonly used to improve the analgesic efects in the Pharmacopoeia of the People's Republic of China [6,7]. It has been proven to have anti-infammatory and analgesic efects [8,9]. Te PER is the stir-fried product of ER macerated with rice wine, which is recorded in the Essential Recipes for Emergent Use Worth a Tousand Gold [10]. Research has shown that the pharmacological actions of raw ER are diferent from those of its processed products [11]. Raw ER may dispel cold to relieve pain, and the efect of dispelling cold and relieving pain is strengthened after wine processing [11,12]. Our prior study demonstrated that raw ER could reduce the number of body twists, prolong the latency of body twisting, and reduce pain in dysmenorrhea mice, and that the efect of PER on dysmenorrhea is better than that of raw ER [11]. Chinese herbal remedies often contain multiple components [13]. However, whether wine processing causes changes in components of raw ER, whether the analgesic action of these components improves dysmenorrhea, and the mechanisms are unknown. Terefore, the components of raw ER and PER and which of them have the potential to exert treatment efects are also a focus of research.
TCM usually contains various chemical compositions; thus, data on their treatment efects, mechanisms of action, and toxicology characteristics are frequently limited [14]. Terefore, it becomes the focus of the research concerning the identifcation of chemical compositions of such medicines, evaluation of biological, and standardization of safety quality [15]. Plant metabolomics is a comprehensive method that can qualitatively and quantitatively analyse metabolite data in a specifc period or condition, and it is combined with chemical informatics analysis to ensure target diferential metabolites [16]. UPLC-Q-TOF-MS is a widely used metabolite analysis method because of its high resolution and sensitivity [17]. ER is closely related to dysmenorrhea. A dysmenorrhea mouse uterine smooth muscle cell model was performed to prove the diferences between raw ER and PER in terms of their curative efect [18].
Tis research aimed to investigate the diference between ER and PER, and whether the diferential components of ER and PER could alleviate pain by inhibiting the contraction of mouse uterine smooth muscle cells. First, metabolomics methods based on UPLC-Q-TOF-MS were used to analyse the diferential components of ER and PER. Afterwards, a dysmenorrhea mouse uterine smooth muscle cell model was established, and the efects of diferential components on cell contraction and the expression of PGE2, ET-1, NO, P2X3, and Ca 2+ in mouse uterine smooth muscle cells were explored ( Figure 1). Te research results demonstrated not only reveal the efect of the diferential components on mouse uterine smooth muscle cells but also provide a theoretical basis for the quality control of ER and clinical application. . ICR mice were fed in the SPF laboratory with the First Hospital of Hunan University of Chinese Medicine for one week. All mice were free to obtain food and water. Te laboratory is maintained at optimum temperature and humidity (20-26°C, 50%-70%).

PER Preparation.
Yellow rice wine (30 mL) was diluted in 200 mL distilled water, and the diluted yellow rice wine was mixed with raw ER at a proportion of 1 : 1. Te mixture was hermetically soaked for two hours until the wine was completely absorbed, after which it was stir-fried over a slow fre in a pan until the colour of raw ER deepened; next, it was removed and cooled to a temperature around 18-25°C.

Raw ER and PER Extract Preparation.
Te three batches of raw ER and PER were weighed for 270 g, respectively. Tese samples were mixed with the water at a proportion of 1 : 10 and soaked in water for 30 min. After soaking, the mixture was decocted over a martial fre in a container until boiling, and then the fre turned into a gentle fre to be decocted for one hour. Te liquid was fltered with gauze. Te remaining residue was mixed with water at a proportion of 1 : 8; this method was repeated once. Twice the fltrates were mixed and condensed by a rotary evaporator. Finally, we obtained 75.57 g of raw ER and PER dry extract; namely, each gram of dry extract was equivalent to 3.62 g of TCM. (3.0 × 100 mm, 1.8 μm) was used, and the fow rate was 0.4 mL/min. Te mobile phase was composed of acetonitrile (A) and water containing 0.1% formic acid (B), with an injection volume of 1 μL. A gradient elution procedure was used: 0-10 min, 5%-15% A; 10-15 min, 15%-20% A; 15-25 min, 25%-45% A; and 25-40 min, 45%-80% A.

Mass Spectrometry Conditions.
Te mass spectrometer was equipped with an ESI. Te scanning detection was approximately M/Z100-1700, the solvent gas was nitrogen at 6.8 L/min, the desolventizing temp was 325°C, and the source temp was 350°C. Te voltage of the capillary was adjusted to 40 V.

UPLC-Q-TOF-MS Data
Processing. Te collected metabonomics data were turned into mass/charge ratio (m/ z) data using MassHunter Profnder software, which can perform peak extraction, recognition, matching, alignment, and normalization. Te datasets, composed of sample name, retention time, mass charge ratio, and peak area, were derived from the software, and all m/z values were normalized. Afterwards, the SIMCA-P 14.1 statistical software was applied to multivariate statistical analysis (PCA and OPLS-DA), which was used to screen diferential compounds. Te screening conditions were as follows: (VIP) value > 1, P (CORR) > 0.05, fold change ≥ 1.4, and t-test P < 0.05. Te information on the diferential metabolites was input into Qualitative Analysis software from the Agilent Technologies Scientifc company for recognition. Determination of the diferential metabolites in ER and PER was done by using accurate mass spectrometry.

Cell Culture, Identifcation, and Grouping.
Te uteri were isolated from dysmenorrhea and normal mice and soaked in 75% alcohol for 30 s. Te fat and connective tissue was removed from the separated uterine tissue surface. Te mouse uterine strips were cut into 3-5 mm pieces with ophthalmic scissors and incubated in DMEM containing 0.02% collagenase II and 15% serum for 30 min at 37°C. Te collagenase digestion solution was removed after centrifugation. Te separated cells were resuspended in a DMED containing fetal bovine serum. On the second day, the separated cells were moved to the cell culture bottle for incubation at 37°C and 5% CO 2 . Half of the medium was then replaced with a culture medium every two days. Primary cultures of dissociated cells were performed as we have described. While these cells were grown to 100% confuence, they will interact with trypsin in the cell culture bottle. Ten, these cells and trypsin in the cell culture bottle were isolated by centrifugal machines. Te separated cells were resuspended in a 96-well cell culture plate with a CO 2 cell culture medium at a ratio of 10,000 per well for 48 h. PBS was then used to wash the cell culture plate twice. All cells were fxed with 4% cell fxative for 45 min, treated with 0.25% triton-100 for 15 min, and blocked with 5% BSA for 10 min. Afterwards, these cells were incubated in antibodies against the α-SMA diluent (1 : 300) at 4°C for overnight and the goat anti-rabbit Alexa Fluor ® 488 Conjugate (1 : 400) for 30 min at 37°C. PBS was used to wash the cell culture plate fve times. Tese cells were then incubated in the dark for 12 min at room temperature with DAPI (1 : 800), and the morphology of dysmenorrhea mouse uterine smooth muscle cells was observed by a laser confocal microscope.
Te isolated dysmenorrhea uterine smooth muscle cells were randomly divided into four groups: model group, 7hydroxycoumarin group (1 mmol/L), chlorogenic acid (1 mmol/L), and limonin (50 μmol/L). Te normal group consisted of the isolated normal uterine smooth muscle cells, which were repeated 3 times in each group. We have performed a series of preliminary concentration experiment, which determined the optimum concentration of these compounds (see supplementary fle (available here)).

ELISA Assay.
After administering for 24 h and collecting the supernatant, the cell supernatant was centrifuged at 3000 ×g for 15 min at room temperature. According to the instructions, the concentration of PGE2, ET-1, and NO in the uterine smooth muscle in each group was measured using the PGE2, ET-1, and NO ELISA kit.

Immunofuorescence Staining.
All cells were fxed with 4% cell fxative for 45 min, treated with 0.25% triton-100 for 15 min, and blocked with 5% BSA for 10 min. Afterwards, these cells were incubated in the cell culture plate containing 100 μL HCS CellMask Deep Red Stain at room temperature in the dark for 30 min. Te level of cell contraction was detected by a laser confocal microscope.
After being administered for 24 h, each group cells were washed twice with calcium-free D-hank's solution and then incubated in the cell culture plate containing 5 μmol/L Furo-3/AM working solution at 37°C in the dark for 45 min. A laser confocal microscope was used for detection of the Ca 2+ level.
We detected the expression of the P2X3 protein using a laser confocal microscope. All cells were fxed with 4% cell fxative for 45 min, treated with 0.25% triton-100 for 15 min, and blocked with 5% BSA for 10 min. Afterwards, these cells were incubated in primary antibody in P2X3 diluent (1 : 200) overnight at 4°C and then the goat anti-rabbit ALEXA Flour ® 488 Conjugate (1 : 400) for 30 min at 37°C. After washing fve times with PBS, all cells were incubated with DAPI (1 : 800) in the dark for 12 min at room temperature.

Statistical Analysis.
All experimental data were processed by Graphpad prism software. Te calculation results were expressed as means ± standard deviations. Te primary outcomes are the uterine smooth muscle cell contraction. Other indicators are secondary results. Student's t-test was used to compare the diferences between the two groups. One-way ANOVA was used for the comparison of multiple groups and Tukey's test was the post hoc test. Shapiro-Wilk test was used for normality assessment test, and all groups satisfy the normal distribution (see supplementary fle (available here)). Tere was a signifcant diference (P < 0.05).

Metabolomics Data Analysis
3.1.1. PCA Analysis of Raw ER and PER. We used PCA to analyse the sample data for raw ER and PER. In the positive ion, the PCA model parameter R 2 X (representing the PCA model interpretation rate) was 0.946, indicating that the model was stable. Te sample points of the raw ER and PER group were completely separated in the t [1] line (Figure 2(a)). Te results suggested that the chemical components before and after ER wine processing had signifcant diferences. As shown in t [2] line, the points, WZY-S-1-1, WZY-S-1-2, WZY-S-2-1, WZY-S-2-2, WZY-J-1-1, and WZY-J-1-2, are gathered in the positive semi-axis. Te points of WZY-S-3-1, WZY-S-3-2, WZY-J-2-1, WZY-J-2-2, WZY-J-3-1, and WZY-J-3-2 are gathered in the negative half axis (Figure 2(a)), which indicates that the chemical constituents of ER were also diferent among batches of the same variety.

OPLS-DA Analysis of Raw ER and PER.
As illustrated in the OPLS-DA score plots (Figure 2(b)), raw ER and PER were completely separated in the positive ion. Meanwhile, we performed permutation tests 200 times and repeated cross-validation (2CV) one time to evaluate the reasonableness of the established OPLS-DA model (Figure 2(c)). In the positive ion, R 2 Y � 0.997 and Q 2 X � 0.99, indicating that 99.7% of the samples were in accordance with pattern recognition, and the model's predictive accuracy was 99% (Figures 2(b) and 2(d)). Tese consequences indicate that the model had preferable ftting precision and predictability.   (Figure 2(e)). Te results may demonstrate that there are diferences between raw ER and PER.

Heatmap Analysis.
Te changes in raw ER and PER were shown visually by Heatmap analysis (Figure 2(f )). Te abscissa axis was diferent batches of raw ER and PER group; the ordinate axis was the compounds of the corresponding groups. Te results showed that the compounds of the raw ER group were diferent from those of the PER group and diferent among batches of the same variety, further considering that the treatment actions of the raw ER group and PER were also diferent.
Among then, the reference substances of chlorogenic acid, 7hydroxycoumarin, laudanosine, limonin, 1-methyl-2-[(z)-4nonenyl]-4 (1H)-Quinolone are shown in Figure 3(c). As shown in Figure 4(a), the chemical structure formulas of 6 diferential compositions were obtained from PubChem. In addition, the comparison of the ionic intensity of the markers in the raw ER and PER is shown in (Figure 4(b)). We found that the diferential composition of ionic intensity were stronger in PER compared to raw ER.

Morphology of Dysmenorrhea Mouse Uterine Smooth
Muscle Cells. We used a laser confocal microscope to observe the morphology of dysmenorrhea mouse uterine smooth muscle cells. Te results of cell imaging showed that the dysmenorrhea mouse uterine smooth muscle cells presented a spindle or irregular triangle with sharp processes. Te cell nucleus was a circle or ellipse. Tese cells were arranged in parallel and partially overlapped in their growth phase. Tey showed a typical structure: "peak valley" (Figure 5(a)). For further observation, immunofuorescence staining was used for dysmenorrhea mouse uterine smooth muscle cells. Te immune reaction product was presented in green, and the expression of α-SMA was positive. More than 98% of the cells were mouse uterine smooth muscle cells in randomly selected visual felds. Te morphological and immunofuorescence staining results showed that the cultured cells were mouse uterine smooth muscle cell ( Figure 5(a)).

7-Hydroxycoumarin, Chlorogenic Acid, and Limonin Inhibited the Contraction of Dysmenorrhea Uterine Smooth Muscle Cells.
To discuss the efect of 7-hydroxycoumarin, chlorogenic acid, and limonin on contraction in dysmenorrhea mouse uterine smooth muscle cell, we observed the dysmenorrhea mouse uterine smooth muscle cells contraction with Furo-3/AM fuorescent probe staining and laser confocal microscopy. Te cell contraction was increased in the model group compared to the normal group, and the cell area was reduced (P � 0.008, Figures 5(b) and 5(c) and Table 2). 7-Hydroxycoumarin could suppress the upregulation of cell contraction when compared to the model group (P � 0.018, Figures 5(b) and 5(c) and Table 2). Te application of limonin could inhibit cell contraction (P � 0.011, Figures 5(b) and 5(c) and Table 2). Te chlorogenic acid group could reduce the cell contraction, although not signifcantly. Notably, the cell contraction was not also found to have a signifcant diference between 7hydroxycoumarin, chlorogenic acid, and limonin groups. Te results indicated that 7-hydroxycoumarin, limonin, and chlorogenic acid could expand cells and inhibit the contraction of mouse uterine smooth muscle cells.

7-Hydroxycoumarin, Chlorogenic Acid, and Limonin Regulated the Levels of PGE2, ET-1, and NO in Dysmenorrhea Uterine Smooth Muscle Cells.
Changes in PGE2, ET-1, and NO were associated with the pathogenesis of dysmenorrhea during menstruation. To investigate the efect of 7-hydroxycoumarin, chlorogenic acid, and limonin on PGE2, ET-1, and NO in dysmenorrhea mouse uterine smooth muscle cells. We measured the concentrations of these indices. Te concentrations of PGE2 and ET-1 were increased in the model group compared to the normal group (P � 0.011, P � 0.006, respectively) (Figures 5(d) and 5(e), Tables 3, and 4), and NO was reduced (P � 0.002, Figure 5(f ) and Table 5). However, the concentration of PGE2 and ET-1 in the limonin group were suppressed compared to the model group (P � 0.033, P � 0.022, respectively) (Figures 5(d) and 5(e), Tables 3 and 4), and NO increased (P � 0.009, Figure 5(f ) and Table 5). 7-hydroxycoumarin had an infuence on ET-1 and NO (P � 0.036, P � 0.025, respectively) ( Figures 5(e) and 5(f ), Tables 4 and 5). Te chlorogenic acid group was not signifcant. Tere was no signifcant diference between the 7-hydroxycoumarin, chlorogenic acid, and limonin group. Te results demonstrate that dysmenorrhea led to changes in PGE2, ET-1, and NO levels, and 7-hydroxycoumarin and limonin efectively regulated the levels of NO, ET-1, and PGE2 in dysmenorrhea mouse uterine smooth muscle cells.

Discussion
Evodia rutaecarpa (ER) was frst recorded in Shen Nong's Herbal Classic and has the features of warming and alleviating pain [19]. Increasing evidence has revealed that ER can alleviate dysmenorrhea by slowing the contraction frequency and intensity of uterine smooth muscle, and wine processing would strengthen the efect of dispelling cold to relieve pain [11,20,21]. However, few investigations have been conducted on the correlation between the analgesic efect and the active components of ER before and after wine processing. Terefore, this study used UPLC-Q-TOF-MS and metabolomics technology to screen and identify the diferential components before and after ER wine processing, after which a dysmenorrhea mouse uterine smooth muscle cell model was established. Te model was intervened with the diferential components to examine its therapeutic efect and assess the underlying mechanism. Our research not only demonstrated wine processing can cause changes in components of raw ER but also found the differential components clearly attenuated dysmenorrhea mouse uterine smooth muscle cell contraction, reduced the levels of PGE2, ET-1, increased the content of NO, and suppressed the expression of P2X3 and Ca 2+ . We fnally identify seven diferential components in the extracts of raw ER and PER: 7-hydroxycoumarin, chlorogenic acid, hydroxy evodiamine, laudanosine, limonin, evollionines A, and 1-methyl-2-[(z)-4-nonenyl]-4 (1H)quinolone. Alkaloids are the main analgesic components of raw ER [22]. During processing, the infltration of millet wine increases the permeability of decoction pieces and internal tissues, which is conducive to the dissolution of alkaloids and improves their dissolution efciency [23]. Terefore, we speculated that the change in hydroxy evodiamine, laudanosine, and 1-methyl-2-[(z)-4-nonenyl]-4 (1H)-quinolone is related to wine processing. Te analgesic efects of 7-hydroxycoumarin, chlorogenic acid, and limonin have been reported in relevant studies. Barros TA found that 7-hydroxycoumarin can produce sustained analgesic efects on chronic infammatory pain induced by acetic acid and formalin [24]. Bagdas suggested that chlorogenic acid has antinociceptive efects on streptozotocin-induced diabetic neuropathic pain in rats [25]. Te limonin derivative V-A-8's analgesic and anti-infammatory efects were more powerful than aspirin and naproxen [26]. Moreover, we found that the analgesic efect of Evodia rutaecarpa is related to 7-hydroxycoumarin, chlorogenic acid, and limonin by molecular docking technology. Terefore, 7hydroxycoumarin, chlorogenic acid, and limonin were selected for validation in the later experimental validation.
ER has a good curative efect on dysmenorrhea [27]. However, there have been few reports on the efect and mechanism of its diferential components before and after wine processing on dysmenorrhea. Terefore, we have cultured dysmenorrhea mouse uterine smooth muscle cells and intend to use this model to evaluate the efect of differential components before and after ER wine was processed (7-hydroxycoumarin, chlorogenic acid, and limonin) on dysmenorrhea mouse uterine smooth muscle cell contraction. Consistent with other reports, our study demonstrated that 7-hydroxycoumarin, chlorogenic acid, and limonin could inhibit the contraction of mouse uterine smooth muscle cells, which suggested that the diferential components may be efective in inhibiting uterine smooth muscle contraction.
In addition, the endocrine factors PGE2, ET-1, and NO play crucial roles in dysmenorrhea [28][29][30]. During menstruation, the increase in PGE2 content destroys endometrial cells and releases a large amount of prostaglandinE2α, which may cause dysmenorrhea as a result of strong contraction of uterine smooth muscle [31]. ET-1 and NO are two vasoactive substances with opposing efects on the endometrium [32]. Te increase in ET-1 content causes blood vessels to contract strongly and produce pain [32]. In dysmenorrhea, diferent doses of NO act on target cells and exert two-way adjusting efects. When the NO content is increased, it can inhibit the release of nociceptive substances and alleviate pain; by contrast, it will promote the contraction of uterine smooth muscle cells and cause pain [32]. As described in Figure 5, the levels of PGE2 and ET-1 were suppressed in the 7-hydroxycoumarin and limonin groups, and the content of NO was upregulated. In addition, the level of PGE2, ET-1, and NO in the chlorogenic acid group changed compared to the model group, but there was no signifcant diference, which suggested that 7-hydroxycoumarin, limonin, and chlorogenic acid regulated the level of endocrine factors in uterine smooth muscle cells and thus improved dysmenorrhea. Te P2X3 receptor plays an essential role in various pain symptoms [33]. Activation of the P2X3 receptor can cause upregulation of Ca 2+ levels in rats with chronic pancreatitis, and the antagonist treatment downregulates P2X3 activity and Ca 2+ signal transduction [34]. Te pathophysiology of dysmenorrhea may involve abnormalities within the critical efector systems responsible for Ca 2+ efects on uterine   smooth muscle cells [35]. Reports have demonstrated that calcium in smooth muscle cells regulates cell contraction as a second messenger, and the cell contraction may be associated with an increase in intracellular calcium concentration [36,37]. In addition, the P2X3 receptor antagonist A-317491 has a positive efect in the treatment of hyperalgesia caused by endometriosis in rats. Furthermore, researches have reported that the P2X3-Ca 2+ pathway is involved in ATP-induced analgesia actions [38]. Our study demonstrated that 7-hydroxycoumarin, chlorogenic acid, and limonin could inhibit dysmenorrhea by inducing downregulation of P2X3 and Ca 2+ . Te limitations of our experiment are as follows: (1) we selected a small animal sample size because the original    intention of the study was to obtain uterine smooth muscle cells by establishing a dysmenorrhea mouse model. (2) "A priori" power calculations in the study were lacking, which may have made the sample size insufcient and infuenced the interpretation of the experimental data.

Conclusion
In conclusion, seven compounds were screened and identifed in the extracts of raw ER and PER, and 7hydroxycoumarin, chlorogenic acid, and limonin were able to inhibit cell contraction and the expression of PGE2, ET-1, P2X3, and Ca2 + in dysmenorrhea mouse uterine smooth muscle cells. Furthermore, these compounds increased NO content. Tese results demonstrated the compounds of the PER were diferent from those of the raw ER, and 7-hydroxycoumarin, chlorogenic acid, and limonin could improve dysmenorrhea mice's uterine smooth muscle cell contraction and relieve pain when combined with endocrine factors and the P2X3-Ca2 + pathway.

Data Availability
Te data used in this study are made available upon reasonable request to the corresponding author.

Ethical Approval
Te

Conflicts of Interest
Te authors declare that there are no conficts of interest.