You are here

Article Text

Article menu

Download PDFPDF

Original article
Roles of fibroblasts from the interface zone in invasion, migration, proliferation and apoptosis of gastric adenocarcinoma
  1. Li-Hui Shan1,
  2. Wen-Guang Sun2,
  3. Wei Han1,
  4. Lei Qi1,
  5. Chun Yang2,
  6. Cui-Cui Chai1,
  7. Ke Yao1,
  8. Qiu-Feng Zhou1,
  9. Hong-Mei Wu1,
  10. Li-Feng Wang3,
  11. Jia-Ren Liu4
  1. 1Department of Pathology, The First Clinical College of Harbin Medical University, Harbin, The People's Republic of China
  2. 2Department of Nutrition Division, The First Clinical College of Harbin Medical University, Harbin, The People's Republic of China
  3. 3Department of Pathology, Xin Hua Hospital affiliated to Shanghai JiaoTong University School of Medicine, Shanghai, The People's Republic of China
  4. 4Department of Anesthesia, Harvard Medical School (CHB), Boston, Massachusetts, USA
  1. Correspondence to Professor Li-Feng Wang, Department of Pathology, Xin Hua Hospital affiliated to Shanghai JiaoTong University School of Medicine, 1665 Kong Jiang Road, Shanghai 200092, The People's Republic of China; hljwlf{at}yahoo.cn

Abstract

Aims Interface zone fibroblasts (INFs) are very important in the progression and metastasis of tumours but their effect on the invasion and migration of gastric cancer cells is still unclear.

Methods Primary fibroblasts were isolated from the distal normal zone (normal zone fibroblasts, NFs), interface zone (INFs) and tumour zone (cancer-associated fibroblasts, CAFs) of 60 human gastric carcinoma tissue samples. The crosstalk between these fibroblasts and human gastric cancer MGC-803 cells was evaluated using an indirect co-culture model in vitro.

Results A high level of fibroblast activation protein (FAP) in the invasion front of gastric cancer was found in the gastric cancer tissue samples and no FAP expression was found in 20 normal gastric tissue samples by immunohistochemistry. High FAP expression was associated with Lauren classification, degree of differentiation, tumour node metastasis stage and depth of tumour invasion (p<0.05 or p<0.01). INFs promoted invasion and migration of MGC-803 cells. The number of invasions in INFs, CAFs and NFs were 120.10±27.53 (95% CI 102.12 to 138.10), 63.00±14.80 (95% CI 53.33 to 72.67) and 14.22±6.20 (95% CI 10.17 to 18.27), respectively; the number of invasions in INFs were 8.45-fold and 1.89-fold higher than those in NFs and CAFs, respectively (p<0.05). The number of migrations in INFs, CAFs and NFs were 118.00±16.83 (95% CI 107.00 to 129.00), 61.00±16.36 (95% CI 50.31 to 71.69) and 24.00±11.52 (95% CI 16.47 to 31.53), respectively; the number of migration in INFs were 4.91-fold and 1.92-fold higher than those in NFs and CAFs, respectively (p<0.05). INFs also significantly promoted cell proliferation and inhibited apoptosis in MGC-803 cells compared with NFs and CAFs (p<0.05).

Conclusions These findings indicate that INFs exhibit a more robust biological modulatory activity than CAFs and NFs. INFs may be a key factor leading to tumour progression and metastasis and may be of use as a tool for post-treatment surveillance.

https://doi.org/10.1136/jclinpath-2012-200909

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Gastric cancer is a public health problem in most developing countries including China, and is the leading cause of cancer deaths worldwide.1 ,2 One of the hallmarks of cancer is the invasion of tumour cells into the adjacent normal tissue.3 ,4 Invasive tumour cells interact with their microenvironment and remodel it to support growth and progression of the tumour cell.4 Given that most tumours grow following the activation of tumour microenvironment remodelling, the normal zone that is remote from the tumour load may be unaffected by the tumour.5 Tumour microenvironment remodelling could occur most rapidly in the interface zone, which is a unique functional and molecular region between the normal zone and the tumour invasion front.6 It has been reported that stromal fibroblasts from the interface zone of human breast carcinomas induce an epithelial–mesenchymal transition (EMT)-like state in breast cancer cells in vitro.7

    There is accumulating evidence to show that the tumour stroma provides a microenvironment that is critical for cancer cell growth and invasion.8 Cancer-associated fibroblasts (CAFs) are one of the most crucial components of the tumour microenvironment. They promote growth and invasion of cancer cells through involvement of the EMT, angiogenesis, the deregulation of antitumour immune responses, progressive genetic instability9 and the synthesis, deposition and remodelling of the extracellular matrix (ECM).10 ,11 CAFs also differ from normal fibroblasts in their phenotypic properties, the expression of growth factors and ECM molecule synthesis.12 Some findings also suggest that CAFs possess biological properties and functions that are distinct from those of normal fibroblasts.13 However, the functional contributions of fibroblasts located in the interface zone between the remote normal zone and the tumour invasion front of cancer cells are still unknown.

    The traditional concept is one in which the tumour margin is defined by the edge between the tumour and the normal zone (figure 1A). The new three-zone concept differs from this traditional concept (figure 1B).7 Gastric cancer tissue from at least 60 patients was examined by macroscopic dissection after the following geographical mapping: (1) distal normal tissue located at least 10 mm from the tumour margin, termed the normal zone; (2) normal-like tissue located up to 5 mm from the invasive front of the tumour, which was termed the interface zone; and (3) tumour tissue located at the epicentre of the tumour tissue, which was termed the tumour zone (figure 1B). We hypothesised that fibroblasts isolated from the interface zone have distinct microenvironmental influences on gastric cancer cells. The objectives of this study were to determine the roles of fibroblasts from the interface zone in the invasion, migration, proliferation and apoptosis of gastric adenocarcinoma.

    Figure 1
    Figure 1

    Division of tumour margins in gastric cancer tissue into three zones. (A) The traditional tumour margin is simply established as the edges between the tumour and the normal zone. (B) The tumour tissue grossly divided into the normal zone, the interface zone and the tumour zone.7 The histological appearance was confirmed by H&E staining. All cases were divided into a definite tumour burden zone, a clear normal region and an interface zone (C, 100×). The black arrow refers to the normal zone, the pink arrow to the interface zone and the red arrow to the tumour zone of gastric cancer tissue. Morphological histological observation showed regular arrangement of glands, regular size of nuclei and no structural heterologies of cells in the normal zone (D, 400×). In the interface zone, a partial regular arrangement of glands, enlarged cell body and irregular size of nuclei, increasing ratio of nuclei and cytoplasma and evident structural heterologies of cells as well as obvious desmoplasia are seen (E, 400×). In the tumour zone, irregular arrangement of glands, enlarging cell body, cell nuclus pleomorphism, increasing ratio of nuclei and cytoplasm and very evident structural heterologies of cells are shown (F, 400×). Scale bar=100 μm.

    Materials and methods

    Tissue specimens

    All tissue specimens used in this study were surgically removed from patients at the First Clinical Hospital of Harbin Medical University from 2008 to 2010 following signed consent. Patients who had received preoperative radiotherapy or chemotherapy were excluded. Of these cases, 60 were gastric carcinoma tissue and 20 were normal gastric tissue. All tumour samples were obtained from the primary tumour site. The clinical diagnosis was confirmed by histopathology. Of the 60 gastric carcinomas, 24 were stage I–II and 36 were stage III–IV. The mean age of the patients was 49.0 years (range 21–65).

    Immunohistochemistry

    Tissue sections (4 μm) were prepared from formalin-fixed paraffin-embedded blocks. Fibroblast activation protein (FAP) was determined in tissue from the tumour invasion front by immunohistochemical staining. Briefly, the deparaffinised section was quenched with 3% H2O2 for 10 min. After washing with phosphate- buffered saline (PBS), the sections were incubated in 5% bovine serum albumin for 20 min, followed by incubation with FAP primary monoclonal antibody (2 μg/ml) (R&D Systems, Minneapolis, MN) overnight at 4°C and secondary antibodies for 40 min at room temperature. The localisation of FAP was visualised by 3,3-diaminobenzidine (DAB) solution for 3–5 min. Finally, the slides were washed with water before being counterstained with haematoxylin. PBS was substituted for primary antibody as a negative control. Three slides for each tissue were independently evaluated by two observers with a light microscope. FAP expression was quantified as the relative percentage of the positive staining stroma area to the selected whole stroma area in slides. The degree of FAP staining in gastric cancer was classified into three groups14: +++, strong staining in >50% of stroma fibroblasts; ++, moderate staining in >50% of stroma fibroblasts; and +, faint or weak staining in >50% of stroma fibroblasts.

    Isolation of primary fibroblasts from different zones of human gastric tumour tissue

    Human gastric tumour specimens were obtained from patients who were undergoing surgery in the First Clinical Hospital of Harbin Medical University. An experienced anatomical pathologist grossly examined and obtained representative samples of tumour tissue from three zones: the normal zone (distal normal tissue at least 10 mm from the outer tumour boundary), the interface zone (adjacent tissue within 5 mm of the outer tumour boundary) and the tumour zone (tissue within the tumour boundary). These zones were confirmed by H&E staining. A representative sample of tissue was collected from each region for subsequent fibroblast isolation. Briefly, tissues from these zones were finely minced into 1–2 mm3 fragments, washed twice in antibiotic-containing PBS and disaggregated overnight in serum-free Dulbecco's Modified Eagle Medium (DMEM) with 0.1% collagenase at 37°C on a rotator. After 24 h the epithelial cells were separated from the stromal cells by differential centrifugation, as described by Speirs et al.15 The stromal cells were washed in sterile PBS and seeded in 35 mm dishes at 37°C in a humidified chamber containing 5% CO2. Interface zone fibroblasts (INFs) were those from the interface zone, normal zone fibroblasts (NFs) from the normal zone and CAFs from the tumour zone. All experiments were performed using fibroblasts that had been cultured for 3–10 passages.

    Characterisation of primary fibroblasts

    These cells were seeded in 24-well glass chambers. After 24 h incubation the cells were rinsed twice with cold PBS, fixed in 4% paraformaldehyde for 20 min at room temperature and permeabilised with 0.1% Triton X-100 for 30 min on ice. Primary antibody solutions (FAP, R&D Systems; cytokeratin, Dako, Glostrup, Denmark; vimentin, Abcam, Cambridge, Massachusetts, USA) were added to each chamber and the slides were incubated overnight at 4°C. After washing with PBS the cells were incubated with a FITC-conjugated second antibody (1 : 100) or a TRITC-conjugated second antibody (1 : 100) for 1 h at room temperature. The cells were then stained with a drop of 4’,6-diamidino-2-phenylindole for 30 min to visualise the nuclei and covered with a coverslip. The slides were observed using an immunofluorescent microscope (Olympus, Japan).

    Migration and invasion assays

    The gastric cancer MGC-803 cell line was purchased from the Center for Culture Collection, Harbin Medical University.

    For cell migration assays, MGC-803 cells (2×l04 cells/well) were seeded on transwell inserts with 8 μm pores (Corning, New York, USA) and cultured in serum-free DMEM. In the lower chamber, 5×l04 NFs, CAFs or INFs in 800 μl DMEM containing 10% FBS were seeded. After 24 h incubation, cells which did not migrate through the pores were removed by scraping the membrane with a cotton swab. Cells that traversed the membrane were fixed in 95% ethanol and stained with 10% Giemsa.

    For cell invasion assays the cells need to migrate through an ECM barrier by enzymatic degradation. Thus, the transwell membranes were coated with a uniform layer of 50 μl of a 1 : 3 dilution of Matrigel basement membrane matrix (BD Biosciences, Bedford, Massachusetts, USA) per well. MGC-803 cells were seeded on transwell inserts (2×l04 cells/well) and cultured in serum-free DMEM and 5×l04 NFs, CAFs or INFs in DMEM containing 10% FBS were seeded in the lower chamber. After 24 h incubation, invasive cells that penetrated through the pores and migrated to the underside of the membrane were stained with Giemsa. The number of cells was counted under a microscope (200×) from five random fields by two independent observers with double-blinded samples.16 ,17

    Preparation of conditioned medium

    The primary cells were cultured in 25 cm2 cell culture dishes with growth medium until they reached 90% confluence. The cells were rinsed twice with PBS and then incubated in 4 ml fresh serum-free DMEM for 24 h. The essential media were collected after 24 h incubation. The conditioned medium (CM) was centrifuged at 1500–1600 rpm for 5 min to remove cell debris and then filtered with 0.45 μm Millipore Ultrafree centrifugal filters (Millipore, Temecula, California, USA).

    Cell proliferation and apoptosis assays

    Cell proliferation and apoptosis assays were performed using a Cell Counting Kit-8 (Dojindo, Rockville, Maryland, USA) and a FITC Annexin V Apoptosis Detection Kit I (BD Pharmingen, San Diego, California, USA). For proliferation assays, MGC-803 cells were plated in 24-well plates and cultured in NFs-CM, CAFs-CM or INFs-CM for 24–48 h and stained with crystal violet. MGC-803 cells were also plated in 96-well plates and cultured in NFs-CM, CAFs-CM and INFs-CM for 24, 48 and 72 h. Cell proliferation was determined according to the instructions of the manufacturer of the Cell Counting Kit-8 assay. For the analysis of apoptosis, MGC-803 cells were also plated in 6-well plates and cultured in NFs-CM, CAFs-CM or INFs-CM for 24 h. MGC-803 cells were routinely harvested and analysed using the FITC Annexin V/PI Apoptosis Detection Kit according to the manufacturer's protocol.

    Statistical analysis

    All data are expressed as mean±SD. The differences were calculated using the two-tailed Student t test and one-way analysis of variance. The intensity of FAP staining in relation to various clinicopathological factors was assessed with the χ2 test. Data analyses were generated using SPSS for Windows V.18.0. Statistical significance was set at p<0.05.

    Results

    Division of tumour margins in gastric cancer tissue into three zones

    As shown in figure 1, three zones were defined in gastric cancer tissue. All cases were divided into a definite tumour burden zone, a clear normal region and an interface zone. The histological appearances of these zones were confirmed by H&E staining and a representative sample of tissue was collected from each region for subsequent fibroblast isolation (figure 1C). In the normal zone there was a regular arrangement of glands, nuclei of regular size and no structural heterologies of cells (figure 1D). In the interface zone, partial regular arrangement of glands, enlarging cell body and irregular size of nuclei, increasing ratio of nuclei and plasma and evident structural heterology of cells as well as obvious desmoplasia were seen (figure 1E). Tissue from the tumour zone showed irregular arrangement of glands, enlarging cell body, cell nucleus pleomorphism, increasing ratio of nuclei and plasma and very evident structural cell heterology (figure 1F).

    FAP expression in gastric cancer invasion front tissues

    In order to understand the overall FAP expression in gastric cancer, 80 gastric tissue samples (60 gastric cancer and 20 normal gastric tissues) were examined by immunohistochemistry. The results showed that no FAP expression was found in stromal fibroblasts of normal gastric tissue (figure 2A) but a high level of FAP expression was seen in stromal fibroblasts of tissue from the gastric cancer invasion front (figure 2B). A negative control without primary antibody is shown in figure 2C. Quantitative analysis of the positive staining area showed that the expression of FAP in the gastric cancer invasion front (31.83%) was more prominent than in other sites of gastric cancer tissue (22.65%, p<0.05; figure 2D).

    Figure 2
    Figure 2

    Fibroblast activation protein (FAP) expression in gastric tissue. FAP expression was determined in 60 gastric cancer tissue samples and 20 normal gastric tissue samples by immunohistochemistry. No FAP expression was detected in stroma cells of normal gastric tissue (A). A large proportion of FAP expression was seen in stroma cells of the gastric cancer invasion front (B). A negative control without FAP antibody is shown (C). Quantitative analysis of the positive staining area showed that FAP expression in the gastric cancer invasion front was more prominent than in gastric cancer tissue (D, p<0.05). The arrow shows positive cells of FAP expression. Scale bar=100 μm.

    Relationship between FAP and clinicopathological characteristics

    The relationship between the quantitative level of FAP in gastric cancer invasion front tissue and the clinicopathological characteristics was assessed. As shown in table 1, the staining area of FAP in gastric cancer invasion front tissue was significantly associated with Lauren classification, the degree of differentiation, tumour node metastases (TNM) stage and depth of tumour invasion but was not related to age and gender. Specifically, FAP staining area in stromal showed a different expression in the I–II stage (19.37±7.98%) and III–IV stage (30.16±5.51%). In addition, analysis of the relationship between FAP staining intensity and clinicopathological factors showed a significant association with Lauren classification, degree of differentiation, TNM stage and depth of tumour invasion. No correlation was found between the staining intensity of FAP and age or gender (table 1).

    View this table:
    Table 1

    Association between FAP expression (invasion front) and clinicopathological variables of gastric cancer

    Identification of INFs, NFs and CAFs

    The fibroblasts isolated from the normal, interface and tumour zones of tissue were designated as NFs, INFs and CAFs. These fibroblasts possessed the basic fibroblast characteristics: a long spindle-like morphology, strong expression of the fibroblastic marker vimentin and negative expression of the epithelial marker cytokeratin. For further characterisation of the NFs, CAFs and INFs, FAP expression as a marker was examined in the three cell types by immunocytochemistry (figure 3, FAP(I)) and immunofluorescence (figure 3, FAP(II)). The results showed that INFs, CAFs and NFs had universally positive FAP expression. However, the fluorescence intensity of FAP expression was higher in INFs than in CAFs and NFs (figure 3I). Importantly, the morphological results showed that INFs had some regular arrangement of fibroblasts and some disorderly arrangement of fibroblasts, NFs had a regular arrangement of fibroblasts and CAFs had a disorderly arrangement of fibroblasts (figure 3I). The negative controls are shown in figure 3II.

    Figure 3
    Figure 3

    Identification of interface zone fibroblasts (INFs), normal zone fibroblasts (NFs) and cancer-associated fibroblasts (CAFs) from human gastric tumour tissue. The characteristics of these cells were identified by immunocytochemical and immunofluorescent methods (I). These cells show a long spindle-like morphology, expression of vimentin and FAP, and negative expression of cytokeratin. NFs, CAFs and INFs showed a universally positive expression of FAP. However, the fluorescent intensity (FAP(I)) and immunohistochemical expression (FAP(II)) of FAP were higher in INFs that in CAFs and NFs. Their negative controls are also shown (II). The nuclei were stained with 4’,6-diamidino-2-phenylindole. Scale bar=100 μm.

    Effects of INFs, NFs and CAFs on invasion, migration, proliferation and apoptosis of MGC-803 cells

    The crosstalk between the three types of fibroblasts with gastric cancer cells was evaluated using an indirect co-culture model in vitro. Our results showed that the number of invasions in INFs, CAFs and NFs were 120.10±27.53 (95% CI 102.12 to 138.10), 63.00±14.80 (95% CI 53.33 to 72.67) and 14.22±6.20 (95% CI 10.17 to 18.27), respectively; the number of invading MGC-803 cells was increased by 8.45-fold and 1.89-fold as a result of co-cultivation with INFs compared with co-cultivation with NFs and CAFs, respectively (p<0.05; figure 4A,C). In the migration assay the number of migrations in INFs, CAFs and NFs were 118.00±16.83 (95% CI 107.00 to 129.00), 61.00±16.36 (95% CI 50.31 to 71.69) and 24.00±11.52 (95% CI 16.47 to 31.53), respectively; the number of migrating MGC-803 cells was increased by 4.91-fold and 1.92-fold as a result of co-cultivation of MGC-803 cells with INFs in comparison with co-cultivation of MGC-803 cells with NFs and CAFs, respectively (p<0.05; figure 4B,D). The results showed that MGC-803 cells cultured with INFs-CM grew more rapidly than those in NFs-CM and CAFs-CM. The same results were also seen in crystal violet staining (p<0.05; figure 5A,B). The proliferation of MGC-803 cells with INFs-CM, NFs-CM or CAFs-CM at 24, 48 and 72 h was evaluated by the Cell Counting Kit-8 assay. As shown in figure 5, INFs promoted more proliferation of MGC-803 cells than NFs and CAFs (figure 5C; p<0.05). The number of proliferations in INFs-CM,CAFs-CM and NFs-CM at 24 h were 0.85±0.17 (95% CI 0.73 to 0.96), 0.54±0.10 (95% CI 0.48 to 0.60) and 0.25±0.06 (95% CI 0.21 to 0.28), respectively; the number of proliferations in INFs-CM, CAFs-CM and NFs-CM at 48 h were 1.45±0.09 (95% CI 1.39 to 1.51), 0.86±0.17 (95% CI 0.75 to 0.97) and 0.45±0.12 (95% CI 0.37 to 0.53), respectively; and the number of proliferation in INFs-CM, CAFs-CM and NFs-CM at 72 h were 1.86±0.18 (95% CI 1.74 to 1.98), 1.50±0.16 (95% CI 1.39 to 1.61) and 0.90±0.29 (95% CI 0.71 to 1.09), respectively. The results from flow cytometry showed that INFs inhibited apoptosis of MGC-803 cells when the cells were cultured with INFs-CM. Apoptosis of MGC-803 cells was inhibited much more in INFs-CM (2.84±0.10, 95% CI 2.77 to 2.91) than in NFs-CM (4.12±0.068, 95% CI 4.08 to 4.16) and CAFs-CM (3.72±0.08, 95% CI 3.67 to 3.77) (p<0.05; figure 5D,E).

    Figure 4
    Figure 4

    Effects of interface zone fibroblasts (INFs), normal zone fibroblasts (NFs) and cancer-associated fibroblasts (CAFs) on invasion and migration of MGC-803 cells. Invasion and migration of MGC-803 cells were determined when co-cultured with INFs, NFs or CAFs (A–D). Cells invading or migrating to the transwell membrane were stained with Giemsa (A, B). Both CAFs and INFs promoted the invasion and migration of MGC-803 cells compared with NFs. (A, C) Invasion of MGC-803 cells co-cultured with INFs, NFs and CAFs. (B, D) Migration of MGC-803 cells co-cultured with INFs, NFs and CAFs. Scale bar=100 μm. *p<0.05 vs NF group.

    Figure 5
    Figure 5

    Effects of interface zone fibroblasts (INFs), normal zone fibroblasts (NFs) and cancer-associated fibroblasts (CAFs) on proliferation and apoptosis of MGC-803 cells. The proliferation of MGC-803 cells with INFs-CM or NFs-CM or CAFs-CM for 24, 48 and 72 h was evaluated by crystal violet staining (A, B) and the Cell Counting Kit-8 assay (C). INFs-CM promoted rapid growth of MGC-803 cells compared with NFs-CM and CAFs-CM (p<0.05). In addition, apoptosis of MGC-803 cells induced by INFs-CM, NFs-CM and CAFs-CM was examined by flow cytometry (D, E). The results showed that INFs-CM inhibited apoptosis in MGC-803 cells more than NFs-CM and CAFs-CM (E). *p<0.05 vs NF and CAF groups.

    Discussion

    In this study, 80 gastric tissue samples including 60 gastric cancer and 20 normal gastric tissues were examined. Our findings showed that stroma fibroblasts from the gastric cancer invasion front had strong positive FAP expression, with no FAP expression in normal gastric tissue. The level of FAP expression was significantly related to Lauren classification, degree of differentiation, TNM stage and depth of tumour invasion. FAP is a cell surface serine protease that is highly expressed in CAFs of human epithelial carcinomas but not in cancer cells or normal fibroblasts. FAP is also a type II membrane-bound serine protease exhibiting dipeptidyl peptidase and collagenase activities.18 ,19 Thus, fibroblasts expressing FAP in tumour tissue have an important role in the development of tumour.

    In a previous study, fibroblasts designated as normal referred to fibroblasts that were isolated from benign tissues, in accordance with normal practice.20 In our study, normal fibroblasts were defined as those that were isolated from a normal region of the same tissue as isolated CAFs and INFs. In addition, we defined an interface zone that corresponds to a region located up to 5 mm from the invasive front of the tumour. Although fibroblasts from NFs, CAFs and INFs exhibited long and spindle-like shapes, they showed different levels of FAP expression. A higher level of FAP expression was detected in INFs than in NFs and CAFs. In another study, patients with colonic tumours with high levels of stromal FAP were more likely to have aggressive disease progression and potential development of metastases or recurrence.21 This indicates that there is a relationship between the expression of FAP and ECM degradation. Thus, these fibroblasts in stroma, especially INFs, may be involved in invasion, migration, proliferation and apoptosis, and in identifying the molecular signal involved in the tumour development process.

    In our study, INFs from the gastric cancer invasion front had a high level of FAP expression compared with CAFs and NFs. The impact of FAP on the ECM is complex, with recent investigations suggesting that its influence on surrounding stroma is differentiated by collagen subtype.22 An important observation in this study was that high FAP expression in the gastric cancer invasion front was associated with the depth of tumour invasion. It is possible that secreted factors from stromal fibroblasts of the gastric cancer invasion front may supply their own optimal microenvironment to support the survival of cancer cells or to promote invasion and migration. Thus, co-culture of the human gastric cancer MGC-803 cell line with INFs, CAFs or NFs was performed to determine the effect on cell migration, invasion and proliferation. INFs promoted more cell invasion and migration of MGC-803 cells than CAFs and NFs. It has been reported that INFs from the interface zone of human breast carcinomas induce an EMT-like state in breast cancer cells in vitro.7 It indicated that INFs exhibited more robust biological modulatory activity in a dynamic region vital to tumour progression. In order to make way for invading cancer cells, INFs initiate both proteolytic and structural modification of the ECM to create the path for the following migrating cancer cells. It has been widely recognised that breaking the basement membrane is the first step for cancer cells to extravagate into the circulation system.23 The remodelled ECM proteins can alter the expression levels of some specific genes that are essential for the structural scaffolding and cytoskeletal organisation.24 In our study, MGC-803 cells cultured with INFs-CM also showed evident cell proliferation compared with CAFs-CM or NFs-CM. It is suggested that INFs-CM has more cytokine factors to promote cell proliferation of MGC-803 cells. In addition, INFs-CM also significantly inhibited apoptosis of MGC-803 cells in comparison with CAFs-CM or NFs-CM. Thus, INFs possess a greater capacity to interact with and to modulate the gastric cancer cell model system than NFs or CAFs. Thus, signal crosstalk between the cancer cells and INFs may direct the modification of the adjacent ECM and basement membrane. The possible mechanisms need to be studied further.

    In summary, high FAP expression of INFs in the stroma of the gastric cancer invasion front was related to Lauren classification, degree of differentiation, TNM tumour stage and depth of tumour invasion in patients. INFs showed significant promotion of cell proliferation and inhibition of apoptosis in MGC-803 cells compared with CAFs and NFs. INFs also showed a higher level of FAP expression than CAFs and NFs. When MGC-803 cells were co-cultured with INFs, INFs significantly increased cell invasion and migration of human gastric cancer cells compared with CAFs and NFs. It is suggested that INFs from the interface zone of the tumour may have a potential dynamic region that is a key factor leading to tumour progression and metastasis of gastric cancer. Targeting INFs may therefore be a therapeutic strategy for the treatment of gastric cancer that warrants further study. INFs may be a good surrogate marker for cancer progression and invasion and serve as a tool for post-treatment surveillance.

    Take-home messages

    • Interface zone fibroblasts (INFs) from gastric cancer tissue possess a modulatory activity in tumor progression and metastasis.

    • They may be good surrogate markers for cancer progression and invasion and serve as a tool for post-treatment surveillance.

    Acknowledgments

    This work was supported by the Natural Science Foundation of HeiLongJiang Province (D201103) and National Natural Science Foundation of China (No. 81072296), the People's Republic of China.

    References

    1. ↵
      1. Anderson C,
      2. Nijagal A,
      3. Kim J
      . Molecular markers for gastric adenocarcinoma: an update. Mol Diagn Ther 2006;10:345–52.
      OpenUrlPubMedWeb of Science
    2. ↵
      1. Siegel R,
      2. Naishadham D,
      3. Jemal A
      . Cancer statistics, 2012. CA Cancer J Clin 2012;62:10–29.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Xing F,
      2. Saidou J,
      3. Watabe K
      . Cancer associated fibroblasts (CAFs) in tumor microenvironment. Front Biosci 2010;15:166–79.
      OpenUrlCrossRefPubMed
    4. ↵
      1. Casey T,
      2. Bond J,
      3. Tighe S,
      4. et al
      . Molecular signatures suggest a major role for stromal cells in development of invasive breast cancer. Breast Cancer Res Treat 2009;114:47–62.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Acharya PS,
      2. Zukas A,
      3. Chandan V,
      4. et al
      . Fibroblast activation protein: a serine protease expressed at the remodeling interface in idiopathic pulmonary fibrosis. Hum Pathol 2006;37:352–60.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Bhowmick NA,
      2. Neilson EG,
      3. Moses HL
      . Stromal fibroblasts in cancer initiation and progression. Nature 2004;432:332–7.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Gao MQ,
      2. Kim BG,
      3. Kang S,
      4. et al
      . Stromal fibroblasts from the interface zone of human breast carcinomas induce an epithelial-mesenchymal transition-like state in breast cancer cells in vitro. J Cell Sci 2010;123:3507–14.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Kalluri R,
      2. Zeisberg M
      . Fibroblasts in cancer. Nat Rev Cancer 2006;6:392–401.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Bauer M,
      2. Su G,
      3. Casper C,
      4. et al
      . Heterogeneity of gene expression in stromal fibroblasts of human breast carcinomas and normal breast. Oncogene 2010;29:1732–40.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Jung DW,
      2. Che ZM,
      3. Kim J,
      4. et al
      . Tumor-stromal crosstalk in invasion of oral squamous cell carcinoma: a pivotal role of CCL7. Int J Cancer 2010;127:332–44.
      OpenUrlPubMedWeb of Science
    11. ↵
      1. Orimo A,
      2. Gupta PB,
      3. Sgroi DC,
      4. et al
      . Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 2005;121:335–48.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Sadlonova A,
      2. Bowe DB,
      3. Novak Z,
      4. et al
      . Identification of molecular distinctions between normal breast-associated fibroblasts and breast cancer-associated fibroblasts. Cancer Microenviron 2009;2:9–21.
      OpenUrlPubMed
    13. ↵
      1. Franco OE,
      2. Shaw AK,
      3. Strand DW,
      4. et al
      . Cancer associated fibroblasts in cancer pathogenesis. Semin Cell Dev Biol 2010;21:33–9.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Sugiura T,
      2. Inoue Y,
      3. Matsuki R,
      4. et al
      . VEGF-C and VEGF-D expression is correlated with lymphatic vessel density and lymph node metastasis in oral squamous cell carcinoma: implications for use as a prognostic marker. Int J Oncol 2009;34:673–80.
      OpenUrlPubMed
    15. ↵
      1. Speirs V,
      2. White MC,
      3. Green AR
      . Collagenase III: a superior enzyme for complete disaggregation and improved viability of normal and malignant human breast tissue. In Vitro Cell Dev Biol Anim 1996;32:72–4.
      OpenUrlCrossRefPubMed
    16. ↵
      1. Li Y,
      2. Sun WG,
      3. Liu HK,
      4. et al
      . Gamma-tocotrienol inhibits angiogenesis of human umbilical vein endothelia cell induced by cancer cell. J Nutr Biochem 2011;22:1127–36.
      OpenUrlCrossRefPubMed
    17. ↵
      1. Liu HK,
      2. Wang Q,
      3. Li Y,
      4. et al
      . Inhibitory effects of gamma-tocotrienol on invasion and metastasis of human gastric adenocarcinoma SGC-7901 cells. J Nutr Biochem 2010;21:206–13.
      OpenUrlCrossRefPubMed
    18. ↵
      1. Tahtis K,
      2. Lee FT,
      3. Wheatley JM,
      4. et al
      . Expression and targeting of human fibroblast activation protein in a human skin/severe combined immunodeficient mouse breast cancer xenograft model. Mol Cancer Ther 2003;2:729–37.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Cheng JD,
      2. Dunbrack RL Jr.,
      3. Valianou M,
      4. et al
      . Promotion of tumor growth by murine fibroblast activation protein, a serine protease, in an animal model. Cancer Res 2002;62:4767–72.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Paland N,
      2. Kamer I,
      3. Kogan-Sakin I,
      4. et al
      . Differential influence of normal and cancer-associated fibroblasts on the growth of human epithelial cells in an in vitro cocultivation model of prostate cancer. Mol Cancer Res 2009;7:1212–23.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Henry LR,
      2. Lee HO,
      3. Lee JS,
      4. et al
      . Clinical implications of fibroblast activation protein in patients with colon cancer. Clin Cancer Res 2007;13:1736–41.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Christiansen VJ,
      2. Jackson KW,
      3. Lee KN,
      4. et al
      . Effect of fibroblast activation protein and alpha2-antiplasmin cleaving enzyme on collagen types I, III, and IV. Arch Biochem Biophys 2007;457:177–86.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Ostman A,
      2. Augsten M
      . Cancer-associated fibroblasts and tumor growth–bystanders turning into key players. Curr Opin Genet Dev 2009;19:67–73.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Gaggioli C,
      2. Hooper S,
      3. Hidalgo-Carcedo C,
      4. et al
      . Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells. Nat Cell Biol 2007;9:1392–400.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors Shan LH and Sun WG contributed equally to the study i.e carried out the primary culture, established indirect co-culture model, cell proliferation and apoptosis assays and analyzing data of this manuscript. Han W, Qi L, Yang C, Chai CC, Yao K, Zhou QF and Wu HM dealed with specimen collection and the immunohistochemistry staining. Wang LF and Liu JR are co-corresponding authors of the study i.e. experimental design, analyzed and interpreted the data, drafted and critical revised the manuscript, and approved the final version of this manuscript.

    • Conflicts of interest None.

    • Patient consent Obtained.

    • Provenance and peer review Not commissioned; externally peer reviewed.