Exploring the correlation and mechanism of natural killer cell cytotoxic sensitivity against gastric cancer

Abstract

Background: Human natural killer (NK) cells have attracted widespread attention as a potential adoptive cell therapy (ACT). However, the therapeutic effects of NK cell infusion in patients with solid tumors are limited. There is an urgent need to explore a suitable new treatment plan to overcome weaknesses and support the superior therapeutic activity of NK cells.

Methods: In this study, the mechanisms underlying the susceptibility of gastric cancer (GC) cell lines AGS, HGC-27, and NCI-N87 to NK cell-mediated cytotoxicity were explored.

Results: Lactic dehydrogenase (LDH) release assays showed that all three GC cell lines were susceptible to the umbilical cord blood NK (UCB-NK) cells, and HGC-27 cells with high CD56 expression were the most sensitive to UCB-NK, followed by NCI-N87 and AGS. When the expression of CD56 in HGC-27 cells decreased, the lytic activity of NK cells in HGC-27 cells was abating. In addition, combining oxaliplatin with NK cells produced additive anti-tumor effects in vitro, which may have resulted from oxaliplatin-induced NK group 2 member D (NKG2DL) upregulation in GC cells. These results of cytotoxicity activity showed that inhibition of CD56 expression might suppress the sensitivity of GC cells to NK cell-mediated cytotoxicity, and upregulation of the expression of NKG2DL on the surface of GC cells by oxaliplatin could enhance the killing sensitivity of NK cells.

Conclusion: Collectively, our study provides a deeper theoretical foundation and a better therapeutic strategy for NK cell immunotherapy in the treatment of human GC.

Keywords: CD56 (neural cell adhesion molecule NCAM); Gastric cancer; Natural killer group 2 member D (NKG2D) ligand (NKG2DL); Oxaliplatin; Umbilical cord blood natural killer (UCB-NK) cells.

Figure 1. Representative flow cytometric scatter of CD56 and CD3 double staining. Percentage of CD3^-CD56⁺-expressing NK cells in umbilical cord blood mononuclear cells cultured for 0 days and 14 days.

Figure 2. Flow cytometry analysis of the UCB-NK cell phenotype. (A) NKG2A, TIGIT, and PD-1 expression on CD3^-CD56⁺ NK cells. (B) NKG2D, CD226, and CD16 expression on CD3^-CD56⁺ NK cells. (C) NKp44, NKp46, and NKp30 expression on CD3^-CD56⁺ NK cells.

Figure 3. The expression of surface molecules (CD56 and HER2) and NK cell receptors-ligands (HLA-E: ligand of NKG2A receptor; CD155: ligands of DNAM-1/TIGIT/CD96 receptor; CD112: ligands of DNAM-1/TIGIT) in gastric cancer cell lines was determined using flow cytometry.

Figure 4. Oxaliplatin inhibited the proliferation of gastric cancer cells. Human gastric cancer cell lines HGC-27 (A), NCI-N87 (C), and AGS (E) were treated with different concentrations of oxaliplatin over a period of 24, 48, and 72h, and total viable cell counts were determined by CCK-8 assay. Cytotoxicity of UCB-NK cells against gastric cancer cell lines and oxaliplatin pretreated gastric cancer cell lines HGC-27 (B), NCI-N87 (D), and AGS (F) was measured by the lactic dehydrogenase (LDH) release assay at different effector to target ratios (E/T). Statistical analysis was done by two-tailed Student’s t-test, error bars denote S.D. *p < 0.05; ***p < 0.001; ns, no significance.

Figure 5. shRNA-mediated inhibition of CD56 expression in HGC-27 cells. (A) The protein expression levels of CD56 in HGC-27 cells were determined by flow cytometry. (B) Normalized protein expression of CD56 in HGC-27 cells and specific shRNA-treated cells; Relative mRNA levels of CD56 in HGC-27 cells and specific shRNA-treated cells determined by quantitative real-time PCR analysis. (C) Cytotoxicity of UCB-NK cells against gastric cancer cell lines HGC-27 and specific shRNA-treated HGC-27 cells was measured by the lactic dehydrogenase (LDH) release assay at an effector to target ratio (E/T) of 2:1. ** p < 0.01; *** p < 0.001; ns, no significance.

Figure 6. Effects of UCB-NK cells and oxaliplatin on xenotransplantation of human gastric cancer HGC-27 cells in NOG mice. (A) Schematic diagram of the treatment program of the HGC-27-bearing mice. (B) At the end of the experiment, resected tumors from each group were collected and imaged. (C) Tumor growth curves during the treatments. *p < 0.05; ns, no significance. (D) Tumor weights for each group were measured on the last day of the experiment; ns, no significance.