Bone marrow mesenchymal stem cells combined with Atractylodes macrocephala polysaccharide attenuate ulcerative colitis.

The aim of the present study was to explore the effects of bone marrow mesenchymal stem cells (BMSCs), combined with Atractylodes macrocephala polysaccharide (AMP), in an experimental model of ulcerative colitis. BMSCs were first isolated, cultured, and identified by flow cytometry. A rat model of colitis was established by trinitrobenzene sulfonic acid (TNBS) injection. Rats were treated with BMSCs with or without AMP for 1 or 2 weeks. H&E staining was performed to assess the extent of histological injury. IEC-6 and BMSCs were co-cultured and treated with AMP. Cell migration was measured using the Transwell assay, whilst the levels of cytokines in the rat blood samples were detected using ELISA. In addition, cytokine levels in the cell supernatant were measured by microarray. The results showed that BMSCs were successfully isolated. BMSCs treatment could markedly alleviate injury according to histological analysis and regulate inflammatory cytokine production in this rat model of TNBS-induced colitis, where a higher number of BMSCs was found in the intestinal tract, compared to the model. AMP not only potentiated the effects of BMSCs on preventing TNBS-induced colitis but also promoted BMSC homing to the injured tissue and regulated cytokines. Furthermore, BMSCs and AMP promoted the migration of IEC in vitro and influenced multiple genes. In conclusion, AMP treatment improved the therapeutic effects of BMSCs on ulcerative colitis, potentially providing a novel clinical treatment strategy for colitis.

Introduction

Inflammatory bowel disease (IBD), which includes ulcerative colitis (UC) and Crohn’s disease (CD), is a chronic remitting or progressive inflammatory condition of the gastrointestinal tract [1]. IBD typically involves a combination of genetic and environmental factors [1]. UC is clinically manifested by several pathophysiological characteristics, including weight loss, abdominal pain, bloody mucous diarrhea, rectal atrophy, and a relapsing-remitting course [2]. Currently available therapies for UC include aminosalicylates, corticosteroids, and immunosuppressants.

Mesenchymal stem cells (MSCs) are cells that can renew themselves and have the unique potential to differentiate into various cell types in the body [3]. In particular, MSCs have been reported to possess immunomodulatory properties, which provides a theoretical basis for the treatment of a number of diseases [3]. Several studies have demonstrated that transplantation with MSCs could alleviate IBD in both clinical trials and in animal models [4,5]. It has also been reported that, following intravenous injection, fluorescently marked MSCs tend to migrate toward the damaged intestinal mucosa and submucosa of rats with UC [6]. In addition, after transplanting MSCs into the intestinal tract, the MSCs either differentiate into intestinal epithelial cells to promote proliferation or repair of cells in this layer, or regulate the abnormal immune response to ameliorate the pathological intestinal immune condition [7,8]. Therefore, these previous findings suggest that MSCs may prove to be a promising therapeutic option for IBD.

However, MSCs transplantation alone has limited therapeutic effects, and a combined treatment occupies a research hotpot. Traditional Chinese medicine (TCM) has been found to be of therapeutic value in experimental colitis, and is therefore worth considering as an adjunct therapy for UC [9]. Atractylodes macrocephala (AM) is a perennial herb that has been cultivated for >700 years in temperate and subtropical regions. It has been used as a traditional Chinese medicine in East Asia for the treatment of osteoporosis, cancer, obesity and gastrointestinal dysfunction. This is primarily due to its reported pharmacological activities, including hepatoprotective, anti-inflammatory, antioxidant, antitumor, and antiviral properties [10-14]. AM has been shown to promote the intestinal healing process following intestinal injury in a number of intestinal disorders [15]. In addition, polysaccharides from AM (AMP) has been demonstrated to effectively alleviate colitis [16]. Thus, considering combined treatment, whether MSCs combined with AMP could improve the therapeutic effect on colitis raised our interest.

Therefore, the aim of the present study was to examine the effects of MSC transplantation on UC and to assess the combined effects of MSCs and AMP, which should hopefully provide a theoretical basis for elucidating the potential mechanism underlying the combined effects of MSC transplantation and TCM in UC treatment.

Materials and methods

Preparation of AMP

Crude polysaccharide isolated from AM was prepared using microwave-assisted extraction. In total, 5.0 g dried sample (40-mesh sieve) was accurately weighed, inserted into a 200-ml flask, and then extracted with 140 ml water for 3 min with the microwave power set at 700 W. The extracted supernatant was concentrated by vacuum distillation using a rotary evaporator and subsequently precipitated with 90% ethanol at 4°C for 12 h. The supernatant was finally lyophilized in a freeze-dry apparatus to obtain AMP.

Animals and experimental design

The male Sprague–Dawley (SD) rats, 6-weeks, weighing 160–180 g, were purchased from Jinan Pengyue Experimental Animal Breeding Co., Ltd. and placed under standard lighting conditions (12-h dark/light cycles), constant humidity, and temperature (22–24°C) with free access to water and food. After an adjustable feeding period of 1 week, the SD rats were randomly assigned into the following six groups (n = 10 rats per group): i) Control group; ii) model group; iii) salicylazosulfapyridine (SASP) group; iv) BMSCs for 1 week (BMSCs-1 W) group; v) BMSCs-2 W group; vi) BMSCs + AMP-1 W group; and vii) BMSCs + AMP-2 W group.

To induce colitis, rats were anesthetized with 4% isoflurane at a delivery rate of 1.0 l/min until loss of movement and 1.5–2% for maintenance. Following anesthesia, 50% ethanol solution containing 25 mg/ml trinitrobenzene sulfonic acid (TNBS; Sigma-Aldrich; Merck KGaA) was injected into the colon of the rat through the anus at a dose of 2 ml/kg using a lavage needle. The rats in the control group were injected with the same volume of normal saline (NS). Subsequently, 3 days later, all rats except for those in the control and model groups were injected with BMSCs (5 × 106 cells/500 μl; the BMSCs were isolated from rats described as follows) suspended in PBS via tail vein injection with or without intragastric administration of AMP (540 mg/kg converted from adult consumption) for 1 or 2 weeks. Rats in the control and model groups were injected with PBS. Rats were anesthetized with isoflurane (4% for induction and 1.5–2% for maintenance; Sumitomo Dainippon Pharma Co., Ltd.). The anesthetized rats were then cervically dislocated. After the heartbeat and respiratory arrest of the rats, the colon tissues and blood samples were collected for analysis. The animals were monitored every day. All the animal experiments were performed in compliance with the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and were approved by the Ethics Committee of Shandong University of Traditional Chinese Medicine (Approval No: SDUTCM202103190003; Jinan, China).

Histopathological analysis

Colon macroscopic damage index (CMDI) examination

The assessment of the CMDI was made based on the area of inflammation and the presence of ulcers according to a previously reported method [17].

CM-Dil staining

The colon tissue was fixed in 10% formalin and embedded in OCT. The frozen OCT-embedded tissues were then cut into three continuous sections (5-μm thick). The ulcerous and non-ulcerous parts of the sections were observed under the AiryScan module of an inverted LSM880 confocal laser scanning microscope (Carl Zeiss AG).

Hematoxylin & eosin (H&E) stain assessment

The tissue sections were fixed in 10% formalin, embedded in paraffin, and then cut into sections (5-μm thick), followed by staining with H&E. Histological microscopic assessment was performed in a blinded fashion by the same pathologist in accordance with Sykes criteria [18].

Enzyme-linked immunosorbent assay (ELISA)

The levels of interleukin (IL)-6, IL-10, IL-17A and transforming growth factor (TGF)-β in the rat blood samples were detected using the corresponding ELISA kits (cat. no. ml064292 for IL-6; cat. no. ml037371 for IL-10; cat. no. ml037365 for IL-17A; cat. no. ml002856 for TGF-β; Shanghai Enzyme-linked Biotechnology Co., Ltd., China) according to the manufacturer’s protocols.

Cultivation and identification of BMSCs

The anesthetized 4-week-old SD rats were sacrificed as aforementioned before their femurs were isolated and their marrow cavity was flushed with low-DMEM (Hyclone; Cytiva) containing 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) in a sterile environment. Cells were cultured at 37 °C in a humidified incubator with 5% CO2 without disturbance for 24 h and were subsequently washed to remove any non-adherent cells. The medium was changed every other day thereafter. After reaching 80–90% confluence, the adherent cells were detached using 0.25% (wt/v) trypsin/0.02% (wt/v) EDTA (Gibco; Thermo Fisher Scientific, Inc.). Passage 3 (P3) BMSCs were identified by flow cytometry analysis, where the phenotype was characterized with the use of anti-rat CD45 and CD90 (BioLegend, Inc.).

To better observe the BMSCs, cells were labeled by a cell tracker. Briefly, BMSCs were suspended in phosphate buffer saline (PBS) at 1 × 109/ml. Subsequently, 5 μl CM-Dil (1 g/l dissolved in dimethylsulfoxide (DMSO; Invitrogen; Thermo Fisher Scientific, Inc.)) was added to the BMSCs and incubated at 37 °C for 5 min in a humidified incubator and then at 4 °C for 15 min. After rinsing with PBS, BMSCs were observed under an Axio Image.A2 microscope (Carl Zeiss AG).

Cell culture

The small intestinal cell line IEC-6 was obtained from Shanghai Jikeiyin Chemical Technology Co., Ltd. and MSC was obtained from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. Both IEC-6 cells and MSCs were cultured in DMEM supplemented with 10% FBS at 37°C in a humidified incubator with 5% CO2.

Cell viability

IEC-6 cells were seeded into a 96-well plate and cultured for 24 h at 37°C in a humidified incubator with 5% CO2. AMP (400 μg/ml) was then added to the cells for 24 h at 37°C. IEC-6 cells were incubated with 10 μl of CCK-8 solution (Dojindo Molecular Technologies, Inc.) for another 3 h at 37°C. The absorbance at 450 nm was measured using a microplate reader.

Co-culture system of BMSCs and IEC-6 cells and assessment of cell migration ability

BMSCs were added to the upper chambers of the Transwell inserts and treated with AMP at 400 μg/ml, whereas IEC-6 cells were added to the lower chambers. After 24 h, a straight line was generated using a 20-μl pipette tip and ruler, before the medium was refreshed with serum-free media. The Transwell insert was cultured in the incubator for another 24 h. Images were captured at 0 and 24 h under an Axio Image.A2 microscope and cell migration was calculated.

Microarray detection

BMSCs and IEC-6 were co-cultured with IEC-6 cells in the upper chambers and BMSCs treated with 400 μg/ml AMP in the lower chamber. After co-culture for 48 h, the cell supernatant of the upper chamber was collected for cytokine analysis with microarray detection (H-Wayen Biotechnologies).

Statistical analysis

All data were analyzed using the SPSS 20.0 software (IBM Corp.) and expressed as the mean ± standard deviation. Experiments were repeated ≥ three times. One-way ANOVA followed by Tukey’s post-hoc test was used to analyze the differences among groups. P < 0.05 was considered to indicate statistically significant differences.

Results

Identification of BMSCs

At first, BMSCs were isolated from SD rats and were cultured for experiments. As shown in Figure 1(a), BMSCs were round and distributed individually, where the cells were gradually deformed and adhered to the wall. Cells could generally reach 80% confluence on days 4 and 5, by which time the cells were arranged in a uniform, spindle-shaped spiral manner. In addition, cells exhibited yellow circular suspension after labeling with CM-Dil for 30 min and were adherent to the wall after 24 h (Figure 1(a)). Flow cytometry assays showed that the P3 cells expressed CD90 (70.98% positive rate) and the expression rate of CD45 was 2.19% (Figure 1(b, c)). Therefore, this suggests that the isolation of BMSCs was successful.

Figure 1.

Identification of BMSCs. (a) BMSCs were isolated from SD rats and were cultured. Morphological changes in the BMSCs at different periods were observed. (b, c) Flow cytometry assay was performed to identify the phenotype of BMSCs. a, 0 h after culture; b, 24 h after culture; c, 72 h after culture; d, third-generation cells; e, cells labeled with CM-Dil for 30 min; f, cells labeled with CM-Dil for 24 h. BMSCs, bone marrow mesenchymal stem cells.

Effects of the BMSCs and AMP combination on colitis

To assess the therapeutic effect of BMSCs combined with AMP, the rats with colitis received BMSCs and AMP (540 mg/kg) for 1 or 2 weeks for treatment, as well as SASP as a positive drug. After the rats were sacrificed, the colon tissues of each group were collected. As shown in Figure 2(a), the surface of the intestinal wall was smooth with clear texture in the control group. However, the intestinal wall was thickened with erosive ulcer bleeding in the model group. In the treatment groups, this severe condition was observed to be attenuated, as the severity of the mucosal ulcer bleeding was reduced. CMDI assessment yielded the highest score in the model group, which was reduced after treatment with BMSCs (Figure 2(b)). In addition, higher reduction of CMDI was calculated following co-treatment with BMSCs and AMP, especially for 2 weeks (Figure 2(b)). Histological analysis was subsequently performed under a microscope after H&E staining. The mucosa and submucosa exhibited severe inflammatory ulceration in the model group, which were accompanied by infiltration of numerous inflammatory cells. However, this form of severe histological injury was also attenuated by treatment with BMSCs and co-treatment with AMP and BMSCs (Figure 2(c, d)). Furthermore, the levels of IL-6, IL-10, IL-17A and TGF-β in the rat blood samples were detected. As shown in Figure 2(e–h), colitis was associated with significantly increased levels of IL-6 and IL-17A, but significantly decreased levels of IL-10 and TGF-β, all of which were markedly reversed by co-treatment with BMSCs and AMP.

Figure 2.

Effects of BMSCs and AMP on colitis. (a) After sacrifice, the colon tissues were collected from the different groups and were observed. (b) Colon macroscopic damage index assessment was performed. (c) Colon tissues were stained with H&E and observed under a microscope. (d) Histological score. (e–h) IL-6, IL-17A, IL-10 and TGF-β levels in the rat blood samples of different groups were detected using ELISA. *P < 0.05, **P < 0.01 and ***P < 0.001 vs. control; #P < 0.05, ##P < 0.01 and ###P < 0.001 vs. model. a, control; b, model; c, SASP; d, BMSCs-1 W; e, BMSCs-2 w; f, BMSCs + AMP-1 W; g, BMSCs + AMP-2 W. BMSCs, bone marrow mesenchymal stem cells; AMP, Atractylodes macrocephala polysaccharide.

Distribution of CM-Dil-labeled BMSCs in vivo

Next, to investigate the distribution of BMSCs in vivo, CM-Dil-labeled BMSCs in the lung and intestinal tract were observed under an inverted confocal laser scanning microscope. As shown in Figure 3(a-b), the difference in the number of BMSCs among the different groups was not significant. Further analysis of the distribution of BMSCs in the intestinal tract was conducted, and the results showed that the CM-Dil-labeled BMSCs were mainly distributed in the mesenchymal layer of the colonic glands around the inflammatory foci, but not in the epithelium of the colonic mucosal surface or in the colonic glandular epithelium. Additionally, compared with that in BMSCs-1 W group, the number of surviving cells was increased in BMSCs + AMP-1 W group (Figure 2(c-d)).

Figure 3.

Distribution of CM-Dil-labeled BMSCs in vivo. (a) Distribution of CM-Dil-labeled BMSCs in the lung. (b) Number of CM-Dil-positive cells in the lung. (c) Distribution of CM-Dil-labeled BMSCs in the intestinal tract. (d) Number of CM-Dil-positive cells in the intestinal tract. ***P < 0.001 vs. BMSCs-1 W. BMSCs, bone marrow mesenchymal stem cells. BMSCs, bone marrow mesenchymal stem cells; AMP, Atractylodes macrocephala polysaccharide.

Effects of AMP on cell migration and cytokine levels in a co-culture system of BMSCs and IEC-6 cells

Next, we further explored the effects of BMSCs and AMP on the small intestinal cells to evaluate their direct function in the intestine. A co-culture system of BMSCs and IEC-6 cells was then established. The results from the CCK-8 assay showed that cell viability was higher in the co-culture system compared with that in IEC-6 cells when cultured alone, which was further potentiated, albeit slightly, upon the addition of AMP into this co-culture system with BMSCs and IEC-6 cells. However, the magnitude of these changes was small, and there was no significant difference among the different groups (Figure 4(a)).

Figure 4.

Effects of AMP on cell migration and cytokine levels in the co-culture system of BMSCs and IEC-6 cells. (a) Cell Counting Kit-8 assay was performed to detect cell viability. (b) Transwell assay was performed to measure IEC-6 cell migration. (c) Cellular supernatant was collected for cytokine analysis with microarray detection. BMSCs, bone marrow mesenchymal stem cells; AMP, Atractylodes macrocephala polysaccharide.

Data from the Transwell assay exhibited a markedly strengthened migratory ability in IEC-6 cells upon co-culturing with BMSCs compared with that in the group where IEC-6 cells were cultured alone (Figure 4(b)). This was increased further by additive treatment with AMP (Figure 4(b)). The cell culture supernatant was then collected for cytokine analysis with microarray detection. As shown in Figure 4(c), the levels of cytokines in the cell supernatant of different groups were measured by microarray detection. Compared with those in IEC-6 group, the levels of activin A, cytokine-induced neutrophil chemoattractant-2α/β, macrophage inflammatory protein (MIP)-2 and vascular endothelial growth factor were markedly increased in the IEC-6 + BMSCs group and were further increased in the IEC-6 + BMSCs + AMP group. By contrast, the level of MMP2 was the lowest in IEC-6 group and the highest in the IEC-6 + BMSCs + AMP group. In addition, the highest levels of some cytokines, including granulocyte macrophage-colony stimulating factor, IL-4 and IL-10, were exhibited in the IEC-6 + BMSCs group. However, the levels of other cytokines, such as MIP-1α, were found to be the lowest in the IEC-6 + BMSCs group, which requires further investigation.

Discussion

The potential use of stem cell therapy for IBD has attracted the attention of the research community. Kashyap et al. [19] previously reported that a patient with CD and non-Hodgkin lymphoma remained in clinical remission for 7 years after autologous hematopoietic stem cell transplantation. BMSCs can exert unique functions on immune regulation, tissue repair and regeneration. Stem cell transplantation can regulate immune system function and the intestinal microenvironment in patients with UC and repair ulceration by directional differentiation en route to restoring normal intestinal function. Therefore, in cases where conventional treatment methods fail to alleviate UC, BMSC transplantation appears to be a promising treatment strategy for UC. However, the potential therapeutic effects of BMSC transplantation are limited due to the low efficacy of cell homing to the injured tissues. Therefore, it remains crucial to improve the homing capability of BMSCs. Chen et al [20] reported that CXC chemokine receptor 4 (CXCR-4) promoted the migration of BMSCs in vitro and the homing of BMSCs to the damaged intestinal mucosa in vivo. In addition, CXCR-4 overexpression in BMSCs conferred therapeutic effects on colitis in vivo, suggesting that overexpressing the CXCR-4 gene in BMSCs may be a potential approach to improving the efficacy of BMSCs on colitis. Similarly, IL-35 and heparin have also been found to improve the therapeutic effects of BMSCs therapy by enhancing the distribution of BMSCs to the targeted injured organs [21,22]. In the present study, AMP not only improved cell viability and migration ability, but also promoted the distribution of BMSCs in the injured intestinal tract following colitis induction in vivo. In addition, BMSCs markedly alleviated histological injury and inflammatory infiltration caused by TNBS-induced colitis in vivo, with the therapeutic effects becoming more potent over time. The protective role of BMSCs against colitis was also found to be improved by AMP treatment. Therefore, these results further suggest the importance of the capability of BMSCs to home to the damaged organs, where AMP may also be a potential candidate for enhancing the therapeutic effects of BMSCs on colitis.

Bioactive polysaccharides have become a field of intense scientific research due to their various reported bioactive activities, including antitumor, anti-inflammatory, antioxidant, and antidiabetic properties [23]. In terms of colitis treatment, only a small number of polysaccharides extracted from traditional Chinese medicines have been reported. For example, the polysaccharide from Hericium erinaceus was found to attenuate dextran sulfate sodium-induced colitis by regulating oxidative stress, inflammation, and the gut microbiota profile [24]. In addition, rhamnogalacturonan, a chemically defined polysaccharide, could alleviate colitis by restoring intestinal barrier function in vivo and in vitro [25]. In another study, Ganoderma lucidum polysaccharides were found to ameliorate acute colitis through the suppression of immune responses [26]. Therefore, bioactive polysaccharides have a potential role in the prevention and treatment of colitis. The anti-colitis function of AM has been previously documented [10], and AMP was also found to promote intestinal epithelial cell migration, which facilitated intestinal injury healing [27]. To the best of our knowledge, the present study is the first to investigate the effects of AMP on BMSC-treated colitis. Translating the results from the present study into clinical application, AMP may be useful as a supplementary agent during maintenance therapy for preventing colitis relapse.

As the pharmacologic actions of BMSCs and AMP have been demonstrated as aforementioned, we further explored the effects of BMSCs and AMP on the small intestinal cells to evaluate their direct functions in intestine. Data obtained from microarray detection exhibited a great change of cytokines, such as Activin A, MIP-2, and VEGF, after BMSCs and AMP treatment. Activin A was shown to promote BMSCs differentiation, which was able to protect against dextran sulfate sodium-induced colitis; on the contrary, Activin A was reported to be elevated in patients with colitis, and was recognized as a novel inflammatory marker in colitis [28]. In addition, Activin A could inhibit cell proliferation of IEC-6 cells, and served as an important role in inflammatory response [29,30]. In the present study, with the addition of BMSCs and AMP, the production of Activin A was elevated, reflecting its close association with BMSCs, but its controversial role needs further precise detection. MIP-2 was also exhibited as an inflammatory marker in RAW264.7 macrophages [31]; however, MIP-2 was reported to play a complex role in the liver diseases progression by mediating liver inflammation at a high concentration and promoting liver regeneration at a low concentration. Here, we observed a change of the level of MIP-2 in cell supernatant of IEC-6 cells upon co-culturing with BMSCs, with or without treatment of AMP, which might partly owe to the cell homing property of BMSCs and the anti-inflammatory property of AMP. A previous study has pointed out that with the higher therapeutic effect on treating colitis by CXCR-4 gene overexpressed BMSCs, the expression level of VEGF was higher, reflecting a positive connection between VEGF and colitis treatment [20]. Consistently, BMSCs and AMP induced an increased expression level of VEGF, which has the potential to attenuate colitis.

However, there are some limitations in the present study. Firstly, we only explored the changed cytokines using microarray detection, but experiments on more precise data about these cytokines or an in vivo verification were needed to further ensure the involvement of these cytokines. Secondly, except cytokines secreted by BMSCs, exosomes, secreted by BMSCs, were previously reported to participate in the regulation of disease progression [32]. Thus, whether BMSC-derived exosomes are also involved in this therapeutic effects is deserved to be studied.

Conclusion

Collectively, the results of the present study show that BMSC treatment, alone or in combination with AMP, has the potential to prevent against TNBS-induced colitis. In the present study, AMP treatment was found to enhance the therapeutic effects of BMSCs on colitis by promoting BMSC homing to the injured tissue and regulating the cytokine profile.