BM aspirates were obtained from the tibia and femur of male Wistar rats using 21-gauge needles containing Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 1% antibiotics and antimycotics. After filtering through a nylon mesh (50 µm) and washing with DMEM, the cells were centrifuged, re-suspended, and plated in a petri dish (100 mm) in DMEM containing 1% heat-inactivated fetal calf serum. Cells were incubated at 37 °C with 5% CO${}_2$ in a humidified chamber. After 72 h, the adherent cells were gently washed using phosphate-buffered saline (PBS) before adding fresh culture medium, and the medium was replaced every three days. At approximately 80–90% confluency, the cells were detached using trypsin (0.25%) containing Ethylene Diamine Tetraacetic Acid (EDTA) (1 mmol/L). The cells were subsequently titrated and passaged. The cells were used for experiments after the P3 generation, when the morphology of the BMMSCs remained consistent.

The appropriate concentration of CellTracker™ CM-Dil dye (1 µg/mL; C7000; Invitrogen Corp., Carlsbad, CA, USA) was established as suitable through preliminary experiments. Cells from the third passage were harvested and counted, and CM-Dil (3 µL) working solution was utilized to stain 1 $\times$ 10${}^6$ cells following the manufacturer’s instructions. The cells were incubated at 37 °C for 5 min, and were then incubated at 4 °C for an additional 15 min. After labeling, the cells were washed with PBS and resuspended in fresh medium.

Wistar rats of the male gender, aged 6 weeks (160–180 g) were provided by the Laboratory Animal Center of the Fifth Medical Center, Chinese PLA General Hospital (Certification number SCXK-JUN 2007-004). The room temperature was set at 20 $\pm$ 2 °C, and the humidity was 60–70%. A 12-h day–night cycle was maintained, and the rats had free access to standard diet and water. All animals were acclimated for 3 days prior to the experiments. All experimental procedures were conducted with the approval of the Ethics Committee of the Animal Facility of the Chinese PLA General Hospital and adhered to the established guidelines for the care of laboratory animals.

Six-week-old male Wistar rats were randomly divided into two groups: control group (n = 4) and model group (n = 22). Rats in the control group were fed standard rat chow and tap water ad libitum. In addition to standard rat chow, rats in the model group also had access to a fresh solution of MMNG (N-methyl-N${}^{\prime }$-nitro-N-nitrosoguanidine; 100 mg/L; MO527, TCI, Tokyo, Japan) ad libitum. The systemic influence of MNNG was initially explored by assessing appetite, body weight, and activity of each rat. When the model was established, three rats were anesthetized and killed every month for macroscopic and microscopic analyses of the MNNG-induced gastric lesions. Ten months after the initiation of the model, increased inflammatory cell infiltration, glandular epithelial atrophy, reduction of intrinsic glands, structural disorder of some glands, and irregular shapes were observed in the gastric mucosa of rats in the model group. Animals in the gastric lesion model group were subsequently randomly divided into two additional groups: the transplantation group (n = 11) and the nontransplantation group (n = 11). Rats in the transplantation group received transplantation of CM-dil labeled MSCs (1 mL) at a concentration of 3 $\times$ 10${}^6$ cells/mL via tail vein injection once per week for three consecutive weeks. Rats in the nontransplantation group received an equal volume of physiological saline through the tail vein.

Rats in the control and nontransplantation groups were killed one week after the last stem cell transplantation, and blood was collected from the heart chambers after intraperitoneal anesthesia with 10% chloral hydrate (0.3 mL/100 g). After the blood was centrifuged for 10 min at 3000 rpm, serum was collected and stored at –20 °C. In addition to blood, the entire glandular stomach was resected and incised rapidly along the greater curvature. For background staining and immunohistochemistry, the gastric tissue was fixed with 4% paraformaldehyde before paraffin-embedding and sequential slices were obtained. Lesions, inflammatory cell infiltration, atrophy, intestinal metaplasia, and dysplasia were assessed based on histological criteria from hematoxylin and eosin-stained gastric mucosa samples. The pathological changes were graded as mild, moderate, or severe based on a pathological grading system [18]. In the transplantation group, rats were anesthetized and killed for macroscopic and microscopic analyses after the last injection of labeled cells. Samples of the glandular stomach that were left after the previously described tissue collection were embedded in Tissue-Tek OCT compound (4583, Sakura Co., Ltd., Tokyo, Japan) and frozen in liquid nitrogen for subsequent cryostat sectioning and immunofluorescence staining.

Continuous frozen sections (7 µm) were permeabilized with Triton X-100 (1%) in PBS (pH 7.4) for 20 min, followed by incubation in blocking buffer containing goat serum (10%) for 30 min. For immunofluorescence staining, the sections were subsequently treated with anti-vimentin antibody (SC-6260, monoclonal mouse, 1:200 dilution, Santa Cruz Biotechnology) or pan-keratin antibody (4545S, monoclonal mouse, 1:300 dilution, Cell Signaling Technology, Danvers, MA, USA) overnight at 4 °C. Fluorescein isothiocyanate-conjugated goat anti-mouse IgG antibody (SC-2010, Santa Cruz Biotechnology, Dallas, TX, USA) was utilized as the secondary antibody (1:100; Jackson ImmunoResearch Laboratories, West Grove, PA, USA) and sections were incubated for 1 h at 37 °C. Nuclei were visualized using 4${}^{\prime }$6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich, St. Louis, MO, USA).

Immunohistochemistry was performed to detect the expression level of vascular endothelial growth factor (VEGF). Briefly, paraffin-embedded sections were deparaffinized and rehydrated. After blocking endogenous peroxidase activity with H${}_2$O${}_2$ (0.3%) for 30 min, the sections were subsequently incubated with rabbit anti-VEGF antibody (ab46154, 1:200 dilution; Abcam, Cambridge, UK) overnight at 4 °C. After washing with PBS (pH 7.4), the sections were incubated with anti-rabbit immunoglobulins (ZDR-5118, ZBGB-BIO, Zhongshan, China) for 30 min, and then allowed to react with streptavidin conjugated to horseradish peroxidase for 30 min. Antibody binding was visualized by incubation with 3.3${}^{\prime }$-diaminobenzidine chromogen (Dako, Carpinteria, CA, USA). Hematoxylin was used for nuclear counterstaining.

The presence of interleukin (IL)-17 (BMS8635FF; eBioscience, San Diego, CA, USA), IL-4 (51-9004113; BD Biosciences, Franklin Lakes, NJ, USA), IL-10 (51-9004112, BD Biosciences), and interferon-gamma (IFN-$\gamma$; 51-9004111, BD Biosciences) in the serum was detected by flow cytometry following the manufacturers’ instructions.

Two pathologists specializing in pathology for more than 5 years assessed gastric tissue lesions under light microscopy. The diagnostic criteria for histological changes in the gastric mucosa refer to the classification and diagnostic criteria of experimental gastric precancerous lesions. The pathological indicators were summarized as follows, encompassing the assessment of gastric mucosal injury, inflammation, atrophy, intestinal metaplasia, and atypical hyperplasia.

Gastric mucosal injury: The degree of gastric mucosal injury was categorized into four levels.

Mild: Superficial injury of gastric mucosal surface epithelial cells, scored 1 point; Moderate: Damage above the gastric pit, manifested as superficial erosion, scored 2 points; Severe: Damage below the gastric pit, manifested as deep erosion, scored 3 points; Ultra Severe: Full-thickness necrosis and desquamation of gastric mucosa, manifested as acute ulcer, scored 4 points.

Gastric mucositis was semi-quantitatively assessed. Five visual fields were observed for each gastric body and gastric antrum slice under low magnification. Based on the degree of inflammatory cell infiltration, it was classified into the following levels.

Grade 0: No inflammation, scored 0 points; Grade 1: Multiple chronic inflammatory cell infiltrates in the gastric mucosal epithelium or at the base of the proper gland, scored 1 point; Grade 2: More inflammatory cells infiltrating from the gastric mucosal epithelium to the mucosal muscularis, scored 2 points; Grade 3: Piles of inflammatory cell aggregates in the gastric mucosa, scored 3 points.

Wistar rats of the male gender, aged 6 weeks (160–180 g), and meeting the criteria of Specific Pathogen-Free (SPF) status were utilized in the study. These rats were obtained from and maintained by the Animal Laboratory Center of the Fifth Medical Center, Chinese PLA General Hospital, under license number SCXK (Military) 2007-004. All experimental procedures were conducted with the approval of the Ethics Committee of the Animal Facility of the Chinese PLA General Hospital and adhered to the established guidelines for the care of laboratory animals [19].

Five different fields of view were randomly selected from each immunohistochemical section under a high-power microscope (400$\times$). According to the observations, the cells with brownish-yellow particles in the cytoplasm were positive for VEGF. Image-pro plus 5.1 software (Media Cybernetics, Bethesda, MD, USA) was used to analyze the optical density and area of immunohistochemical positive signals in the selected visual field, and to calculate the average optical density value of VEGF (average optical density value = cumulative optical density/area). A higher average optical density value indicates a stronger expression.

All the data were expressed as the mean $\pm$ standard deviation and analyzed using SPSS 16.0 software (IBM Corp., Armonk, NY, USA). For making comparisons among multiple groups, the repeated measures analysis of variance (ANOVA) test was utilized. A p-value less than 0.05 was considered statistically significant.

In this study, labeled MSCs were injected into the tail vein of rats to assess healing in a model of gastric lesions. Pathological changes of the gastric mucosa were observed in the transplantation group (transplant) and compared to the non-transplantation group (model) and the untreated normal group (control). Both morphological and pathological changes were investigated. The percentages of gastric mucosal pathological changes observed in the three groups of rats are presented in Table 1.

In the model group, atrophic gastritis, intestinal metaplasia, and dysplasia were all observed along with inflammatory cell infiltration. In the transplant group, a significant reduction in gastric lesions was found, with atrophic gastritis being the sole identified lesion. This finding was confirmed through both morphological and histological assessments (Fig. 1).

Fig. 1.

Gastric precancerous lesions were less severe in mice treated with bone marrow-derived mesenchymal stem cells (BMMSCs) that were administered through the tail vein of rats. (A–C) representative images from morphological observations of the rat gastric mucosa in normal (A), gastric lesion model (B) and BMMSC transplant (C) groups. (D–F) Hematoxylin and eosin-stained images from frozen pathological section of rat gastric mucosa in normal (D), gastric lesion model (E), and BMMSC transplant (F) groups.

Macroscopically, the normal rat gastric mucosa exhibited good elasticity, a smooth surface, and regular, light-red folds (Fig. 1A). In contrast, the gastric mucosa in rats of the gastric lesion disease model appeared pale with diminished folds (Fig. 1B). The gastric mucosa in the transplant group showed improved morphological changes compared with the model group, and signs of mucosal repair, such as increased mucosal folds were observed (Fig. 1C). Microscopically, the normal rat mucosa exhibited integrated epithelium with regular cell size, well-organized gland cells, and a small number of scattered lymphocytes (Fig. 1D). The gastric lesion model group exhibited clear indications of mucosal atrophy, intestinal metaplasia, and dysplasia (Fig. 1E). Exfoliation of epithelial cells, reduced glands in the lamina propria, and glandular structure damage with noticeable inflammatory cell infiltration were characteristic pathological changes in this group. However, intestinal metaplasia and dysplasia were not found in the transplant group (Fig. 1F). Further investigation also revealed that MSC transplantation reduced the severity of gastric mucosal inflammation (Fig. 2). An intensified inflammatory response was detected in both the antrum and body of the gastric mucosa (Fig. 2).

Fig. 2.

Inflammation was reduced in the rat stomach during a gastric lesion model after the injection of bone marrow-derived mesenchymal stem cells (BMMSCs). The severity of inflammation within the gastric antrum (left) and gastric body (right) of the control, model, and BMMSC transplant groups. *p 0.05, **p 0.01, ***p 0.001.

Representative images from frozen gastric tissues following MSC transplantation are illustrated in Fig. 3. Cells with red fluorescence (CM-Dil-labeled cells) were MSCs and were observed at both the gastric body and antrum, while no cells with red fluorescence were found at the gastric fundus. Experiments on control rats were performed in parallel and no cells with red fluorescence were identified in any gastric tissues (Fig. 3). These results demonstrated that MSCs could colonize gastric lesions in rats at the gastric body and antrum.

Fig. 3.

Injection of bone marrow-derived mesenchymal stem cells (BMMSCs) could be colonized in gastric precancerous tissues. Representative images from frozen pathological sections of rat gastric body (A–D) and gastric antrum (E–H). (A,E) Hematoxylin and eosin (H&E)-stained image. (B,F) CM-Dil-labeled BMMSCs in precancerous tissues. (C,G) DAPI nuclear staining. (D,H) Merged images. Exposure time, ms; magnification rate, 400$\times$.

Using frozen pathological section, two antigens were labeled to identify epithelial and stromal cells, pan-keratin (Fig. 4A–D) and vimentin (Fig. 4E–H), respectively. The results revealed that some MSCs that colonized gastric tissues were differentiated into epithelial cells, while some MSCs retained the expression level of vimentin, an MSC marker.

Fig. 4.

Bone marrow-derived mesenchymal stem cells (BMMSCs) were colonized in gastric precancerous lesions and differentiated into epithelial cells and mesenchymal cells. Frozen pathological section showing that BMMSCs colonized in gastric precancerous tissues could differentiate into epithelial cells (A–D) and mesenchymal cells (E–H). (A,E) CM-Dil-labeled BMMSCs. (B,F) Pan-keratin (B) and vimentin (F). (C,G) DAPI nuclear staining. (D,H) Merged images.

The growth of blood vessels can facilitate the repair of lesions. A low expression level of VEGF was identified in normal gastric mucosa (Fig. 5A), and the expression level of VEGF was significantly upregulated in gastric mucosal cells within gastric lesions (Fig. 5B) compared with the normal group (p 0.001). The VEGF expression level in the MSC transplant group was significantly elevated compared with that in both the normal group (p 0.001) and the model group (p 0.01) (Fig. 5C,D). This observation suggested that angiogenesis is evident during the occurrence of gastric lesions, and MSC transplantation further enhances the level of angiogenesis in the injured tissue.

Fig. 5.

Bone marrow-derived mesenchymal stem cells (BMMSCs) promoted vascular endothelial growth factor (VEGF) expression in gastric precancerous tissue. (A–C) Immunohistochemical staining of frozen pathological sections from normal, model, and BMMSC transplant groups. (A) Low VEGF expression in control tissue. (B) Enhanced VEGF expression in model tissue. (C) Highest expression of VEGF in transplant group. (D) Optical density was compared among the three groups. n $\ge$ 4. **p 0.01; ***p 0.001.

Interleukins and interferons are important cytokines mediating the cellular immune response and are closely associated with carcinogenesis. In the present study, four cytokines potentially related to gastric carcinogenesis were selected and their levels were measured in the serum of rats, including IL-4, IL-10, IL-17, and IFN-$\gamma$ (Table 2). The results revealed that only serum levels of IL-17 and IFN-$\gamma$ in the model group significantly increased compared with those in the control group. In contrast, serum levels of IL-4 and IL-10 in the MSC transplant group were significantly higher than those in the control group (p 0.05). Additionally, serum levels of IL-17 and IFN-$\gamma$ in the MSC transplant group were significantly higher than those in the control group (p 0.01 for IL-17, p 0.001 for IFN-$\gamma$), while significantly lower than those in the model group (p 0.05 for IL-17, p 0.01 for IFN-$\gamma$) (Fig. 6). IL-4 and IL-10 may reflect the level of the humoral immune response, while IL-17 and IFN-$\gamma$ are more related to the cellular immune response and inflammation severity of gastric lesions.

Fig. 6.

Bone marrow-derived mesenchymal stem cell (BMMSC) transplantation regulated the serum levels of interleukins and interferon-gamma. Serum levels of IL-4 (A), IL-10 (B), IL-17 (C), and IFN-$\gamma$ (D) were compared among control, model, and BMMSC transplant groups. n $\ge$ 4. NS: no significance. *p 0.05, **p 0.01, ***p 0.001.