Decreased level of RASSF6 in sporadic colorectal cancer and its anti-tumor effects both in vitro and in vivo

Abstract

Ras-association domain family protein 6 (RASSF6) is a member of tumor suppressor RASSFs family with a wide range of function from RAS interaction, Hippo signaling involvement to cell cycle and apoptosis regulation. RASSF6 is reported inactivated in various types of cancer. However, whether RASSF6 is associated with colorectal cancer and the underlying mechanisms have yet to be investigated. In our previous exome sequencing study, we found a somatic loss-of-function (LoF) mutation in RASSF6 in one sporadic colorectal cancer (sCRC) patient, and two missense mutations in deep sequencing group of sCRC samples, implying the possibility that RASSF6 may be involved in the pathogenesis of sCRC. In this study, we demonstrate that RASSF6 acts as a tumor suppressor in colon cancer cells. Decreased level of RASSF6 was observed in adenocarcinoma compared to normal tissues, especially in advanced tumor cases. Further experiments showed exogenous introduction of RASSF6 into LoVo cells suppressed cell proliferation, migration, invasion, and induced apoptosis in vitro as well as tumor growth in vivo. In contrast, knockdown of RASSF6 in HT-29 cells showed the opposite effects. Taken together, our results suggest, in addition to epigenetics changes, functional somatic mutations may also contribute to the downregulation of RASSF6 and further participate in the pathogenesis of sCRC. RASSF6 may serve as a novel candidate against tumor growth for sCRC.

Keywords: RASSF6; apoptosis; loss-of-function mutation; proliferation; sporadic colorectal cancer.

Figure 1. Somatic mutations detected in whole exome sequencing sample and deep sequencing samples

(A) Resequencing of RASSF6 mutation in one of the whole exome sequencing samples. A heterozygous frameshift mutation (c.367insA (T), p.R123fs*20) in exon 5. (B) A heterozygous missense mutation (c.746A > T, p.D249V) in exon 9 identified in one of the deep sequencing samples. (C) A heterozygous missense mutation (c.992T > A, p.I331K) in exon 11. (D) Overview of 3 somatic mutations in protein level. RASSF6 contains two domains: RA domain and SARAH domain. △ represents frameshift mutation while ☆ represents missense mutation.

Figure 2. RASSF6 expression in colon tissues and CRC cell lines

(A) Immunohistochemical staining of RASSF6 expression in tissue microarray. Weak or negative staining for RASSF6 was detected in adenocarcinoma. (B) Adjacent benign glands were showing strong RASSF6 staining. (A and B) are representative images from the same patient. (C) Comparison of staining score between 52 cases of adenocarcinoma and adjacent normal tissue. Staining score is classified as weak or negative (0), mild (1), moderate (2), and strong (3) (***P < 0.001, paired t test). (D) Transcriptional level of RASSF6 in colorectal cancer cell lines (SW480, LoVo, and HT-29) was analyzed using quantitative real time-PCR. RASSF6 was expressed in all cell lines, relatively high in HT-29 cells and low in LoVo cells, with GAPDH as the normalized reference gene. (E) Western blot analysis of RASSF6 protein levels that normalized to GAPDH expression. The result is consistent with transcriptional level analysis. All data are shown as mean ± SE.

Figure 3. RASSF6 inhibits cancer cell proliferation and induces apoptosis

(A) The overexpression and knockdown efficiency were confirmed by western blot, with anti-flag antibody and specific anti-RASSF6 antibody used respectively. (B) LoVo and HT-29 cell proliferation was determined by CCK8 assays after transfection, on day 1, 2, 3, 4. (C) The effects of RASSF6 on the colony forming ability of LoVo and HT-29 cells. Images are representative of three independent experiments. (D and E) The effects of RASSF6 on LoVo and HT-29 cells apoptosis was detected by flow cytometry. (F) Tumor volume was assessed once a week with a total period of 5 weeks (n = 5 or 4). (G) Images of nude mice bearing the tumors and isolated tumors from injected mice. (H) Weights of the isolated tumor (n = 5 or 4). All data are showed as mean ± SE. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 4. RASSF6 inhibits the migration and invasion ability of colorectal cancer cells

(A) Wound scratch assays were performed in LoVo cells with transfection of empty vector as control or RASSF6 (Left, A), and HT-29 cells with transfection of scrambled control siRNA or RASSF6 siRNA respectively. Images were visualized at 0 h, 24 h, and 48 h at a magnification of 100 ×. RASSF6 highly-expressed cells showed lower migration ability than RASSF6-low cells. (B) The migration ability was also assessed using Transwell chamber without Matrigel (18 h for LoVo and 48 h for HT-29 cells). (C) The effects of RASSF6 on the invasion ability of cancer cells were assessed by Matrigel-invasion assay (24 h for LoVo and 48 h for HT-29 cells). Representative images were visualized at a magnification of 100 ×. All data are showed as mean ± SE of three independent experiments, and for every single experiment, the data are mean of at least three fields, *P < 0.05.