LINC00037 Inhibits Proliferation of Renal Cell Carcinoma Cells in an Epidermal Growth Factor Receptor-Dependent Way

Clear cell renal cell carcinoma (ccRCC) accounts for approximately 70% of renal cancer, which is one of the most common cancers, with approximately 202, 000 cases and 102, 000 deaths per year around the world [1, 2]. In clinical, RCC is characterized by anonymous symptoms and early metastasis [3, 4]. Early detection and selection of accurate targets for ccRCC could be beneficial to improve of prognosis of ccRCC patients [5, 6].

Long noncoding RNAs (lncRNAs) are non protein coding RNAs ranging from 200 nt to 100 kb in length [7]. The past decade has witnessed the discovery of diverse IncRNAs involved in physiological and pathological processes [8-12]. Many IncRNAs have been reported as important regulators in ccRCC, and several were also verified as biomarkers for early diagnosis and prognostic prediction for ccRCC [13, 14].

It is of vital significance to clarify the underlying mechanisms of molecular changes in ccRCC thus can contribute new strategies for the diagnosis and therapies to improve the prognosis. Whereas the functional roles and clinical value of the greatmass of IncRNAs in the occurrence and progression of ccRCC remain largely unknown.

Sequencing technologies and bioinformatics analyses are widely used to analyze the expression patterns of IncRNAs in diverse diseases and to identify potential disease associated genes [15-17]. Here we studied a IncRNA, LINC00037, which was high expressed in ccRCC tissues comparing to corresponding normal tissues [18-20]. We further explored its functional role in RCC cells and the underlying mechanism, which demonstrated that LINC00037 may be a therapeutic target for ccRCC.

The ccRCC cell lines A498 and 786-O cells were conventionally maintained in Dulbecco's improved Eagle's medium (DMEM) (Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, Carlsbad, CA, USA), L-glutamine (2 mM), penicillin (100 U/ml), and streptomycin (100 mg/ml). The cells were grown in an environment of 37 º C, 5% C02. We purchased ccRCC cell lines A498 and 786-O cells from the Shanghai Institute of Biochemistry and Cell Biology (Chinese Academy of Sciences, Shanghai, China).

4-8 weeks old male BALB/c nu/nu mice from the Experimental Animal Center of Yangzhou University were fostered in the absence of pathogen conditions. All the animals were subjected to humanitarian care, and all experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals.

As mentioned above, Lentivirus plasmids constructed with the short hairpin RNA (shRNA) of the LINC00037 sequence (lentiviral-short hairpin RNA, Lv-shRNA) were used to transfectA498 and 786-O cells. The lentiviral plasmids were constructed using the interference sequence (named Lv-NC) as a control. The lentiviral vector (pll3.7) was inserted with the synthetic and purified LINC00037 gene fragment, termed Lv-LINC00037. The packaged recombinant lentivirus thentransfected LINC00037 knockdown cells. The shRNA of EGFR was used for the targeting sequence: GGCTGGTTATGTCCTCATT. Construction of lentivirus plasmids using shRNA of EGFR (EGFR-shRNA), was designed for transforming ccRCC cells as mentioned above.

For total RNAs extraction from fresh liver tissues and cells, TRIzol reagent (Invitrogen) was used. The SurePrep Nuclear or Cytoplasmic RNA Purification Kit (Life Science SOURCE, Biovision, Milpitas, CA, USA) was used to extract RNA from the cytoplasm and nuclear. Reverse transcriptase (TaKaRa) kit was used for reverse transcription of total RNA (500ng) in order to detect mRNA. Has-5S was used for internal control. SYBR Green Mastermix kit (TaKaRa, Tokyo, Japan) was used to detect LINC00037 and mRNA expression level by quantitative real-time PCR and triplicate assays were used for analysis on the ABI Prism 7900HT (Applied Biosystems, Foster City, CA, USA) according to the instructions. Primer 3.0 software was used for designing primer sequences (http://www.simgene.com/Primer3). Primer sequences were as follows:: Has-5S: Forward: 5'-GGAGAGGGAGCCTGAGAAACG-3' and Reverse: 5'-TTACAGGGCCTCGAAAGAGTCC-3', human LINC00037: Forward: 5'-CACGAGTGTAGTGCCCAGTT-3' and Reverse: 5'-GGTCAGGGACCTTTGTCGTT-3' and human EGFR: Forward: 5'- TCCTCTGGAGGCTGAGAAAA-3' and Reverse: 5'-GGGCTCTGGAGGAAAAGAAA-3'.

Total proteins were extracted from cultured cells using radio-immunoprecipitation assay buffer added with protease and phosphatase inhibitors (Beyotime, Nantong, China) and then were quantified by the Bradford assay (Bio-Rad Laboratories, Hercules, CA, USA). Each lane was added with the same amount of protein samples (30 µg). Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was used for protein separation and then the separated protein was transferred to a polyvinylidene fluoride membrane. Antibodies against EGFR (Cell Signaling Technology, Danvers, MA, USA) and glyceraldehyde-phosphate dehydrogenase (GAPDH) (Cell Signaling Technology) were used for immunoblotting. The integrated density of the bands was quantified by ImageJ software (NIH, Bethesda, MD, USA).

For apoptosis detection, the Annexin V-FITC/PI Apoptosis Detection Kit (Vazyme Biotech, China) was used and cells were treated with 0.05 mM H₂O₂ for 2 hours to stimulate apoptosis. For cell cycle measurement, a Cell cycle Assay Kit (Vazyme Biotech) was used. Finally, the FACS Calibur flow cytometer served as cell analysis with CellQuest software (BD Biosciences, New York, NY, USA).

Cell proliferation was assessed with a CCK8 kit (Vazyme Biotech, China) and a Cell-Light EdU Apollo567 In vitro Kit (RiboBio, Guangzhou, China). For CCK8 detection, transfected cells (2×10³) were transferred to 96-well plates, incubated for 24,48, 72 and 96 hours, then added with CCK8 reagent and finally incubated at 37°C for 2 hours. For absorption measurement, microplate reader at 450 nm (ELX-800; Bio-Tek, Winooski, VT, USA) was used. For 5-ethynyl-2'-deoxyuridine (EdU) detection, the transfected cells (2×10⁵) were first transferred to Glass Botttom Cell Culture Dishes (Nest Biotechnology, NJ, USA) and then treated as described. Finally, the treated cells were measured by laser confocal scanning microscopy. For invasion assay, the Transwell units (Corning Costar, Tewksbury MA, USA) percolated with Matrigel (BD Biosciences, New York, NY, USA) were used to evaluate. Cells (2 ×10⁴cells/well) were transferred to the upper chamber in DMEM without FBS and the lower chamber was full of DMEM containing 10% FBS as a chemokine. After 48 hours of incubation at room temperature, the filters were gathered and fixed with 4% paraformaldehyde and then dyed with 0.1% crystal violet. Remove the non-invasive tumor cells from the top of the filters with a cotton swab and then count the cells through the filter under an optical microscope.

Cells (5×10⁶) knocked down or overexpressed LINC00037 in ccRCC and were inoculated into the bilateral armpits of each BALB/C nude mice in a subcutaneous manner. After five weeks of inoculation, the mice were all killed and tumors were weighed. Tumor tissues were stripped as a whole.

The agarose powder (0.5 g) was dissolved in the 0.5% Tris-acetate-EDTA (TAE) buffer of 50 ml and heated to near boiling state followed by 2.5 µL of GoldView (Beyotime, China, Nantong, China) and then mixed. The cDNA was made by reverse transcription of the total RNA extracted from the cell lines, and the cDNA was mixed with the loading buffer (Beyotime, Nantong, China) and the mixture was added to the wells. 0.5% TAE was used as the running buffer and electrophoresis was performed at 80V for 40 minutes. Image Lab software with an ultraviolet (UV) transilluminator was performed to analyze data.

Polyacrylamide gel was made according to the standard protocol. Twenty microliters of each sample were mixed with 10 × running buffer (CapitalBio Corp., Beijing, China) and the mixture was added. The gel was operated at 120 V for 2.5 hours and then dyed by Fast Silver Stain Kit (Beyotime). Finally, the lanes were cut into ten pieces and placed in Eppendorf tubes respectively. NH₄HCO₃ (50 mM) was loaded and using a pipette broke the gel into pieces. The gel solution was then digested overnight at 37°C using trypsin (Promega, Fitchburg, WI, USA). Tubes were shaken, added methyl cyanide (HPLC grade) and then centrifuged at room temperature for 2 minutes. After centrifugation, the supernatant was sucked into a clean tube and dried at 60°C by means of a vacuum concentrator. The peptides were analyzed by LC-MS/MS of nano-LC together with Orbitrap Q Exactive mass spectrometer (Thermo Scientific, Waltham, MA, USA) at a scan range of m/z 400–1500. The original files were analyzed using the Thermo Proteome Discoverer (1.4.0.288) software (http://www.thermoscientific.com/en/product/proteome-discoverer-software.html) platform. MS/MS spectra of protein identified were retrieved from the protein database (human-refseq-20140303-71465s.fasta, National Center of Biotechnology Information).

RIP was conducted with a Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Bedford, MA, USA). The co-precipitated RNAs were determined by quantitative real-time PCR after pulling down assay with anti-EGFR antibody (Cell Signaling Technology).

The results of quantitative real-time PCR were showed in the way of mean ± S.E.M. The statistical differences between clinical and demographic characteristics were checked by Student's t test and χ² test. Statistical analysis was conducted by STATA version 9.2 (Stata Corp., College Station, TX, USA) and SPSS version 18.0 (SPSS Inc, Chicago, IL, USA) and presented with the GraphPad prism software (GraphPad Software, San Diego, CA, USA). Finally, P <0.05 was recognized statistically significant in all cases.

LINC00037 is located on chromosome 22 (22qll.21) and it is not a conserve gene as indicated by the homologous analysis (Fig. 1A). Also, it is also reported as a non protein coding RNA in the LNCipedia database and in HUGO Gene Nomenclature Committee (HGNC) database (Fig. 1B). LINC00037 is widely found in various tissues of homo sapiens (Fig. 1B). So we searched the several databases and found studies on LINC00037 in ccRCC consistently presented that LINC00037 was highly expressed in ccRCC compared with that of heathy controls [18-20] (Fig. 1C-G); however no studies investigated its function and the underlying mechanism in ccRCC.

To detect the functional role of LINC00037 in regulatingthe biological behaviors of cells, shRNAs plasmids packaged into lentivirus plasmid, termed Lv-shRNA were used to knockdown LINC00037 in 786-O and A498 cells (Fig. 2A and 2B). The cell proliferation rate assay was performed with a CCK8 kit and an EdU kit, which showed that LINC00037 knockdown resulted in significantly inhibited proliferative ability of 786-O and A498 cells. When the expression of LINC00037 was rescued by LINC00037 overexpression with lentivirus plasmids containing LINC00037 cDNA sequence (Lv-LINC00037) in 786-O and A498 cells, the reduced cell proliferation rate was restored and improved (Fig. 2C-G). The role of LINC00037 in regulating cell invasion was also explored, however, no significant difference was observed in Transwell assay (data not shown).

We further potential role of LINC00037 in regulating ccRCC cell apoptosis and cell cycle with flow cytometry assay. It showed that when LINC00037 was knockdown, ccRCC cells presented higher level of apoptosis, and this effect was reversed by re-expression of LINC00037 in ccRCC cells (Fig. 3A). The cell cycle assay showed that LINC00037 knockdown resulted in significant cell cycle arrest. Over-expression of LINC00037 in LINC00037 knockdown cells have re-established the cell ability with reduced cell cycle arrest (Fig. 3B).

To study the effect of LINC00037 on tumor growth in vivo, BABL/c nude mice were used to construct the xenotransplantation model through subcutaneously injection with Lv-NC, Lv-shRNA or Lv-LINC00037 transfected ccRCC cells. It showed that ccRCC cells after LINC00037 depletion presented significantly attenuated tumor growth, which can be restored and improved by LINC00037 re-expression (Fig. 4A and 4B).

In order to determine the underlying role of LINC00037 in ccRCC, we firstly detected the subcellular location of LINC00037 transcript by real-time PCR after amplifying with separating nuclear and cytoplasm RNA in 786-O cells. The result showed that LINC00037 was located primarily in the cytoplasm of 786-O cells (Fig. 5A). This indicated that LINC00037 may function via interacting with components in the cytoplasm, based on which, we conducted the RNA pull-down assay with the biotin-labeled IncRNA. Through rapid silver staining, we found that protein band at the ∼170 kDa location indicated most significant difference in amount (Fig. 5B). Then protein bands at the ∼170 kDa were cut for mass spectrum screening and we found that peptide of EGFR was most differently enriched between two groups (Fig. 5C).

Then we detected the expression of EGFR by both quantitative real-time PCR and western blotting in ccRCC cells with LINC00037 depletion or overexpression. We found that there was significant difference at the protein level (Fig. 5D and 5E), however, no significant change was found in mRNA level of EGFR (data not shown). Therefore, we hypothesized that LINC00037 may function through interacting with EGFR protein in ccRCC.

Then we used anti-EGFR antibody to performed the protein pull-down assay. The pulldown products were collected for RNA extraction. Through real-time PCR and agarose gel electrophoresis, we found that LINC00037 was enriched in anti-EGFR pull-down products (Fig. 6A-E), which suggested that LINC00037 could directly bind to EGFR protein in ccRCC.

Studies on the expression patterns and regulatory mechanisms of functional lncRNAs in ccRCC may provide promising therapeutic targets and novel diagnostic or prognostic biomarkers for ccRCC. In this study, we discussed the functional role of LINC00037 in ccRCC. We further explored its functional roles and the underlying mechanism in RCC cells, which demonstrated that LINC00037 is a potential therapeutic target for ccRCC.

As described in the NCBI database, LINC00037, also called DiGeorge syndrome critical region gene 5 (DGCR5), is a human noncoding RNA validated by the Human Gene Nomenclature Committee. It is located at chromosome 22q11.21 with 2113 bp and has six exons. Previous study by Huang et al. [21], reported that LINC00037 was low expressed in serum of patients with HCC, and may act as a potential biomarker for the diagnosis and prognosis in HCC. Liu et al. [22] performed a study with data from The Cancer Genome Atlas - Cancer Genome (TCGA) database, which suggested that LINC00037 was low expressed in and correlated with better OS in lung squamous cell carcinoma. And Yong et al. [23] found that that the down-regulation of LINC00037 predicted poor prognosis in pancreatic ductal adenocarcinoma (PDAC). These implied the tumor suppressive role of LINC00037. However, in ccRCC, several studies have reported that LINC00037 was highly expressed in ccRCC tissues compared to normal renal tissues, which consistently indicated the oncogenic role of LINC00037 in ccRCC [18-20]. In consideration of the contrasting roles predicted by studies of LINC00037 expression patterns, the underlying mechanism of LINC00037 in cancer should be investigated and clarified.

To study the function of LINC00037 in ccRCC, we firstly knockdown LINC00037 in two ccRCC cell lines, 786-O and A498 cells, then we also re-expressed LINC00037 in the former LINC00037 knockdown ccRCC cells. In vitro study showed that absence of LINC00037 resulted in attenuated cell proliferation, however, the invasive ability of ccRCC cells was not significantly changed. Re-expression of LINC00037 effectively restored and enhanced the proliferative ability but not affect the invasive ability of ccRCC cells. Moreover, cell apoptosis and cell cycle arrest was obviously prompted by LINC00037 depletion, and the former effects can be reversed by LINC00037 re-expression. From the in vivo study, we observed that knockdown LINC00037 in ccRCC cells significantly inhibited the tumor growth and re-expressing LINC00037 observably reversed the growth inhibition. The mechanism of IncRNA is closely associated with its location. LncRNAs in cell nucleus often act as a transcription factor, whereas those in cytoplasm usually function through binding to certain proteins. To explore the subcellular location of LINC00037 in ccRCC cells, real-time PCR analysis with separated nuclear and cytoplasm RNA was performed. The result showed that LINC00037 was located primarily in the cytoplasm of 786-O cells, therefore may function through directly binding to certain proteins. Subsequently, we carried out the RNA pull-down assay and MS assay in 786-O cells with LINC00037 depletion or not, which showed that EGFR was most differently enriched between two groups.

EGFR is vitally involved in regulating cell survival and proliferation [24, 25]. And in model systems of autocrine stimulation of EGFR, or EGFR overexpression and mutation, consequent transformation has been observed [26, 27]. EGFR may function through activating a network of signaling pathways, involving Ras, phosphoinositide 3-kinase and the signal transducer, activator of transcription family [28-30], and so forth in ccRCC [31, 32].

Previous studies have found that IncRNAs may increase the level of certain proteins by directly binding to and stabilizing the later [33-35]. However, whether IncRNAs modify ccRCC development and progression via EGFR dependent pathway remains unknown. Several IncRNAs have been reported to bind to EGFR protein, such as LINC00125 [28] and LINC-EGFR [36]. In our study, we observed that the protein level but not the mRNA level of EGFR was significantly reduced as a result of LINC00037 knockdown in ccRCC cells, which can be reversed by LINC00037 overexpression. These suggested that LINC00037 may affect EGFR signaling primarily via regulating EGFR protein. To investigate whether the regulation is directly we performed EGFR protein pull-down assay with anti-EGFR antibody. It showed that LINC00037 was significantly enriched in the pull-down products as a compound with EGFR protein.

Our study showed the aberrantly up-regulation of LINC00037 in ccRCC, and further demonstrated that LINC00037 played a crucial role in ccRCC through binding to and increasing the level of EGFR protein as a consequence. This may provide a promising therapeutic target for ccRCC. However, more studies are warranted for further investigation of the function of LINC00037 in the future.

Authors involved have contributed adequately to this work and have no conflicts of interest to disclose.

None.

The details of the study design and main results were presented in the flowchart (Fig. 7).

X. Gong and X. Du are co-first authors.