Salvia miltiorrhiza Ameliorates Disease Progression in Dextran-Sodium-Sulfate Induced Colitis in Mice

Salvia miltiorrhiza (SM, or Danshen) extract has been approved by China FDA for the treatment of cardiovascular and cerebrovascular diseases owing to its potent anti-inammatory effects. Whether SM may be used to treat inammatory bowel disease (IBD) remains elusive. In the current study, Dextran-Sodium-Sulfate (DSS) induced colitis in mice was used as a model of IBD, and SM was given orally for 7 days. SM administration has signicantly reduced the disease activity index (DAI) score and weight lost and colon shortening in the DSS-induced colitis mice. The macrophage inltration was signicantly reduced in the SM treatment group. To explore the mechanisms, macrophage processor cell line Raw 264.7 was used to verify the anti-inammatory effect of SM. SM treatment inhibited lipopolysaccharide (LPS)-induced macrophage activation in RAW264.7 cells and signicantly reduced the production of proinammatory factors. The current study provided evidence that oral administration of SM ameliorates pathological deterioration of IBD in mice, and warrants future clinical application of SM for the management of IBD.


Introduction
In ammatory bowel diseases (IBD), including ulcerative colitis, and Crohn's disease, is malfunction of gastrointestinal system with multiple causes. It is characterized with abnormal immune-mediated intestinal chronic and recurrent in ammation [1] and is signi cantly associated with colorectal cancer risks [2]. In recent decades, the global prevalence of IBD keeps increasing [3]. The pathogenesis of IBD remains elusive, environmental factors that act on genetic susceptibility may be the main cause. With the participation of intestinal ora, IBD initiates natural intestine immunity and active immune response.
Immune hyper activation results in pathological changes such as intestinal mucosal barrier damage, ulcer healing and in ammatory hyperplasia. Various immune cells and cytokines are involved in the pathogenesis and treatment of IBD [4][5][6]. DSS damages intestinal epithelial cells and tight junction barriers and ultimately increases intestinal permeability to induce acute colitis. In our study, the DSS-induced mouse acute colitis model was used to mimic the clinical symptom of IBD.
Indeed, anti-in ammatory agents and immunomodulators have been employed to treat IBD. In the longterm, both two types of drugs have limited effectiveness with side effects, such as gastrointestinal reactions, osteoporosis, and myelosuppression. Recently, two anti-tumor necrosis factor (anti-TNF) drugs, in iximab and adalimumab, have been proved their clinical e cacy [7,8]. However, 50% of patients lose the effective treatment response within 1 year [9]. The IBD is a chronic process, and most patients have recurrent episodes, so a supplemental and alternative treatment is needed.
Recently herbal therapy as a potential alternative medical therapy have drawn attention. Many herbs have been shown to treat mice colitis effectively [10][11][12][13][14]. Salvia miltiorrhiza (SM, or Danshen), a traditional Chinese herb, has been widely used in Asian countries for cerebral ischemia-reperfusion injury, blood rheology, platelet function, anti-hypertensive, anti-in ammatory and even protecting the cardiovascular system [15]. Studies indicate SM have anti-tumor, antioxidant, anti-diabetic and anti-brotic effects, which plays a positive role in promoting butyric acid metabolism and regulating intestinal ora [16,17]. In addition, a study has been reported that salvianolic acid B, which was an effective component of SM, could restore impaired barrier function in rat colitis model [18]. Other studies have shown that salvianolic acid A does not affect normal coagulation [19]. Compared with heparin, SM may be safer as it has not been found to increase the risk of blood thinning, no report on hemorrhage in the animal studies. Despite anticoagulant treatment in IBD management is still debatable [20,21], using heparin for treatment of thrombotic UC achieved good results and it is generally accepted [22][23][24]. It has been found that IBD patients have an increase in peripheral platelet count and abnormal activation which in turns to. release a variety of in ammatory mediators, leading to aggregation of in ammatory cells and worsening the in ammatory responses of IBD [25][26][27]. Hence platelets activation is related to the severity of IBD, whereas SM has anti-platelet aggregation effects, therefore SM was selected to test their roles in DSSinduced colitis in mice.

Chemicals and Reagents
Medical grade SM extract was purchased from Chiatai Qingchunbao Pharmaceutical Co., Ltd (National Medicine Standard Z33020177, Hangzhou, Zhejiang, China). Dextran sulfate sodium (DSS, molecular weight 36-50 kDa) was purchased from MP Biologicals (US). Lipopolysaccharide (LPS) was purchased from Sigma-Aldrich (US). Fetal bovine serum (FBS), penicillin, and alpha-MEM medium were purchased from Life Technologies. Griess reagent (modi ed) from Sigma was used. Rat anti-mouse F4/80 was purchased from BIO-RAD (US). Goat Anti-Rat IgG H L was purchased from Abcam (US).

Animals
Male C57BL/6 mice, 6-10-weeks-old, were provided by the Laboratory Animal Research Centre of the Chinese University of Hong Kong with animal experimental ethical approvals. Animal experiment was carried out under the animal license issued by the Hong Kong SAR government and the approval of the institutional animal care and use committee (AEEC: 20-191-MIS). The efforts were made to minimize the suffering of the animals.

Ulcerative colitis (UC) mouse model and treatment of SM
Acute colitis was induced in C57BL/6 mice by oral administration of 3.0% DSS in drinking water for 7 days. They were randomly assigned to 3 groups: (a) Control group (n=4), mice received a drinking water only. (b) DSS group (n=4), mice were given 3.0% DSS in drinking water for 7 days. (c) SM group (n=4), mice were given 3.0% DSS in water and SM (5ml/kg) gavage feeding once daily for 7 days. Mice were terminated at the end of the experiment, the colons (from cecum to anus) and spleens were harvested. The length of the colons was measured using a ruler, and the samples were xed in 4% formaldehyde for 24 hours, and prepared for para n embedding, sectioning and histological examinations.

Evaluation of disease activity index (DAI).
DAI was used for evaluation of the severity of colitis [28]. The scores were determined by combining scores of bleeding, diarrhea, and body weight loss every two days.

Immuno uorescent Analysis
To assess the extent of in ltration of mucosal macrophages by immuno uorescence, the para nembedded colon tissue sections were labeled with F4/80 antibody. First, the sections were depara nized.
After exposing the antigens, the colon sections were incubated with 1:50 primary antibodies overnight at 4℃, then co-incubated with 1:1000 secondary antibodies for 1 hour at room temperature. Finally, sections were sealed with DAPI and observed under a microscope at magni cation, x20, x40.
2.7 RAW264.7 cell culture and nitric oxidase (NO) production assay RAW264.7 cells (Beyotime, China) were grown in alpha-MEM containing 10% FBS and 100 U/ml penicillin. Cells were incubated at 37℃ in a humidi ed incubator with 5% CO2. These RAW264.7 cells are often used as an in vitro model of macrophage activation. Raw 264.7 cells (4 × 10 4 /well) were cultured in 96-well plates. Cells were pretreated in the presence or absence of SM (0.2%) in for 1h and then stimulated with LPS (10μg/ml) for 24 h. The NO levels in the culture media were evaluated by using modi ed Griess reagent to measure the nitrite accumulation. The culture supernatant and Griess reagent were mixed with equal volume and incubated for 10 min. The absorbance at 540 nm was measured with a microplate reader. Sodium nitrite (NaNO2) was used to generate a standard curve for quanti cation.

3.Statistical Analysis
All values were presented as means ± SD. Statistical signi cance was assessed by Student's t-test or oneway ANOVA using Prism GraphPad 5.0 (San Diego, CA, USA). Probabilities (p) <0.05 were considered signi cant.

SM improves the symptoms of DSS-induced colitis in mice.
Acute colitis is commonly induced by administration of 2-5% DSS for 4-9days [30,31]. Our mouse colitis model was induced by 3.0% DSS in drinking water for 7 days, and the treatment group received daily SM via gavage (Fig.1A). Bodyweight presented as the percentage of original body weight, diarrhea and bleeding were assessed every two days. The results showed that SM treatment group attenuated the body weight loss and signi cantly reduced the disease activity index (DAI), compared to the DSS control group (Fig. 1B-C). In the DSS only group, the colonic length of mice was markedly shortened than that of the SM treatment group (Fig.1 D-E). The control group (normal) mice had an average colon length of 8.38 ± 0.76 cm, while the DSS treatment led to a signi cant reduction of colon length to 4.9 ± 0.84cm. SM treatment reduced the extent of colon shortening to6.35 ± 0.39 cm). It has been reported that DSS induced splenomegaly was associated with increased splenic macrophages in ltration [32]. The DSS only group had signi cantly increased the size and weight of the spleen comparing to other groups (Fig.1 F-G).

SM treatment attenuated the histological changes of DSS-induced colitis in mice.
We next evaluated the severity of colonic in ammation by histopathological analysis using H&E staining. The higher histological scores indicate more severer damages. Damage of crypts, loss of goblet cells, in ltration of mononuclear cells, and formation of serious ulcers in the colons were examed. Comparing to the SM treatment group, the DSS group had signi cantly increased the in ltration of in ammatory cells, signs of epithelial loss and crypt destruction ( Fig. 2A) with higher histological scores (Fig. 2B).

SM treatment suppressed in ltration of macrophages in DSS-induced colitis in mice
Immuno uorescence analysis of F4/80, a marker of macrophages, was employed to evaluate the in ammatory phenotype of SM treatment in DSS-induced colitis mice. An increased number of F4/80 in ammatory cells was observed in the DSS only mice whereas SM treatment signi cantly reduced the number of macrophages in mucosa tissues (Fig. 3A-B).

SM inhibited NO production and downregulated proin ammatory cytokine expression in LPSstimulated RAW 264.7 cells
A hallmark of macrophage activation is the production of NO in response to LPS [33]. Therefore, we investigate the effect of SM on the production of NO in LPS-stimulated RAW 264.7 cells. Data showed that LPS induced a signi cant increase in NO production while SM treatment suppressed NO production in LPS-activated RAW 264.7 cells (Fig. 4A). We further tested the effects of SM treatment on some proin ammatory factors expression (IL-1β, IL-6 and IL-10) by q-PCR. Following SM treatment, the mRNA expressions of IL-1β, IL-6 and IL-10 in LPS-stimulated RAW267.4 cells were signi cantly reduced comparing to the control groups ( Figures 4B-D).

Discussion
This study evaluated the effects of SM on DSS-induced colitis model in mice. This model has been used to mimic many manifestations of human UC, including mucosal ulcers and systemic symptoms such as colon shortening, diarrhea, bloody stools, splenomegaly [34,35]. DSS damages intestinal epithelial cells and tight junction barriers and ultimately increases intestinal permeability to induce acute colitis. The current study found that SM treat mental leviated clinical symptoms of IBD in mice, including maintaining colon length, body weight, and reduction of in ammatory factors production.
Increased platelets, oxygen free radicals and in ammatory cells are associated with occurrence of UC or IBD [36][37][38]. The effect of SM in promoting microcirculation is widely recognized in the cardiovascular eld, but its anti-in ammatory effect is underestimated. Its role in gastrointestinal in ammation is largely unexplored, especially for the treatment of UC or IBD [15]. Cao et al. found that Crytotanshinone (CTN), one of the main constituents of SM, has anti-in ammatory effects. CTN inhibits prostaglandin E2 production and COX-2 expression via suppression of TLR4/NF-κB signaling pathway in LPS-stimulated Caco-2 cells [39]. Jin et al. also proved that Rosmarinic acid (RA), a phenolic ester also found in SM, resulted in the reduction of the in ammatory-related cytokines, such as IL-6, IL-1β, and IL-22, and protein levels of COX-2 and iNOS in mice with DSS-induced colitis [42]. Furthermore, RA effectively and pleiotropically inhibits the activation of NF-κB and STAT3, and subsequently reduces the activity of prosurvival genes that depend on these transcription factors [40]. These previous published results support the current ndings that SM does possess anti-in ammation potential in IBD conditions. The gut microbiota of IBD patients is imbalanced [41]. In recent years, probiotics are found to be one important therapeutic target for maintaining the normal immune response and the integrity of the intestinal mucosal [42,43]. SM has been reported to increase the number of intestinal probiotics and reverse the intestinal disordered microbiota. SM extracts signi cantly regulated Peptococcus, Peptostreptococcaceae and Ruminococcus and reduced in ammation in guts [44]. Dihydrotanshinone I (DHTS), a liposoluble extract of SM, e ciently restored disordered fecal microbiota community and increased the abundance of g_Akkermansia in chemotherapy-induced intestinal mucositis mice. g_Akkermansia, a genus in the phylum Verrucomicrobia, a potential probiotic, was reported to use their mucous to produce acetate and propionate. DHTS also enriched bacterial species which promote butyric acid metabolism or negatively correlated with in ammatory factors [45]. Whether SM treatment could stabilize gut microbiota via inhibiting intestinal in ammation, is the subject for future investigation, to explore the impact of SM treatment on gut microbiota diversity and severity of intestinal in ammation.
SM extract (SME) has been reported to halt the disease progression in DSS-induced colitis mice, in that 25 and 50 mg/kg SME intraperitoneal injection inhibited colonic shortening in a dose-related manner [46]. In the current study, we gave SM orally which is in medical grade injection from at a dose of 5ml/kg, and we have obtained comparable results in slowing down disease progression as previous studies [46] with much low dose SM. We also analyzed the spleen/body weight ratio and in ammatory factors level in plasma. When tanhinone IIA, a main component of SM, was given by intraperitoneal injection at a dose 20 mg/kg) [47], there was no statistical difference in body weight, colon length and plasma IL-6 levels in the DSS-induced colitis mice [28], while the above indices were signi cantly reduced in our study. The level of iNOS and IL-1β in the SM treatment group was also lower than that of tanhinone IIA treatment group [28] or cryptotanshinone treatment group (50 mg/kg, oral administration) [48] , indicating that SM is a more potent anti-in ammation modulator. Histological results further con rmed that SM treatment protects the intestinal mucosa and structure as well as reduced in ammatory cells in ltration. Our study also suggests that oral administration of SM may be more effective than intraperitoneal injection for the management of UC/IBD. These results taken together explain partially the underlying mechanisms why SM could alleviate symptoms of UC (IBD).
The current study reported a potential new application of a widely used herbal drug SM for the management of UC/IBD through oral administration. SM may be used clinically as a supplementary drug for the treatment of Crohn's disease or colorectal cancer. In future studies, different doses of SM combined with glucocorticoids or other biological reagents for the management of IBD shall be investigated through animal studies and clinical trials.

48.
Min, X., et al., Cryptotanshinone protects dextran sulfate sodium-induced experimental ulcerative colitis in mice by inhibiting intestinal in ammation. Phytother Res, 2020. Figure 1 SM treatment ameliorates the symptoms of DSS induced colitis. (A) Experimental schematic. Body weights (B), disease activity indexes on diarrhea, bleeding, and body weight loss (C) of all mice in groups were assessed on days 0, 2, 4, 6. All mice were terminated on day 7, colons and spleens were collected to measure the colon length and spleen weight (D-G). The values were presented as the means ± SE. *p<0.05, **p <0.01, ***p< 0.001, n = 4.