Fig 2: Lentivirus-mediated knockdown and overexpression of NCOA5 in CRC cells(A) Western blot analysis of NCOA5 in a panel of CRC cell lines. The lysates of HT29, HCT8, HCE8693, SW620 and SW480 CRC cells were immunoblotted with anti-NCOA5 or anti-GAPDH (a loading control) antibody. **, P<0.01 compared withHT29, HCT8, HCE8693 and SW480 group; *, P<0.05 compared withHT29, HCT8 and HCE8693 group, one-way repeated measures ANOVA, n=3 replicates per sample. (B) Fluorescence microscopic analysis of GFP expression. The GFP expression in NCOA5-silenced SW620 and NCOA5-overexpressed SW480 CRC cells was examined under fluorescence microscopy. Representative pictures of GFP and differential interference contrast (DIC) were shown. (C) Flow cytometric analysis of GFP expression. (D) Western blot analysis of lentivirus-mediated NCOA5 knockdown in SW620 CRC cells. The lysates of SW620-shNCOA5 1#, SW620-shNCOA5 2#, SW620-shNCOA5 3# and SW620-shNTC CRC cells were immunoblotted with anti-NCOA5 or anti-GAPDH (a loading control) antibody. **, P<0.01 compared withSW620-shNTC group; *, P<0.05 compared withSW620-shNCOA5 1#group, one-way repeated measures ANOVA, n=3 replicates per sample. (E) Western blot analysis of lentivirus-mediated NCOA5 expression in SW480 cells. The lysates of SW480-LV-NCOA5, SW480-LV and SW480 CRC cells were immunoblotted with anti-NCOA5 or anti-GAPDH (a loading control) antibody. **, P<0.01 compared withSW480-LV and SW480 group, one-way repeated measures ANOVA, n=3 replicates per sample. The expression level of NCOA5 in these Western blot assays was normalized to GAPDH and expressed as a NCOA5/GAPDH ratio. Data shown were representative of three independent experiments.

Fig 3: Knockdown of NCOA5 suppresses CRC cell migration and invasion, whereas forced expression of NCOA5 enhances these processes(A and B) Wound healing assay. The SW620-shNCOA5 2#, SW620-shNCOA5 3# and SW620-shNTC; SW480-LV-NCOA5 and SW480-LV CRC cells were scratched in the monolayer with a pipette tip. Wound closures were photographed at 0, 24 and 48 hours after wounding. The representative figures of wound healing assay were shown (A). The relative migratory ability of control group was presented (B). SW620-shNCOA5 2# compared with SW620-shNTC group: *, P<0.05 at hour 24, and **, P<0.01 at hour 48; SW620-shNCOA5 3# compared with SW620-shNTC group: **, P<0.01 at hour 24 and 48; SW480-LV-NCOA5 compared with SW480-LV group: *, P<0.05 at hour 24, and **, P<0.01 at hour 48, two-way repeated measures ANOVA, n=6 replicates per condition. (C and D) Transwell invasion assay. The representative figures of Transwell invasion assay were shown (C). The number of invaded tumor cells was counted (D). SW620-shNCOA5 2# and SW620-shNCOA5 3# compared with SW620-shNTC group: **, P<0.01, one-way repeated measures ANOVA; SW480-LV-NCOA5 compared with SW480-LV group: **, P<0.01, Student t test, n=3 replicates per condition, n=10 observations per replicate. Data shown were representative of four independent experiments.

Fig 4: Knockdown of NCOA5 inhibits CRC xenografted tumor growth, whereas overexpression of NCOA5 promotes its growthThe SW620-shNCOA5 3# and SW620-shNTC; SW480-LV-NCOA5 and SW480-LV CRC cells s.c. injected into nude mice. The tumor volume (A) was measured after implantation of tumor cells.SW620-shNCOA5 3# compared with SW620-shNTC group: *, P<0.05 at week 2, and **, P<0.01 at week 3 and 4; SW480-LV-NCOA5 compared with SW480-LV group: *, P<0.05 at week 2, and **, P<0.01 at week 3 and 4, two-way repeated measures ANOVA, n=6 replicates per condition. The xenografted tumors were removed (B) 4 weeks after tumor cell's implantation and tumor weight (C) was then measured. SW620-shNCOA5 3# compared with SW620-shNTC group: **, P<0.01; SW480-LV-NCOA5 compared with SW480-LV group: **, P<0.01, Student t test, n=6 replicates per condition. Data shown were representative of three independent experiments.

Fig 5: NCOA5 is highly expressed in CRC clinical tissues(A) Immunohistochemistry analysis of NCOA5 in CRC tissue specimens. The representative pictures (Case 3, +++; Case 9, ++; Case 8, +; and Case 4, -) of immunohistochemical staining were shown. T, colorectal tumor tissue; and N, adjacent non-tumor normal tissue. (B) Western blot analysis of NCOA5 in CRC tissue specimens. The representative pictures of Western blot analysis of NCOA5 in 12 pairs of T and N tissues derived from 12 CRC cases were shown. GAPDH was used as a loading control. (C) Real-time RT-PCR analysis of NCOA5. The RNAs derived from 70 pairs of CRC T and N tissues were subjected to SYBR Green I-based real-time quantitative PCR analysis of transcriptional expression of NCOA5. The level of NCOA5 mRNA was expressed as −ΔCTNCOA5=−(mean CTNCOA5−mean CTGAPDH). **, P<0.01 compared with N tissue, Student t test, n=3 replicates per sample. Data shown were representative of two independent experiments.