Fig 2: Activation of the Erk pathway is involved in the DUSP9 silencing-mediated tumor growth of CRC. (A,B) Colony formation assays for cell growth ability of HCC cells upon treatment with the specific Erk signaling inhibitor PD98059 or activator Curcumin. (C) Left panel: representative IHC images of DUSP9 and p-Erk in tumor tissues of CRC. Right panel: relationship between the expression of DUSP9 and p-Erk was analyzed based on IHC staining. Scale bars, 50 μm. (D) MTS assays for cell growth ability of HCC cells upon treatment with the specific Erk signaling inhibitor PD98059 or activator Curcumin. (E) Western blot analyses for expressions of cell apoptosis-related markers in both SW480 and LoVo cells with treatment as indicated. *P < 0.05, **P < 0.01.

Fig 3: DUSP9 promoter hypermethylation contributes to DUSP9 silencing in human CRC. (A) MethHC (the human pan-cancer methylation database) was used to predict the methylation status of DUSP9 gene promoter in a variety of cancers, including bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), colon adenocarcinoma (COAD), head and neck squamous cell carcinoma (NHSC), kidney renal clear cell carcinoma (KIRC), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), thyroid carcinoma (THCA), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), pancreatic adenocarcinoma (PAAD), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), sarcoma (SARC), skin cutaneous melanoma (SKCM), stomach adenocarcinoma (STAD), and uterine corpus endometrial carcinoma (UCEC). *P < 0.05, **P < 0.01. Red represents tumor and green represents normal tissue. (B) MethHC was used to predict the CpG island of DUSP9, which extends from –58 to –480 from TSS. Each red tick mark represents one CpG site. The arrows indicate the TSS. CpG island prediction criteria used: island size >100, GC percent >50.0, Obs/Exp >0.6. (C) To determine the methylation level, BSP was carried out on 11 CpG sites extending from –230 to –337 from TSS (underlined). (D) Bisulfite sequencing evaluation of CpG island methylation of the 11 CpG sites of DUSP9 promoter in CRC (T = 12) and normal intestinal mucosa (N = 12). White spots, unmethylated CpG; black spots, methylated CpG. (E) Bisulfite sequencing was carried out on 11 CpG sites in SW480 cells following treatment with 5-aza-dC for 72 h to determine the methylation level. *P < 0.05, **P < 0.01. (F) Treatment with 5-aza-dC can lead to DNA demethylation and restore DUSP9 expression in SW480 cells. TSS, transcription start site.

Fig 4: The schematic diagram of DUSP9 in CRC. In normal colorectal cells, the promoter of DUSP9 is hypomethylated, and the expression level of DUSP9 is relatively high, which inhibits cell proliferation, migration, and invasion by dephosphorylating ERK. However, the hypermethylation status of CpG island in promoter of DUSP9 could lead to a significant decrease of DUSP9 expression levels in human CRC. Moreover, the increased miR-1246 in CRC further reduced the expression level of DUSP9, which promoted tumor cell proliferation, migration, and invasion by phosphorylating ERK. CRC, colorectal cancer.

Fig 5: DUSP9 suppresses tumor growth in vivo. (A) DUSP9 knockdown promoted growth of SW480 xenograft tumors, which were generated by injecting SW480-shDUSP9 cells or SW480-Control cells. The growth of xenograft tumors was measured by volume, and the tumor weight was measured. (B) DUSP9 overexpression inhibited growth of LoVo xenograft tumors, which were generated by injecting LoVo cells overexpressing DUSP9 or carrying a control vector. The growth of xenograft tumors was measured by volume, and the tumor weight was measured. (C,D) HE, IHC, and PCNA staining of the xenograft tumors are shown. Scale bars, 50 μm. *P < 0.05, **P < 0.01, and ***P < 0.005.