miR-150-5p suppresses tumor progression by targeting VEGFA in colorectal cancer

miR-150-5p is downregulated in CRC tissues and cell lines

First, to determine the clinical significance of miR-150-5p in patients with CRC, we used the StarBase database (http://starbase.sysu.edu.cn/) to reveal that miR-150-5p expression was downregulated in CRC tissues compared with normal tissues (Fig. 1A) and the database is accompanied with gene expression data of 32 types of cancers which are derived from 10,882 RNA-seq and 10,546 miRNA-seq data. Then, we measured the miR-150-5p expression in 112 pairs of CRC tissues using qRT-PCR. The results showed that miR-150-5p was significantly downregulated in CRC tissues compared with that in matched adjacent normal tissues (ANTs) (Fig. 1B). In addition, the miR-150-5p expression was markedly downregulated in all CRC cell lines in comparison with colonic epithelial cell (FHC) (Fig. 1C). Next, we examined the potential clinical significance of miR-150-5p in CRC. We used the median miR-150-5p expression as a cut-off value. As shown in Table 1, low miR-150-5p expression was markedly associated with TNM stage (p=0.013) and pM stage (p=0.031).

Figure 1. Downregulation of miR-150-5p in CRC and its association with poor OS. (A) miR-150-5p was found to be downregulated compared with normal tissues in the Starbase database. (B) Quantitative RT-PCR was performed to detect the relative expression of miR-150-5p in CRC tissues compared with ANTs. (C) The relative expression of miR-150-5p was assessed in different CRC cell lines and FHC using qRT-PCR. (D) Kaplan-Meier survival analysis was employed to detect the correlation between the levels of miR-150-5p expression and overall survival of CRC patients. Data were shown as mean±SD. **p<0.01.

The Kaplan-Meier survival analysis indicated that CRC patients with low miR-150-5p had an obviously shorter overall survival (OS) (Fig. 1C). Furthermore, the results of univariate and multivariate analyses of potential prognostic factors for overall survival are shown in Table 2. The univariate analysis indicated that TNM stage (p=0.001), pN stage (p=0.006), pM stage (p=0.002) and miR-150-5p (p=0.004) were predictors of OS. The multivariate analysis also showed that higher TNM stage (p=0.006), pN stage (p=0.017) and pM stage (p=0.008), as well as lower miR-150-5p (p=0.011), were significantly correlated with poor OS (Table 2). Based on these results, we suggested that miR-150-5p might play a key role in CRC progression.

miR-150-5p inhibits cell proliferation, migration, invasion and angiogenesis in vitro

To determine the biological function of miR-150-5p in CRC, we selected HCT116 and HCT8 cells for transfection of agomiR-150-5p and antagomiR-150-5p or their corresponding negative controls (agomiR-NC and antagomiR-NC). The efficiency of transfection was confirmed by qRT-PCR (Fig. 2A). First, a CCK-8 assay performed to assess CRC cell growth showed that upregulation of miR-150-5p resulted in decreased cell proliferation, while the inhibition of miR-150-5p expression obviously increased cell proliferation in HCT116 and HCT8 cells (Fig. 2B). A colony formation assay further substantiated the role of miR-150-5p in inhibiting the proliferation of CRC cells (Figs. 2C and Fig. S1A). Next, wound-healing and transwell invasion assays were performed in HCT116 and HCT8 cells. As shown in Figs. 2D and and Fig. S1C, the migratory and invasive abilities were significantly reduced when miR-150-5p expression was increased, while miR-150-5p downregulation significantly increased cell migration and invasion in HCT116 and HCT8 cells. Additionally, to explore whether miR-150-5p was a key angiogenesis suppressor in CRC, we investigated the influence of miR-150-5p on HUVEC tube formation and showed that miR-150-5p upregulation obviously reduced the tube-formation capacity, while silencing miR-150-5p increased the tube-formation ability of HUVECs (Fig. 2F and Fig. S1D).

Figure 2. miR-150-5p suppressed CRC cell proliferation, migration, invasion and HUVEC tube formation in vitro. (A) qRT-PCR was used to detect the expression of miR-150-5p in CRC cells transfected with antagomiR-150-5p or agomiR-150-5p and their negative control (antagomiR-NC, agomiR-NC). (B, C) CCK-8 (B) and colony formation (C) were performed to evaluate the proliferation abilities of CRC cells transfected with antagomiR-150-5p, agomiR-150-5p or negative control (antagomiR-NC, agomiR-NC). (D, E) Wound-healing (D) and transwell (E) assays were employed to detect the ability of migration and invasion of antagomiR-150-5p or agomiR-150-5p-transfected CRC cells and their negative control. (F) HUVECs were cultured in TCM derived from CRC cells transfected with antagomiR-150-5p or agomiR-150-5p and their negative control in 24-well Matrigel-coated plates. Each bar represents the mean ± SD of three independent experiments. *p<0.05, **p<0.01, ***p<0.001.

miR-150-5p ameliorates CRC growth, metastasis and angiogenesis in vivo

To evaluate the biological functions of miR-150-5p in vivo, CRC cells were subcutaneously injected into nude mice. First, we investigated the effects of miR-150-5p on the xenograft growth in vivo and found that the volume of tumors injected with agomiR-150-5p was significantly smaller than that with agomiR-NC (Fig. 3A, 3B). Furthermore, to examine the influence that miR-150-5p had on metastasis in vivo, CRC cells were injected into nude mice and agomiR-150-5p or agomiR-NC via the tail vein. After 8 weeks, computed tomography (CT) scanning was used on each mouse, and results revealed that the metastatic lesions in the agomiR-150-5p group were less abundant than that in the agomiR-NC group (Fig. 3C). Following CT scanning, these mice were sacrificed, and hematoxylin-eosin (HE) was employed to stain mouse lungs. Compared with low miR-150-5p-expressing CRC cells, upregulation of miR-150-5p significantly decreased the number of metastatic nodules (Fig. 3C, 3D). We then detected CD31 (a marker for microvessels denoting enhanced angiogenesis) expression using immunohistochemistry (IHC) in xenografts tissues and showed that CD31 expression in the agomiR-150-5p group was significantly lower than those in agomiR-NC group (Fig. 3E).

Figure 3. miR-150-5p inhibited tumor growth, metastasis and angiogenesis in vivo. (A) Photographs of nude mice and xenograft tumors of CRC cells with miR-150-5p overexpression or not after 20 days of implantation. (B) Growth curves of xenografts of CRC cells with or without miR-150-5p overexpression. (C) The representative images of lungs from nude mice (left), representative CT scans of lung metastatic foci of the nude mice (middle) and the microscopic images of lung tissue sections stained by hematoxylin and eosin (right). (D) A statistical plot of the average number of lung metastases in the tail vein injection. (E) These tumor tissues were stained using anti-CD31 antibody, data are shown as the mean ±SD. n=8. ***p<0.001.

VEGFA is a direct target gene of miR-150-5p

Many miRNAs have been reported to be dysregulated and exert their functions mainly through their target genes [13,14]. The PicTarSites, miRandaSites and Tarbase databases were used to search for potential target genes of miR-150-5p. As shown in Fig. 4A, eight potential genes were predicted in all four databases. VEGFA, which was selected as a target gene of miR-150-5p, is closely associated with metastasis and angiogenesis of human cancers. Next, dual-luciferase reporter assays were used to determine whether VEGFA was a direct target gene in CRC cells. The 3’-UTR of VEGFA wild-type (WT-VEGFA-3’UTR), as well as VEGFA mutant type (Mut-VEGFA-3’UTR) were constructed and cloned into pmirGLO luciferase reporter, then cotransfected with agomiR-150-5p or agomiR-NC into HCT116 and 293T cells. Luciferase reporter assay confirmed that miR-150-5p overexpression obviously reduced the luciferase activity of pmirGLO-VEGFA-3’UTR-WT but showed no significant effect on the luciferase activity of pmirGLO-VEGFA-3’UTR-Mut (Fig. 4B). Additionally, qRT-PCR and Western blot analyses showed that both VEGFA mRNA and protein levels are dramatically decreased after miR-150-5p upregulation in HCT116 and HCT8 cells (Fig. 4C, 4D).

Figure 4. A. VEGFA was a direct target of miR-150-5p in CRC. (A) The direct target genes of miR-150-5p were predicted using the PicTarSites, miRandaSites and Tarbase databases. (B) Wild-type and mutant VEGFA-3’UTR sequences were cloned into luciferase reporter. Luciferase activity was determined in HCT116 and 293T cells cotransfected with agomiR-150-5p or agomiR-NC and pmirGLO-VEGFA-3’UTR-WT or pmirGLO-VEGFA-3’UTR-Mut. Luciferase activities were normalized to that of renilla luciferase. C, D. qRT-PCR (C) and western blot (D) analyses showed that both VEGFA mRNA and protein expression levels were dramatically suppressed by agomiR-150-5p in HCT116 and HCT8 cells, GAPDH was used as the internal control. **p<0.01.

VEGFA upregulation could reverse the inhibitory effect caused by miR-150-5p

To investigate whether miR-150-5p suppressed proliferation, migration, invasion, angiogenesis and EMT process through VEGFA, we cotransfected agomiR-150-5p and pcDNA 3.1-VEGFA or empty vector, and VEGFA upregulation was verified by western blotting (Fig. 5A). Through CCK-8, colony formation, wound-healing, transwell invasion and tube formation assays, we found that ectopic VEGFA expression could partially reverse the suppression of proliferation (Fig. 5B, 5C), migration (Fig. 5D), invasion (Fig. 5E) and angiogenesis (Fig. 5F) caused by miR-150-5p overexpression. These data indicated that VEGFA was not only a direct target gene but also a functional mediator of miR-150-5p in CRC.

Figure 5. miR-150-5p inhibited CRC progression by targeting VEGFA. (A) VEGFA protein expression was determined in HCT116 and HCT8 cells transfected with agomiR-150-5p with VEGFA expression plasmid or empty vector using western blot; GAPDH was used as the internal control. (B-D) Cell proliferation (B, C), migration (D)and invasion (E) were evaluated in HCT116 and HCT8 cells transfected with agomiR-150-5p with VEGFA expression plasmid or empty vector. (F) HUVECs were cultured in TCM derived from HCT116 and HCT8 cells transfected with agomiR-150-5p plus VEGFA expression plasmid or empty vector. Data are shown as the mean±SD. *p<0.05, **p<0.01, ***p<0.001.

The expression of VEGFA is negatively correlated with miR-150-5p

qRT-PCR performed to detect VEGFA mRNA in CRC tissues clearly showed that VEGFA mRNA was overexpressed in CRC tissues compared with matched adjacent normal tissues (ANTs) (Fig. 6A), moreover, we analyzed VEGFA mRNA in CRC tissues at different miR-150-5p levels. The low versus high miR-150-5p levels were defined at the median of miR-150-5p level. The levels of VEGFA mRNA expression in high-miR-150-5p CRC tissues were markedly lower than that in low-miR-150-5p CRC tissues (Fig. 6B). Meanwhile, an obvious negative correlation was found between miR-150-5p and VEGFA mRNA (Fig. 6C).

Figure 6. VEGFA was reversely correlated with miR-150-5p in CRC tissues. (A) qRT-PCR analysis of VEGFA mRNA expression in CRC tissues and matched adjacent normal tissues. (B) Scatterplots of the average VEGFA mRNA expression in patients with low or high expression of miR-150-5p in CRC tissues. (C) The correlation between miR-150-5p levels and VEGFA mRNA levels. (D) Representative images of IHC in adjacent normal tissues, low or high miR-150-5p CRC tissues. (E) The statistical graph showed that the IHC score of VEGFA in CRC tissues was significantly higher than in matched adjacent normal tissues. (F) Scatterplots of the average staining scores for VEGFA expression in patients with low or high expression of miR-150-5p. G. The correlation between miR-150-5p levels and VEGFA protein levels. Data were shown as the mean±SD. **p <0.01, ***p<0.001.

In addition, IHC was performed to assess the protein expression of VEGFA in CRC tissues, and as shown in Fig. 6D and 6E, the IHC staining intensity indicated that the VEGFA protein expression levels in CRC tissues were obviously higher than in adjacent normal tissues. The VEGFA was relatively upregulated in the low-miR-150-5p expression CRC tissues compared with in high-miR-150-5p CRC tissues (Fig. 6D and 6F); a negative correlation between VEGFA and miR-150-5p was also noted in CRC tissues (Fig. 6G).

VEGFA knockdown significantly inhibits CRC progression

To further verify that miR-150-5p exerts its inhibitory effect via VEGFA, we knocked down VEGFA in HCT116 and HCT8 cells (Fig. 7A) and found that this markedly suppressed CRC cell growth, migration, invasion and HUVEC tube formation (Fig. 7B-7E). These data further proved that miR-150-5p exerted its biological function by inhibiting VEGFA expression.

Figure 7. VEGFA knockdown significantly inhibited CRC progression. (A) VEGFA expression was downregulated in HCT116 and HCT8 cells transfected with siVEGFA-1 or siVEGFA-2. (B) VEGFA knockdown inhibited CRC cell proliferation (B), migration (C), invasion (D) and HUVECs tube formation (E). Data are shown as the mean±SD of three independent experiments. *p<0.05, **p<0.01, ***p<0.001.

miR-150-5p inactivates the Akt/mTOR signaling pathway by directly inhibiting VEGFA

Accumulating evidence has shown that VEGFA could bind to its receptor VEGFR2, then activate PI3K and its downstream targets Akt and mTOR, which play crucial roles in cancer progression and survival [15–17]. To further explore whether miR-150-5p inhibited CRC development through VEGFA-mediated Akt/mTOR signaling pathway, we found that miR-150-5p overexpression significantly decreased the level of VEGFA, and then significantly suppressed the phosphorylation of VEGFR2 (p-VEGFR2), Akt (p-Akt) and mTOR (p-mTOR) in HCT116 and HCT8 cells. Next, we treated miR-150-5p in HCT116 and HCT8 cells with VEGFA expression plasmid and found that VEGFA treatment could at least partially increase the expression of p-VEGFR2, p-Akt and p-mTOR (Fig. 8). These results revealed that miR-150-5p exerted its inhibitory function via the VEGFA/VEGFR2/Akt/mTOR signaling pathway.

Figure 8. miR-150-5p inhibited VEGFA/VEGFR2/Akt/mTOR signaling pathway in CRC. Western blot was used to measure the expression of VEGFA, VEGFR2, p-VEGFR2, Akt, p-Akt, mTOR, p-mTOR in transfected HCT116 and HCT8 cells. GAPDH was used as a loading control. Data are shown as the mean±SD of three independent experiments. *p<0.05, **p<0.01, ***p<0.001.

Patient samples

One hundred-twelve fresh CRC tissues and matched adjacent normal tumor tissues were obtained from CRC patients who received radical surgery at Nanjing First Hospital, Nanjing Medical University, between January 2001 and December 2007. The patients who received systemic treatment were excluded. All sample tissues were stored at -80°C until RNA extraction and embedding in paraffin. Informed consents were obtained from all CRC patients, and this study was approved by the Institutional Review Board of Nanjing First Hospital, Nanjing Medical University. Clinicopathologic characteristics of these patients are listed in Table 1.

Cell culture

The human CRC cell lines including HCT116, HCT8, HT29, SW620, SW480 and DLD-1 and normal colonic epithelial cells (FHC) were obtained from ATCC. These cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Gibco, USA) with 100 U/ml penicillin, 0.1 mg/ml streptomycin and 10% fetal bovine serum (FBS, Gibco, USA) at 37°C in 5% CO₂ atmosphere.

Cell transfection

AgomiR-150-5p, antagomiR-150-5p, small interference RNA targeting VEGFA (siVEGFA-1, siVEGFA-2) and their corresponding negative controls (agomiR-NC, antagomiR-NC and scrambled sequence) were obtained from GenePharma (Shanghai, China), and their sequences are listed in Supplementary Materials and Methods. pcDNA3.1-VEGFA and empty vector were obtained from GeneCreate (Wuhan, China). Cell transfections were performed using Lipofectamine^TM 2000 (Invitrogen, USA) following the manufacturer’s protocol.

Cell proliferation assays

Cell proliferation assays were performed using Cell Counting Kit (CCK-8, KeyGEN BioTECH, China) in accordance with the manufacturer’s instructions. Approximately 1×10⁴ transfected cells were seeded into a 96-well plate. After 12, 24, 48 and 72 h, 10 µl CCK-8 test solution was added into a 96-well plate and incubated for 3 h. The absorbance at 450 nm was measured in a microplate reader (Infinite M200 Pro, Tecan).

Wound-healing assays

HCT116 and HCT8 cells were seeded into 6-well plates and grew to reach 90% confluency. Then, cells were scraped using 200-µl pipette tips, and the wound closure was detected using an inverted microscope (Nikon, Japan) at 100× magnification.

Transwell invasion assays

The cell invasion was evaluated by transwell assay. For this, 3×10⁵ cells were cultured in 200 µl of serum-free media in the upper chamber precoated with 100 µl 2% Matrigel (BD Biosciences, USA). The lower chamber was filled with 500 µl of medium containing 10% FBS. After incubation for 36 h, the noninvaded cells were wiped off in the upper chamber with cotton swabs, and the cells in the lower surface were fixed by methanol, then stained with 0.1% crystal violet (Beyotime, China). The invaded cells were counted under inverted microscope (Nikon, Japan) at 200× magnification.

Tube formation assays

HUVECs cells (7×10⁵) were suspended in the mixture of tumor-conditioned medium (TCM, 300 µl) and DMEM containing 10% FBS (300 µl), and then plated into a 24-well plate precoated with Matrigel (200 µl per well, BD Biosciences, USA). Tube formation was observed after 6 h incubation at 37°C and was imaged with a computer-assisted inverted microscope (Nikon, Japan), the number of tube branches was counted using Image J software.

Dual-luciferase reporter assays

The 293T and HCT116 cells were cotransfected with pmirGLO-VEGFA-3’UTR-WT and pmirGLO-VEGFA-3’UTR-Mut reporter plasmids and agomiR-NC and agomiR-150-5p. Twenty-four hours after transfection, dual-luciferase reporter assay (Promega, Madison, WI, USA) was performed to measure the relative luciferase activity normalized to renilla luciferase activity.

Western blot

The total proteins were extracted from cells with RIPA lysis buffer mixed with phenylmethyl sulfonylfluoride (PMSF), protein inhibitors and phosphatase inhibitors (KeyGEN BioTECH, China). Equal amounts of proteins were separated with 10% SDS-PAGE gel and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, USA). The membranes were blocked with 5% bovine serum albumin (BSA) for 1.5 h and then incubated with primary antibodies: rabbit polyclonal anti-VEGFA (1:1000, ab46154, abcam, UK), anti-VEGFR2 (1:1000, ab39256, abcam, UK), anti-phospho(Y1175)-VEGFR2 (1:1000, ab194806, abcam, UK), anti-Akt (#9272, Cell Signaling Technology, USA), anti-phospho (Ser473)-Akt (1:2000, #4060, Cell Signaling Technology, USA), anti-mTOR (1:1000, ab2732, abcam, UK), anti-phospho(S2448)-mTOR (1:1000, ab84400 abcam, UK) and rabbit anti-GAPDH (1:10000, ab9485, abcam, UK). Proteins were then detected by enhanced chemiluminescence system (ECL) reagent (KeyGEN BioTECH, China) after incubation with secondary antibodies for 1 h at room temperature.

RNA isolation and quantitative RT-PCR

TRIzol reagent (Invitrogen, USA) was used to extract total RNA in accordance with manufacturer’s instructions. MiR-150-5p expression was detected by a Hairpin-it^TM microRNA and U6 snRNA normalization RT-PCR quantitation kit (Genepharma, China). For VEGFA mRNA, complementary DNA (cDNA) was synthesized using PrimeScript^TM reagent kit with gDNA Eraser (Takara, Dalian, China) and qRT-PCR was analyzed using SYBR Premix Ex Taq kit (Takara, Dalian, China). The relative expression of miRNA or mRNA was analyzed using the 2^-ΔΔCT method. All results were normalized to GAPDH or U6 gene. The primers for VEGFA and GAPDH mRNA are listed below (Table 3):

Immunohistochemistry

The CRC tissues, adjacent normal tissues and xenograft tissues were collected, paraffin-embedded and cut into 4-µm-thick sections. Sections were incubated with rabbit anti-VEGFA and anti- CD31 antibodies at 4°C overnight, followed by incubation with secondary antibodies. The sections were visualized under an inverted microscope (Nikon, Japan) at 200× or 400× magnification. The staining intensity was scored by two independent pathologists into the following four categories: no staining=0, weak staining=1, moderate staining=2 and strong staining=3. The stain-positive tissues were scored into four grades: 0 (0%), 1 (1-33%), 2 (34%-66%), and 3 (67%-100%). The final ICH score was calculated by multiplying the percentage of positive cells by the intensity score.

In vivo experiments

All animal experiments were approved by the Animal Care Committee of Nanjing First Hospital, Nanjing Medial University (acceptance No. SYXK 20160006). For xenografted tumor model, 1×10⁷ CRC cells in 0.2 ml PBS were subcutaneously injected into BABL/c nude mice, which were randomly divided into two group (n=8 per group). After tumor formation (day 6), 2 nmol agomiR-150-5p or agomiR-NC were injected into tumors, and the injections were performed 7 times at an interval of 2 days between each injection (day 6, 8, 10, 12, 14, 16, 18). The mice were sacrificed 2 days after their last injection, and the tumor volumes were calculated with the following equation:

V=0.5 × (length × width²)

For metastasis experiments, 2×10⁶ cells in 0.2 ml PBS were injected into the tail vein of nude mice which were randomly divided into two groups (n=8 per group). After 8 weeks of injection, computed tomography (CT) scans were performed to detect the lung metastasis in mice. After CT scans, mice were sacrificed, and their lungs were removed and stained by hematoxylin and eosin (HE) Staining.

Statistical analysis

All experiments were performed more than three times. The data are expressed as the mean ± SD and analyzed by SPSS 22.0 software. The Student’s t-test and one-way analysis of variance (ANOVA) were performed to detect the differences between groups. Pearson’s, Mann-Whitney U-test or χ2-test was used to analyze the relationship between miR-150-5p expression and clinicopathological features. The Kaplan-Meier method was applied to assess overall survival (OS). The survival curves were compared with log-rank test. The follow-up time was censored if the patient was lost` during follow-up. Cox proportional hazards model was used to perform multivariate analysis and calculate the 95% confidence interval (95% CI). A statistically significant difference was considered at P < 0.05.