Bone marrow‐derived mesenchymal stem cells promote Helicobacter pylori‐associated gastric cancer progression by secreting thrombospondin‐2

Abstract Objectives Bone marrow‐derived cells (BMDCs), especially mesenchymal stem cells (MSCs), may be involved in the development of Helicobacter pylori‐associated gastric cancer (GC) in mice, but the specific mechanism remains unclear, and evidence from human studies is lacking. Materials and Methods To verify the role of BM‐MSCs in H pylori‐associated GC, green fluorescent protein (GFP)‐labelled BM‐MSCs were transplanted into the subserosal layers of the stomach in a mouse model of chronic H pylori infection. Three months post‐transplantation, the mice were sacrificed, and the gastric tissues were subjected to histopathological and immunofluorescence analyses. In addition, we performed fluorescence in situ hybridization (FISH) and immunofluorescence analyses of gastric tissue from a female patient with H pylori infection and a history of acute myeloid leukaemia who received a BM transplant from a male donor. Results In mice with chronic H pylori infection, GFP‐labelled BM‐MSCs migrated from the serous layer to the mucosal layer and promoted GC progression. The BM‐MSCs differentiated into pan‐cytokeratin‐positive epithelial cells and α‐smooth muscle actin‐positive cancer‐associated fibroblasts (CAFs) by secreting the protein thrombospondin‐2. FISH analysis of gastric tissue from the female patient revealed Y‐chromosome‐positive cells. Immunofluorescence analyses further confirmed that Y‐chromosome‐positive cells showed positive BM‐MSCs marker. These results suggested that allogeneic BMDCs, including BM‐MSCs, can migrate to the stomach under chronic H pylori infection. Conclusions Taken together, these findings imply that BM‐MSCs participate in the development of chronic H pylori‐associated GC by differentiating into both gastric epithelial cells and CAFs.


| INTRODUC TI ON
Gastric cancer (GC) is one of the most common malignant tumours worldwide. According to the latest global cancer data released in 2020, GC ranks 6th in incidence and 4th in mortality among all malignancies. 1 Risk factors for GC include chronic Helicobacter pylori infection, alcohol consumption, 2,3 tobacco smoking 4,5 and consumption of foods preserved by salting, 6,7 among others. Chronic H pylori infection is considered the principal cause of non-cardia GC 1 ; approximately 90% of new cases of non-cardia GC are associated with H pylori infection. 8,9 Nearly half of the global population is infected with H pylori, making it the most common infection in the world. 10,11 Chronic H pylori infection is thought to cause gastric epithelial hyperplasia and mitotic error, leading to metaplasia, dysplasia and ultimately adenocarcinoma. 12 However, beyond these histological characteristics, the mechanisms by which H pylori-associated GC originates and progresses are not fully understood. 13 Previous studies suggested that GC is caused by malignant transformation of gastric mucosal epithelial cells, [14][15][16] but recent cell lineage tracing studies have proposed that bone marrow-derived cells (BMDCs) are the cell source of GC. 17,18 Bone marrow-derived mesenchymal stem cells (BM-MSCs) have multi-lineage differentiation potential and play an important role in tissue repair. 19 The tropism of BM-MSC for sites of tissue damage and the tumour microenvironment has been confirmed, 20,21 and evidence indicates that the tissue-regenerative function of MSCs may go awry in malignant tumours. [22][23][24][25][26] Nonetheless, the exact role of BM-MSCs in H pyloriassociated GC and the underlying mechanisms remain unclear.  [27][28][29] α-smooth muscle actin (α-SMA) is a robust CAF marker that is commonly used to identify CAFs with myofibroblast morphology. 30,31 CAFs can also be further induced to secrete cytokines that promote tumour cell growth and invasion. [32][33][34][35] Multiple lines of evidence suggest that a significant proportion of CAFs in tumours originates from BM-MSCs. 26 For instance, in a mouse model of pancreatic ductal adenocarcinoma, BM-MSCs are recruited into the tumour microenvironment, where they differentiate into CAFs and secrete VEGF to promote tumour progression. 36 BM-MSCs promote tumour growth by secreting interleukin-6 as CAFs in a murine ovarian carcinoma xenograft model. 37 Therefore, in this study, we aimed to investigate whether BM-MSCs can promote the progression of H pylori-associated GC as CAFs and the underlying mechanism. To do so, we transplanted BM-MSCs into a mouse model of chronic H pylori infection.

| Establishment of BM-MSCs transplantation in chronic H pylori-infected mice
Four-to six-week-old male BALB/C mice were purchased from Beijing HFK Bioscience Co., LTD., and raised in a specific pathogen-free (SPF) animal feeding room at the Animal Center of Tongji Medical College in a constant-temperature (21-25℃), constant-humidity (50%-60%) environment. The mice had free access to standard rodent diet and water before the experiment. All procedures were conducted strictly in accordance with the Guide for the Care and To establish the model of chronic H pylori infection, mice were inoculated orally with a suspension of H pylori strain SS1 (0.1 ml, 1-2 × 10 9 CFU/mL) thrice over a 5-day period using a mouse gavage needle. 38 The SHAM and BM-MSCs groups were mock-inoculated with H pylori liquid culture medium. Three months after successful infection with H pylori strain SS1, green fluorescent protein (GFP)labelled BM-MSCs (2 × 10 6 cells in 0.1 ml of PBS) were transplanted into the antrum area of the greater curvature into the subserosa.
Three months after transplantation, the mice were sacrificed, stomach samples were collected, and gastric tissues were subjected to histological analysis. The distribution of GFP-labelled BM-MSCs in the stomach was detected by flow cytometry. Laser confocal immunofluorescence microscopy was used to observe the migration and the co-expression of GFP and the CAF marker α-smooth muscle actin (α-SMA) or gastric epithelial cell marker pan-cytokeratin (pan-CK).

| In vivo tumorigenesis in nude mice
Male BALB/c nude mice (HFK BIOSCIENCE CO., LTD, Beijing, China) were bred in a licenced SPF laboratory at the Animal Center of Tongji Medical College. To assess the effect of BM-MSCs on tumour growth in vivo, 5-week-old nude mice (20 g body weight) were randomly allocated to the following groups (n = 6): 1) control mice injected with PBS (Control group); 2) mice injected with

MSCs on mice with chronic H pylori infection
To

MSCs on gastric cancer xenografts
To investigate the effect of THBS2-deficient BM-MSCs on tumour growth in vivo, mice were allocated to the following groups (n = 5): sh-THBS2-BM-MSCs (SGC +sh-THBS2-MSCs). The mice were sacrificed 2 weeks after injection, and the subcutaneous tumours were analysed by histological staining.   Table S1.

| BM-MSCs promote the development of H pylori-associated gastric cancer in vivo
To explore the role of BM-MSCs in GC, BM-MSCs were isolated ( Figure S1) and locally transplanted into the stomach in mice that had been infected with H pylori for 3 months. At 3 months post-transplantation, histological analysis showed that the incidence of high-grade gastric intraepithelial neoplasia (HGIN) and GC was significantly higher in the Hp +BM-MSCs group than in the Hp group ( Figure 1A; 33.3% vs. 0%, P <.01). In addition, the incidence of lowgrade gastric intraepithelial neoplasia (LGIN) was significantly higher in the Hp +BM-MSCs group ( Figure 1A; 66.7% vs. 5%, P <.001).
In the Hp group, the incidence of inflammation was 70%, the incidence of intestinal metaplasia was 25%, and the incidence of LGIN was 5%. BM-MSCs transplantation did not promote the incidence of GC in non-H pylori-infected mice ( Figure 1A;

| H pylori enhances the proliferation and migration of BM-MSCs in vitro
Next shown in Figure 2A) increased in culture in vitro (P <.01; Figure 2B).

The effect of H pylori on BM-MSC migration was evaluated by the
Transwell migration assay. The results showed that H pylori promoted the migration ability of BM-MSCs in vitro (P <.01; Figure 2C).

| BM-MSCs promote the progression and metastasis of gastric cancer in nude mice xenografts
The

| Depletion of the THBS2 gene reduces the tumour-promoting ability of BM-MSCs
Next, we explored the effects of THBS2 expression in BM-MSCs on GC progression. To this end, we transfected BM-MSCs with THBS2specific shRNA ( Figure 4A) and verified that THBS2 gene/protein expression was downregulated compared with the control group transfected with scrambled shRNA (P <.001; Figure 4B-C). After exposure to H pylori, THBS2 mRNA and protein expression were significantly upregulated in BM-MSCs (P <.001; Figure 4D) but not THBS2-depleted BM-MSCs (P >.05; Figure 4D).
In addition, our data showed that THBS2-depleted BM-MSCs had lower tumour-promoting ability in GC xenografts in nude mice compared with BM-MSCs without THBS2 knockdown ( Figure 4E-H). Similarly, in chronic H pylori-infected mice, the incidence of GC was significantly lower in the Hp +sh-THBS2-MSCs group than in the Hp +sh-NC-MSCs group (0% vs. 28%, P <.01; Figure 4I-J).

| Gastric cancer patients with high THBS2 expression have worse prognosis
Data from The Cancer Genome Atlas further confirmed that THBS2 gene expression is significantly higher in GC tissues than in normal tissues (P <.05, Figure 5J). Survival analysis showed that GC patients with high THBS2 expression had poorer prognosis and a lower survival rate [HR =1.55 (1.29-1.85), P = 1.4e-06; Figure 5K]. Previous studies have shown that BM-MSCs are recruited to various types of tumour tissues. 23 We found that transplanted Consequently, we examined the specific role and mechanism of BM-MSCs in H pylori-associated GC. We found that BM-MSCs differentiated into both epithelial cells and CAFs in mice with chronic H pylori infection, thus both supporting and explaining the conflicting results found in the previous literature. 17,18 We also explored the mechanism of BM-MSCs as CAFs in H pylori-associated GC. THBS2 has been linked to a variety of diseases. 39 F I G U R E 5 Migration of allogeneic bone marrow-derived cells to the stomach of a patient with H pylori-associated chronic gastritis. A, HE staining of the gastric biopsy of a female patient with acute myeloid leukaemia who received a bone marrow transplant from a male donor. B, Fluorescence in situ hybridization (FISH) analysis of the gastric biopsy tissue from the same patient using CEPY (orange)/ CEPX (green) dual-colour probes. C, Quantitative chromosome analysis by FISH. D-I, FISH analysis of localized areas of the gastric gland with immunofluorescent staining of CD105 (red) or CD45 (red). Y chromosome-positive, CD105-positive and CD45-negative cells (white arrows) around epithelial glands correspond to BM-MSCs. J, Data from The Cancer Genome Atlas showed that THBS2 gene expression is significantly higher in gastric tumour tissues (T, n = 408) than in normal tissues (N, n = 211). K, Survival analysis showed that patients with gastric cancer have poorer prognosis and a lower survival rate when THBS2 expression is high [HR =1.55 (1.29-1.85, P = 1.4e-06)]. *P <.05

ACK N OWLED G EM ENTS
Supported by the National Natural Science Foundation of China (Nos. The H pylori SS1 strain was kindly provided by the First Affiliated Hospital of Nanchang University.

CO N FLI C T S O F I NTE R E S T
None.

AUTH O R S CO NTR I B UTI O N S
Rong Lin designed and supervised the study and data analysis; Huiying Shi, Cuihua Qi and Lingjun Meng performed most of experiments, analysed the data, wrote and revised the manuscript; Hailing Yao, Chen Jiang and Mengke Fan helped the experiments and analysed the data; Qin Zhang provided pathological assessment and analysis; Xiaohua Hou supervised the study. All the authors approved the final manuscript and agreed for the publication.

E TH I C A L A PPROVA L
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions. Informed consent was obtained from the patient included in the study. All institutional and national guidelines for the care and use of laboratory animals were followed.

DATA AVA I L A B I L I T Y S TAT E M E N T
Main data generated or analysed during this study are included in this published article, and detailed data are available from the corresponding author on reasonable request.