Spectrum-Effect Relationship between HPLC Fingerprint and Anti-Inflammatory and Analgesic Activities of Chloranthus fortunei (A. Gray) Solms-Laub

The predominant objective of the research is to establish the anti-inflammatory and analgesic spectrum-effect relationship of Chloranthus fortunei (A. Gray) Solms-Laub (CF), to reveal the pharmacodynamic basis of the anti-inflammatory and analgesic effects of CF. The fingerprints of ten batches of CF from various origins were established by high-performance liquid chromatography (HPLC) and evaluated for similarity, hierarchical cluster analysis (HCA), and principal component analysis (PCA). The anti-inflammatory and analgesic effects of CF were evaluated with the xylene-induced ear swelling in mice and the acetic acid torsion test, while the anti-inflammatory and analgesic spectrum-effect relationship of CF was evaluated by gray relational analysis (GRA) and partial least squares regression analysis (PLSR) to effectively elucidate the anti-inflammatory and analgesic substance basis of CF. The ten batches of CF HPLC fingerprints established in this work successfully identified a total of 13 common peaks that refer to 4 components, with peak 1 being neochlorogenic acid, peak 3 being chlorogenic acid, peak 5 being cryptochlorogenic acid, and peak 10 being rosmarinic acid. The HCA results presented that the ten batches of CF samples were clustered into 3 categories, which was consistent with the PCA results. Simultaneously, the results of the spectrum-effect relationship also indicated that neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, and rosmarinic acid were the possible anti-inflammatory and analgesic substances of CF. In order to better understand the anti-inflammatory and analgesic substance basis of CF, this experiment established the anti-inflammatory and analgesic spectrum-effect relationship of CF, which can provide a scientific foundation for the quality evaluation and further research as well as the usage of CF herbs.


Introduction
Chloranthus fortunei (A. Gray) Solms-Laub (CF) is a perennial herb which can be mainly discovered in the eastern and southern parts of China. At the same time, it is also a relatively common medicinal herb for the Yao people in China. It possesses the efcacy of dispersing wind and cold, activating blood circulation, relieving pain, detoxifying, and reducing swelling [1,2]. Concurrently, it has been commonly employed clinically in the treatment of rheumatoid arthritis, bruising, and swelling pain as well as chronic gastroenteritis. At present, the research on this herb has been primarily focused on its chemical composition, but research on the material basis of its anti-infammatory and analgesic efects has not been comprehensively examined [3,4].
Infammation represents a general, complex physiological response of the immune system to external challenges or tissue damage. Te occurrence of an infammatory response assists in the healing of wounds. However, when infammation is excessive or uncontrollable, an abnormal and prolonged infammatory response will inevitably result in a variety of diseases [5][6][7]. Pain has been one of the predominant features of infammation and is defned by the International Association for the Study of Pain as an unpleasant sensory and emotional experience which has been often associated with actual or potential tissue damage [8][9][10]. Similarly, there have also been numerous studies that have revealed a defnite link between infammation and pain. As such, the research on the comorbidity of infammation and pain has unavoidably become increasingly widespread worldwide [11][12][13].
NSAIDs such as ibuprofen and paracetamol, which have been proven to possess satisfactory anti-infammatory and analgesic efects and yet also have their own limitations, are the most popular anti-infammatory and analgesic medications available today [14,15]. One of their most common shortcomings is gastrointestinal tract injury which is mainly due to the irritation of the drug itself that can cause damage to the gastrointestinal tract which comprises indigestion, nausea and vomiting, and severe pain in the abdomen [16,17]. As such, this is defnitely not what we wish to achieve. Hence, it is necessary to research and develop new highly efective low-toxic anti-infammatory and analgesic drugs. Simultaneously, the pharmacological basis of its action should also be elucidated.
In recent years, high-performance liquid chromatography (HPLC), which has the advantages of excellent separation efciency, swift analysis speed, and outstanding detection sensitivity, has been extensively employed in the quality assessment of herbal medicines. Tis technology allows for quick qualitative and quantitative analysis of the components of the samples analyzed [18][19][20]. In order to more accurately refect the complex chemical information contained in traditional Chinese medicine (TCM), fngerprinting technology is utilized in TCM to systematically characterize the chemical components of TCM from the perspective of the overall chemical substance base of TCM, primarily by the HPLC method. By combining fngerprinting with pharmacological efects through relevant statistical methods, the active ingredients of TCM that function can be efectively screened [21,22]. In the present study, the above methods were organically combined to reveal the antiinfammatory and analgesic potent substances of CF in a multidimensional manner.
Based on early screening for anti-infammatory and analgesic efcacy, the aqueous extract of CF with the most optimal anti-infammatory and analgesic efects was selected for the fngerprinting investigation, and the fngerprinting of the CF aqueous extract was established using the HPLC technique. Te fngerprints of the CF aqueous extract from ten diferent origins were evaluated for similarity, hierarchical cluster analysis (HCA), and principal component analysis (PCA); the anti-infammatory and analgesic activities of the CF aqueous extract were investigated by constructing a xylene-induced ear swelling model in mice and an acetic acid torsion test in mice. Gray relational analysis (GRA) and partial least squares regression analysis (PLSR) were employed to examine the anti-infammatory and analgesic spectrum-efect relationships. Tis study provides a scientifc basis for revealing the anti-infammatory and analgesic substances of CF. Table 1, the ten batches of CF herbs had originated from various regions of Guangxi, China, and were verifed by experimentalist Zhu Yilin of the Guangxi University of Traditional Chinese Medicine, Guangxi, China. Rosmarinic acid (batch number: RP190718), chlorogenic acid (batch number: RP190505), cryptochlorogenic acid (batch number: RP2101101), and neochlorogenic acid (batch number: RP200416) were purchased from Chengdu Madison Technology Co., Ltd. with purity ≥99.97%; methanol, ethanol, and glacial acetic acid were analytically pure, ultrapure water (laboratory homemade), acetonitrile (chromatographic alcohol, Fisher), and methanol (chromatographic alcohol, Tedia).

Preparation of the Standard Solution and Sample Solution.
Chlorogenic acid, neochlorogenic acid, cryptochlorogenic acid, and rosmarinic acid were accurately weighed and dissolved in methanol to prepare mixed standard solutions that contain 219.2 μg/mL of neochlorogenic acid, 282.8 μg/ mL of chlorogenic acid, 244 μg/mL of cryptochlorogenic acid, and 343.2 μg/mL of rosmarinic acid.
In the fngerprinting section, 2 g of CF powder was weighed precisely; 25.00 mL of water was added precisely, placed in a fask, weighed and refuxed for 1 h, and then cooled and weighed again to make up the weight. Subsequently, the CF solution to be measured is obtained by thoroughly shaking the mixture, fltering it, and then passing the fltrate through a 0.22 μm microporous membrane.
Te CF herbs were utilized in the anti-infammatory and analgesic experiment in which they were cut into 1 cm pieces, weighed 500 g, added to distilled water (solid-liquid ratio of 1 : 10), and soaked for 30 minutes. Tey were then twice extracted at refux for 1.5 hours each time, fltered through medical gauze, combined and concentrated into an infusion, and used after being diluted with distilled water to the corresponding concentration.

Validation of Methodology.
Te sample solution of CF (S1) was accurately prepared in accordance with the sample preparation method in "2.3" for fngerprinting. Next, the sample was injected six times according to the chromatographic conditions in "2.4," and the precision was checked. Similarly, the stability of the CF (S1) solution was investigated by sample preparation and chromatographic conditions in "2.3" and "2.4" at 0 h, 2 h, 4 h, 8 h, 12 h, and 24 h, respectively. Subsequently, the six replicates of the CF (S1) solution were injected, and the reproducibility of the CF (S1) solution was investigated.

Establishment and Evaluation of Fingerprint.
Te sample preparation procedure described in "2.3" was implemented to prepare ten batches of CF samples, and the chromatographic conditions described in "2.4" were applied to determine and obtain the relevant profles. Te "Chromatographic Fingerprint Evaluation System for Traditional Chinese Medicine (2012 version)" was utilized to analyze the HPLC fngerprints of the ten batches of CF samples. At the same time, the superimposed spectra and common patterns of the ten batches of CF samples were obtained using the S1 pattern as the reference pattern, and the generated control pattern R was used as the reference. Te evaluation of general similarity was performed. Finally, HCA and PCA were applied to the fngerprint profles of each source [23].

Anti-Infammatory
Test. 120 Kunming mice, weighing 18-22 g each, were randomly assigned to 12 groups (n � 10, including 5 males and 5 females). Subsequently, ten CF aqueous extract groups of various sources (7.28 g/kg), the positive group (dexamethasone 0.006 g/kg), and the model group (distilled water) were individually set up, and each group shall be administered with 20 mL/kg of the drug once daily for seven days. Te mice's right ear was uniformly treated with 20 μL of xylene 45 minutes after the last administration, whereas the left ear was left untreated. After 30 min of infammation, the mice were executed, the round earpieces were punched with a 6 mm diameter perforator in the same part of both ears and weighed, and the diference in weight between the left and right earpieces was utilized as the degree of swelling. Te swelling degree and the swelling inhibition rate were calculated according to the following formula, respectively [24]: swelling de gree � weight of the right ear − weight of the left ear. (1) Swelling inhibition rate (%) � (model group ear swellingadministration group ear swelling)/model group ear swelling × 100%.

Analgesic
Test. 120 Kunming mice, weighing 18-22 g each, were randomly assigned to 12 groups (n � 10, including 5 males and 5 females). Each group was set up with the following: the model group (distilled water), the positive group (0.05 g/kg rhododendron), and the aqueous extract group (7.28 g/kg) of ten various sources of CF. Each group shall be administered with 20 mL/kg of the drug once daily for seven days. Simultaneously, 0.6% acetic acid solution (10 mL/kg) was administered intraperitoneally 1 hour after the last administration. Te mice torsion times within 15 minutes were counted, and the analgesic inhibition rate was determined using the following formula [25].
Analgesic inhibition rate (%) � (torsion times of the model group-torsion times of administration group)/ writhing times of the model group × 100%.

Evaluation of the Anti-Infammatory Analgesic Spectrum-
Efect Relationship 2.9.1. Gray Relational Analysis (GRA). Gray relation analysis is a research method that aids in decision-making by examining the magnitude of data correlation (the correlation degree between parent sequence and subsequence) and measuring the degree of association between data through correlation. Te inhibition rate of mouse ear swelling and the inhibition rate of mouse torsion were taken as the parent sequences, and the peak areas of the common peaks of the fngerprint profles of ten batches of CF were taken as the subsequences. At the same time, the respective processed data were recorded as Y(k) and Xi(k). Te raw data were processed and calculated according to formula (2) to successfully obtain the absolute value of the diference between Y(k) and Xi(k), respectively. Te correlation coefcient is usually utilized to indicate the degree of association between the 2 compared sequences, whereby it can be attained by substituting the absolute value data of the calculated diference series into formula (3), and the discrimination coefcient ρ (0∼1) was chosen as 0.5. Te correlation degree (r) is the degree of association between the two correlations, and the calculated correlation coefcient was brought into formula (4) to calculate the degree of association between each common peak and the antiinfammatory and analgesic efects of CF [26]. When the correlation degree is greater than 0.6, it indicates that the chemical compounds represented by the chromatographic peaks are correlated with the pharmacological index; when the correlation degree is greater than 0.8, it indicates that the chemical compounds represented by the chromatographic peaks have a great correlation with the pharmacological index; when the correlation degree is greater than 0.9, it indicates that the chemical compounds represented by the chromatographic peaks are very much correlated with the pharmacological index:

Partial Least Squares Regression Analysis (PLSR).
PLSR is a statistical method which combines linear regression analysis, typical correlation analysis, and principal component analysis. By establishing regression equations and comparing regression coefcients and VIP values, the combined contribution of the common peaks to efcacy can be accurately refected. Te trend of the efect of each common peak on efcacy can be determined from the positive and negative values of the regression coefcients, where a positive value means that the common peak is positively correlated with efcacy and a negative value means that it is negatively correlated with efcacy. Te variable importance in projection (VIP) of PLSR is an important index refecting the explanatory ability of the independent variable on the dependent variable, and the larger the value, the stronger the explanatory ability of the independent variable on the dependent variable. It is generally accepted that when VIP > 1, the independent variable has signifcant importance in explaining the dependent variable. Te peak area of each common peak in the fngerprint was utilized as the independent variable X. Te inhibition rates of CF aqueous extract on ear swelling in mice as well as the inhibition rate of analgesia in mice were utilized as the dependent variables Y. Concurrently, the PLSR of the anti-infammatory and analgesic spectrum-efect relationship of CF was successfully performed using the SIMCA 14.1 software [27].

Method Validation for HPLC Fingerprint.
Peak 10 (rosmarinic acid), chosen as the reference peak to assess the HPLC fngerprinting methodology, has excellent stability and an appropriate peak area. Te results demonstrated that the approach has better precision, stability, and repeatability and is acceptable for CF fngerprinting because the RSDs of the relative retention times and relative peak areas of the other 12 peaks were less than 3%. Figures 1 and 2 display the HPLC fngerprints and reference chromatograms of ten batches of CF samples. 13 peaks that were well separated were designated as common peaks. Four peaks were identifed in the fngerprint, with peak 1 being neochlorogenic acid, peak 3 being chlorogenic acid, peak 5 being cryptochlorogenic acid, and peak 10 being rosmarinic acid (Figure 3). Eventually, peak 10 (rosmarinic acid) was selected as the reference peak (S), and the RSD of the relative retention time of each common peak in the ten batches of CF samples was calculated to be less than 0.64% (n � 10), indicating that the reproducibility among samples was satisfactory; the results have been presented in Table 2. However, the RSD of their relative peak areas ranged from 23.13% to 50.65% (n � 10) with large fuctuations, indicating that the content of coconstituents in the samples of CF from diferent sources varied substantially. Te relevant results have been displayed in Table 3.

Similarity Analysis of Fingerprints.
Te overall similarity was evaluated with the reference fngerprint (R) ( Table 4). Te similarity between the fngerprint profles of each origin and the reference fngerprint (R) was greater than 0.900, indicating that the preparation method of CF was stable and that the deviation between the origins was small. Hence, it was able to satisfy the requirements of fngerprinting studies.

Hierarchical Cluster Analysis (HCA).
By utilizing the SPSS26.0 software, the peak areas of 13 common peaks in ten batches of CF samples were employed as variables, and the intergroup linkage clustering method was selected, followed by clustering with the metric of squared Euclidean distance. Te results have been presented in Figure 4. When the classifcation distance was 10, ten batches of CF samples could be classifed into three classes. Among them, samples S1, S3, S5, S7, and S9 were clustered into class 1 and samples S2, S4, and S6 were clustered into class 2, while the remaining samples were clustered into class 3. Te clustering results refected that there were certain diferences among the ten batches of CF samples, indicating that the diferences in environmental factors among diferent origins might be the predominant infuencing factors for their quality diferences.

Principal Component Analysis (PCA).
Te eigenvalues and variance contribution rates were determined by PCA with SPSS26.0 software using the peak areas of 13 common peaks from ten batches of CF samples as the variables. Te outcomes are displayed in Table 5. Using eigenvalues >1 as the criterion, three principal components were obtained with a cumulative variance contribution of 83.40%, representing most of the information of the samples. Terefore, the ten batches of CF samples were evaluated using the frst three main components as indicators. According to Table 6, which depicts the common factor loading matrix, the common peaks 1, 2, 3, 5, 6, and 7 have obviously positive loadings in principal component 1, while peaks 4 and 9 have obviously positive loadings in principle component 2. From Figure 5, it can be observed that the ten batches of CF samples can be divided into 3 categories. Among them, samples S1, S3, S5, S7, and S9 were clustered into class 1 and samples S2, S4, and S6 were clustered into class 2, while samples S8 and S10 were clustered into class 3, which was consistent with the results of cluster analysis.

Results of the Analgesic Test.
In comparison with the model group, the number of torsion was signifcantly lower (P < 0.01) in the CF aqueous extract group of ten diferent sources, and their analgesic inhibition rates were all greater than 37%, with the inhibition rates in the aqueous extract groups of S2, S3, and S6 being greater than 48%. In comparison with other sources, the analgesic inhibition rate was lower in S8 and S10, with 37.53% and 38.08%, respectively. In comparison with the positive   group (rotundine), the analgesic inhibition rate of the CF aqueous extract group was lower than that of the positive group, except for the CF aqueous extract group in S6. At the same time, the experimental results also demonstrated that the ten batches of the CF aqueous extract group would be able to efectively reduce the efect of acetic acid-induced pain in mice, as presented in Table 8 and Figure 7.

Gray Relation Analysis (GRA) between the CF Fingerprint and Its Anti-Infammatory and Analgesic Activities.
Te contribution of the common peaks of the CF aqueous extract to the efect of mouse ear swelling can be observed from Table 9 in the order of chromatography: 12 > 3>10 > 7>11 > 5>1 > 4>6 > 2>8 > 9>13, and the correlation of all peaks except peak 13 is greater than 0.6. Te contribution of the common peaks of the CF aqueous   extract to the efect of mouse torsion can be observed in Table 10 in the order of chromatography: 10 > 4>3 > 9>11 > 8>12 > 7>1 > 5>6 > 2>13, and the correlation of all peaks except peak 13 is greater than 0.6. Te results indicated that the anti-infammatory and analgesic efects of CF were the result of the combined action of multiple compounds. Te compounds represented by the common peaks, with the exception of peak 13, were closely related to the anti-infammatory and analgesic efects of CF.     Swelling degree (mg) ** ** ** ** ** ** ** ** ** ** ** Figure 6: Efects of ten batches of CF aqueous extract on xylene-induced ear swelling in mice (n � 10) (compared with the model group * : P < 0.05 and * * : P < 0.01).    Figure 9, the VIP values of peaks 1, 2, 3, 5, 6, and 7 were greater than 1, which indicated that these peaks had a signifcant efect on the anti-infammatory efect in mice, among which peak 1 was identifed as neochlorogenic acid, peak 2 as chlorogenic acid, and peak 5 as cryptochlorogenic acid, and the antiinfammatory ability of the CF aqueous extract was signifcantly enhanced when its content was increased.

PLSR between the CF Fingerprint and Its Analgesic
Activity. By using the peak area of each common peak in the fngerprint as the independent variable X and the analgesic inhibition rate of the CF aqueous extract in mice as the dependent variable Y, the regression equation obtained from the PLSR analysis of the analgesic spectrum-efect relationship by utilizing the SIMCA 14.1 software can be represented by Y � 0.1629X1 + 0.1104X2 + 0.1654X3 − 0.0885X4 + 0.1538X 5 + 0.0985X6 + 0.175X7 − 0.0437X8 + 0.1758X9 + 0.0715X10 − 0.0657X11 + 0.1001X12 + 0.1123X13. Te values of the model parameters R2X, R2Y, and Q2 are 0.761, 0.987, and 0.940, respectively. As presented in Figure 10, peaks 8 and 11 are negatively correlated with analgesic efects, while the rest of the peaks are positively correlated with analgesic efects. In Figure 11, the VIP values of peaks 1, 2, 3, 5, 6, and 7 are all greater than 1, indicating that these peaks had a more signifcant efect on the analgesic efcacy in mice, among which peak 1 was identifed as neochlorogenic acid, peak 2 as chlorogenic acid, and peak 5 as cryptochlorogenic acid, and the analgesic ability of the CF aqueous extract was also signifcantly enhanced when its content was increased.

Discussion
In the fngerprint section, various columns, wavelengths, fow rates, temperatures, and mobile phases were examined, and the chromatographic conditions that could most comprehensively characterize the information of CF samples were selected in order to produce the HPLC fngerprint with rich chemical information [28]. Te common peaks of ten batches of CF were identifed at the same time. A total of four components were identifed, with peak 1 being neochlorogenic acid, peak 3 being chlorogenic acid, peak 5 being cryptochlorogenic acid, and peak 10 being rosmarinic acid. Te results of similarity analysis presented that the fngerprint profles of ten batches of CF were compared with the      reference profles, and the similarity calculation results were above 0.928, indicating that there were diferences in the chemical composition contents between CF aqueous extracts, which might be due to the diferences in origin, season, climate, harvesting time, growth condition, and maturity between samples. Te results of both CA and PCA indicated that all the ten batches of CF samples were classifed into three categories with S1, S3, S5, S7, and S9 as one category, S8 and S10 as one category, and S2, S4, and S6 as one category. Hence, this clearly indicates that the method can be utilized for the quality evaluation of CF samples.
In the study of the analgesic spectrum-efect relationship, the results of the GRA presented that the correlation of the common peaks that contributes to analgesia was in the order of 10 > 4> 3 > 9> 11 > 8> 12 > 7> 1 > 5> 6 > 2> 13, and the correlation of all peaks was greater than 0.5. Te PLSR also presented that peaks 8 and 11 were negatively correlated with analgesia, and the rest of the peaks were positively correlated with analgesia. Te VIP values of the peaks were ranked as 7 > 3> 5 > 1 > 2 > 6> 8 > 9> 4 > 10 > 13 > 12> 11, among which the VIP values of peaks 1, 2, 3, 5, 6, and 7 were all greater than 1.
By combining the above two analytical methods, the anti-infammatory and analgesic efects of CF may be the consequence of the combined action of several components. Peaks 1, 2, 3, 5, 6, and 7 may be the basis of the antiinfammatory and analgesic substances of CF. Among them, peak 1 is neochlorogenic acid, peak 2 is chlorogenic acid, and peak 5 is cryptochlorogenic acid. All of these have been reported to possess satisfactory antiinfammatory and analgesic activities in the relevant literature [29][30][31]. In addition, the results of this data processing indicated that the VIP value of peak 10 (rosmarinic acid) was not greater than 1, but the literature has shown that rosmarinic acid is an active antiinfammatory and analgesic component in herbs and is also one of the quality control indicators of Sarcandrae Herba of the CF family plants [32,33]. As such, rosmarinic acid should also be the CF anti-infammatory and analgesic potent substance. Lastly, the other chromatographic peaks were not successfully identifed, and further in-depth studies are required in the future.

Conclusion
CF is a widely used medicinal plant employed by the Yao ethnic group in China. It has been successfully utilized in clinical settings to treat several infammatory disorders, demonstrating the herb's potent anti-infammatory and analgesic efects. In this study, the fngerprints of ten batches of CF from diferent sources were established, and a total of 13 common peaks were chosen, among which 4 components were identifed. Te results of similarity analysis displayed that the similarity between the fngerprint profles of each origin and the reference profle (R) was greater than 0.900. Hence, this indicates that the preparation method of CF was stable and that the deviation between the origins was relatively small, which was in line with the requirements of fngerprinting research. Simultaneously, the antiinfammatory and analgesic activities of CF were also comprehensively evaluated with the xylene ear swelling test and acetic acid torsion test, and the results presented that CF possesses satisfactory anti-infammatory and analgesic activities. Te anti-infammatory and analgesic spectrumefect relationship of CF was established by combined GRA and PLSR, whereby it was revealed that neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, and rosmarinic acid may be the anti-infammatory and analgesic potent substances of CF. As such, this study acts as a basis and provides a scientifc basis for the establishment of CF  quality standards as well as their further development and utilization.

Data Availability
Te data used to support the fndings of this study are included within the article.

Conflicts of Interest
Te authors declare that there are no conficts of interest regarding the publication of this paper.