GINS complex subunit 4, a prognostic biomarker and reversely mediated by Krüppel-like factor 4, promotes the growth of colorectal cancer

Abstract

GINS complex subunit 4 (GINS4) is essential for DNA replication initiation and elongation in the G₁ /S phase cell cycle in eukaryotes, however, its functional roles and molecular mechanisms remain unclear in many aspects. Our study was designed to investigate the clinical significance, biological function, and molecular mechanism of GINS4 in colorectal cancer (CRC). First, we confirmed that GINS4 expression was significantly overexpressed in CRC cells and tissues. The immunohistochemical results in tissue microarray from 106 CRC patients showed that a high level of GINS4 expression was positively correlated with advanced T stage, higher tumor TNM stage, and poor differentiation. The results from univariate and multivariate survival analysis models based on 106 CRC patients revealed that GINS4 might serve as an independent prognostic indicator for overall survival and disease-free survival of CRC patients. Moreover, downregulated GINS4 can inhibit growth and the cell cycle and accelerate cell apoptosis progression in vitro as well as inhibit tumorigenesis in vivo. Besides, our results also indicated that Krüppel-like factor 4 (KLF4) can negatively regulate GINS4 expression at the transcriptional level and the KLF/GINS4 pathway might play a vital role in the growth and prognosis of CRC.

Keywords: GINS4; KLF4; biomarker; colorectal cancer; growth.

Figure 1

Overexpression of GINS complex subunit 4 (GINS4) indicates poor clinical outcome of colorectal cancer (CRC). A, Oncomine data showed that GINS4 expression was elevated in CRC tissues from the Kaiser Colon dataset and colon adenomas from the Sabates‐Bellver Colon dataset. B, Relative expression of GINS4 in 63 paired CRC tissues was significantly higher than in adjacent normal CRC tissues, and was higher in 49 (77.78%) fresh CRC clinical samples detected by quantitative real‐time PCR. C, Representative images of GINS4 levels in CRC tissues and matched normal mucosa (magnification, 100× and 400×)from the tissue microarray (including 106 CRC tissues and matched normal tissues). D, Level of GINS4 in CRC tissues was determined by western blotting. E‐G, GINS4 expression between T1 and T2 vs T3 and T4 (E, P = .022), grade I vs grade II and III (F, P = .024), TNM stage I and II vs TNM stage III and IV (G, P < .0001). H, Kaplan‐Meier survival analysis (log‐rank test) showed that CRC patients with strong positive expression of GINS4 had a lower overall survival and disease‐free survival than those with negative and weak positive expression.*P < .05, **P < .01, ***P < .001

Figure 2

Effect of shRNA on expression of GINS complex subunit 4 (GINS4). A, B, Expression of GINS4 and Krüppel‐like factor 4 (KLF4) in 8 CRC cell lines was detected by quantitative real‐time (qRT)‐PCR and western blot analyses. C, D, GINS4 knockdown in HCT8 and RKO cells was determined by qRT‐PCR and western blot analyses. **P < .01

Figure 3

GINS4 complex subunit 4 (GINS) knockdown inhibits colorectal cancer cell growth in vitro. A, CCK‐8 growth assay was used to examine the knockdown of GINS4 expression on HCT8 and RKO cells. B, EdU assay was used to analyze the proliferative activity of HCT8 and RKO cells. C, Colony formation assay was used to determine whether knockdown of GINS4 expression inhibited the ability of colony formation in HCT8 and RKO cells. Three independent experiments were carried out for each group. *P < .05, **P < .01, ***P < .001. OD, optical density

Figure 4

GINS4 complex subunit 4 (GINS) knockdown inhibits colorectal cancer growth in vivo. A, Tumors were removed 38 days after HCT8‐shGINS4‐1/HCT8‐scramble and RKO‐shGINS4‐1/RKO‐scramble cells were s.c. injected. B, GINS4 mRNA and protein level and Ki‐67 protein level in tumors with HCT8‐shGINS4‐1/RKO‐shGINS4‐1 was significantly lower than that in tumors with HCT8‐scramble/RKO‐scramble. C, Tumor weight of RKO and HCT8 cells. Tumor volumes of RKO and HCT8 cells measured every 7 days (5 times in total) were analyzed. D, Representative images of GINS4 immunohistochemical staining in tissues of xenografts (magnification, 200× and 400×). *P < .05, **P < .01

Figure 5

GINS complex subunit 4 (GINS4) knockdown suppresses colorectal cancer cell cycle progression and promotes cell apoptosis by inhibiting PI3K/AKT and MAPK/ERK pathways. A, Flow cytometry analysis of cell cycle of HCT8 and RKO cells and related data statistics. B, Flow cytometry analysis of cell apoptosis and cell cycle of HCT8 and RKO cells and related data statistics. C, Representative images of Ki‐67 immunohistochemical staining in tissues of xenografts from HCT8 and RKO cells (magnification, 200× and 400×). D, Different expression of pathways, cell cycle‐related, apoptosis‐related and growth marker proteins (pERK1/2, pAKT, cyclin D1, Bcl‐2, and Ki‐67) in HCT8 and RKO cells. All experiments were independently repeated in triplicate in each group. All data are represented as mean ± SD. *P < .05, **P < .01, #P > .05

Figure 6

Krüppel‐like factor 4 (KLF4) downregulates GINS complex subunit 4 (GINS4) expression by directly binding to its promoter. A, Two putative KLF4‐binding sites in the GINS4 promoter region and the 2 mutated binding sites. B, The Cancer Genome Atlas datasets show GINS4 was negatively correlated with KLF4 from 493 colon cancer and 171 rectal cancer patients (r = −0.23, P < .0001). C, mRNA and protein levels of GINS4 were reduced when KLF4 expression was overexpressed in HCT8 and RKO cells. D, Luciferase reporter assay indicated that KLF4 could suppress the promoter activities of GINS4 both in HCT8 and RKO cells. E, In HCT8 and RKO cells, ChIP assay indicated KLF4 binding sites were across the GINS4 promoter region. Immunoprecipitated DNA products were subjected to PCR analysis to amplify a 230‐bp region and presented in 2% agarose gel. *P < .05, **P < .01

Figure 7

Combination of Krüppel‐like factor 4 (KLF4) and GINS complex subunit 4 (GINS4) levels is useful for predicting clinical prognosis in colorectal cancer (CRC) patients. A, Relative mRNA expression of KLF4 in 63 CRC tissues compared to paired adjacent normal tissues. B, Correlation between KLF4 and GINS4 mRNA in 63 CRC tissues was also negative (r = −0.329, P = .0084). C, D, Relationship between KLF4 and GINS4 in CRC tissues (200×) (C; r = −0.210, P = .0308) and normal mucosa (D; r = −0.230, P = .0177), detected by immunohistochemistry, were both weakly negatively related. E, Prognostic values of GINS4 combined with KLF4. *P < .05, **P < .01, ***P < .001