Lentivirus-mediated RASSF1A expression suppresses aggressive phenotypes of gastric cancer cells in vitro and in vivo

Abstract

Loss of Ras association domain family protein 1 isoform A (RASSF1A) expression is associated with the development of a variety of human cancers and the expression of carcinoembryonic antigen (CEA) frequently occurs in gastric cancer. This study investigated the effects of RASSF1A expression restoration using a hypoxia-inducible CEA promoter-driven vector on xenograft tumor growth in nude mice and on the in-vitro regulation of gastric cancer cell viability, cell cycle distribution, apoptosis, colony formation and invasion capacity. The data showed that the level of CEA mRNA and protein was much higher in gastric cancer SGC7901 cells than in a second gastric cancer cell line, MKN28, or in the MCF-10A normal epithelial breast cell line. RASSF1A expression was restored in SGC7901 cells compared with the negative control virus-infected SGC7910 cells. RASSF1A expression restoration significantly inhibited gastric cancer cell viability, colony formation and invasion capacity, but induced cell cycle arrest and apoptosis in vitro, especially under hypoxic culture conditions. At the gene level, restoration of RASSF1A expression under hypoxic culture conditions significantly suppressed matrix metalloproteinase-2 expression and prevented cyclinD1 expression. A nude mouse xenograft assay showed that the restoration of RASSF1A expression reduced gastric cancer xenograft formation and growth. In conclusion, the restoration of RASSF1A expression using a hypoxia-inducible and CEA promoter-driven vector suppressed aggressive phenotypes of gastric cancer cells in vitro and in vivo. These results suggest that LV-5HRE-CEAp-RASSF1A gene therapy may be a promising novel approach to treat advanced gastric cancer.

Figure 1

Expression of endogenous CEA and RASSF1A mRNA or protein in SGC7901, MKN28 and MCF-10A cells. (a) Quantitative reverse-transcription (qRT-PCR) analysis of CEA mRNA levels ***P<0.001 vs MKN28 and MCF-10A. (b) Quantified data for CEA protein levels in SGC7901, MKN28 and MCF-10A cells. (c) Western blot analysis of RASSF1A protein levels in SGC7901, MKN28 and MCF-10A cells. ***P<0.001 vs MCF-10A; ^##P<0.01 vs MKN28.

Figure 2

Overexpression of RASSF1A mRNA in SGC7901 and MKN28 cells. (a) Quantitative reverse transcription (qRT-PCR) analysis of RASSF1A mRNA in RASSF1A stably transfected cells. ***P<0.001 vs SGC7901, SGC7901/NC, MKN28 and MKN28/5HC-RASSF1A cells; ^•••P<0.001 vs SGC7901/5HC-RASSF1A cells cultured under normoxic conditions. (b) Western blot analysis of RASSF1A protein in RASSF1A stably transfected cells under normoxic (N) and hypoxic (H) conditions. (c) Luciferase activity assay. 5HRE-CEAp activity was measured in SGC7901, MKN28 and MCF-10A cells. *P<0.05 vs MKN28 and MCF-10A cells under hypoxic conditions; **P<0.001 vs MKN28 cells under normoxic conditions; ***P<0.001 vs SGC7901 under normoxic conditions.

Figure 3

Effects of RASSF1A expression on SGC7901 and MKN28 cell viability. (a) Cell viability CCK-8 assay. SGC7901, SGC7901/NC and SGC7901/5HC-RASSF1A cells were grown in monolayer with or without CoCL₂ for 5 days and then subjected to cell viability analysis. (b) Growth inhibition of SGC7901, SGC7901/5HC-RASSF1A, MKN28 and MKN28/5HC-RASSF1A cells with or without CoCL₂ on the fourth day after treatment. **P<0.01 vs SGC7901/5HC-RASSF1A and SGC7901-CoCL₂; *P>0.05 vs MKN28/5HC-RASSF1A.

Figure 4

Effect of RASSF1A expression on the regulation of SGC7901 cell apoptosis and cell cycle distribution under normoxic or hypoxic conditions. (a) Flow cytometry apoptosis assay. SGC7901 cells were grown and subjected to tumor cell apoptosis analysis. **P<0.01 vs SGC7901 and SGC7901/NC cells under normoxic or hypoxic conditions and SGC7901/5HC-RASSF1A cells under normoxic conditions. (b) Flow cytometric cell cycle analysis. Cells were grown under normoxic or hypoxic conditions and subjected to cell cycle analysis.

Figure 5

Effect of RASSF1A expression on the regulation of tumor cell colony formation and migration capacity in normoxic (N) or hypoxic (H) conditions. (a) Colony formation assay. SGC7901, SGC7901/NC and SGC7901/5HC-RASSF1A cells were grown under normoxic or hypoxic conditions and subjected to colony formation assay. (b) Summarized data of a. *P<0.05 vs SGC7901 or SGC7901/NC cells under hypoxic conditions. (c) Transwell tumor cell migration assay. SGC7901, SGC7901/NC and SGC7901/5HC-RASSF1A cells were grown under normoxic or hypoxic conditions and subjected to a tumor cell migration assay. (d) Summarized data of c. ^★★P<0.01 vs SGC7901 or SGC7901/NC cells under normoxic conditions; **P<0.01 vs SGC7901/5HC-RASSF1A cells under normoxic conditions.

Figure 6

Effect of RASSF1A expression on the regulation of HIF-1α, caspase 3, MMP-2 and cyclinD1 protein expression in SGC7901 cells under normoxic (N) or hypoxic (H) conditions. (a) Western blot analysis of HIF-1α, caspase 3, MMP-2, cyclinD1 and β-actin in SGC7901, SGC7901/NC and SGC7901/5HC-RASSF1A cells. β-Actin was used as the internal control. (b) Summarized data of MMP-2 protein in cells. **P<0.01 vs SGC7901 or SGC7901/NC cells under hypoxic conditions and SGC7901/5HC-RASSF1A cells under normoxic conditions; (c) Summarized data of caspase 3 protein in cells. **P<0.01 vs SGC7901 or SGC7901/NC cells under hypoxic conditions; (d) Quantification of cyclinD1 protein expression in cells, **P<0.01 vs SGC7901/5HC-RASSF1A cells under normoxic conditions.

Figure 7

Effect of RASSF1A expression on the regulation of tumorigenicity in nude mice. (a) The mean tumor volume at the end of experiments. **P<0.01 vs NS and LV-NC groups. (b) The mean tumor weight at the end of experiments. **P<0.01 vs NS and LV-NC groups. (c) Image of xenograft tumors from each group. (d) Immunohistochemical analysis of RASSF1A expression in xenograft tumors and normal mouse liver tissues.