NF-kappaB-dependent MicroRNA-425 upregulation promotes gastric cancer cell growth by targeting PTEN upon IL-1β induction

The human embryonic kidney cell line HEK293 (ATCC® CRL-1573™), the human breast cancer cell line MDA-MB361 (ATCC® HTB-27™), the human gastric adenocarcinoma cell line AGS (ATCC® CRL-1739™), SNU-1 (ATCC® CRL-5971™), SNU-5 (ATCC® CRL-5973™), SNU-16 (ATCC® CRL- 5974™), Hs746T (ATCC® HTB-135™), NCI-N87 (ATCC® CRL-5822™), and KATO III (ATCC®HTB-103™) were maintained in DMEM containing 10% fetal bovine serum. All cell lines were maintained in media containing penicillin (100 IU/ml) and streptomycin (100 mg/ml) at 37°C with 5% CO₂. The miRNA mimics and anti-miRNA were purchased from Ambion (Austin, TX, USA). The IKK inhibitor TPCA-1 (Cat. No. S2824), the p38 MAPK inhibitor BIX02188 (Cat. No. S1574) and the JNK inhibitor SP600125 (Cat. No. S1460) were purchased from Selleckchem (Houston, TX, USA). Recombinant human IL-1β were purchased from Sigma-Aldrich (Cat. No. H6291, Shanghai, China).

Total RNA was extracted from cells using TRIzol (Invitrogen, Carlsbad, CA). For microRNA analysis, poly(A) tails were added to total RNA using poly(A) polymerase (Ambion, Carlsbad, CA) prior to reverse transcription. The MiRcute miRNA qPCR detection kit (TIANGEN, Beijing, China) was used to quantitate the expression levels of mature miR-425 according to the provided protocol, and GAPDH was used as an internal control. Real-time PCR was performed under the following conditions: 95°C 10 m, 1 cycle; 95°C 10 s, 55°C 34 s, 40 cycles.

For all results obtained by real-time PCR methods, we used the delta delta CT method to calculate the fold change in gene expression between different groups. The amount of target (PTEN/miR-425), normalised to the endogenous housekeeping gene GAPDH and relative to a reference sample, is given by the following equation: amount of target =2^-△△CT.

Proteins were separated on a 10% SDS-PAGE gel and subsequently transferred to a PVDF membrane. After blocking with 5% nonfat milk, the membrane was incubated with a mouse monoclonal anti-PTEN antibody (1:500, Santa Cruz, sc-7974) and a NF-kappaB p65 Phospho (pS536) (RELA) antibody (1:10000, EPITOMICS, Cat.#: 2220–1). IRdye-labeled secondary antibodies were used for quantitation of the immunoblotting signal, and the signals were analyzed using an Odyssey scanner (LI-COR Biosciences, Lincoln, NE, USA).

HEK293 cells and AGS cells were transfected with miR-425 and pGL3 luciferase reporter constructs harboring the miR-425 target sequence. After 24 h, the activities of firefly luciferase and renilla luciferase in the cell lysates were measured with the Dual-Luciferase Assay System (Promega, Madison, WI, USA). For the luciferase transcription reporter assay, miR-425 gene promoter sequences (WT or site deletion) were cloned into the promoter region of the pGL3-Basic vector, and luciferase activity was measured as described above.

Briefly, treated cells were cross-linked with 1% formaldehyde, sheared to an average size of 400 bp, and subsequently immunoprecipitated with antibodies against NF-kappaB (Santa Cruz, sc-166588). The ChIP-PCR primers were designed to amplify the promoter regions containing putative NF-kappaB binding sites within miR-425 as illustrated. A positive control antibody (RNA polymerase II) and a negative control non-immune IgG were used to demonstrate the efficacy of the kit reagents (Epigentek Group Inc, P-2025-48). Immunoprecipitated DNA is then cleaned, released, and eluted. Eluted DNA can be used for downstream applications ChIP-PCR. Fold enrichment (FE) was calculated by using a ratio of amplification efficiency of the ChIP sample over that of non-immune IgG. Amplification efficiency of Polymerase RNA II was used as a positive control. FE% = 2^{(IgG CT – Sample CT)} × 100%.

A cell proliferation assay was performed using the Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) according to the manufacturer's instructions. Before the addition of CCK-8, the cells were washed with warm culture media by spinning the plate at 500 rpm for 3 m and then discarding the supernatant.

The cancer cells were harvested and resuspended in 500 μl of a binding buffer. The cell suspension (100 μl) was incubated with 5 μl annexin-V and propidium iodide at room temperature for 20 minutes. The stained cells were analyzed with fluorescent-activated cell sorting (FACS) using BD LSR II flow cytometry.

For the flow cytometry analysis, cells were trypsinized and fixed in 70% ethanol overnight. The cells were then incubated in 0.5 ml of propidium iodide solution containing 25 μg ml⁻¹ RNase for 15 minutes at 37°C and measured.

The NCI-N87 cells (3 × 10⁶) were injected into the right flanks of athymic nu/nu mice. One week after the injections, mice with comparably sized tumors were treated for 4 weeks with anti-miR-425. The anti-miR-425 (2 nmol) were injected directly into the tumors twice weekly for 4 weeks.

To characterize the miRNAs responsible for IL-1β induction, we profiled miRNA expression in PBS-treated AGS cells and IL-1β-induced AGS cells using the Exiqon miRCURY™ LNA Array System (v.14.0). The miRNA levels differed greatly between the PBS-treated group and the IL-1β-induced group, as illustrated in the heat map shown in Figure 1A. Among the 1,891 capture probes, 46 miRNAs were differentially expressed in IL-1β-induced AGS cells compared with paired PBS-treated AGS cells; of these miRNAs, 29 were increased and 17 were decreased in the IL-1β-induced AGS cells (Table 1), indicating that a specific miRNA pattern is associated with IL-1β induction.

Among these miRNAs, miR-425 was the most highly upregulated upon IL-1β induction. Using real-time PCR analysis, we analyzed miR-425 expression in 36 paired samples (tumor and adjacent normal tissues from the same patient). We found a significantly higher level of miR-425 expression in the tumor samples relative to the levels in the adjacent normal tissues (p < 0.0001) (Figure 1B). We examined the expression level of miR-425 in a set of gastric cancer cell lines and six normal gastric mucosa cells. As shown in Figure 1C, we picked up the AGS cells with down-regulated miR-425 and the NCI-N87 cells with up-regulated miR-425 for further study. Although the activation of miR-425 has been reported to have a fundamental impact on cancer initiation and progression of cancer cells by reducing the expression of an extensive network of genes [15], the role of miR-425 in human cancers has not been elucidated. We therefore chose miR-425 for further investigation.

To identify the targets of miR-425, we employed a commonly used algorithm, miRecords (http://​mirecords.​biolead.​org/​), which is an integrated resource for animal miRNA-target interactions. To increase the accuracy of this prediction, genes that were predicted by at least five of eleven databases (Diana, microinspector, miranda, mirtarget2, mitarget, nbmirtar, pictar, pita, rna22, rnahybrid and targetscan) were selected as putative targets. Among these putative targets of miR-425, gene ontology analysis revealed that the expression levels of 9 candidate genes were altered; thus, this alteration could contribute to the malignant phenotype. Using 3’ UTR luciferase reporter assays, we found that overexpression of miR-425 significantly inhibited luciferase activity in HEK293 cells and AGS cells expressing the wild type PTEN 3’ UTR reporter (Figure 2A). We confirmed that PTEN is a putative direct target of miR-425. To illustrate the specificity of miR-425, we showed that anti-miR-425 specifically abolished the inhibition of luciferase activity induced by miR-425 in HEK293 cells and NCI-N87 cells (Figure 2B). Mutations in the miRNA binding sites (Figure 2C) rendered the constructs unresponsive to miR-425 induction (Figure 2D), further confirming that the PTEN gene is a direct target of miR-425.

Furthermore, mutation of the miR-425 target sequence also significantly attenuated IL-1β-induced repression of PTEN 3’ UTR luciferase reporter activity in HEK293 cells and AGS cells (Figure 2E). Overexpression of miR-425 was sufficient to downregulate PTEN expression at both the protein and mRNA levels in AGS cells (Figure 2F). Accordingly, IL-1β-induced PTEN repression was rescued by expressing anti-miR-425 in AGS cells (Figure 2G). Anti-miR-425 was able to up-regulate PTEN expression in NCI-N87 cells without IL-1β stimulation (Figure 2H).

Our data also indicated that the 3’ UTR is required for miR-425-mediated PTEN downregulation because expression of a PTEN coding region construct (PTEN-CDS) was insensitive to miR-425 overexpression and IL-1β induction in AGS cells (Figure 2I). Taken together, these results indicate that miR-425 plays a critical role in repressing PTEN expression by targeting its 3’ UTR upon IL-1β induction.

To determine the mechanism involved in miR-425 transactivation upon IL-1β induction, we examined the impact of various kinase inhibitors on miR-425 induction in IL-1β-treated AGS cells. IL-1β-induced miR-425 upregulation was significantly inhibited by the IKK inhibitor TPCA-1 but not by the p38 MAPK inhibitor BIX02188 or the JNK inhibitor SP600125 (Figure 3A). Previous studies have demonstrated that IKK is an essential kinase required for NF-kappaB signaling; therefore, this result indicated the critical role of NF-kappaB signaling in the regulation of miR-425 transcription upon IL-1β induction.

Consistently, induction of pri-miR-425 upon IL-1β treatment was remarkably inhibited in the presence of the IKK inhibitor or siRNAs for NF-kappaB (Figure 3B and C). We also observed that IL-1β-induced PTEN repression was attenuated in the presence of the IKK inhibitor or siRNAs for NF-kappaB (Figure 3D). To determine whether NF-kappaB activity was present in AGS cells treated with IL-1β, we used a western blot to determine the level of phosphorylated NF-kappaB p65 (serine 536). The level of phosphorylated NF-kappaB p65 (serine 536) was high in AGS cells treated with IL-1β (10 ng/ml; 24 hours) (Figure 3E). In addition, silencing of NF-kappaB inhibited miR-425 expression in NCI-N87 cells without IL-1β treatment (Figure 3F). These results suggesting that IKK-dependent NF-kappaB activation upon IL-1β treatment is required for PTEN downregulation, most likely via its enhancement of miR-425 transcription.

To determine whether NF-kappaB directly regulates miR-425 transcription, we analyzed the upstream sequences of miR-425 using the WeightMatrix library (matrixTFP60.lib) and identified three potential NF-kappaB binding sites in the promoter region of miR-425 (cut-offs: matrix similarity = 0.9 and core similarity = 0.95) (Figure 4A). We performed chromatin immunoprecipitation (ChIP) assays with AGS cancer cells using monoclonal anti-NF-kappaB antibodies. As shown in Figure 4B, only primer-B of miR-425 produced strong PCR products, which suggested that the NF-kappaB protein formed complexes with the B binding site in the miR-425 promoter. The results of luciferase reporter assays suggested that the potential B binding site in the miR-425 promoter is required for transactivation of the downstream gene upon IL-1β induction (Figure 4A and C).

It was shown that PTEN is among the most frequently inactivated tumor suppressor genes. Overexpression of PTEN in different mammalian tissue culture cells affects various processes including cell proliferation, cell death and cell migration [16]. We also found that inhibiting PTEN decreased the activation of caspase-3 in cells treated with IL-1β (Figure 5A). It is plausible that miR-425 induction may inhibit apoptosis via the downregulation of PTEN in IL-1β-treated cells. Indeed, overexpression of miR-425 inhibited caspase-3 activation in cisplatin-treated AGS cells (Figure 5B). Moreover, in cisplatin-treated AGS cells, cotransfection of a construct containing only the PTEN coding region (PTEN-CDS), which is insensitive to miR-425, bypassed the antiapoptotic effect of miR-425 overexpression (Figure 5C). Accordingly, transfection of anti-miR-425 in AGS cells significantly enhanced caspase-3 activation and apoptosis in response to IL-1β treatment (Figure 5D). In addition, transfection of anti-miR-425 in NCI-N87 cells significantly enhanced caspase-3 activation and apoptosis without IL-1β stimulation (Figure 5E).

Consistent with its role in inhibiting caspase activation, upregulation of miR-425 substantially enhanced AGS cell proliferation, whereas the pro-survival effect was completely blocked by co-transfection with exogenous PTEN (PTEN-CDS) (Figure 6A). Anti-miR-425 decreased the percentage of proliferating cells for NCI-N87 cells (Figure 6B). We also found that inhibiting PTEN had a protective effect similar to that observed in cells overexpressing miR-425 (Figure 6C), suggesting that PTEN repression may play a major role in miR-425-dependent protection in cells treated with IL-1β.

We investigated the effect of miR-425 on tumorigenicity in vivo. The tumors treated with anti-miR-425 showed increased levels of the PTEN protein (Figure 6D). Also, anti-miR-425 reduced the tumor weight of the mice compared with the miR-NC-treated group (Figure 6E). Using non-parametric tests, we found a significant inverse correlation between PTEN mRNA and miR-425 expression in the gastric cancer samples (Figure 6F). The expression levels of PTEN were also determined in six normal gastric mucosa cells and gastric cancer cell lines using real-time PCR. As shown, the cells with “down-regulated” miR-425 have higher amounts of PTEN compared to cell lines with “up-regulated” miR-425 levels (Figure 6G). In conclusion, our results have proven that miR-425 plays a causal role through targeting PTEN in gastric cancers.