Mir-1287 suppresses the proliferation, invasion, and migration in hepatocellular carcinoma by targeting PIK3R3

2.1 TCGA data analysis

The miRNA expression data was retrieved from The Cancer Genome Atlas (TCGA) Research Network (https://cancergenome.nih.gov). The miRNA-seq data of 375 HCC patient samples and 50 normal tissue samples were normalized and transformed in log₂ form. The TCGA IDs of the samples were listed in the supplementary materials Table S1.

2.2 Gene chip for miRNA expression

The miRNA microarray used was miRCURY LNA microRNA Array, sixth generation—hsa, mmu, & rno (EXIQON). The gene chip was checked and analyzed by UCLA Department of Pathology and Laboratory Medicine David Geffen School of Medicine. The cell samples were QGY-7703 which we chose as representative hepatoma cells, and a subpopulation hepatoma cells with suppressed malignant phenotypes which we treated as normal cells for control.13

2.3 Cell culture

Human HCC cell lines, QGY-7703 and SMMC-7721 are cultured with RMPI-1640, supplemented with 10% FBS and 100 U/mL of penicillin-streptomycin, at 37°C in a humidified chamber supplemented with 5% CO₂.

2.4 Xenograft assay on nude mice

The nude mice used were the male BALB/c athymic nude mice. The cells stably expressing mir-1287 or empty vector were suspended with phosphate-buffered saline and subcutaneously injected nude mice at a total cell number of approximately 2 × 10⁶ each mouse. Sixteen days after injection, when tumors were palpable, the tumor volumes in mice were measured with a slide caliper every 4 days until the scarification. And all the tumors were excised and weighed in the end.

2.5 Soft agar colony formation assay

The cells stably expressing mir-1287 or empty vector were seed (300 cells each) onto the plate with 0.7% low melting point agarose in growth media. After incubated at 37°C for 2 weeks, the cells were stained with 1% neutral red.

2.6 Quantitative real-time PCR

Total RNA samples were extracted from cancer cells with TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Then total RNA was reverse transcripted with M-MuLV reverse transcriptase kit (Shanghai Pufei Biotech, Shanghai, China). Quantitative real-time polymerase chain reaction (PCR) was operated on SYBR Green PCR kit (Bio-Rad, Hercules, CA). The primers was purchased from Genepharm (Shanghai, China).

2.7 Cell proliferation assay

QGY-7703 and SMMC-7721 was seeded on 96-well plates 24 hours before transfection of miRNA mimics or inhibitors together with negative control. At 0, 24, 48, and 72 hours, the culture medium was removed and then replaced by fresh medium with 10% concentration of CCK-8 (Dojindo, Kumamoto, Japan) for each well. And the cells were incubated for 2 hours at 37℃, before the absorbance of suspension was measured the at 450 nm by microplate reader (Thermo Scientific Multiskan MK3, Waltham, MA).

2.8 Cell cycle analysis with flow cytometer

Cell cycle was assessed with Flow Cytometer (BD FACS Calibur, San Jose, CA). All the cells were cultured in serum-free medium for 48 hours to synchronize the cell cycle before transfection of miRNA mimics, inhibitors or negative control. The cells was collected, washed with phosphate-buffered saline and fixed with 70% ethanol at 4℃ overnight. Then the cells were stained with propidium iodide at room temperature for 30 minutes before analysis on flow cytometry.

2.9 In vitro cell invasion and migration assay

Cell invasion and migration was evaluated by transwell assay respectively with matrigel and without matrigel in each well. After transfection of miRNA mimics, inhibitors or negative control, cells were cultured in serum-free medium for 24 hours. Then the cells were resuspended and seeded 1 × 10⁵ cells each in the upper chamber of the transwell chamber (Corning Costar, Corning, NY), and culture medium with 10% FBS was added in the lower chamber. The incubation time was 24 hours for migration and 48 hours for invasion. And the migratory or invasive cells were stained with crystal violet and manually counted at ×200 magnification.

2.10 Luciferase report assay

The 3′-UTR sequence of PIK3R3 containing the predicted target site of mir-1287 was cloned onto psiCHECK-2 vector (C8021; Promega, Madison, WI). And a mutant plasmid on the target site was constructed with the Site-directed Gene Mutagenesis kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions. The luciferase constructs and the mir-1287 mimics or negative control were cotransfected into QGY-7703 cell with lipofectamine 2000. Forty-eight hours after transfection the cells were collected and examined the luciferase activity using Dual-Luciferase Reporter Assay System (E1910; Promega).

2.11 Western blot

Cell lysis were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred on to poly(vinylidene difluoride) membranes (Millipore, Billerica, MA). The membranes were blocked with nonfat milk at room temperature for 2 hours, followed by incubation with primary antibodies for 1.5 hours. After washing with 2‰ TBST 3 times, the membranes were incubated with secondary antibodies for 45 minutes and washed another 3 times before exposure with ECL (Thermo Fisher Scientific, Waltham, MA). The intensity of the protein bands was analyzed by Image J system (NIH, Bethesda, MD). The primary antibodies used were as follows: PIK3R3 (Abgent, Wuxi, China) Akt, p-Akt, NF-κB pathway sampler kit (Cell Signaling Technology, Danvers, MA).

2.12 Statistical analysis

All data were evaluated with the software R version 3.3.2 and Origin 9.1 (OriginLab, Northampton, MA). A 2-tailed Student t test was performed to compare the data. Difference was considered significant at P < .05.

3.1 The expression of mir-1287 is suppressed in HCC cells

To understand miRNA homeostasis in HCC cells, we analyzed miRNA levels derived from microarray profiles and found that a number of miRNAs are aberrantly expressed in human HCC cells (Figure 1A). Among these, mir-190, mir-10a, and mir-34a are reported to be involved in tumorigenesis.14-16 While mir-1287 is also largely suppressed in tumor cells, indicating that the normal expression level of this microRNA is potentially important for tumor suppression.

To determine whether the suppression of mir-1287 is a universal phenomenon in HCC, we further compared the miRNA expression levels in HCC tumor and adjacent normal tissues using data downloaded from TCGA database (Figure 1B). The normalized data was log 2 transformed to demonstrate the fold-change of the mir-1287 level between tumor tissues and normal tissues. The results indicated that the expression of mir-1287 was downregulated in HCC tumor tissues than in normal tissues (6.14 ± 1.61 and 7.00 ± 0.79, respectively; P < .001).

To confirm the aberrant level of mir-1287 in a cell model, we further used real-time PCR to validate the expression level of mir-1287 in 7 types of human liver cell lines (Figure 1C). Among them, the normal liver cells QSG-7701 and L02 showed a remarkably high expression of mir-1287, while in the rest 5 types of HCC cell lines, the expression levels were relatively low, indicating that mir-1287 was downregulated in HCC cells.

3.2 Mir-1287 suppresses the tumor growth and the colony formation

We next sought to determine whether the overexpression of mir-1287 would suppress the tumor growth. We exogenously expressed mir-1287 to increase the levels of this miRNA in the QGY-7703 and SMMC-7721 tumor cells. These cells formed fewer colonies in soft agar assays (Figure 2A,B). Furthermore, they formed much smaller tumors after being xenografted into nude mice (Figure 2C,D).

3.3 Mir-1287 suppresses the cell proliferation in QGY-7703 and SMMC-7721 and arrest the cell cycle at G1/S phase

Given the robust phenomenon that mir-1287 suppressed the tumor growth, we next identified the cellular mechanism underlying this miRNA. We further analyzed the proliferation of the QGY-7703 and SMMC-7721 tumor cells by CCK-8 assays (Figure 3A,B). The cells transfected with miR-1287 mimics showed apparently reduced proliferation rates. This was associated with increased numbers of cells at the G1 phase (Figure 3E,F), indicating a possible impact of miR-1287 on cell cycle arrest. In contrast, the cells transfected with miR-1287 inhibitors (antisense RNA of miR-1287) proliferated faster and fewer cells were arrested at the G1 phase (Figure 3C,D,G,H).

3.4 Mir-1287 inhibits the migration and invasion of HCC cell lines

To determine the function of mir-1287 in tumor metastasis, we used transwell assays to test the migration and invasion of the QGY-7703 (Figure 4A,B) and SMMC-7721 (Figure 4C,D) tumor cells. The cells transfected with the miR-1287 mimics showed reduced migration and invasion. In contrast, transfection of the miR-1287 inhibitors led to enhanced cell migration and invasion.

3.5 PIK3R3 is the specific target of miR-1287

To identify the direct target of mir-1287, a number of mRNAs, the expression levels of which are linked to HCC, have been predicted as targets of miR-1287 mediated regulation. We picked out 11 candidates, which were most relative with HCC, from 140 multiple predicted mRNAs. We further checked their levels and binding sites (see Supplementary Material Figure S2), and found one direct target PIK3R3, a regulatory subunit of PI3K complex, as well as a novel oncogene in many cancers including HCC.17-19

To validate the targeting site of mir-1287 on PIK3R3, we cloned the predicted miR-1287’s binding site on PIK3R3 onto the dual-luciferase report assay, and checked the luciferase level after transfection of miR-1287 mimics. A mutant on binding site was also designed as contrast. The luciferase decreased in the wild-type group, while the mutant group showed no significant difference (Figure 5A).

We further confirmed the protein level of PIK3R3 by Western blot. With the increased volume of miR-1287 transfected, the expression of PIK3R3 protein showed a decrease (Figure 5B). The expression of PIK3R3 was upregulated by the transfection of miR-1287 inhibitors (Figure 5C). These results confirmed that PIK3R3 is the direct target gene of miR-1287.

3.6 Overexpression of miR-1287 leads to the inhibition of PIK3R3 involving cell signaling pathways

We next aimed to determine whether PIK3R3 is the biological effectors of mir-1287, and the molecular mechanism underlying the mir-1287 regulation. PIK3R3 can directly bind with the Rb protein, an important regulator in cell cycle progression.20, 21 We defined the phosphorylation level of Rb protein is decreased by the increased amount of transfected miR-1287 mimics through Western blot (Figure 6A).

Furthermore, PIK3R3 is a regulatory subunit of PI3K complex, which activates PI3K/Akt pathway and the downstreaming NF-κB pathway.22 IKK-α/β links the activation of Akt and NF-κB. The NF-κB, when phosphorylated, transcriptionally regulates cyclin D1.

Western blot is used to check the phosphorylation levels of these proteins targeted by PIK3R3, including AKT, IKK-α/β, NF-κB, and cyclin D1. With the increased amount of transfected miR-1287 mimics, all the proteins presented loss of phosphorylation (Figure 6B), indicating that a cascade reaction occurred through downregulation of PIK3R3 in the PI3K/Akt and NF-κB pathway.

The transfection of miR-1287 inhibitors increased the amount of cyclin D1, and phosphorylation of Akt and NF-κB (Figure 6C).

3.7 Overexpression of PIK3R3 rescues the phenotypes aroused by miR-1287 in HCC cells

We further examine whether ectopic expression of PIK3R3 would rescue the tumor suppressive effects of miR-1287. We exogenously expressed the CDS sequences of PIK3R3 gene in both QGY-7703 and SMMC-7721 cell lines. After transfecting with miR-1287 mimics, both cells showed upregulated proliferation rates compared to those expressing empty vectors (Figure 7A,B). The transwell assay also showed increased invasion and migration in contrast to the cells without PIK3R3 ectopic expression (Figure 7C–F).