Chimonanthus salicifolius extract alleviates DSS‐induced colitis and regulates gut microbiota in mice

Abstract Ulcerative colitis is a chronic and recurrent gastrointestinal intestinal disease accompanied by inflammatory disorders, immunologic inadequacy, and intestinal flora dysbiosis, and current therapeutic pharmaceuticals have limited side effects. In this study, we revealed the extraction method of Chimonanthus salicifolius, analyzed the main component, compared the effect of its extract, Lactobacillus, and conventional drugs with different properties on DSS (dextran sodium sulfate)‐induced colitis, and indicated extract regulatory properties of inestinal flora. A colitis model was established on experimental design, and BALB/c mice (male, 7 weeks old) were randomly assigned to five groups (n = 10): control, DSS model, Chimonanthus salicifolius extract (CSE), Lactobacillus rhamnosus GG (LGG), and 5‐aminosalicylic acid (5‐ASA) groups. The three treatments could alleviate the symptoms and remit inflammation induced by DSS, in which CSE and LGG groups could both decrease the proinflammatory cytokine IL‐6, IL‐8, and TNF‐α levels and increase anti‐inflammatory cytokines IL‐10 and TGF‐β. The CSE intervention significantly promoted the higher production of butyric acid than LGG and 5‐ASA groups (p < .05) after DSS challenge. Analysis of intestinal flora showed that CSE administration remarkably decreased the relative abundance of pathogenic bacteria Heliobacteriaceae and Peptococcaceae and exhibited higher abundance of Lactobacillaceae and Bifidobacterium than LGG in intestinal tract of mice (p < .05). These findings indicated that Chimonanthus salicifolius extract may have been beneficial for preventing and treating colitis.

one of the most common diseases in the world (Gwee et al., 2017;Krogsg et al., 2013). Routine medications for IBD include antibiotic and aminosalicylic acids, glucocorticoids, and immunosuppressants, in which 5-ASA is the main treatment for maintenance and remission of IBD (Aljammaz et al., 2020;Sugaya et al., 2021). However, 5-ASA derivatives are easily absorbed by the duodenum and upper jejunum and cannot reach the lesion site which lost its topical antiinflammatory in the intestinal mucosa, meanwhile, the vast majority of patients present with alternating recurrence and remission and adverse reactions to drugs may cause headache, vomiting, and nausea (Arokiadoss & Weber, 2021;Fatahi et al., 2022). Probiotic was reported to have therapeutic properties, such as improving microbiota and regulating immune response. Administration of Lactobacillus rhamnosus GG, significantly relieved DSS-induced colitis, its effect was related to altering the expression of the MUC and TFF genes.
The soluble-protein HM0539 derived from LGG, showed significant protective effects against murine colitis, the protective function involved in the modulation of inflammatory responses (Li et al., 2020;Macdonald et al., 2011). Therefore, exploring safe therapeutic targets with higher efficacy and lower incidence of adverse reactions is still a focus of attention in the treatment of IBD.
Although the etiology and etiopathogenesis of UC are not completely elucidated, intestinal mucosal barrier impairment is considered to be an important pathogenesis of IBD; with the long-term intestinal inflammation, the disruption of intestinal mucosal barrier function can lead to the displacement of intestinal bacteria and endotoxins and the release of inflammatory mediators, thus resulting in persistent tissue damage (Ozair et al., 2021;Staudacher et al., 2021;Wójcik et al., 2020). It is becoming increasingly clear that a combination of susceptibility genes, environmental factors, and the immune system plays important roles in production of inflammatory cytokines, and mediators cause inflammation of the intestinal mucosa, in which intestinal tract environmental factors are primary Luiza et al., 2020;Tilg & Kaser, 2011).
The gut microbiome constitutes a vast and complex ecosystem in intestinal tract environment, in which the bacteria are mainly composed of Bacteroidetes, Firmicutes, Proteobacteria, and Actinobacteria (Agnello et al., 2020;Labus et al., 2017). Most of these bacteria belong to obligate anaerobe; Firmicutes and Bacteroidetes accounted for more than 90% of intestinal bacteria (Belizário & Faintuch, 2018).
The species and quantity of intestinal flora vary with intestine location. When the structure and quantity of intestinal flora change, it can affect the homeostasis of the intestinal environment and the abnormal function of intestinal flora, which then leads to intestinal tract diseases (Lin et al., 2021;Zhou et al., 2020). According to previous research, in colitis patients, compared with healthy people, the number of microorganisms in the intestinal tract is severely reduced, especially for the benefit of bacteria (Xing et al., 2020). With the decreasing abundance and diversity of the bacteria, the stability of the dominant bacteria is damaged in the intestinal tract of patients.
The imbalance of intestinal flora has a certain relationship with the incidence and aggravation of colitis (Tang et al., 2021).
Chimonanthus salicifolius (C. salicifolius) is a unique medicinal plant that belongs to the traditional antiviral of She nationality in China, which is classified as Calycanthaceae family and applied for food (Liang et al., 2016;Liu & Fu, 2018). According to the Chinese Pharmacopeia, it is a kind of food herbal tea used to treat stomach pain and indigestion caused by dyspepsia and eating injury, as well as to prevent infantile malnutrition and diarrhea . The genome of C. salicifolius was sequenced as 820.1 Mb containing 36,651 annotated protein-coding genes (Lv et al., 2020). The chemical constituents mainly focused on volatile oil (monoterpenoids and sesquiterpenoids), and its stems and leaves contained coumarins, flavonoids, anthraquinones, and other components, which have antioxidant and anti-inflammatory function (Wang et al., 2016). The current studies mainly researched its chemical constituents, but are rarely reported on its bacteriostasis application, especially for gastrointestinal regulation. In our previous study, C. salicifolius reduced defecation time and significantly increased gastrointestinal transit rate, fecal particles, and water content, which effectively improved loperamide-induced constipation in mice. The aim of this study was to identify the main components of C. salicifolius extract, verify the extract treatment effects on DSSinduced colitis comparison with Lactobacillus and 5-ASA evidenced by SCFA and cytokine levels, and reveal the regulatory mechanisms of intestinal microbiome.

| Preparation of C. salicifolius aqueous extract (CSE)
Chimonanthus salicifolius leaves were obtained from Zhejiang TactArtiste Biotechnology Group Co. Ltd. The fresh leaves, which were prepared for crushing in burnisher, were dehydrated at room temperature with constant aeration. Three liter of sterile water was added into 500 g dry material and then boiled at 95°C for 3 h in a water bath. The supernate was collected and filtered by kieselguhr.
The extract was centrifuged (8,000 g for 15 min), supernatant was collected, and freeze-dried by freeze dryer (Christ Delta LSC). Then, the extracted water was redissolved with aquae sterilisata into 0.5 g/mL of crude drug. Mice were administrated intragastrically by gavage needle with C. salicifolius extract twice a day by 10 mL/kg bodyweight (equivalent to 10 g/kg).

| High-performance liquid chromatography (HPLC) analysis
Main component of C. salicifolius extract was determined by highperformance liquid chromatography (HPLC) equipment. CSE (0.1 g/ mL) was prepared in a 10 mL volumetric flask, where methanol was added to dissolve and dilute the solution to the scale. After mixing the solution, the sample was filtered by 0.45 μm microporous membrane (Sigma). A ZORBAX Extend-C18 liquid chromatography column (250 mm × 4.6 mm, 5 μm) was used for separation and the temperature was maintained at 25°C. Ten microliter of the sample was injected into the system with a 1 mL/min flow rate. The mobile phase consisted of acetonitrile (A) and 0.2% glacial acetic acid water (B). The linear gradient elution started at 14.5%-18.0% (A), changed to 16.5%-35% (A) after 20 min, and changed to 50%-16.5% (A) after 30 min. The extract was monitored by a diode array detector at 365 nm.

| Induction of colitis and experimental design
A total of 50 male BALB/c mice (7-week-old, bodyweight 30-32 g), provided by the Shanghai Laboratory Animal Center, were housed in a standard rodent cage at a constant temperature and humidity under conventional standard laboratory conditions. All protocols for this study were approved by the Laboratory Animal Ethics of Zhejiang Chinese Medical University, China. Lactobacillus rhamnosus GG was cultured and propagated twice in MRS broth at 37°C for 16 h, and then the bacterial suspension was collected and adjusted to 3.0 × 10 9 CFU/mL with normal saline which would be used for oral gavage to mice.
The mice (BALB/c mice, male, 7 weeks old, inbred strain) were randomly divided into five groups (n = 10 for each group) and classified into a normal control group (control), DSS-induced model group (2.5%, w/v, 36-50 k Da), 10 g/kg CSE group, LGG (3.0 × 10 9 CFU/mL), and 5-ASA (100 mg/kg) group (Murray et al., 2020;Wen et al., 2021). Apart from the control group, DSS solution (2.5%, w/v) was prepared for drinking water to the mice daily in the other four groups for 2 weeks, while the mice in the control group were given a saline solution. After administering, in the model groups were detected changes in colon length, occult blood, and hematoxylin-eosin (HE) staining sections. When the model groups were conducted, the treatment with CSE, LGG, and 5-ASA was intragastrically administered daily to the mice for 14 days, meanwhile, the control and DSS groups were given 0.85% normal saline. At the end of the experiment, animals were anesthetized to death and the colon of each mouse was collected. The upper part of the colon below 1 cm was taken 2 cm and cut into five sections for inflammatory cytokines, short-chain fatty acid, and microbiome assay.

| Histological analysis of colon
The distal colon was fixed at 1 cm in 10% formalin for 16-18 h, then rinsed with water for 24 h to remove the fixative fluid. Different gradient ethanol was used as a dehydrating agent to dehydrate colon tissue. The dehydrated colon tissue was placed in an equal mixture of alcohol and xylene for 15 min by EG1160 paraffin-embedding machine (Leica) for paraffin embedding. The standard for evaluation of histological injury was identified by HE-staining sections.

| Determination of cytokine levels in colon tissue by real-time PCR
The sample isolated from 30 mg colon tissue was used for extracting RNA using the TRIzol™ reagent. The total RNA of mouse tissue was reversely transcribed into cDNA using the PrimeScript™ First Strand cDNA Synthesis Kit (Thermo Scientific). The reverse transcription system was as follows (10 μL): 6 × buffer 2 μL, prime mix 1 μL, enzyme 1 μL, and RNA + RNase Free dH2O 6 μL. Specific primers of cytokine are shown in Table 1. Real-time fluorescence quantitative PCR (RT-QPCR) system was prepared according to the requirements, and the BIO-RAD real-time system (Bio-Rad Laboratories) was used for RT-QPCR. The expression level of the target gene was calculated using the 2 −ΔΔCt method (Johnson et al., 2005;Schmittgen & Livak, 2008).

| Determination of short-chain fatty acid (SCFA)
The tissue sample was precisely weighed 50 mg, added into 1 mL of 6% phosphoric acid solution homogenate, and then transferred to a 20 mL headspace sample bottle. GC-MS 7000D (Agilent Technologies) was used to measure the changes of short-chain fatty acids in the colon of mice for each intervention group. Static headspace injection was used with an incubation temperature of 85°C, incubation time of 30 min, injection needle temperature of 95°C, and injection volume of 1 mL. The selective ion monitoring mode was adopted. The mass-to-charge ratio of acetic acid, propionic acid, isobutyric acid, butyric acid, and valeric acid were 60, 74, 743, 60, 60 respectively. The collected peak area ratio of each peak was substituted into the standard curve to calculate the content of each component in the sample.

| Analysis of intestinal microflora
The colon sample was taken from the fresh colon tissue treated with liquid nitrogen, which ground with a sterile mortar. Total  et al., 2018). Lastly, the PCR purified products were sequenced on an Illumina HiSeq 2500 platform (Illumina) according to the protocol described. Sequences with 97% similarity cutoff were assigned to the same OTUs. OTUs abundance information was normalized using a standard of sequence number corresponding to the sample with the least sequences. Principal coordinate analysis (PCoA) was used to analyze the diversity between groups.
The alpha diversity was calculated to analyze the diversity of the microbial community by using QIIME (Version 1.7.0). R software (Version 2.15.3) was used to determine the beta diversity and compare the differences between treatment groups (Bokulich & Spiller, 2016).

| Statistical analyses
Statistical analysis of the data was calculated and conducted with GraphPad 9.3 Software. Significant differences among groups were performed using one-way ANOVA. The data were expressed as the mean value ± SD. p < .05 was regarded as statistically significant.

| Main components of CSE
The HPLC results was represented in Figure 1. The main components of CSE were rutin, hyperin, isoquercitrin, afzelin, and kaempferol.
The instrument precision inspection showed that the standard deviations were 0.56%, 0.23%, 0.62%, and 0.56%, respectively. The data were less than 2%, which indicates the precision of the instrument.

| Chimonanthus salicifolius extract improved the colonic damage induced by DSS
DSS-induced colitis model mice were measured by bodyweight of mice, the fecal occult blood, and gross blood stool. Changes in mice weight are observed in Table 2.
During the first 7 days, there was no significant difference in bodyweight between the treatment groups. On day 14, mice weight of DSS treatments reduced significantly, meanwhile accompanied with mice's curly hair and lazy phenomenon. The stool of the mice with weight loss was irregular and adhered to the anus; stool test was positive with fecal occult blood test paper, which showed that the DSS-induced colitis model was successfully constructed. After gavage by 10 g/kg CSE, LGG (3.0 × 10 9 CFU/mL), and 5-ASA (100 mg/ kg), respectively, for 14 days, significantly reduced weight after DSS treatment compared with control group. There were no significant differences in bodyweight between the DSS treatment groups.
Histopathological observation was an effective method to diagnose the severity of colitis. In order to evaluate the protective function of different treatments, the histological scores of colonic tissues were based on inflammatory cells, goblet cells, crypt arrangement, epithelial villous, and glandular structure in mice colon. The histological score was defined as the sum of score for five indexes (Table 3). In the control group (Figure 2), the colon epithelium, villi structure, and crypts were arranged orderly, and there were abundant goblet cells. The glands were intact, and no inflammatory cell infiltration was observed.
However, histological score of the DSS model was 11.78 ± 1.75, ulceration appeared in the mid-intestinal tissue, intestinal glandular, epithelial structures, and villi and crypts completely disappeared with extensive connective tissue proliferation in the mucosal layer. In the CSE group (histological scores 9.13 ± 1.06), a spot of connective tissue proliferation appeared, but villi, crypts, and intestinal glandular were relatively complete, which showed a significant decreased scores from LGG and 5-ASA groups. The intestinal glandular of LGG group was disorderly and inflammatory cell emerged in 5-ASA group.
Colon length could reflect DSS-induced colonic damage. After DSS exposure, colon lengths of treatments were shorter than control group, but the colon length of mice supplemented with CSE was a significant increase compared to LGG and ASA groups, which indicated that C. salicifolius extract could alleviate DSS-induced colitis.

| Expression levels of cytokine genes in colon tissue
Effects of different treatments on the relative expression of cytokines in colonic tissue were analyzed by real-time PCR. IL-6 and

IL-8 can stimulate the proliferation of T and B cells and secrete
antibodies to participate in the inflammatory response. TNFα is a recognized proinflammatory cytokine, which activates and induces the production of inflammatory cells. According to previous studies, these cytokines are commonly used as proinflammatory factors to evaluate immunomodulatory function (Kobayashi et al., 2017;Liu et al., 2020).
We detected levels of changes in proinflammatory cytokines (IL-6, IL-8, and TNFα) and anti-inflammatory cytokines (IL-10 and TGFβ) in colon tissues of mice (Table 4)  (IL-6, IL-8, and TNFα) and increase anti-inflammatory (IL-10 and TGFβ) compared to model group. There was no significant difference in TNFα levels between the control and treatment groups. Notably, the LGG group showed the lowest expression level of IL-6, and administration of the CSE group could significantly reduce IL-8 expression level comparison with LGG and 5-ASA groups. The CSE group increased the expression of anti-inflammatory cytokines significantly in comparison with other two groups and presented the highest expression at IL-10 and TGFβ. Overall, the administration of CSE suggested the most significant inflammatory modulation effect.

| Chimonanthus salicifolius extract regulated the production of SCFAs
Short-chain fatty acids (SCFAs) degraded undigested carbohydrates and small amounts of protein, which are produced primarily by microorganisms that are essential for gut health. In the model group, the level of acetic acid (AA), propionic acid (PA), isobutyric acid (IA), butyric acid (BA), and valeric acid were (VA) significantly lower than control group (

| Effect of C. salicifolius extract on modulation of intestinal microbiota
The V3-V4 region of the 16 S rRNA gene was sequenced with Illumina Miseq to indicate the modulatory effects of different treatments on gut microbiota of DSS-induced mice. The annotation results of species evaluated the relative abundances of each group at the genus level ( Figure 5). A total of 1076 gene sequences were identified, of which 10 genera were classified to have relative abundances.
The relative abundance showed different diversity of treatments ( Figure 6). A significant decrease in the relative abundance of Heliobacteriaceae was observed following treatment with all treatments compared with the model group. The LGG and CSE groups exhibited extremely significant lower abundance (p < .05).

Heliobacteriaceae was usually classified into Firmicutes and
Peptococcaceae belonging to Bacteroidetes. The ratio between the two phyla was considered for certain relations with intestinal flora (Meng et al., 2020). Lactobacillaceae and Bifidobacterium were regarded as probiotic, which was beneficial for intestinal adjustment.
At Bacteroidaceae genus level, there was no difference among LGG, 5-ASA, and model groups, however, the abundance of CSE group was significantly reduced. After the administration of CSE and LGG groups, the relative abundances of Bifidobacterium increased significantly (p < .05); noteworthily the CES group reached the highest abundance at 39.52%. For the Lactobacillaceae, only the CES group markedly increased the relative abundance compared with model group and presented a maximum abundance of 1.98%.
Colitis is a common inflammatory disease that is involved with abdominal pain, diarrhea, and constipation, which seriously affects human health (Min et al., 2020). Although 5-ASA as therapeutic drug could treat colitis, the treatment process is accompanied by side effects such as stomach distension, nausea, vomiting, and abnormal liver function. Previous studies have suggested that intestinal tract disorder and intestinal flora change was important factor for colitis.
Most modern medical research on colitis has been concerned with probiotics, supplemented to relieve diarrhea, which is reported to be due to their metabolites such as bacteriocins and SCFAs (Bharucha et al., 2017). However, there have been less reports focusing on plant extract provided by natural resources in the degree of alleviation of colitis or the resulting changes in the fecal flora in mice.
Chimonanthus salicifolius distributed in the mountain areas, which is considered a unique plant in Eastern China, has been applied for both medicine and food. isoquercitrin, afzelin, and kaempferol, respectively, which is consistent with the previous study that the different chemical substances existed between aqueous and alcohol extract (Liu et al., 2013).
Notably, these components belonged to flavonoids and widely ex-  In this study, DSS-induced colitis model was established successfully accompanied by fecal occult blood and histological scores.
The treatment groups could improve colon length and colonic epithelial structure. Remarkably, the CES group showed significantly decreased scores and increased colon length compared with LGG and 5-ASA groups, and villi, crypts, and intestinal glandular were relatively complete. DSS upregulated the expression levels of IL-6, IL-8, and TNFα in the colon, which may lead to the immune disorder of the colon, aggravated inflammation, and damaged colon tissue.
According to the inflammatory cytokines, although the LGG, 5-ASA, and CSE group could inhibit inflammatory cytokines and in-

| CON CLUS IONS
In this study, the main components of C. salicifolius extract were analyzed and identified as rutin, hyperin, isoquercitrin, afzelin, and kaempferol. We investigated the effects of the administration of the extract on colitis symptoms compared with LGG and 5-ASA.
The treatment of C. salicifolius could significantly improve colonic mucosal injury and colon length, inhibit inflammatory cytokines, increase anti-inflammatory levels (p < .05), and promote metabolic production of SCFAs. Moreover, the intervention of C. salicifolius extract on colitis mice effectively regulated intestinal flora induced by DSS and restored their relative abundance. Furthermore, the

CES group significantly increased the abundance and diversity of
Lactobacillaceae compared with other treatments. Therefore, the results indicate that C. salicifolius extract presents the potential to be developed as a candidate for treating colitis.

ACK N OWLED G M ENTS
We would like to thank Key Laboratory for Food Microbial Technology of Zhejiang Province for experimental help. We are grateful to Zhejiang Tact Artiste Biotechnology Group Co. Ltd for their assistance in sample collections.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors have no conflict of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.