ISL1, a novel regulator of CCNB1, CCNB2 and c-MYC genes, promotes gastric cancer cell proliferation and tumor growth

Abstract

Islet-1 (ISL1) belongs to the LIM homeodomain transcription factor family, which is specifically expressed in certain tissue types only. Previously, we reported that ISL1 is aberrantly overexpressed in gastric cancer (GC). However, its role in GC is not clear. Here, we report that ISL1 is aberrantly upregulated not only in human gastric carcinoma tissues but also in some GC cell lines. Upregulated ISL1 expression enhanced xenografted gastric carcinoma development, while ISL1 knockdown inhibited GC growth in nude mice. ISL1 overexpression promoted GC cell proliferation, colony formation, and cell growth in soft agar, and facilitated cell cycle transition in GC cells, demonstrated an increase in the proportion of cells in the G2/M and S phases and a decrease in the proportion of cells in the G1 phase. Furthermore, we provide evidence that ISL1 is a novel regulator of the cyclin B1 (CCNB1), cyclin B2 (CCNB2) and c-myc (c-MYC) genes. ISL1 activated the expression of these genes in GC cells by binding to the conserved binding sites on their promoters or enhancers. The expression levels of the genes were decreased in response to ISL1 knockdown. Therefore, ISL1 may serve as a potential therapeutic target in GC.

Keywords: CCNB; ISL1; c-MYC; gastric cancer; proliferation.

Figure 1. Aberrant expression of ISL1 in GC tissues

(A) Representative IHC stainings of ISL1 expression in normal gastric mucosa (top, n = 12) and GC samples (bottom, n = 24). (B) Grayscale scanning analysis results of ISL1 expression in the tested cell lines. Grayscale scanning was performed on the western blots of three independent experiments for quantitative analysis. Representative western blots of three experiments are shown at the bottom. β-Actin served as the internal control. GES1, normal gastric cell line; others, GC cell lines. Bars represent the means ± SD (**p < 0.01 vs. GES1 cells).

Figure 2. ISL1 promoted colony formation in vitro

(A) Western blotting analysis of ISL1 levels in MKN28 and MGC803 cell lines with stable ISL1 overexpression (ISL1) or knockdown (si-ISL1) by transfection with pcDNA3.1-ISL1 and pLL3.7-ISL1-siRNA plasmids (cells transfected with pcDNA3.1 vector were used as the negative control), respectively. GAPDH levels served as the internal control. (B) PCF assay performed to determine the proliferation of MKN28 or MGC803 cells stably transfected with ISL1, si-ISL1, or the vector control (a–f). The quantification results are presented on the right. (C) SACF assay of MKN28 and MGC803 cells stably transfected with ISL1, si-ISL1, or vector (a–f). The quantification results are presented on the right. The data are the means ± SD from three independent experiments, each performed in triplicate. (**p < 0.01, *p < 0.05 vs. control).

Figure 3. ISL1 promoted tumor growth in nude mice

(A, B) MGC803 cells stably transfected with pcDNA3.1 (control), pcDNA3.1-ISL1 (ISL1), (C, D) MGC803 cells stably transfected with pLL3.7-Non-silencer (Non-silencer), or pLL3.7-ISL1-siRNA (si-ISL1). A total 1 × 10⁷ appropriate cells suspended in 200 μl PBS were injected subcutaneously into the right oxter flank of BALB/c nude mice. Tumors isolated at day 21 (ISL1 and control) or day 24 (si-ISL1 and Non-silencer) are shown in A and C Tumor size was measured at the indicated days post-injection and the results are shown in B and D (n = 3 at every time point in each group, *p < 0.05 vs. control or Non-silencer). (E, F)Western blot detection of ISL1 levels in tumor samples. GAPDH levels served as the internal control.

Figure 4. ISL1 promoted GC cell proliferation

(A–C) The cell proliferation rate as determined using CCK-8 assay. The values at each time point are normalized to that of the relative cells at time 0, which was set as 1. The data are the means ± SD from three independent experiments, each performed in triplicate (**p < 0.01, *p < 0.05 vs. control or Non-silencer at each time point). (D, E) Flow cytometry analysis of MKN28 and MGC803 cell cycle distributions. Values are the means ± SD of three independent experiments, each performed in triplicate (*p < 0.05 vs. control or Non-silencer). Control, ISL1, Non-silencer and si-ISL1 represent cells transfected with pcDNA3.1, pcDNA3.1-ISL1, pLL3.7-Non-silencer and pLL3.7-ISL1-siRNA plasmids, respectively.

Figure 5. ISL1 bound to CCNB1, CCNB2 and c-MYC

(A) MatInspector software analysis of the consensus binding site (TAAT box) for ISL1 on the human CCNB1 promoter, the CCNB2 promoter, and the c-MYC enhancer. Sequences containing mutant base pairs (boxes) were used to construct the luciferase reporter constructs mutant-CCNB1-luc, mutant-CCNB2-luc, and mutant-c-MYC-luc. (B, C) Luciferase reporter assay of ISL1 transcriptional activity on the luciferase reporter constructs (-luc) of CCNB1, CCNB2 and c-MYC in MKN28 cells (WT, wild-type plasmid; M, mutant plasmid). Luciferase activity on the reporter constructs was normalized to Renilla luciferase activity. (D) ChIP assay was performed with anti-ISL1 antibody using chromatin harvested from MKN28 cells. Extracted DNA was amplified using primers that cover the ISL1 binding sites on the CCNB1 and the CCNB2 promoters and the c-MYC enhancer by real-time PCR; normal IgG was used as the control. Data represent three independent experiments, each performed in triplicate. Bars represent the means ± SD (*p < 0.05, **p < 0.01 vs. control).

Figure 6. ISL1 regulated CCNB1, CCNB2 and c-MYC expression

(A, B) qRT-PCR analysis of CCNB1, CCNB2 and c-MYC mRNA levels in MGC803 cells transfected with pcDNA3.1-ISL1 (ISL1) or pLL3.7-ISL1-siRNA (si-ISL1). The cells transfected with pcDNA3.1 (control) or pLL3.7-Non-silencer (Non-silencer, control) were used as the controls, whose values were set as 1. The data represent three independent experiments, each performed in triplicate. 18S rRNA levels served as the internal control. Bars represent the mean ± SD (*p < 0.05, **p < 0.01 vs. control). (C) Representative images of western blots of CCNB1, CCNB2 and c-MYC expression in the cells. The grayscale scanning was performed on the images from three independent experiments and the results are presented at the bottom. GAPDH levels served as the internal control. The values of the controls were set as 1. Bars represent the mean ± SD (*p < 0.05 vs. control or Non-silencer).