Tripterygium glycoside fraction n2: Alleviation of DSS-induced colitis by modulating immune homeostasis in mice

Abstract

Background: The Tripterygium glycosides (TG) is the main active extractive of Tripterygium wilfordii Hook F and is widely used in clinical practice to treat inflammatory diseases (including inflammatory bowel disease). However, due to its severe toxicity, TG is restricted to the treatment of many diseases. Therefore, it is necessary to study a new method to obtain the attenuated and synergistic extracts from TG.

Purpose: Tripterygium glycosides-n2 (TG-n2) was obtained from TG by a new preparation method. In this study, we aimed to investigate the difference in the chemical compositions between TG and TG-n2, further explored its toxicity and therapeutic effects on DSS-induced colitis in mice.

Methods: The major chemical compositions of TG and TG-n2 were analyzed by ultra-performance liquid chro- matography (UPLC). Subsequently, acute toxicity test was applied to evaluate the toxicity difference between TG and TG-n2. Dextran sulfate sodium (DSS)-induced acute colitis model was used to explore the therapeutic effect of TG and TG-n2 and their potential mechanisms of action.

Results: We found that the chemical compositions of TG-n2 is different from TG. The main difference is the ratio of triptriolide (T11) / triptolide (T9). Acute toxicity test proved that TG-n2 was less toxic than TG. Base on this, further studies showed that TG-n2 has a similar therapeutic effect as compared to TG on attenuating the symptoms of colitis, such as diarrhea, bloody stools, body weight loss, colonic atrophy, histopathological changes, inhibiting cytokines secretion and reducing absolute lymph number. In addition, TG and TG-n2 can increase the apoptosis of T lymphocyte in vivo. Further investigated showed that TG and TG-n2 could increase the expressions of Bax and p62 on CD3-positive T cells.

Conclusion: This study showed that oral administration of TG-n2 is safer than TG. Moreover, the attenuated TG- n2 has the similar therapeutic effect on treating experimental colitis in mice when compared to TG. Its me- chanism may be related to activating the expression of Bax in T cells and inducing T cells autophagy to regulate the survival of T lymphocytes in colitis mice, thus reducing inflammation in colon.

Introduction

Inflammatory bowel disease (IBD) is a chronic, recurring and ag- nogenic intestinal diseases, which includes two main subtypes: Crohn’s disease (CD) and Ulcerative colitis (UC). The clinical manifestations of IBD patients are characterized by abdominal pain, colonic mucosal ul- ceration, diarrhea with mucus and blood as well as body weight loss (Carroccio et al., 2003). Currently, the clinical therapies for IBD, such as anti-inflammatory drugs, immunosuppressants and TNF blockers have achieved some success but have obvious side effects (Qiao et al., 2017). Therefore, it is necessary to explore novel and effective ther- apeutic methods or drugs for the treatment of IBD.

Although the pathogenesis of IBD is still unclear, recent studies have shown that the abnormal activation of T lymphocytes is closely related to the occurrence and development of IBD (Matsuura et al., 1993). As the key mediator, dysregulated T lymphocytes can cause chronic in- flammation in IBD patients or IBD model (Siegmund et al., 2004). In IBD patients or DSS-induced colitis model, activated CD4+ T cell sub- sets, including Th1, Th2, Th17 and Treg cells, can secrete large amounts of cytokines (such as IL-1β, IL-6 and TNF-α) in intestinal mucosa tissues (Sethi et al., 2013). These could further aggravate the intestinal im- mune inflammatory response, and induce the ulceration and necrosis of colon (Liang et al., 2017). In addition, the auto-reactive CD8+T cells can also cause the damage to colonic mucosal epithelium tissue and increase the exposure of intestinal antigen, thereby expanding the CD4+ T cells subsets and aggravating intestinal mucosal inflammation (Smids et al., 2017). Therefore, some medicines which can inhibit the proliferation of T lymphocytes or induce the apoptosis of T lymphocytes may provide an effective therapeutic strategy for IBD treatment.

Tripterygium wilfordii Hook F (TWHF), a traditional Chinese medi- cine, has been widely used to treat kinds of diseases including rheu- matoid arthritis (RA) and autoimmune diseases (Zhuang et al., 2013). Accumulated researches have reported that TWHF possesses multiple pharmacological activities, such as anti-inflammation, immunoregula- tion and anti-tumor effects (Kuchta et al., 2017). The Tripterygium glycosides (TG) is an active extract of TWHF, and is widely applied in clinical practice to treat inflammatory diseases and autoimmune dis- eases (Xue et al., 2010). Particularly, recent studies have shown that TG could effectively improve colitis induced by DSS (Han et al., 2010). However, due to its severe toxicity to the liver, the clinical application of TWHF in the treatment of IBD is restricted (Wang et al., 2016). Thus, it is necessary to study a new method to obtain the attenuated and synergistic extracts from TG. A novel TG-n2 was obtained from TG, but its toxicity is much lower than of TG. However, it is still unclear whe- ther TG-n2 has similar pharmacological activity to TG in the treatment of IBD.

Therefore, in the present study, we first to analyze the difference between TG and TG-n2 on major chemical constituents. Based on this, we further investigated the differences in the therapeutic effect and underlying mechanism between TG and TG-n2 on DSS-induced acute colitis mice. Finally, we provided a new insight into the application of TG for clinical treatment of IBD.

Material and methods

Chemicals and reagents

TG was kindly provided by Hunan Qianjin synergy Pharmaceutical Co., Ltd. Dextran sodium sulfate (DSS, MW; 36–50 kDa) was purchased from MP Biomedicals (MP Biomedicals, Solon, OH, USA). IL-1β, IL-6, IL-10 and TNF-α ELISA kits were obtained from eBioscience (eBioscience, San Diego, CA, USA). Anti-bodies against Ly6G were purchased from Santa (Santa, USA). Anti-bodies against CD3 were purchased from GeneTex (San Antonio, USA). Anti-bodies against Bax were purchased from Abcam (Cambridge, MA, USA). Anti-bodies against p62 were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-mouse CD3 (APC) anti-body, anti-mouse CD4 (PE/Cy7) anti-body, and anti-mouse CD8 (PE/Cy7) anti-body were gained from eBioscience (eBioscience, San Diego, CA, USA), RPMI 1640 basic Medium and phosphate buffer saline (PBS) were purchased from GIBCO Laboratories (Grand Island, NY, USA). Apoptosis Assay Kit was ob- tained from Lianke Biology Inc (Hangzhou, China).

Animals

7–8 weeks old Male BALB/c mice (20 ± 2 g) were purchased from the Laboratory Animal Services Center, Guangzhou University of Chinese Medicine (Guangzhou, China). Mice were raised under specific pathogen-free conditions where they received standard food and ster- ilized water ad libitum on a constant temperature (20–25 °C) and hu- midity (65%–70%) with a 12 h light/dark cycle. All mice studies were performed according to institutional and National Institutes of Health (NIH) guidelines for humane animal use. Experimental protocols were approved by the Animal Ethics Committee of Guangzhou University of Chinese Medicine.

Preparation of TG-n2

The preparation method of TG-n2 was performed as described by Musser (Musser, 2000), with modification. Briefly, TG was suspended in phosphate buffer (pH 4.0) at room temperature and heated to reflux in the oil bath at 120 °C for 30 h, then added triptriolide (T11, purity >98% obtained from Bo Liu’s laboratory (Yang et al., 2018)) at the end. After the solvent was removed by rotary evaporation, the residual solid was TG-n2, and stored at −20 °C.

Chemical components analysis of TG and TG-n2

The fingerprint of TG and TG-n2 were detected according to the following condition: the sample was diluted with ethanol and purified by neutral alumina column chromatography, then evaporated solvent (≤60 °C), finally resolved in methanol and passed through the 0.22 μm membrane filter, and analyzed by UPLC system (Waters Acquity) with BEH Shield RP18 column (2.1 × 100 mm, 1.7 μm) at 35 °C. Acetonitrile and Q-water were used as mobile phases, flow rate was 0.3 ml/min, detection wavelength was 220 nm, injection volume: 3 μl, the UPLC gradient was showed in Table 1.

Acute toxicity test

Acute toxicity test was conducted according to the OECD guideline (Li et al., 2014). Briefly, BALB/c mice were randomly divided into three groups, each group contains 10 mice (5 males and 5 females). Mice were given TG-n2 orally at graded doses of 760, 920, 1160, 1400, 1800 and 2400 mg/kg. TG was administered to mice at graded doses of 280, 360, 480, 600, 720 and 860 mg/kg. Each animal was continuously observed for 4 h after single-dose administration. All mice were ob- served once a day and this process maintained for 14 days. The mor- tality and toxic symptoms (such as bradykinesia, convulsion and palsy) were recorded daily.

Establishment of experimental colitis

Acute ulcerative colitis was induced by drinking 4% dextran sodium sulfate (DSS, M.W: 36,000–50,000, dissolved in sterile water) freely for 7 days. 60 mice were randomly assigned to five groups (n = 12 for each group): control group, DSS group, the other three groups were orally
treated with TG (10 mg/kg/day), TG-n2 (10 mg/kg/day) and 5-ami- nosalicylic acid (5-ASA, 100 mg/kg/day), respectively. The control and DSS groups were given sterile water (0.1 mL/10 g, p.o.) once a day; the former only received distilled water, but the latter was given 4% DSS randomly instead of distilled water. Meanwhile, the treatment groups were intragastrically administered with TG (10 mg/kg/day), TG-n2 (10 mg/kg/day) and 5-ASA (100 mg/kg/day) once daily, respectively, and allowed free access to 4% DSS (Fig. 2A). The experiment lasted for 7 days, and the behavior of animals, body weight, food intake and water intake, fecal appearance and fecal occult blood, disease activity index (DAI) and mortality were recorded daily. After 7 days induced by DSS, the mice were killed humanely by pentobarbital sodium on the 8th day.

Histological scoring

The colon tissue was separated and cut into cantlet, then fixed in 4% paraformaldehyde for histological examination. In brief, fixed colon tissues were dehydrated using serial dilutions of alcohol, embedded in paraffin, then cut into 4 μm slices. Sections were hydrated and then stained with hematoxylin and eosin (H&E) followed by microscopic observation. The histological score was assessed according to the cri- teria previously reported by Liang (Liang et al., 2017).

General morphology scoring

The colon was removed, freed from surrounding tissues, rinsed and processed with stroke-physiological saline solution for macroscopic evaluation. General morphology scoring was determined according to the criterion (Liang et al., 2018) as follows: 0, normal. 1, local edema, no ulceration and tissue necrosis. 2, ulceration and tissue necrosis at one site. 3, ulceration and tissue necrosis at two sites, and the ulceration length >1 cm. 4, ulceration and tissue necrosis at more sites, and the ulceration length >2 cm.

Hematological analyze

Peripheral blood was collected from the orbit after the last admin- istration. The hematological indicators, such as white blood cell (WBC#), Monocyte (Mon#), Granulocyte (Gran#) and lymphocyte (Lymph#), were detected by the automated hematology analyzer.

Organ coefficient

At the end of the experiment, all animals were killed, spleen and thymus were separated quickly, photographed and weighed. The organ coefficient was calculated by formula (2).In the formula, Ws and Wb were stand for the organ weight and body weight, respectively.

Measurement of cytokines

The levels of inflammatory cytokines (including, IL-1β, IL-6, IL-10 and TNF-α) in colonic tissue were detected by enzyme linked im- munosorbent assay (ELISA) kits (eBioscience, USA) according to the manufacturer’ instructions. The data were analyzed by using the ELISA Calc software and quantified by standard curves.

Immunofluorescence

Immunofluorescence staining were performed according to the manufacturer’ instruction. Briefly, the sections (4 μm thickness) were deparaffinized and rehydrated, then antigen was repaired and per- meabilized with 0.5% Triton X-100 for 10 min, blocked with 5% goat serum for 30 min at 37 °C. The sections were incubated with primary antibodies in PBS at 4 °C overnight, after that the sections were rinsed with PBS and incubated with different fluorescent-labeled secondary antibodies for 2 h at RT and washed with PBS for 3 × 5 min. The sections were counterstained with DAPI for 5 min, then washed with PBS for 3 × 5 min. All sections were observed and analyzed by Laser Scanning Confocal Microscope (ZEISS, Germany).

Apoptosis analyze in vivo

The splenic lymphocytes were separated from colitis mice. The se- paration method was described previously by Liang (Liang et al., 2017). Splenic lymphocytes (1 × 106 cells/ml) were stained with surface markers, including CD3-APCeFlour780, CD4-PE/CY7 or CD8-PE/CY7 (eBioscience) in RPMI1640 medium for 20 min at room temperature in the dark. After that, the cells were centrifuged and washed twice with PBS, then suspended in binding buffer and stained by Annexin V/PI kit according to the instruction of manufacturer. Annexin V+/PI− splenic lymphocytes were considered as early apoptotic cells. Annexin V+/PI+ splenic lymphocytes were considered as late apoptotic cells. Apoptosis analysis was analyzed by FACS canto™ flow cytometer.

Statistical analysis

All data were expressed as mean ± S.E.M. Statistical analysis was performed using SPSS version 17.0 (Chicago, IL, USA). Group com- parisons were implemented by one-way analysis of variance (ANOVA) followed by post hoc Tukey’s test or Student’s t-test when appropriate. p < 0.05 was considered as statistically significance.

Results

Chemical components analysis of TG and TG-n2

We analyzed the fingerprints of TG and TG-n2 (Fig. 1A), and com- pared the differences between TG and TG-n2 (Fig. 1B). The greatest difference between TG and TG-n2 was the ratio of T11/T9. As shown in Fig. 1C and Table 2, the content of T9 in TG-n2 was 227.47 ± 1.98 µg/ g, and the content of T11 was 2450.01 ± 11.56 µg/g. While the content of T9 in TG was 442.50 ± 5.54 µg/g, and the content of T11 was 204.44 ± 3.21 µg/g. Therefore, the different ratio of T11/T9 controls the different properties of TG and TG-n2 as well as pharma- cological activities.

Comparison of acute toxicity between TG and TG-n2

With respect to the acute toxicity of TG and TG-n2, we found that TG-n2 did not cause any obvious toxic reaction and animals death at the highest dose of 2400 mg/kg (Supplementary Table S1). While oral administration of TG, mice deaths occurred at the lowest dose of 360 mg/kg, and some obvious toxic reactions such as bradykinesia, convulsion and palsy were also occurred after treatment with TG (Supplementary Table S1). These results showed that the toxicity of TG-n2 was lower than TG. However, as compared to TG, whether TG-n2 with reduced toxicity has the same protective effect on acute colitis was still unknown.

TG and TG-n2 could ameliorate the symptoms of acute UC mice

Previous studies have evidenced that acute colitis can be induced by drinking 4% DSS freely. Based on this, we further investigated whether attenuated TG-n2 has a similar therapeutic effect on colitis mice as compared to TG. In this study, we found that the body weight, food intake and water intake in colitis mice were decreased significantly as compared to the control group (Fig. 2A–E). The body weight of mice in DSS group mice began to decrease rapidly on day 4 and the animal’s condition deteriorated rapidly until the end of the experiment (Fig. 2C). Compared to DSS group, these signs were improved significantly after treatment with TG, TG-n2 and 5-ASA (Fig. 2A–E). TG, TG-n2 and 5-ASA began to significantly improve the condition of body weight loss on day 4, day 6 and day 6, respectively (Fig. 2A–C). Meanwhile, we also found that oral administration of TG, TG-n2 and 5-ASA could increase food intake and water intake in DSS-induced colitis mice (Fig. 2D and E).

DAI score contains body weight loss score, loose stools score and bloody stools score, which can reflect the severity of colitis symptoms in mice. After giving DSS, colitis mice have obvious loose stools and bloody stools. The DAI score in DSS group was also increased sig- nificantly when compared with the control group (Fig. 2F). However, treatment with TG, TG-n2 and 5-ASA could significantly decrease the DAI score and relieve symptoms of diarrhea and bloody stool (Fig. 2F), suggesting that TG-n2 has a similar therapeutic effect as TG on DSS- induced colitis.

TG and TG-n2 could alleviate colon injury in IBD mice model

Oral DSS can cause severe damage to the colon, which was char- acterized by colon shortening, edema and mucosal necrosis (Hundorfean et al., 2012). To further investigated the protective effect of TG-n2 on colon, colon length, histopathological examination and colonic morphology score were used to evaluate colonic changes in colitis mice. After giving DSS, the length of colon was significantly reduced (Fig. 3A and B) and the thickness of colon was markedly in- creased in colitis mice (Fig. 3C). However, a remarkable increase in colon length and a decrease in colon thickness were observed after
treatment with TG, TG-n2 and 5-ASA (Fig. 3A–C). In addition, DSS-induced colitis can cause obvious ulceration and necrosis in the colonic mucosal tissue, and the morphology score was also significantly in- creased as compared to the control group (Fig. 3D and E). TG, TG-n2 and 5-ASA administration groups were effective in alleviating mucosal ulceration and the morphology score of DSS-induced colitis mice was also significantly reduced (Fig. 3D and E).

Oral DSS could also cause damage to colonic structure, such as in- flammatory cells infiltration, loss of goblet cells, distortion of crypts and mucosal necrosis. Moreover, DSS caused damage to the intestinal mu- cosa of mice, leading to edema, ulceration, tissue necrosis (Fig. 3G). The histopathologic score was also significantly increased in DSS group (Fig. 3F). However, treatment with TG, TG-n2 and 5-ASA effectively alleviated DSS-induced colonic injury, restored the structure of mucosa and significantly reduced the infiltration of inflammatory cells in colon tissues (Fig. 3G). The histopathologic score was also significantly re- duced when compared with the DSS group (Fig. 3F). These results showed that attenuated TG-n2 has a similar effect on colon protection as compared to TG.

TG and TG-n2 have similar anti-inflammatory effect on DSS-induced colitis mice

DSS-induced colitis can lead to a large number of neutrophil in- filtrations in colon tissues, which could damage the colonic mucosal barrier (Deng et al., 2016). As shown in Fig. 4, we observed that many Ly6G+ cells infiltrated in the colonic mucosa tissues of DSS-induced colitis mice (Fig. 4A and B). While treatment with TG, TG-n2 and 5-ASA could significantly reduce the number of Ly6G+ cells (Fig. 4A and B).

The result showed that TG and TG-n2 could reduce the inflammatory infiltration in colon.

Excessive secretion of pro-inflammatory cytokines in the colon is another remarkable feature of DSS-induced colitis (Islam et al., 2010). To confirm whether TG-n2 and TG have the similar anti-inflammatory effect in DSS-induced colitis mice, the ELISA kits were used to test the inflammatory levels of colon tissue. Our results showed that the levels of pro-inflammatory cytokines IL-1β, IL-6 and TNF-α were significantly increased (Fig. 4C–E) and the level of anti-inflammatory cytokine IL-10 was decreased in DSS group (Fig. 4F). Surprisingly, TG, TG-n2 and 5-ASA treatment could significantly reduce the levels of above-mentioned inflammatory cytokines and increase the level of anti-inflammatory cytokine IL-10 (Fig. 4C–F). By comparing TG-n2 and TG, we found no significant difference in the anti-inflammatory effects between TG-n2 and TG.

The hematological analyses of TG and TG-n2 in DSS-induced colitis mice

Hematological analyses of DSS-induced acute colitis mice confirmed that there was a significant increase in the absolute numbers of WBC, Lymph#, Mon# and Gran# when compared with control group (Fig. 5A–D). As compared to DSS group, treatment with TG, TG-n2 and 5-ASA could significantly reduce the absolute numbers of WBC, Lymph#, Mon# and Gran# (Fig. 5A–D), and the absolute numbers of these cells was gradually restored to normal levels.

The effect of TG and TG-n2 on immune organs in DSS-induced colitis mice

DSS-induced acute colitis often causes splenomegaly and thymic atrophy, which are closely related to the immune dysfunction and in- flammatory imbalance in DSS-induced colitis mice (Liang et al., 2017). In this study, we also observed significant splenomegaly and thymic atrophy in DSS-treated mice (Fig. 6A and B). However, treatment with TG and TG-n2 could significantly inhibit splenomegaly and relieve thymic atrophy in colitis mice (Fig. 6A and B). 5-ASA treatment can reduce the spleen coefficient and alleviate thymic atrophy, but there was no significant difference as compared to DSS group. In addition, the correlation between absolute lymph number and spleen coefficient was analyzed by Pearson correlation. The results of Fig. 6C showed that the absolute lymph number was positively correlated with spleen coeffi- cient, indicating that the increase in absolute lymph number is closely related to splenomegaly.

TG and TG-n2 induced the apoptosis of peripheral blood and spleen T lymphocytes in DSS-induced colitis mice

In TG and TG-n2 administration groups, the significant decrease in the number of lymphocytes may be associated with the effect of TG and TG-n2 on promoting the apoptosis of lymphocytes. To analyze the apoptosis-inducing effects of TG and TG-n2 on peripheral blood (PB) and spleen T cells in colitis mice, we employed apoptosis kits, CD3, CD4 and CD8 anti-bodies to detect the apoptosis ratio of PB and spleen T cell subsets. We found that both the early and late apoptosis ratio of PB T cells (CD3+), helper T cells (CD3+CD4+) and cytotoxic T cells (CD3+CD8+) in DSS group were reduced, and the early apoptosis ratio of PB T cells (CD3+) in DSS group was significantly lower than that in control group (Fig. 7). Treatment with 5-ASA did not obviously increase the apoptosis of PB T cells in colitis mice as compared to DSS group. However, TG and TG-n2 could increase the apoptosis ratio of CD3+T cells, CD3+CD4+T cells and CD3+CD8+T cells in DSS-induced colitis.

Particularly, the early apoptosis ratio of CD3+T cells was significantly increased after treatment with TG and TG-n2 as compared to the DSS group (Fig. 7). We further investigated the effect of TG and TG-n2 on the apoptosis of splenic lymphocyte. The results showed that the early apoptosis of splenic CD3+ T cells and CD3+CD4+ T cells were significantly reduced in DSS group as compared to the control group (Fig. 8A–C), and the early apoptosis of CD3+CD8+ T cells was also decreased in DSS group but with no significant difference when compared with the control group (Fig. 8D). In addition, the late apoptosis of CD3+T cells and CD3+CD4+T cells were decreased in DSS-treated mice. 5-ASA treat- ment could significantly increase the late apoptosis of CD3+CD4+T cells, but has no significant effect on CD3+T cells and CD3+CD8+T cells in DSS-induced colitis mice. Interestingly, both the early and lately apoptosis of splenic CD3+T cells, helper T cells and cytotoxic T cells were significantly increased in TG group (Fig. 8), and the early apop- tosis of splenic CD3+T cells, helper T cells and cytotoxic T cells were also significantly increased in TG-n2 group (Fig. 8). These results showed that TG and TG-n2 could significantly induce the apoptosis of PB and spleen T lymphocytes in mice with acute colitis. The apoptosis- inducing action of TG-n2 on PB and spleen T lymphocytes is similar to TG in DSS-induced colitis mice.

Next, in order to preliminarily explore the effect of TG-n2 on inducing T cells apoptosis, we detected the expressions of CD3, Bax and p62 (autophagosome marker) in the spleen by immunofluorescence. We found that treatment with TG and TG-n2 could significantly increase the expressions of Bax and p62 on CD3-positive T cells (Figs. 9 and 10). We also found that the accumulation of CD3-positive T cells was ob- viously decreased in TG and TG-n2 treatment groups (Figs. 9 and 10), indicating that the apoptosis-promoting effect of TG-n2 may be asso- ciated with activating the expression of Bax and inducing T cell au- tophagy.

Discussion

IBD is a chronic, non-infectious and recurrent intestinal disease af- fecting millions of people around the world. At present, many drugs have been used to treat IBD by inhibiting macrophage-mediated in- flammation (Biancheri et al., 2015). However, long-term use of these drugs can cause serious side effects (Massironi et al., 2015). Therefore, there is an urgent need for safe and effective medicines. Accumulated evidences have shown that TG can obviously relieve colitis and has a significant treatment effect on IBD mice (Han et al., 2010). However, TG also has obvious side effects, such as hepatotoxicity, which limits its application in intestinal diseases (Wang et al., 2015). In the present study, we used a novel method to reduce the toxicity of TG, the atte- nuated extract was named as TG-n2. We first investigated the difference on toxicity between TG and TG-n2 in normal mice and further eval- uated whether TG-n2 has the similar protective effect on colitis mice as compared to TG.

The clinical results showed that TG can cause obvious toxic symp- toms, and higher dose of TG can lead to death (Wang et al., 2016). However, TG-n2 was obtained from TG by a novel method, which has significantly reduced toxicity. Based on this, we further studied the difference of pharmacological effects between TG and TG-n2 on colitis mice.

DSS-induced colitis model was used in this study, which has the similar pathological lesions in the colon of IBD patients (Horino et al., 2008). Our results revealed that TG and TG-n2 could markedly inhibit body weight loss, increase food consumption and water intake in colitis mice. DAI is calculated by body weight loss scores, loose stool scores and blood stool scores, which could reflect the severity of enteritis (Liang et al., 2017). As the results show, TG and TG-n2 treatment could significantly improve the condition of diarrhea, blood stools, and the DAI score was significantly decreased after administration with TG and TG-n2 in colitis mice. These results indicated that detoxified TG-n2 could exert similar protective effect on improving the symptom of colitis mice as compared to TG.

Colon atrophy is an important characteristic to reflect the severity of intestinal injury (Hundorfean et al., 2012). It has been reported that TG has a therapeutic effect on colon (Han et al., 2010). Interesting, treat- ment with TG-n2 could also effectively inhibit colon shortening. In addition, the colon thickness was obviously decreased after TG-n2 treatment. Histopathologic change is an indispensable indicator to evaluate the protective effect of drugs. DSS can induce colonic struc- tural damage, including ulcerations, necrosis and infiltration of neu- trophils. TG-n2 could prominently alleviate colonic lesion, such as goblet cell loss, crypt damage and neutrophils infiltration in colonic tissue, thereby reducing the histopathologic score of mice with colitis. Moreover, TG-n2 could significantly reduce the morphology score of colon tissue by relieving colonic ulceration and necrosis, suggesting that the protective effect of TG-n2 on colon is similar to that of TG.

Clinically, IBD is closely associated with immune-mediated in- flammatory responses (Sethi et al., 2013). Based on this, many im- munosuppressive drugs, such as azathioprine, cyclosporin and metho- trexate, can be used to treat IBD and show good results (Qiao et al., 2017). Similarly, TG has been reported to have a prominent therapeutic effect on colitis mice due to its immunosuppressive and anti-in- flammatory effect. We further investigated whether the detoxified TG- n2 owned similar immunosuppressive effect and anti-inflammatory ef- fect on colitis mice.

T lymphocytes are crucial for immune-inflammation homeostasis (Thompson-Chagoyán et al., 2005). Excessive immune inflammatory response could lead to massive proliferation of splenic T lymphocytes and further aggravate the body’s inflammatory response (Liang et al., 2017). Accumulated evidences have shown that over-expansion of ac- tivated T lymphocytes is closely related to the mechanism of IBD (Woodworth et al., 2010). Many studies about T lymphocytes has fo- cused on the effect of activated CD4+T effector lymphocytes and CD8+T cytotoxic lymphocytes on IBD inflammation response. CD4 subsets cells are involve in cytokines expression in the intestinal mucosa of IBD patients or animals model (Lutz et al., 2015). Meanwhile, CD8+T cells make the epithelial mucosal tissue damage, then attract and ex- pand CD4 lineage to aggravate inflammation (Cheroutre, 2006). In- hibiting T lymphocytes proliferation or increasing T lymphocytes apoptosis provide a successful therapeutic strategy to treat IBD (Smids et al., 2017). In the present study, TG and TG-n2 treatment significantly reduced the absolute number of lymphocytes, Monocyte and Granulocyte in colitis mice. More importantly, administration of DSS to mice produce a decrease in the apoptosis of T lymphocytes. However, treatment with TG and TG-n2 could significantly promote the apoptosis of CD3+CD4+T cells and CD3+CD8+T cells in spleen, as well as inhibit the splenomegaly in colitis mice. Bax is a pro-apoptotic protein which can increases cells apoptosis to reduce the number of cells (Lutz et al., 2015). In addition, autophagy also plays an important role in controlling cells death (Fan et al., 2018). Based on the above results, we further explored the possible mechanism by which TG-n2 regulates the survival of T lymphocytes in colitis mice. We found that TG and TG-n2 treatment could markedly increase the expressions of Bax and p62 on splenic CD3-positive T cells. And the expression of CD3- positive T cells was also significantly decreased after treatment with TG and TG-n2. Taken together, these results indicated that TG and TG-n2 could induce the apoptosis of T lymphocytes, thereby regulating the activation of large numbers of abnormal T lymphocytes in colitis mice, and the pro-apoptotic mechanism of TG-n2 on T cells may be associated with the activation of Bax expression in T cells and the induction of T cell autophagy.

In conclusion, TG-n2, obtained from TG by a novel technology, has a lower toxicity than TG (Commercial product). We found that TG-n2 owns a similar therapeutical effect as compared to TG on the treatment of IBD. Further research indicated that TG and TG-n2 could inhibit the secretion of pro-inflammatory cytokines in colonic tissues. The pro- tective mechanism of TG and TG-n2 on colitis mice largely depend on promoting the apoptosis of T lymphocytes and restoring the physiolo- gical homeostasis of lymphocytes in the body.