S 
were adjusted to 56105 cells/mL, seeded in the bottom chambers of 6-well plates and incubated for 8 h to allow attachment. Inserts containing 56105 cells/mL of cultured PBMCs were then transferred to the top chambers and cocultured for another 24 h in FBS-free conditional medium or complete medium. As negative controls, inserts with PBMCs were placed on wells with the same culture medium in the absence of cancer cells, and wells with GC cells were left without inserts. The cell count in the monoculture group was double that in the coculture group, to ensure similar cell numbers in all groups. Supernatants and cells were collected separately after 24 h for further use.Ethics StatementPatients who received radiochemotherapy, suffered from other cancers, or who had a family history of GC were excluded from the study. Written informed consent was obtained from all the subjects. The project was approved by the Research Ethics Committee of Zhongshan Hospital [28].Immunohistochemistry (IHC)TGF-b1 and TGF-b2 protein levels were examined by IHC in 4-mm-thick paraffin sections cut from a single selected block containing neoplastic and non-neoplastic gastric tissues. Samples were routinely dewaxed and hydrated. After blocking of endogenous peroxidase activity, antigens were retrieved by heating with ethylenediamine tetraacetic acid (pH = 9.0). Antigens were subsequently detected using a standard staining procedure (EnVisionTM Detection Kit, Dako, CA, USA). Rabbit polyclonal antibodies were used to detect TGF-b1 and TGF-b2 (all dilutions 1:100; Santa Cruz Biotechnology, CA). For antibody-negative controls, the primary antibodies were substituted with normal rabbit serum. Cases were regarded as positive if at least 5 of dysplastic or cancer cells displayed cytoplasmic staining for TGFb1 or TGF-b2 at 6100 magnification.Cell Proliferation AssayA Cell-IQ cell culturing platform (Chip-Man Technologies, Tampere, Finland), equipped with a phase-Lecirelin manufacturer contrast microscope (Nikon CFI Achromat phase contrast objective with 610 magnification, Nikon, Japan) and a camera, was used to detect the growth of tumor cells, as described previously [30]. Briefly, GC cells were cultured on 24-well plates (16104 cells/well) for 24 hTGF-b Roles in Tumor-Cell Interaction with PBMCsand then treated with TGF-b1 (Peprotech, USA) at 25 ng/mL. Control groups were left untreated. Cells were then incubated for a further 72 h in the Cell-IQ system. Images were captured at 30min intervals for 72 h, controlled by Image software (Chip-Man Technologies), and analyzed using freely-distributed image software (McMaster Biophotonics Facility, Hamilton, ON), using the manual tracking plug-in created by Fabrice Cordelieres (Institut Curie, Orsay, France). The Cell-IQ system automatically discriminates the dividing and stable cell stages, and calculates the total cell numbers during proliferation. Eight images were analyzed for each group. The mobility of Calyculin A lymphocytes makes it difficult to monitor and calculate cell numbers accurately using the Cell-IQ system. We therefore used a Cell Counting Kit-8 (CCK-8) assay (Dojindo, Kumamoto, Japan) to assess the viability of PBMCs, according to the supplier’s instructions. Briefly, cultured PBMCs were seeded at 56103 cells/well on 96-well plates. Cultures were treated with TGF-b1 or left untreated as controls. After 72 h, CCK-8 reagents were added to each well and the plates were incubated for 4 h. Cell counts were then determined for five wells per exper.S were adjusted to 56105 cells/mL, seeded in the bottom chambers of 6-well plates and incubated for 8 h to allow attachment. Inserts containing 56105 cells/mL of cultured PBMCs were then transferred to the top chambers and cocultured for another 24 h in FBS-free conditional medium or complete medium. As negative controls, inserts with PBMCs were placed on wells with the same culture medium in the absence of cancer cells, and wells with GC cells were left without inserts. The cell count in the monoculture group was double that in the coculture group, to ensure similar cell numbers in all groups. Supernatants and cells were collected separately after 24 h for further use.Ethics StatementPatients who received radiochemotherapy, suffered from other cancers, or who had a family history of GC were excluded from the study. Written informed consent was obtained from all the subjects. The project was approved by the Research Ethics Committee of Zhongshan Hospital [28].Immunohistochemistry (IHC)TGF-b1 and TGF-b2 protein levels were examined by IHC in 4-mm-thick paraffin sections cut from a single selected block containing neoplastic and non-neoplastic gastric tissues. Samples were routinely dewaxed and hydrated. After blocking of endogenous peroxidase activity, antigens were retrieved by heating with ethylenediamine tetraacetic acid (pH = 9.0). Antigens were subsequently detected using a standard staining procedure (EnVisionTM Detection Kit, Dako, CA, USA). Rabbit polyclonal antibodies were used to detect TGF-b1 and TGF-b2 (all dilutions 1:100; Santa Cruz Biotechnology, CA). For antibody-negative controls, the primary antibodies were substituted with normal rabbit serum. Cases were regarded as positive if at least 5 of dysplastic or cancer cells displayed cytoplasmic staining for TGFb1 or TGF-b2 at 6100 magnification.Cell Proliferation AssayA Cell-IQ cell culturing platform (Chip-Man Technologies, Tampere, Finland), equipped with a phase-contrast microscope (Nikon CFI Achromat phase contrast objective with 610 magnification, Nikon, Japan) and a camera, was used to detect the growth of tumor cells, as described previously [30]. Briefly, GC cells were cultured on 24-well plates (16104 cells/well) for 24 hTGF-b Roles in Tumor-Cell Interaction with PBMCsand then treated with TGF-b1 (Peprotech, USA) at 25 ng/mL. Control groups were left untreated. Cells were then incubated for a further 72 h in the Cell-IQ system. Images were captured at 30min intervals for 72 h, controlled by Image software (Chip-Man Technologies), and analyzed using freely-distributed image software (McMaster Biophotonics Facility, Hamilton, ON), using the manual tracking plug-in created by Fabrice Cordelieres (Institut Curie, Orsay, France). The Cell-IQ system automatically discriminates the dividing and stable cell stages, and calculates the total cell numbers during proliferation. 
Eight images were analyzed for each group. The mobility of lymphocytes makes it difficult to monitor and calculate cell numbers accurately using the Cell-IQ system. We therefore used a Cell Counting Kit-8 (CCK-8) assay (Dojindo, Kumamoto, Japan) to assess the viability of PBMCs, according to the supplier’s instructions. Briefly, cultured PBMCs were seeded at 56103 cells/well on 96-well plates. Cultures were treated with TGF-b1 or left untreated as controls. After 72 h, CCK-8 reagents were added to each well and the plates were incubated for 4 h. Cell counts were then determined for five wells per exper.