DBF4 is upregulated in human gastric cancer.
To explore gene expression profiles in GC, we performed gene set enrichment analysis (GSEA) with data from The Cancer Genome Atlas (TCGA; Fig. 1A, 1B). The analysis showed that the CELL_CYCLE signatures were significantly enriched in gastric tumor cells compared with adjacent tissues cells. We chose one of the most differentially expressed genes, DBF4, as our candidate for further experiments. DBF4 is a principle regulator of DNA replication, but its significant has not been previously explored in GC. Statistical analysis of DBF4 expression in 32 adjacent tissues and 375 tumor tissues from TCGA showed that DBF4 was dramatically upregulated in human GC (Fig. 1C). To further confirm the results from the TCGA analysis, an additional cohort with four pairs of GC tissues and adjacent normal tissues were collected to analyze DBF4 expression by western blots. As shown in Fig. 1D, DBF4 was increased significantly in GC tumor tissues, in accordance with the results from TCGA database. qRT-PCR was then employed to evaluate DBF4 expression in GC cells and normal cells, finding that DBF4 was potently up-regulated in GC cell lines (Fig. 1E).Collectively, these findings evinced that DBF4 was highly expressed in GC.
DBF4 promotes the proliferation of gastric cancer cells.
To investigate the biological role of DBF4 in GC, we designed siRNAs to specifically target DBF4 and transfected the siRNAs into MGC-803 and AGS cells. The qRT-PCR and western blot analyses showed that the expression of DBF4 was significantly decreased in DBF4-siRNA-transfected cells (Fig. 2A, 2B). Next, we used CCK-8 assays and colony formation assays to detect proliferation of GC cells with decreased expression of DBF4. These assays showed that GC cell proliferation was decreased in the si-DBF4 group compared with the control group (P < 0.01; Fig. 2C, 2D). Expression of PCNA and cyclin-A was also decreased in GC cells transfected with si-DBF4 (Fig. 2E). To determine whether DBF4 overexpression had the opposite effect, we constructed the DBF4 overexpression vector, PEX-DBF4, and transfected this vector or the control vector (PEX) into GC cell lines (MGC-803 and AGS). The transfection efficiency was validated by qRT-PCR and western blotting (Fig. 2F, 2G). Functionally, the CCK-8 assay and colony formation assay suggested that overexpression of DBF4 significantly enhanced GC cell proliferation (Fig. 2H, 2I). Expression of PCNA and cyclin-A was increased in GC cells with DBF4 overexpression, as detected by western blotting (Fig. 2J). Together, these data demonstrated that DBF4 significantly enhanced GC cell proliferation.
Expression of DBF4 weakens the sensitivity of MGC803 and AGS cells to 5-Fu
5-fluorouracil (5‐Fu) is considered as one of the first‐line chemotherapy drugs in advanced gastric cancer(GC). First, the effect of 5‐Fu treatment on cellular proliferation‐associated MGC803 and AGS GC cells was assessed by CCK‐8 kit. As shown in Fig. 3A, MGC803 and AGS cells were exposed to a series of concentrations of 5‐Fu for 48hours, taking the cell viability of the control sample (without 5‐Fu) as 100%, viability of MGC803 and AGS cells was inhibited significantly after 5‐Fu treatment with the half‐maximal inhibitory concentration (IC50) values of 0.4111 and 0.3952 mM respectively. Then, DBF4 overexpression was performed in vitro based on MGC803 and AGS cells by plasmid transfection and IC 50 values of DBF4‐overexpressing MGC803-DBF4 and AGS-DBF4 cells were detected by the same way with 0.821 and 1.269 mM separately (Fig. 3B). Meanwhile, short interference siRNA for DBF4(siDBF4) were applied to knockdown the expression of DBF4 inMGC803 and AGS cells. IC 50 values of MGC803-siDBF4 and AGS-siDBF4 cells were detected by the same way with 0.1453 and 0.1677 mM separately (Fig. 3C). Taken together, these results demonstrate that as an oncogene, DBF4 weakens the sensitivity of MGC803 and AGS GC cells to 5-Fu.
DBF4 promotes the migration of gastric cancer cells.
To investigate the biological function of DBF4 in GC, transwell assays and woud healing assay were used to determine the migration of MGC-803 and AGS cells. As shown in Fig. 4A, 4B, silencing of DBF4 repressed the migration of both gastric cancer cells. To further confirm the results, DBF4 was overexpressed in MGC-803 and AGS cells and the migration of both GC were detected by transwell assays and woud healing assay. Consistent with the results above, the migratory ability of MGC-803 and AGS cells was strongly enhanced following DBF4 overexpression (Fig. 4C,4D). Collectively, these experiments indicated that DBF4 also perform an important role in migration of GC cells.
The miR-30a inhibits the expression of DBF4.
The miRNAs are involved in the regulation of various biological processes such as cell proliferation [19], they regulate expression of many oncogenes and tumor suppressor genes, and they have been reported to play key roles in various human cancers, including GC [16, 17]. We first analyzed the miRNA expression profile in 436 GC tissues and 41 adjacent tissues tissues from TCGA. A total of 1881 miRNAs were expressed in these tissues and 139 showed differential expression. Of these, 79 miRNAs were downregulated and 60 were upregulated in GC (fold change ≥ 2 or ≤ 0.5, P < 0.05; Fig. 5A). A heat map was created showing differentially expressed microRNAs in GC tissues, relative to matched normal tissues (Fig. 5A). To determine the target regulatory miRNA for DBF4, two databases were screened, including TargetScan (https://www.targetscan.org) and TarBase (http://starbase.sysu.edu.cn/starbase2/). The only overlapping result from the two databases was miR-30a-5p (Fig. 5B). As shown in Fig. 5A and 4B, miR-30a-5p also shows significant differential expression in GC. Statistical analysis showed that miR-30a expression was significantly decreased in GC tumor tissue (Fig. 5C). In addition, the expression level of DBF4 was negatively correlated with the expression level of miR-30a in GC tissues, based on the data from TCGA (Fig. 5D). To further confirm the effect of miR-30a on DBF4 expression, both MGC-803 and AGS cells were transfected with miR-30a mimics and negative controls. DBF4 expression was analyzed by qRT-PCR and western blotting assays, which showed that DBF4 expression was significantly downregulated in GC cells transfected with miR-30a mimics. This suggested that DBF4 may be the target of miR-30a in GC (Fig. 5E). Moreover, bioinformatics analysis predicted a putative 8-mer-binding site with miR-30a in the 3'-UTR of the DBF4 transcript (Fig. 5F). Dual-luciferase reporter assays were performed to determine whether miR-30a directly targets DBF4. Overexpression of miR-30a significantly decreased the luciferase activity of wild-type DBF4 in GC cells, but had no effect on luciferase activity of mutant DBF4 in GC cells (Fig. 5G), demonstrating that miR-30a specifically binds to the 3’ UTR of miR-30a. Taken together, these data suggested that miR-30a, one of the most differentially downregulated miRNAs in GC, directly regulated DBF4 expression.
The miR-30a-DBF4 axis regulates the proliferation of gastric cancer cells.
Having confirmed the interaction of miR-30a and DBF4, next we performed rescue experiments to determine whether miR-30a functions as a tumor suppressor gene via regulation of DBF4. Previous reports suggested that miR-30a may inhibit the proliferation and invasion of GC [22]. In agreement with these previous findings, MGC-803 and AGS gastric cell lines exhibited reduced proliferation after transfection with miR-30a mimics (Fig. 6B, 6C). qRT-PCR analysis showed that DBF4 expression was suppressed in the miR-30a mimic group and upregulated in the DBF4 overexpression group (Fig. 6A). CCK-8 and colony formation assays showed that the suppressive effect of miR-30a overexpression on cell proliferation was abrogated by overexpression of DBF4 (Fig. 6B,6C). In order to further determine the effect of miR-30a-DBF4 axis on cell migration, woud healing assay and t Transwell assays were used, and the results indicated that suppressive effect of miR-30a overexpression on cell migration was abrogated following DBF4 overepression. (Fig. 6D and 6E). Taken together, these results suggested that miR-30a restored the effects of DBF4 on proliferation and migration of AGS and MGC-803 cells.
Lactate in the tumor microenvironment induces aberrant expression of miR-30a and DBF4.
In addition to dysregulated cell proliferation, tumors exhibit another dimension of complexity in that they contain a repertoire of recruited immune cells, metabolites such as lactate, and inflammatory factors such as interleukin-6 (IL-6). To determine the primary factor that induced the aberrant expression of miR-30a and DBF4, we stimulated GC cells with lactate and IL-6 in MGC-803 and AGS cells. As shown in Fig. 7A, lactate suppressed miR-30a expression and IL-6 showed a weak effect on miR-30a expression. By qRT-PCR analysis, we determined that lactate was the main factor inducing the aberrant expression of DBF4 (Fig. 7B). Together, our findings demonstrated that accumulation of lactate in the tumor microenvironment inhibited miR-30a expression and increased DBF4 expression.