Cisplatin resistance in gastric cancer cells is associated with HER2 upregulation-induced epithelial-mesenchymal transition

Abstract

Cisplatin remains to be primary chemotherapeutic drug for gastric cancer patients, especially for advanced stage ones. However, primary or acquired resistance often occurs with the mechanisms being not well understood, which results in relapse of the cancer and poor survival. Herein, we found that HER2 upregulation was associated with cisplatin resistance. We observed that cisplatin-resistant gastric cancer cells underwent a morphological change similar to epithelial-mesenchymal transition (EMT) which is mediated by HER2 overexpression. When specific monoclonal antibody Herceptin, small molecular targeted drug CP724714, or small interfering RNA against HER2 was applied, the EMT-like phenotypic change was dramatically reversed. More importantly, the IC50 and Resistance Index of resistant gastric cancer cells to cisplatin were also decreased by any of these treatments.We demonstrated that expression and amplification of HER2 positively correlated with expression of EMT-related transcription factor Snail in gastric cancer tissues. Furthermore, for the first time, we found that HER2/Snail double positive gastric cancer patients had poorer survival than single positive or double negative counterparts, which provided experimental evidence for the necessity of HER2/Snail double testing in gastric cancer. In conclusion, this study provides some clues of the association of cisplatin resistance with HER2 upregulation-induced EMT in gastric cancer cells.

Figure 1. Cisplatin-resistant gastric cancer cell models display higher motility.

(A) Cell growth curve of cisplatin-resistant MGC803 and AGS cells, versus parental MGC803 and AGS cells. Values represented the mean ± SD from two independent experiments with triplicate samples. DDP, cisplatin. (B) IC₅₀ and RI for DDP in drug resistant and parental gastric cancer cells. Values represented the mean ± SD from three independent experiments with triplicate samples. RI, resistance index. (C) Cisplatin-resistant gastric cancer cells exhibit higher wound healing ability. Photos were taken at 0 and 24 h after wounding. (D) Summary of wound healing assays. Results were presented as the mean ± SD of two independent experiments, **P < 0.01 compared with parental cells. (E) Cisplatin-resistant gastric cancer cells possess higher cell migratory capacity in vitro. The results were mean ± SD of three independent experiments with triplicate wells. **P < 0.01.

Figure 2. Cisplatin-resistant gastric cancer cells exhibit an EMT-like phenotypic change.

(A) Representative images of cellular morphology of cisplatin-resistant MGC803/DDP and AGS/DDP cell lines, plus parental MGC803 and AGS cells. (B) FITC-phalloidin staining of F-actin in cisplatin-resistant MGC803/DDP and AGS/DDP cell lines, and parental MGC803 and AGS cells. (C) Epithelial marker E-cadherin, ZO1 and EMT-related transcription factor Snail were examined by Western blot.

Figure 3. HER2 is overexpressed and both Herceptin incubation and siRNAs against HER2 reversed the EMT in cisplatin-resistant gastric cancer cells.

(A) Western blot of HER2 in MGC803/DDP and AGS/DDP cell lines, and parental MGC803 and AGS cells. GAPDH was used as loading control. (B) Relative amplification of HER2 in parental and cisplatin-resistant cells. Results were presented as the mean ± SD of two independent experiments, *P < 0.05 compared with parental cells. (C) Herceptin incubation abrogated EMT morphology in MGC803/DDP cells. Representative images of cellular morphology of MGC803/DDP cells, plus parental MGC803 cells treated with Herceptin for 24 h were shown. Hercep represents Herceptin. (D) Herceptin incubation reversed EMT morphology in MGC803/DDP cells. Representative images of FITC-phalloidin staining of F-actin after 100 μg/ml Herceptin treatment for 24 h were shown. (E) Effects of Herceptin on levels of ZO1 and Snail. (F) CP724714 treatment abolished EMT-like cell morphology. Cells were incubated with 10 μM CP724714 for 24 h. Representative images of cellular morphology of were captured. (G) CP724714 treatment abrogated EMT-like cell morphology. Cells were treated as in Fig. 4F, FITC-phalloidin staining of F-actin was performed and typical images were shown. (H) Effects of CP724714 on ZO1 and Snail. (I) CP724714 inhibited the increased cell migration in cisplatin-resistant gastric cancer cells. Representative images were presented here. (J) CP724714 inhibited the increased cell migration in cisplatin-resistant gastric cancer cells. Values represented the mean ± SD from three independent experiments with triplicate samples. *P < 0.01.

Figure 4. SiRNAs against HER2 could abrogate the EMT phenotype and cell invasiveness in cisplatin-resistant gastric cancer cells.

(A) Validation of silencing efficiency of HER2. Cells were transiently transfected with 50 nM HER2-specific siRNAs #1-#3 or a control siRNA for 48 h. (B) Knock-down of HER2 abrogated EMT morphology. Cells were transiently transfected with 50 nM specific siRNAs targeting HER2 for 48 h, Representative images of cellular morphology were shown. (C) Knock-down of HER2 abolished EMT phenotype. Cells were treated as in Fig. 4B. FITC-phalloidin staining of F-actin were performed and representative images were shown. (D) Effects of HER2 knock-down on the levels of ZO1 and Snail. (E) Effects of HER2 knock-down on cell migration. Representative images of transwell cell migration assay were shown. (F) Summary of Fig. 4E. Values represented the mean ± SD from three independent experiments with triplicate samples. *P < 0.01.

Figure 5. HER2 inhibition significantly reverses cisplatin resistance in MGC803/DDP cells.

(A) Cell growth curve of cisplatin-resistant MGC803 cells with different treatments. Values represented the mean ± SD from two independent experiments with triplicate samples. (B) Effects of different treatments on IC₅₀ and RI. Values represented the mean ± SD from three independent experiments with triplicate samples.

Figure 6. HER2/Snail double positive gastric cancer patients have unfavorable outcome.

(A) Representative immunohistochemical staining of HER2 and Snail by respective antibody in human gastric tissues. Original magnification, 100 × . (B) Representative FISH staining of HER2 by probe in human gastric tissues. Original magnification, 400 × . HER2 amplification was determined as a ratio of HER2 (red dots) and chromosome 17 centromere (green dots) signal counts. Ratios < 2 were determined as no amplification with FISH, and those >2 as positive amplification with FISH. “a” represents no amplification; “b” represents positive amplification. (C) Chi-square analysis of the correlation of HER2 expression or amplification with Snail expression. (D) Kaplan-Meier Curve analysis of HER2 amplification with OS for patients with gastric cancer by the log-rank test. (E) Kaplan-Meier estimation of Snail overexpression with OS for patients with gastric cancer by the log-rank test. (F) Kaplan-Meier estimation of HER2/Snail double positive expression versus single positive or both negative counterparts with OS for patients with gastric cancer by the log-rank test.