Introduction
Gastric cancer is the fifth most common cancer and the fourth leading cause of cancer-related death worldwide [
1]. Although the incidence of gastric cancer has declined over the past 50 years globally, it is still a considerable health burden in developing counties. The global 5-years survival rates of gastric cancer patients remain unsatisfactory [
2‐
4]. Most patients are diagnosed with an advanced stage of gastric cancer, and the recurrences and metastases of patients are frequently observed [
5]. The most important risk factor for gastric cancer is
Helicobacter pylori (
H. pylori), classified as a class I carcinogen for gastric cancer by IARC in 1994 [
6]. Although numerous studies have been carried out on the association between
H. pylori infection and gastric cancer, the mechanism underlying their relationship remains unclear [
7‐
9]. In recent years,
H. pylori have been found to disrupt the tight junctions between gastric epithelial cells and penetrate the deeper intercellular spaces to colonize the gastric epithelial cells, intercellular spaces, and the underlying lamina propria [
10].
H. pylori at these sites, directly and indirectly, interact with fibroblasts, connective tissue, and other extracellular matrix components. Recently, the direct interaction between
H. pylori and rat stomach fibroblasts has been proposed [
11‐
13].
The tumor microenvironment (TME) favors the growth and metastasis of cancer cells in many solid tumors. Fibroblasts, especially cancer-associated fibroblasts (CAFs), are a critical stroma component and probably contribute to tumor initiation and progression via direct contact with cancer cells or paracrine manner [
14,
15]. CAFs are a functionally heterogeneous population composed of tumor-promoting CAFs, tumor-restraining CAFs, and quiescent CAFs with different marker expressions [
16]. CAFs can be identified through a series of markers such as fibroblast-activating protein (FAP), alpha-smooth muscle actin (α-SMA), fibroblast specific protein 1 (FSP-1), and vimentin [
17]. Based on these markers, Ohlund and colleagues identified two distinct subtypes of CAFs in human pancreatic cancer: α-SMA
hiIL-6
low myofibroblastic CAFs and α-SMA
lowIL-6
hi inflammatory CAFs [
18]. The heterogeneity of CAFs is further reflected in its origin of several cell types, including normal resident fibroblasts (NFs), bone marrow mesenchymal stem cells, pericytes, epithelial cells, and endothelial cells [
19,
20]. Among these cells, tissue-resident NFs activation in response to numerous stimuli from the TME is the primary source of CAFs and exists before tumorigenesis [
21]. Moreover, the transition of NFs to CAFs can induce gene expression and phenotype changes between them [
22]. However, the multifaceted interplay between
H. pylori-induced inflammation, fibroblasts, and cancer cells are poorly understood in gastric tumorigenesis.
Serpin E1, also known as plasminogen activator inhibitor-1, is a structurally well-studied serine protease inhibitor that acts in diverse pathologies such as thrombosis and fibrosis by inhibition of urokinase-type plasminogen activators and tissue-type plasminogen activators. Subsequent studies suggest a pro-tumorigenic role of Serpin E1 [
23]. Multiple cells, including tumor and stromal cells such as endothelial cells, fibroblasts, macrophage cells, and adipocytes, produce Serpin E1 [
24]. Its expression is also regulated by various growth factors, cytokines, and chemokines [
25]. However, Serpin E1 expression in gastric cancer cells and its interaction with
H. pylori infection and fibroblasts remain unknown.
In the current study, we obtain the transcriptome expression profiles of CAFs and NFs isolated from paired gastric cancer and adjacent normal tissues and demonstrate the interplay of H. pylori, fibroblasts, and gastric cancer cells to promote the conversion of NFs to CAFs via cytokine release, especially Serpin E1. H. pylori and CAFs induce gastric epithelial and cancer cells to express Serpin E1, leading to cell migration, invasion, and tumor formation in vitro and in vivo. Our findings thus offer a novel insight into gastric cancer tumorigenesis during H. pylori infection. Targeting Serpin E1 can disrupt the interaction of H. pylori, CAFs, and cancer cells to provide a promising target for gastric cancer therapy, which will become an attractive option for future research.
Materials and methods
H. pylori strain and cell culture
H. pylori strain GZ7 (GenBank accession ID: KR154737.1) was isolated from clinical gastric cancer tissue with informed consent and the approval of the Ethics Committee of Guizhou Medical University and confirmed as a typical East Asian strain (cagA
+) by sequencing in our previous research. Its whole-genome sequence has been submitted to the National Microbiology Science Data Center, China (Accession ID: NMDC60014578,
https://nmdc.cn/). This bacterium was grown on a Columbia blood agar plate (Oxoid Ltd, Cambridge, UK) containing 10% sheep blood and 100 U/ml
H. pylori selective supplement (Oxoid Ltd, Cambridge, UK) at 37 °C under microaerobic conditions (5% O
2, 10% CO
2, 85% N
2).
Human gastric cancer cell lines AGS cells (ATCC, VA, USA), MNK45 cells, HUVECs (CBCAS, Shanghai, China), and primary GC cells (isolated from gastric cancer tissues in our previous study) were cultured in RPMI-1640 medium (Hyclone, Logan, UT, USA) containing 10% fetal calf serum (Gibco, Carlsbad, CA, USA) and incubated in a humidified incubator with 5% CO2 at 37˚C.
Samples of human gastric cancer tissues
Twelve pairs of clinical gastric cancer and para-cancer normal tissues (at least 5 cm from the outer margin) from the surgical resection were collected at the Affiliated Hospital of Guizhou Medical University, China, from January 2020 to December 2021. Half of the tissues were used for pathological analysis, and the other half were transferred to the laboratory for IHC analysis within two hours. The diagnoses were confirmed by two pathologists. Ten patients were male, and two were female. Their median age was 60 (range 27–86) years. Eight cancers were adenocarcinoma, and four were the diffuse type. Nine cancers occurred in the gastric corpus, and three were in the gastric antrum. All patients were positive for the C13 urea breath test (a diagnosis method of H. pylori infection). All subjects gave their informed consent for inclusion before participating in the study. The study was conducted following the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Guiyang Medical University (Approved number: 2017(43)).
Stomach tissue sections of Mongolian gerbils infected with H. pylori
In our previous study,
H. pylori NCTC 11637 (ATCC 43504,
cagA-positive) was used to infect intragastrically Mongolian gerbils for 24 months to establish
H. pylori-related gastric disease models successfully. In these gerbils, chronic superficial gastritis, erosive gastritis, atrophic gastritis, intestinal metaplasia, and well-differentiated gastric cancer were pathologically observed at 3, 6, 12, and 24 months after
H. pylori infection [
26]. At different time points, the gerbils (3 gerbils per time point) were sacrificed after anesthesia, and their stomach tissues were removed, fixed, and embedded in paraffin. In the present study, the paraffin-embedded gerbil stomach tissues were cut into 5-μm sections for IHC staining.
Isolation and culture of primary fibroblasts
Three paired primary fibroblasts (CAFs and NFs) were obtained from three patients with poorly differentiated gastric adenocarcinomas (ages: 52–54 years) undergoing cancer resection at the Affiliated Hospital of Guizhou Medical University, China. The CAFs and NFs were isolated from cancer tissue and matched normal tissue at least 5 cm from the outer tumor margin [
27]. Fresh samples were washed with RPMI-1640 medium containing 100 U/ml penicillin and 100 μg/ml streptomycin (Hyclone, Logan, UT, USA).), cut into small pieces, and incubated in a 5 ml solution containing 2 mg/mL collagenase IV (Sigma-Aldrich, MO, USA) for 1.5–2.5 h at 37 °C. The digested cells were passed through a 200-mesh cell sieve and centrifuged at 1500 rpm for 10 min. The single-cell suspension was cultured in a fibroblast medium (ScienCell, CA, USA) consisting of 2% fetal bovine serum, 1% fibroblast growth supplement, and 1% penicillin/streptomycin. After that, the suspension was incubated in a humidified incubator with 5% CO
2 at 37˚C for 48 h, allowing fibroblasts to attach to the culture plate. The adherent cells were further passaged and cultured. CAFs and NFs from the 3rd to 5th passages were used for our experiments. Three subjects gave their informed consent, and the protocol was approved by the Ethics Committee of Guiyang Medical University [No: 2017(43)].
RNA sequencing (RNA-seq) analysis
Total RNA was isolated from three paired primary CAFs and NFs and reversely transcribed into cDNA to generate an indexed Illumina library, then sequenced at the GENEWIZ Biotechnology Co., Ltd. (Suzhou, China) using Illumina HiSeq 2000 platform. After normalizing the expression of genes, the differentially expression genes (DEGs) with fold-change > 2 and FDR < 0.05 were screened out by comparing the Fragments Per Kilobase per Million (FPKM) between CAFs and NFs using an online analysis tool (
http://www.omicsbean.cn/). The DEGs common to the three pairs of CAFs and NFs were selected for follow-up bioinformatic analysis. A heatmap of the DEGs was generated by the online tool (
http://www.omicsbean.cn/). Subsequently, the DEGs were analyzed by Gene Ontology (GO) using the Gene Ontology Consortium (
http://geneontology.org/), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways using KOBAS (
http://kobas.cbi.pku.edu.cn/), and the protein–protein interaction (PPI) networks using the STRING tool (
https://string-db.org/). Cytoscape software (version 3.7.2) was used to visualize KEGG pathways and PPI networks. Gene set enrichment analyses were performed by Gene Set Enrichment Analysis (GSEA) (v4.1.0) based on the gene expression (FPKM). NOM p-value < 0.01 and FDR q-value < 0.25 were considered as significant gene set.
Human cytokine array
NF and AGS cells were infected with
H. pylori for six hours at an MOI of 50, and then
H. pylori were removed. The culture supernatants were collected after another seven-day culture. The cytokine levels in the supernatants were determined by a Proteome Profiler Human Cytokine Array Kit (ARY005B, R&D Systems, MN, USA) according to the manufacturer's instructions. Briefly, membranes were incubated with blocking buffer at room temperature for one hour. Simultaneously, the cell supernatants (0.5 ml) were mixed with the biotinylated detection antibody cocktail (15 μl) at room temperature for one hour, and then the mixture was incubated with the membrane overnight at 4 °C. Afterward, the arrays (membranes) were incubated with horseradish peroxidase-conjugated streptavidin for 30 min at room temperature. The arrays subsequently were exposed to ECL substrate (Amersham, Bucks, UK), and the images were obtained using ECL chemiluminescence. The array images were analyzed with Image J. Cytokine levels are expressed as the percentage of the average density of two Cytokine Spots divided by the average density of six Reference Spots. Proteome Profiler Human Cytokine Array list was shown in Additional file
1: Table S1.
Direct and transwell co-culture models
NFs or CAFs and AGS cells were co-incubated with H. pylori in a 6-well plate for 6 h at an MOI of 50, and then free H. pylori were removed via extensive washing with PBS to construct the direct co-culture models. NFs or CAFs were seeded into the lower chamber, and H. pylori and/or AGS cells were added to the upper chamber of the Transwell (0.4-μm pores). After the cultures were incubated for six hours, H. pylori were removed via washing with PBS to construct the Transwell co-culture models. The co-cultures in the two models were further cultured for 72 h and seven days for in vitro experiments.
Lentivirus infection
A lentiviral expression vector containing cDNA to code Serpin E1 and control lentiviral were acquired from Jikai Co. (Shanghai, China) and were used to infect AGS, MNK45, and primary GC cells using Lipofectamine 2000 (Invitrogen, CA, USA). Subsequently, three stable cell lines overexpressing Serpin E1 or empty vector were generated by puromycin selection and used for in vitro experiments.
Cell viability assay
Cell viability was evaluated by the Cell Counting Kit (CCK-8) (Dojindo, Kumamoto, Japan). In brief, CAFs or NFs were seeded in a 96-well plate at a density of 1000 cells /well and cultured for six days. Similarly, AGS cells with Serpin E1 overexpression or empty vector were also seeded in a 96-well plate at a density of 500 cells /well and cultured for six days. Then, 10 μl of CCK-8 solution were added into the culture medium in each well on days 1, 2, 3, 4, 5, and 6 of culture, respectively. After incubation for two hours, the absorbance at 450 nm was measured by a microplate reader. The growth curve of fibroblasts and AGS cells was plotted based on the A450 values. Data presented as mean ± SD (n = 6 per group).
NFs or CAFs were co-cultured with AGS cells for ten days in a 24-well plate at a ratio of 10:1 (NFs/CAFs: AGS cells). Similarly, AGS cells with Serpin E1 overexpression or empty vector were seeded in a 24-well plate for ten days. The medium was removed, and the colonies were washed gently with PBS, fixed in 4% paraformaldehyde for 30 min, and stained with 0.1% crystal violet for 30 min. The colonies were counted under a microscope.
Cell cycle assay
AGS cells were collected, washed with cold PBS, and then fixed in cold 70% ethanol overnight at 4 °C. The next day, the cells were centrifugated at 1000 rpm for 5 min and resuspended in 450 μl 1 × PBS containing 50 μl RNase A. After incubation at 37 °C for 30 min, the cells were stained with 10 μl Propidium Iodide (50 μg/ml, BD Biosciences, San Jose, CA) for another 30 min away from light. Cell cycle distribution was determined by flow cytometry, and the results were analyzed with the FloJo software.
Apoptosis assay
Three stable cell lines with Serpin E1 overexpression or empty vector were cultured overnight in a serum-free medium to synchronize the cell cycle. The medium was replaced with a complete medium containing arsenic trioxide (As2O3, 20 and 40 μM, Sigma-Aldrich, MO, USA) for 24 h. Then, 105 cells were resuspended in 500 μl of 1 × binding buffer and stained with 5 μl of Annexin V-APC and 10 μl of 7-AAD (Biosciences, San Jose, CA) for 10 min, protected from light. The cell suspension was filtered through a nylon mesh (400 mesh), and cell apoptosis was detected by flow cytometry. Data were analyzed with the FloJo software.
Transwell migration and Matrigel invasion assay
Transwell assay was performed using an 8 μm pore size 24-well Transwell Chambers (Costar, Cambridge, MA, USA) without and with Matrigel (Biosciences, San Jose, CA). AGS cells or NFs/CAFs in 6-well plates were infected with H. pylori for six hours at an MOI of 50, and then H. pylori were then removed. After the cultures continued for seven days, 1 × 104 AGS cells in 200 μl of 1% FBS medium were seeded into the upper chamber, and 5 × 104 NFs or CAFs in 800 μl medium containing 10% FBS were seeded into the lower chambers. After 24 h (migration assay) and 72 h (invasion assay), the Transwell inserts were moved out, fixed in 4% paraformaldehyde for 30 min, and stained with 0.1% crystal violet for 30 min. The non-migrated or invaded cells were removed from the upper surface of the chambers. The number of migrated and invaded cells was counted under a microscope.
For stably transfected cancer cell lines, 104 cells were seeded in the upper chamber in a 1% FBS medium, and an 800 μl medium containing 10% FBS was added into the lower chambers. After 24 h (migration assay) and 48 h (Matrigel invasion assay) incubation, the Transwell inserts were moved out for follow-up experiments according to the method described above.
Immunocytochemistry, immunohistochemistry, and immunofluorescence
The cell climbing slices, including CAFs, NFs, AGS cells, AGS cells co-cultured with CAFs for three days at a 1:1 ratio, and AGS cells infected with H. pylori for seven days at an MOI of 50, were fixed in 4% paraformaldehyde for 30 min at room temperature. After washing with PBS, the slices were permeabilized with 0.5% Triton X-100 for 30 min and blocked with 5% bovine serum albumin for 60 min. The slices were incubated with different primary antibodies in a humid box overnight at 4 °C and then corresponding HRP-conjugated secondary antibodies (CST, MA, USA) for two hours at room temperature. Subsequently, a DAB reaction was performed using a DAB substrate kit (Abcam, Cambridge, UK), and the reaction was terminated by adding a 30% H2O2 solution. DAB-visualized slices were counter-stained successively with hematoxylin, 1% acid alcohol, and ammonium water. Finally, the slices were dehydrated in graded ethanol, followed by clearing in xylene. Images were acquired microscopically.
For immunohistochemistry, the tissue sections, including gerbil stomach tissues, tumor tissues of mouse xenografts, and human gastric cancer tissues, were deparaffinized in xylene and rehydrated in ethanol with an increased concentration. Then, antigens were retrieved using high pression in 0.01 M citrate buffer (pH 6.0), and the endogenous peroxidases were blocked using 3% H
2O
2. After that, the sections were incubated with primary and secondary antibodies. The following experiments were performed by the same methods as mentioned above. The quantification of staining density was analyzed using IHC Profiler from Image J software, and the IHC scores were calculated as follows: IHC scores = intensity score × percentage score [
28].
For immunofluorescence, a fluorescent secondary antibody was incubated with the slices in a humid box at room temperature for one hour. Cell nuclei were counter-stained with DAPI (Sigma-Aldrich, MO, USA). Images were acquired on a confocal microscope (Olympus, Japan).
The following primary antibodies were used: rabbit polyclonal anti-vimentin, anti-FSP1, anti-Serpin E1, anti-CD31, anti-Ki67, and anti-H. pylori antibodies (Abcam, Cambridge, UK), rabbit polyclonal anti-α-SMA antibody (Proteintech, Chicago, USA), rabbit polyclonal anti-FAP antibody (GeneTex Inc., CA, USA).
Western blotting assay
Cells were lysed in RIPA buffer containing a complete protease and phosphatase inhibitor cocktail (Sigma-Aldrich, MO, USA). The protein concentration of the cell lysates was quantified by a BCA Protein Assay Kit (Pierce, Rockford, IL, USA). The same protein samples were resolved onto 10% SDS-PAGE and then transferred onto PVDF membranes (Millipore Q, Billerica, MA, USA). After blocking with 5% nonfat milk at 37℃ for two hours, the membranes were incubated with primary antibodies overnight at 4℃, followed by incubation with the HRP-conjugated secondary antibody for two hours at room temperature. GAPDH antibody was used as a loading control. Finally, the protein band images were captured by a Gene Detection System with an ECL reagent (Thermo Fisher Scientific, MA, USA). The primary antibodies used in the experiments were rabbit polyclonal anti-FAP antibody (GeneTex Inc., CA, USA), rabbit polyclonal anti-lumican antibodies (Absin, Shanghai, China), rabbit polyclonal anti-Serpin E1, anti-Vimentin, and anti-Urease B antibody (Abcam, Cambridge, UK), rabbit monoclonal anti-GAPDH antibody (CST, MA, USA), and rabbit polyclonal anti-α-SMA antibody (Proteintech, Chicago, USA).
Enzyme-linked immunosorbent assay (ELISA)
ELISA kit of Serpin E1 (Invitrogen, CA, USA) was used to detect Serpin E1 levels in cell culture supernatants according to the manufacturer's instructions. The experiment was repeated three times.
Matrigel was melted at 4℃ overnight and diluted with serum-free medium (1:4). After Matrigel polymerization, human umbilical vascular endothelial cells (HUVECs, 5 × 104) were suspended in a medium and seeded on the Matrigel in the 24-well plate or the lower chamber of Transwell. Human recombinant Serpin E1 (PeproTech, USA) was added to the 24-well plate at a concentration of 2 ng/ml, or 1 × 105 cells, including AGS, MNK45, and primary gastric cancer cells overexpressing Serpin E1 and control vector, were added to the upper chamber of Transwell. After culturing for 24 h, tube formation of HUVECs was observed and photographed.
Tumor xenograft assay
Ten six-week-old male BALB/c nude mice were purchased from Chongqing Tengxin Biotechnology Co., LTD. (Chongqing, China) and raised under specific pathogen-free conditions at the Animal Center of Guizhou Medical University. 1 × 107 primary GC cells with stable Serpin E1 overexpression or empty vector were suspended in 200 µl PBS and injected subcutaneously into the right flanks of each mouse. Tumor size was monitored once every two days for 24 days. The tumor volume (V) was calculated by the formula: V = 1/2 × length × width2 (mm3). Twenty-four days later, all mice were sacrificed after anesthesia, and the tumor tissues were stripped, weighted, and fixed in 4% paraformaldehyde for further IHC staining. The animal study was approved by the Animal Care Welfare Committee of Guizhou Medical University (Approved number:1702155).
Analysis of The Cancer Genome Atlas (TCGA) dataset for Serpin E1
The human gastric cancer genomics dataset and the related information of patients, including 408 tumors and 211 normal tissues, were downloaded from The Cancer Genome Atlas-Stomach Adenocarcinoma (TCGA-STAD) database (
https://gdc-portal.nci.nih.gov/projects/TCGA-STAD) and analyzed for Serpin E1 mRNA expression.
Statistical analysis
All statistical analysis was performed using SPSS 22.0 (SPSS, Inc). GraphPad Prism 7 was used to generate figures. Two groups were compared using a two-tailed paired or unpaired Student’s t-test. More than two groups were compared using one-way ANOVA or two-way ANOVA. Spearman correlation analysis was used to analyze the association between Serpin E1 and FAP. The images shown are representative of at least three independent experiments. All data are presented as means ± standard deviation (SD) from at least three independent samples or experiments. A P value < 0.05 was considered statistically significant.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.