MicroRNA-365 promotes lung carcinogenesis by downregulating the USP33/SLIT2/ROBO1 signalling pathway

This study was approved by the Ethics Review Board of Nanfang Hospital, Southern Medical University (Guangzhou, China). Twenty pairs of primary lung cancer tissues and their corresponding adjacent non-tumour tissue samples were collected in the Thoracic Surgery Department of Nanfang Hospital from March to June 2017. All experiments were performed according to relevant guidelines; informed consent was obtained from each patient. No patients received prior radiotherapy or chemotherapy. All specimens were removed during surgery and immediately stored at − 80 °C in liquid nitrogen for subsequent extraction of total RNA.

A549 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). SPC-A-1 and H1299 cells were purchased from the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China). All cells were cultured in RPMI-1640 medium or DMEM supplemented with 10% FBS (Gibco, Gaithersburg, MD, USA). All cells were maintained in a humidified incubator at 37 °C and 5% CO₂.

Total RNA was extracted from cell lines or tissues using a TRIzol Kit (Takara Bio, Shiga, Japan) according to the manufacturer’s instructions. cDNA was synthesized using Takara RT reagent (Takara Bio). qRT-PCR was performed on a Light Cycler 480 system (Roche Diagnostics, Basel, Switzerland) using a SYBR Green I Master kit (Roche). We used glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal reference. Mature miR-365a-3p expression was measured by qRT-PCR according to the Taqman MicroRNA Assay protocol (Takara Bio) and normalised using U6 small nuclear RNA with the 2^−ΔΔCt method.

Equal amounts of protein were separated by 10% SDS-PAGE and blotted onto PVDF membranes (Millipore, Bedford, MA, USA) probed with the following primary and secondary antibodies: monoclonal rabbit primary antibodies against SLIT2 and ROBLO1 (1:1000; Cell Signaling Technology, Boston, MA, USA) and β-tubulin (1:10,000; Bioworld Technology Inc., St. Louis Park, MN, USA); polyclonal rabbit primary antibody against USP33 (1:500; Abcam, San Francisco, CA, USA); and secondary fluorescent goat anti-rabbit antibody (LI-COR, Lincoln, NE, USA). Primary antibodies were applied overnight at 4 °C, and secondary antibody treatment was performed for about 1 h at 25 °C. An Odyssey Infrared Imaging System (LI-COR) was used to analyse immunoreactive bands. Western blotting was performed three times.

A549 or SPC-A-1 cells were seeded into a 6-well culture plate (200 cells/well) and incubated for 12 days. Cells were stained with Giemsa solution. Plates were scored by determining the number of colonies containing ≥ 50 cells.

5-Ethynyl-2ʹ-deoxyuridine incorporation assays were conducted using the EdU assay kit (RiboBio Co., Guangzhou, China) according to the manufacturer’s instructions. Cells were incubated with 50 nM EdU for 2 h at 37 °C. Cells were then fixed with 4% formaldehyde for 15 min at 25 °C and treated with 0.5% Triton X-100 for 20 min at 25 °C to permeate cell membranes. After washing with PBS three times, cells were incubated with 1× Apollo reaction cocktail (100 µL/well) for 30 min. DNA was stained with 10 µg/mL of Hoechst 33342 stain (100 µL/well) for 20 min, and staining was visualised with fluorescence microscopy. Five fields of view were randomly selected for each sample. EdU-positive cells were stained with red dye, and the relative proliferation-positive ratios were calculated from the average cell count of the five visualised fields.

The migratory and invasive abilities of cells were assessed using Transwell inserts (Corning, Inc., Corning, NY, USA) in 24-well plates. For invasion assays, each group of cells (5 × 10⁴ cells/100 µL) was resuspended in FBS-free RPMI-1640 medium and seeded into the upper chamber containing a Matrigel-coated membrane. After incubation for 24 h at 37 °C with 5% CO₂, the incubation medium and non-invading cells were removed from the upper surface of the membrane with cotton swabs. Invading cells that adhered to the lower surface of the chamber were fixed in 4% paraformaldehyde for 20 min and stained with 0.1% crystal violet for 30 min. Invading cells were photographed and manually counted at 200× magnification using a microscope (Olympus, Tokyo, Japan). For the Transwell migration assays, the process was the same, except the Transwell membrane was not precoated with Matrigel. Each assay was performed at least three times independently.

When cells had grown to approximately 90% confluency (after 48 h), an artificial wound was created with a 10-µL pipette tip. Cells were then cultured in fresh medium without FBS. Images were taken at 0 and 36 h to visualise wound healing. The relative percentage of the wound healed was calculated using the following formula: (width of wound at 0 h − width of wound at 36 h)/width of wound at 0 h.

AntagomiR-365, antagomiR-negative control (antagomiR-NC), agomiR-365, and agomiR-negative control (agomiR-NC) were designed and synthesized by GenePharma (Shanghai, China). pcDNA3.1-USP33 and pcDNA3.1 vectors were designed and synthesized by Obio Technology (Shanghai, China).

A549 and SPC-A-1 cells were seeded in 6-well plates at a density of 30–50%. Transient transfection was performed with Lipofectamine 2000 reagents (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. For all experiments, cells were collected 24–28 h after transfection.

The dual-luciferase reporter plasmids psi-CHECK2-USP33 (containing the wild-type USP33 3ʹ UTR binding site) and psiCHECK2-mGPC3 (containing a mutant USP33 3ʹ UTR) were constructed. A549 or SPC-A-1 cells were added to 24-well plates at 70–80% confluence 24 h before transfection. A mixture of 50 nM miR-365a-3p agomiR or antagomiR and 0.5 µg psi-CHECK2 reporter plasmid (psiCHECK2-wUSP33 or psiCHECK2-mUSP33) was co-transfected into cells using Lipofectamine 2000 reagent. At 48 h after transfection, luciferase activity was analysed using a dual-luciferase reporter assay system (Promega, Madison, WI, USA) according to the manufacturer’s instructions. Each experiment was performed in triplicate.

All animal experimental protocols were approved by the Animal Research Ethics Committee of Nanfang Hospital and complied with the rules of the specific pathogen-free (SPF) animal laboratory of the Nanfang Medical University. 4- to 6-week-old male mice were purchased from the Animal Laboratory Center of Nanfang Medical University (Guangdong, China). A549/antagomiR-365 cells and A549/antagomiR-NC cells were injected subcutaneously into the left and right axilla of five nude mice (4 × 10⁶ cells on each side) beginning 8 days after the injection of tumour cells. The mice were sacrificed 26 days post-injection to observe changes in tumour volume. Tumour size was measured every 3 days using the same protocol, and tumour volumes were calculated with the following formula: V = (L × W²)/2, where V is the volume (mm³), L is the biggest diameter (mm), and W is the smallest diameter (mm).

The Cancer Genome Atlas (TCGA) data portal provides a platform for researchers to query, download, and analyse data sets generated by TCGA (http://​cancergenome.​nih.​gov/​).

Data are expressed as mean ± SD. Each experiment was repeated at least three times unless otherwise indicated. Statistical analyses were performed using SPSS 22.0 software (SPSS, Chicago, IL, USA). Significant differences were analysed using Student’s t tests for continuous variables. Spearman’s correlation was used to analyse the relationship between miR-365a-3p and USP33 mRNA expression; P < 0.05 indicated significance.

To evaluate the expression of miR-365a-3p in lung adenocarcinoma cells, qRT-PCR assays were performed in 20 patients with lung adenocarcinoma and matched paracancerous tissues. U6 was used as an internal control. Results showed that the average expression level of miR-365a-3p in cancer tissues was higher than that in the corresponding paracancerous tissues (fold difference = 2.560, P = 0.0329; Fig. 1a). Next, the expression levels of miR-365a-3p were analysed in various lung cancer cell lines. qRT-PCR results demonstrated that higher expression levels of miR-365a-3p were present in A549, SPC-A-1, H1299, and PC9 lung cancer cell lines than in 16HBE normal lung epithelial cells. This was consistent with our clinical findings, indicating increased miR-365a-3p expression in lung adenocarcinoma tissues (Fig. 1b). Additionally, among the lung cancer cell lines, miR-365a-3p exhibited the highest expression in A549 cells and the lowest expression in SPC-A-1 cells; therefore, we selected these two cell lines for subsequent experiments. Taken together, the above results indicated that the abnormal expression of miR-365a-3p may be involved in lung carcinogenesis.

Our findings suggested that the dysregulated expression of miR-365a-3p may result in lung cancer progression. To explore the function of miR-365a-3p, A549 and SPC-A-1 cells were transfected with agomiR-365 or antagomiR-365 (Additional file 1). To investigate the effect of high expression of miR-365a-3p on the proliferation of lung cancer cells, we used plate clone formation and EdU assays. The results of the plate clone formation assay showed that the overexpression of miR-365a-3p increased the number of colonies (P < 0.05), whereas the opposite result was observed following the inhibition of miR-365a-3p expression (P < 0.05; Fig. 2a). The EdU assay results revealed that cellular fluorescence intensity was increased after transfection with agomiR-365 (P < 0.05) and decreased after transfection with antagomiR-365 (P < 0.05; Fig. 2b). Thus, the EdU assay results were consistent with those of the plate clone formation assay, further confirming that miR-365a-3p promotes the proliferation of lung cancer cells.

Next, we performed wound healing and Transwell migration and invasion assays. The wound healing assay demonstrated that wound healing rates were elevated following transfection with agomiR-365 (P < 0.05) and reduced following transfection with antagomiR-365 (P < 0.05; Fig. 2c). In the Transwell migration and invasion assay, the numbers of cells that migrated and invaded the Matrigel were elevated following transfection with agomiR-365 (P < 0.05) and reduced following transfection with antagomiR-365 (P < 0.05; Fig. 2d). Thus, the results of the Transwell migration and invasion assay were consistent with those of the wound healing assay, indicating that miR-365a-3p promotes the migration and invasion activities of lung cancer cells.

The above in vitro experiments demonstrated that miR-365a-3p is closely correlated with the proliferation, migration, and invasion of lung cancer cells. Analysis using multiple databases showed that USP33 may be a target gene of miR-365a-3p. TCGA data showed a negative correlation between miR-365a-3p and USP33 expression, and survival analysis demonstrated that high expression of USP33 was associated with good prognosis (Additional file 2). Additionally, it was previously reported that USP33 exhibits low expression in lung adenocarcinoma tissues and cells [30, 34]. Thus, we sought to determine whether miR-365a-3p interacts with USP33. A western blot assay showed that the expression of USP33 was significantly downregulated following the overexpression of miR-365a-3p (Fig. 3a). Next, qRT-PCR was used to detect the expression of miR-365a-3p and USP33 in tissues derived from 20 cases of lung adenocarcinoma. Consistent with our hypothesis, we observed a negative correlation between the expression of miR-365a-3p and that of USP33 in these tissues (R² = 0.2467, P < 0.05; Fig. 3b). To further verify that miR-365a-3p directly targets and downregulates USP33, the wild-type 3′ UTR of USP33 (WT USP33 3′ UTR) or a mutated USP33 3′ UTR (Mut USP33 3′ UTR) was cloned into a dual-luciferase UTR vector and then co-transfected with agomiR-365 or a negative control (agomiR-NC) into A549 and SPC-A-1 cells. At 48 h after transfection, cells were harvested and lysed to detect luciferase activity (Additional file 3, Additional file 4: Table S1). The results showed that relative luciferase activity was significantly decreased in cells co-transfected with WT USP33 3′ UTR and agomiR-365. However, there was no significant change in luciferase activity after co-transfection with Mut USP33 3′ UTR and agomiR-365 (Fig. 3c). This suggested that miR-365a-3p directly targets and binds the USP33 3′ UTR.

The above experiments demonstrated that miR-365a-3p targets USP33 and downregulates its expression. Next, functional rescue assays were performed to further verify that the miR-365a-3p-mediated downregulation of USP33 promotes the proliferation, migration, and invasion of lung cancer cells. The pcDNA3.1-USP33 plasmid was constructed to overexpress USP33, with the empty pcDNA3.1 vector used as a negative control (Additional file 1b). The plate clone formation assay results showed that the numbers of cell colonies derived from A549 and SPC-A-1 cells co-transfected with pcDNA3.1-USP33 and agomiR-NC were significantly reduced (P < 0.05), whereas these numbers increased markedly compared with those in negative controls after co-transfection with pcDNA3.1 vector and agomiR-365 (P < 0.05). In contrast, the simultaneous overexpression of USP33 and miR-365a-3p by co-transfection with pcDNA3.1-USP33 and agomiR-365 resulted in no significant changes in the number of cell colonies compared to that in the control group (Fig. 4a). The EdU assay results revealed that the overexpression of USP33 in combination with the overexpression of miR-365a-3p mitigated an increase in the cellular fluorescence intensity caused by the overexpression of miR-365a-3p alone, consistent with the results of the plate clone formation assay (Fig. 4b). This further suggested that miR-365a-3p promotes the proliferation of lung cancer cells by downregulating USP33. Similarly, the wound healing assay showed that the wound healing rate increased in cells overexpressing miR-365a-3p alone and decreased in the presence of pcDNA3.1-USP33 (Fig. 4c). The same patterns were observed in the Transwell migration and invasion assay (Fig. 4d).

The above series of in vitro experiments suggested that miR-365a-3p promotes lung cancer progression by downregulating USP33; however, the signalling pathways involved in this process remain unknown. It has been reported that the USP33/SLIT2/ROBO1 signalling pathway inhibits the migration of cancer cells; therefore, we hypothesised that miR-365a-3p downregulates this pathway to promote lung cancer progression. To verify our hypothesis, a western blot assay was performed to observe changes in the expression of SLIT2 and ROBO1 in A549 and SPC-A-1 cells overexpressing USP33 or miR-365a-3p alone or in combination. The results showed that the expression levels of SLIT2 and ROBO1 were upregulated following the overexpression of USP33 (Fig. 5a, b) but downregulated following the overexpression of miR-365a-3p (Fig. 5c). However, when both miR-365a-3p and USP33 were overexpressed, there was no significant change in the expression of SLIT2 or ROBO1 compared to levels in the control group (Fig. 5d). These results indicated that miR-365a-3p promotes lung cancer progression by downregulating the USP33/SLIT2/ROBO1 signalling pathway.

The results of this in vivo experiment showed that tumourigenic ability began to differ on day 8 post-injection. Beginning on day 8, tumour size changes were measured every 3 days using a slide rule, and tumour growth curves were plotted (Fig. 6). The tumour volumes on day 26 in mice injected with A549/antagomiR-NC cells were significantly larger than those in mice injected with A549/antagomiR-365. Our findings demonstrated that tumours derived from A549/antagomiR-365 cells grew more slowly than those derived from A549/antagomiR-NC cells. Thus, the results of the subcutaneous implantation tumour model further demonstrated that miR-365a-3p promotes tumour formation in vivo.