CHK1 expression in Gastric Cancer is modulated by p53 and RB1/E2F1: implications in chemo/radiotherapy response

Abstract

Radiation has a limited but relevant role in the adjuvant therapy of gastric cancer (GC) patients. Since Chk1 plays a critical function in cellular response to genotoxic agents, we aimed to analyze the role of Chk1 in GC as a biomarker for radiotherapy resistance. We analyzed Chk1 expression in AGS and MKN45 human GC cell lines by RT-QPCR and WB and in a small cohort of human patient's samples. We demonstrated that Chk1 overexpression specifically increases resistance to radiation in GC cells. Accordingly, abrogation of Chk1 activity with UCN-01 and its expression with shChk1 increased sensitivity to bleomycin and radiation. Furthermore, when we assessed Chk1 expression in human samples, we found a correlation between nuclear Chk1 accumulation and a decrease in progression free survival. Moreover, using a luciferase assay we found that Chk1's expression is controlled by p53 and RB/E2F1 at the transcriptional level. Additionally, we present preliminary data suggesting a posttranscriptional regulation mechanism, involving miR-195 and miR-503, which are inversely correlated with expression of Chk1 in radioresistant cells. In conclusion, Chk1/microRNA axis is involved in resistance to radiation in GC, and suggests Chk1 as a potential tool for optimal stratification of patients susceptible to receive adjuvant radiotherapy after surgery.

Figure 1. CHK1 mRNA levels are high in GC cells resistant to radiation (IR) and bleomycin (BLM).

(A) Survival of AGS and MKN45 cells after CDDP, 5-FU, BLM or IR treatment. AGS (○) and MKN45 (●) cells were treated with increasing amounts of CDDP (0–10 μg/ml), 5-FU (0–1000 μM), BLM (0–100 μg/ml) or IR (0–8 Gy). 48 h after treatment, the percentage of viable cells was quantified by the MTS method. Data represent the mean values obtained in two experiments performed in quadruplicate. (B) Clonogenic assay in AGS and MKN45 cells 13 days after irradiation with different doses of Gy (0–8); the graph shows the percentage of CFU (colony forming units). Representative images of AGS and MKN45 cells, 13 days after irradiation (4 Gy) are shown. The arrows point to abnormal morphology in AGS cells. Cleavage of PARP-1 was detected by western blot (WB) in cells harvested 24 h after 4 and 8 Gy IR. Activation of Caspase 3 was detected in the same extracts as above, running under the same experimental conditions. (Full length blot is included in supplementary information). α-Tubulin was used as a loading control. (C) RT-QPCR analysis of CHK1 expression in asynchronous cultures of AGS and MKN45 cells. The graph shows the relative levels of CHK1’s mRNA compared to normal tissue (NT), and using GAPDH as endogenous control. WB for both Chk1 isoforms (Chk1 and Chk1-S) in whole cell extracts from asynchronous cultures of AGS and MKN cells. α-Tubulin was used as a loading control.

Figure 2. Chk1 inhibition sensitizes GC cells to BLM or IR.

(A) Survival rates in AGS and MKN45 cell lines, after treatment with BLM (0–100 μg/ml) and in the presence (●) or absence (○) of 100 nM UCN-01. Histogram: Viability percentage for AGS and MKN45 cells treated with increasing doses of UCN-01 (0–600 nM) in the presence or absence of BLM (3 μg/ml and 10 μg/ml for AGS and MKN45 cells respectively). (B) Cell cycle profile after inhibition of Chk1 by treatment with the inhibitor UCN-01 (100 nM for AGS or 300 nM for MKN45 cells) or after silencing Chk1’s expression by transient transduction with a lentivirus carrying shRNA- Chk1 for 72 hours. One hour after, cells were treated with vehicle or with BLM (3 ug/ml) for 24 hours. Plots are representative of an experiment performed twice in duplicate. AP: Apoptotic Cells, G2/M: cells in G2 or Mitosis. (C) Percentage of apoptotic cells in both cell lines after irradiation, UCN-01 treatment and UCN-01 plus IR. Table containing G2/M and S accumulation index in both cell lines after IR and with or without UCN-01 treatment.

Figure 3. Chk1’s promoter activity is regulated by E2F1 and p53 in GC cell lines.

(A) Cells were transfected with increasing doses (0–1 μg) of an E2F1 expression vector (pCMV-E2F1), the corresponding empty control expression vector, and 200 ng of pGL3-F0. The graph shows activity levels, relative to that of PGL3-F0. (B) AGS and MKN45 cells were transfected with 200 ng of PGL3-F0, 500ng pCMV-E2F1 and increasing doses of p53’s expression vector (0–1 μg). Results are presented as activity level, relative to that of the empty pGL3-Luc reporter in the presence of E2F1. (C) Left graph: MKN45 cells were transfected with 200 ng pGL3-F0 and expression vectors for E2F1 (250 ng), E2F1 plus p53 or E2F1 plus DNp53. Right graph: Both MKN45 and AGS cells were transfected with 200 ng pGL3-F0 and expression vectors for E2F1 (250 ng) or HEY1 (200 ng). All data are presented as the average of at least three independent experiments assayed in triplicate ± SEM. p < 0,05 versus empty vector control (Student’s t test).

Figure 4. The RB1/E2F1 axis controls activation of Chk1 promoter.

(A) Schematic representation of the constructs used in the transfection experiments. AGS and MKN45 were transfected with 250 ng of the indicated construction, and luciferase activity was measured 24 h later. The histograms show relative activity normalized to the empty vector in both AGS and MNK45 cells. (B) Expression of both RB1 and E2F1 was detected by WB using specific antibodies in AGS and MKN45 cells. α-Tubulin was used as loading control. The experiments were repeated three times with similar results. The graph on the center represents relative expression level (log 2-copy number) of RB1 in control (Hs 738St/Int) and tumoral AGS and MKN45 cells obtained from the Oncomine database. The graph on the right shows luciferase activity measurements in AGS and MKN45 cells after transfection with 250 ng of F0 and 100 ng of PCEFL-E1A. The data are presented as activity relative to the values found for the empty vector in transfected cells. The experiment was performed twice in triplicate.

Figure 5. Nuclear Chk1 protein levels correlate with poor clinical outcome in human gastric cancer.

(A) Immunohistochemical staining of Chk1 in representative carcinoma GC specimens: Nuclear positive and negative staining at 20 HPF (a,b respectively); nuclear positive and negative staining at 40HPF (a’,b’ respectively). The arrows point to strong Chk1 staining in the nucleus. (B) Kaplan-Meier curves of progression-free survival (PFS) in patients with high or low expression of nuclear Chk1 in their gastric tumors. Survival curves were statistically different when analyzed by the Breslow. IBM SPSS Statistics 22 software.