MicroRNA-4651 targets bromodomain-containing protein 4 to inhibit non-small cell lung cancer cell growth

Background: Bromodomain-containing (BRD4) overexpression (NSCLC) is important for cancer cell progression. The aim of the present study is to silence BRD4 through expression of its targeted microRNAs in NSCLC cells. Methods: Expression of BRD4 and its targeting miRNA, microRNA-4651 (miR-4651), was tested by qPCR and Western blotting assays. Genetic strategies were utilized to exogenously alter miR-4651 expression. NSCLC cell growth, proliferation and migration were tested. Results: miR-4651 selectively targets and negatively regulates BRD4 in A549 and primary human NSCLC cells. The Ago-2 immunoprecipitation experiments further confirmed that miR-4651 directly binds to BRD4 mRNA. Significantly, in A549 cells and primary NSCLC cells ectopic overexpression of miR-4651 downregulated BRD4’s 3-UTR activity and its expression, both were however elevated with miR-4651 inhibition. Functional studies demonstrated that NSCLC cell growth, proliferation and migration were significantly inhibited with miR-4651 overexpression, but enhanced with miR-4651 inhibition. BRD4 re-expression, by an 3’-UTR mutant BRD4, reversed miR-4651 overexpression-induced inhibitions on A549 cells. Additionally, miR-4651 overexpression or inhibition failed to affect the functions of BRD4-KO A549 cells. In vivo, miR-4651-overexpressed A549 xenografts grew significantly slower than control A549 xenografts in severe combined immunodeficient mice. At last we show that miR-4651 is downregulated in human NSCLC tissues, correlating with BRD4 elevation. Conclusions: miR-4651 targets BRD4 to inhibit NSCLC cell growth in vitro and in vivo.


Background
Lung cancer has become one leading cause of cancer-related human mortalities globally, with estimated over two million new cases and 1.75 million deaths in 2018 [1]. Non-small cell lung cancer (NSCLC) is the primary type of lung cancer, accounts for over 80% of all lung cancers [1,2]. In the clinical practices, surgical resection is one preferred therapy for NSCLC [2][3][4]. Yet the high recurrent rate as well as the relatively resistant to adjuvant chemotherapy/targeted therapies preclude the surgical treatment as the curative option in the fast majority of NSCLC patients [2][3][4]. Therefore, there is an urgent need to further explore the underlying pathological mechanisms for NSCLC progression, and to develop possible novel targeted strategies [3,5].
Bromodomain-containing protein 4 (BRD4), the most-studied and one primary BET (bromodomain and extraterminal) family protein, contains two N-terminal bromodomains recognizing the acetylated lysine residues [6]. Published literatures have demonstrated that BRD4 binds directly to acetylated histones and transcription factors, and recruits positives transcription elongation factor b (P-TEFb) and possible other transcriptional regulators [6]. BRD4 is overexpressed in NSCLC, correlating with the poor prognosis and other clinical features of NSCLC patients [7,8]. Studies have proposed that BRD4 is a major contributor to the invasive phenotype of NSCLC, represents as an important therapeutic oncotarget [7,8]. It is critical for the translation and expression of several key oncogenes, including c-Myc, Bcl-2 and cyclin D1 [6,[9][10][11]. Inhibition or silencing of BRD4 can potently inhibit NSCLC cell progression in vitro and in vivo [7,8].
MicroRNAs (miRNAs) are a large family of conserved, short and single-strand noncoding RNAs (ncRNAs), with 21-25 nucleotides long [12][13][14][15]. By sequence-specific binding to the 3'-UTR (untranslated region) of targeted mRNA, miRNA could suppress mRNA translation and/or induce mRNA degradation, thereby regulating gene expression at the post-transcriptional level [12][13][14][15]. In cancer cells miRNAs could exert regulatory effects to almost all cancerous behaviors, including cell proliferation, cell cycle progression, apoptosis, metastasis and epithelial-mesenchymal transition (EMT) [16,17]. Dysregulation of miRNA has been commonly detected in NSCLC and other lung cancers, plays a pivotal role in cancer formation, tumorigenesis, progression, therapy-resistance and prognosis predication [16,18]. Ulivi et al., revealed that miR-328 is significantly higher in NSCLC patients than that in the healthy donors, represents a potential diagnostic biomarker of NSCLC, especially for early-stage tumors [19]. Similarly the study by Wang et al., demonstrated that miR-21 expression is higher in NSCLC patients' serum samples, which has significant prognostic significance [20]. Previous functional literatures have shown that miR-152 negatively regulated ADAM17 and possible other targets to suppress NSCLC cell progression [21,22].
Studies have suggested that expression of BRD4-targeting miRNAs could be a novel and efficient strategy to inhibit human cancer cells with BRD4 overexpression [23][24][25]. The results of this study 4 indentified microRNA-4651 (miR-4651) as a novel BRD4-targeting miRNA, suppressing NSCLC cell progression in vitro and in vivo. complementary DNA (cDNA), which was subjected to qPCR analyses using a SYBR Premix Ex Taq™ Kit (TaKaRa, Tokyo, Japan). GAPDH was tested as the reference gene and internal control for testing BRD4 and its target genes. Expression of miR-4651 was tested through the TaqMan microRNA qPCR assay kit, using U6 as the internal control. qPCR quantification was through 2 −ΔCt method using the following formula: 2 −(Ct of target gene−Ct of reference gene) . The RNA primers of this study were listed in Table-1. BRD4 luciferase reporter assay. As previously described [28], the BRD4's 3'-UTR sequence containing the predicted miR-4651 binding sites (at position 1715-1722) was amplified by Shanghai Genechem Co. (Shanghai, China), which was sub-cloned into the pMIR-REPORT miRNA Expression Reporter vector (Ambion; Thermo Fisher Scientific, Shanghai, China), establishing the luciferase reporter plasmid, pMIR-BRD4-3ʹ-UTR. The latter was transfected to NSCLC cells using the Lipofectamine 2000 protocol.

Materials And Methods
Cells were then subjected to applied genetic modifications, with luciferase activities detected at 48 h by the Dual-Luciferase Reporter assay system (Promega).
Transfection of miR-4651 mimic. NSCLC cells were initially seeded into the six-well tissue-culturing plates (2 × 10 5 cells per well). Lipofectamine 2000 protocol was carried out for the transfection of 500 nM of the wild-type ("WT") or the mutant ("Mut") miR-4651 mimics (synthesized by Shanghai Genechem Co.). After 48 h miR-4651 expression levels were determined by qPCR.
Ago2 immunoprecipitation (Ago2-IP). For each condition, six million NSCLC cells were trypsinized, 6 homogenized (using the commercial lysis buffer from Beyotime, Wuxi, China), and cell lysates subjected to the Ago2-IP (with biotinylated miR-4651, generated by Shanghai Genechem Co.) under the previously described protocol [29]. As the negative control IP experiments were also carried out using the non-immune IgG beads (Sigma). The IP-pull down RNA was used as template for qPCR assay, and results were quantified to input controls ("Input").
Cell-counting kit 8 (CCK-8) assay. NSCLC cells with the applied genetic treatments were trypsinized, counted, and inoculated into the 96-well tissue-culturing plates (at the initial density of 4,000 cells per well), incubated at 37 °C for 72 h. In each well 10 µL of CCK-8 reagent (Dojindo, Kumamoto, Japan) was added and incubated for 2 h, with CCK-8 absorbance examined at the wavelength of 450 nm.
Colony formation. A549 cells with applied genetic treatments were initially seeded at 3,000 per well into the 12-well tissue-culturing plates. After seven days colonies were stained, with large colonies (> 100 cells/per colony) counted manually.
Transwell migration assay. NSCLC cells with the applied genetic treatments were trypsinized and suspended into serum-free medium. For each treatment 5 × 10 4 cells were added to the upper surfaces of Matrigel-coated Transwell chambers (BD Biosciences, Shanghai, China) with 12 µm poresize. The lower chambers were filled with complete medium with 15% FBS. Cells were allowed to migrate for 24 h. Cells in the upper chambers were removed, with the invasive cells in the lower chambers fixed (by 4% paraformaldehyde), stained and counted.
EdU assay. NSCLC cells with the indicated genetic modifications were trypsinized and initially seeded into the six-well tissue-culture plates (at 2 × 10 5 cells per well), and cultured for 48 h. Using an EdU UTR mutant BRD4. The in vitro site-directed mutagenesis system (Genechem, Shanghai, China) was applied to generate 3'-UTR mutant BRD4 vector (at the miR-4651's binding site, 1716C > A), mut-BRD4. The construct was sub-cloned into the GV248 lentiviral vector, and added to A549 cells.
Expression of the UTR mut-BRD4 was verified by Western blotting.
In vivo tumor growth assay. A549 cells with the applied genetic treatments were collected and injected subcutaneously into the right flanks of the severe combined immunodeficient mice (SCID) mice (all female, 18-19.5 grams in weights, purchased from Soochow University Animal facility, Suzhou, China). When the volume reached approximately 100 mm 3 for each tumor ("Day-0"), the recordings were started. Tumor volumes were calculated through a previously-described protocol Statistical analysis. Data were presented as the mean ± standard deviations (SD), analyzed by SPSS 19.0 software (SPSS Co. Chicago, CA). The one-way ANOVA followed by Student-Newman-Keuls post hoc test was performed to determine statistical significance between multiple groups. When comparing difference between two groups the two-tailed Student's t-test (Excel 2007) was utilized.
Values of P < 0.05 were considered as statistically different.
To support that BRD4 is the direct target of miR-4651, two mutant miR-4651 mimics ("Mut1/2"), containing mutations within the binding sites of BRD4's 3-UTR (see Figure 1G), as well as the wild-9 type (WT) miR-4651 mimic, were transfected to A549 cells. As shown, only the WT miR-4651 mimic, but not the mutants, inhibited BRD4's 3-UTR activity ( Figure 1H) and downregulated BRD4 mRNA levels ( Figure 1I). We also tested whether miR-4651 can exert similar activity in primary cancer cells.

miR-4651 overexpression inhibits lung cancer cell growth, proliferation and migration, while activating cell apoptosis
We next tested whether ectopic overexpression of miR-4651 could affect the function of lung cancer cells. Results of cell counting assay, in Figure 2A, demonstrated that LV-pri-miR-4651-expressing A549 cells (see Figure 1) grew significantly slower than the parental control cells. Furthermore, miR-4651-overexpressed stable cells presented with reduced cell viability (CCK-8 OD, Figure 2B). A549 cell colony formation was potently inhibited as well with miR-4651 overexpression ( Figure 2C).
Additionally, EdU incorporation in the two stable cell lines with LV-pri-miR-4651 was largely suppressed, as compared to that in the control cells ( Figure 2D). Analyzing cell migration, by the "Transwell" assays, showed that miR-4651 overexpression decreased the number of migrated A549 cells ( Figure 2E). These results show that ectopic overexpression of miR-4651 inhibited A549 cell growth, proliferation and migration.
Testing the potential function of miR-4651 on cell apoptosis, we show that LV-pri-miR-4651 significantly increased the number of the Annexin V-positive A549 cells (Figure 2F), and induced cleavages of caspase-3 and poly (ADP) ribose polymerase (PARP) (Figure 2G). These results indicated the pro-apoptosis activity by miR-4651 overexpression. The non-sense control miRNA vector ("miRC") had no significant effect on the functions of A549 cells (Figure 2A-G). In the primary human lung cancer cells, Pri-Ca-1/2/3, LV-pri-miR-4651-induced miR-4651 overexpression (see Figure 1) significantly inhibited CCK-8 viability (Figure 2H), EdU incorporation ( Figure 2I) and cell migration ( Figure 2J). A significant apoptosis activation, evidenced by an increased Annexin V-staining, was detected in miR-4651-overexpressed primary cancer cells ( Figure 2K). Together, we show that miR-4651 overexpression inhibited lung cancer cell growth, proliferation and migration, while activating cell apoptosis.

miR-4651 overexpression inhibits A549 cell growth in SCID mice
To study the potential role of miR-4651 on NSCLC cell growth in vivo, LV-pri-miR-4651-expressing stable A549 cells (see Figure 1 and 2) and miRC-expressing control A549 cells were inoculated into the right flanks of the SCID mice. Testing tumor growth by recording tumor volumes, Figure 5A, demonstrated that miR-4651-overexpressed A549 xenografts grew significantly slower than control A549 xenografts expressing miRC. Calculating daily tumor growth, by the formula (tumor volume at Day-35-tumor volume at Day-0)/35, showed that A549 xenograft growth was significantly inhibited with miR-4651 overexpression (Figure 5B). The mice body weights between the two groups were not significantly different (Figure 5B), neither did we notice any signs of apparent toxicities.
At Day-7 and Day-14, one tumor of each group (total four xenograft tumors) was individually isolated, and tissue lysates subjected to qPCR and Western blotting assays. As shown, the mature miR-4651 levels were significantly elevated in LV-pri-miR-4651-expressing A549 xenografts (about ten folds vs. control tumors, Figure 5D). Contrarily, BRD4 mRNA ( Figure 5E) and protein ( Figure 5F) levels were 13 decreased. Expression of BRD4-dependent genes, c-Myc, Bcl-2, cyclin D1, was also significantly downregulated in miR-4651-overexpressed A549 xenograft tissues (Figure 5E and F). These results together show that miR-4651 overexpression inhibited BRD4 expression and A549 xenograft growth in SCID mice.

miR-4651 is downregulated in human lung cancer tissues, correlating with upregulation of BRD4 and its pathway genes
At last we tested expression of miR-4651 in human lung cancer tissues. A total of ten (10) pairs of NSCLC tissues ("Ca") and paracancer lung epithelial tissues ("N") tissues were obtained. Testing miR-4651 expression, by qPCR, demonstrated that miR-4651 levels were significantly downregulated the "Ca" tissues (P <0.05 vs. the "N" tissues, Figure 6A). Contrarily, BRD4 mRNA levels in the "Ca" tissues were significantly higher than those in the "N" tissues ( Figure 6B). mRNA levels of BRD4dependent genes, c-Myc, Bcl-2, cyclin D1, were also significantly elevated in the "Ca" tissues.
These results together show that miR-4651 is downregulated in human NSCLC tissues, correlating with upregulation of BRD4 and its target genes.

Discussion
The physiological function of miR-4651 is still largely elusive. Recent studies reported a higher serum specifically and directly targets BRD4 in human lung cancer cells. The Ago-2 IP experiments confirmed that miR-4651 directly binds to BRD4 mRNA in A549 cells. Significantly, in A549 cells and primary human NSCLC cells ectopic overexpression of miR-4651, by LV-pri-miR-4651, inhibited BRD4's 3-UTR activity and its expression, both were however significantly elevated in NSCLC cells with miR-4651 inhibition (by LV-antagomiR-4651). Expression of BRD4-regulated genes, c-Myc, Bcl-2, cyclin D1 [6,10,11,33], was also regulated by miR-4651 in A549 and primary human NSCLC cells. Therefore, miR-4651 targets and negatively regulates BRD4 in human NSCLC cells.
Recent studies have demonstrated that BRD4 is overexpressed in human lung cancer tissues/cells, serves as a valuable and novel oncotarget for potential cancer therapy [7,8]. The BRD4 pharmacological inhibitors (JQ1 and I-BET151) or BRD4 silencing (by targeted siRNAs) potently inhibited lung cancer cell progression in vitro and in vivo [7,8]. Here we show that miR-4651-induced silencing of BRD4 suppressed NSCLC cell progression. In vitro, forced overexpression of miR-4651, by LV-pri-miR-4651, inhibited NSCLC cell growth, proliferation and migration, while inducing apoptosis activation. In vivo, miR-4651-overexpressed A549 xenografts grew significantly slower than control A549 xenografts in SCID mice. Contrarily, LV-antagomiR-4651-mediated miR-4651 inhibition further promoted NSCLC cell growth, proliferation and migration. Therefore, miR-4651-mediated BRD4 silencing could be a novel strategy to inhibit NSCLC cell progression.
MiR-4651 could have other potential targets, as suggested by studies in other human cells [34,35].
The results of this study show that BRD4 should be the main target of miR-4651 in NSCLC cells. BRD4 re-expression, by an 3'-UTR mutant BRD4, reversed miR-4651 overexpression-induced inhibitions on A549 cells. Furthermore, BRD4 KO, using CRISPR/Cas9 strategy, mimicked miR-4651 overexpressioninduced activity in A549 cells. More importantly, miR-4651 overexpression or inhibition was completely ineffective in the BRD4-KO A549 cells. Therefore, we conclude that miR-4651 inhibits NSCLC cell progression by silencing BRD4.

Conclusion
The efficiency of the current clinical NSCLC treatment options, including the traditional cancer resection, platinum-based chemotherapy and radiation as well as the latest targeted therapies, is far from satisfactory [2,36]. The results of this study show that miR-4651, being downregulated in human NSCLC tissues, could significantly inhibit proliferation of NSCLC cells by negatively regulating 15 BRD4. It should represent as a novel therapy strategy for NSCLC.

Declarations
Ethical Approval and Consent to participate. This study was approved by the Ethics Committee of Soochow University.

Consent for publication. Not applicable.
Availability of data and material. All data generated during this study are included in this published article. Data will be made available upon request.
Competing interests. The authors declare that they have no competing interests.       miR-4651 overexpression inhibits A549 cell growth in SCID mice. Stable A549 cells, with the lentivirus-packaged pri-miR-4651 construct ("LV-pri-miR-4651") or the non-sense control miRNA vector ("miRC"), were inoculated s.c. to SCID mice (10 mice per group, n=10); Figure 6 miR-4651 is downregulated in human lung cancer tissues, correlating with upregulation of BRD4 and its target genes. Total RNA and proteins were extracted from a total of ten (10) pairs human NSCLC tissues ("Ca") and surrounding normal lung epithelial tissues (N"), relative expression of miR-4651 (A), listed mRNAs (B) and proteins (C and D) were tested by qPCR and Western blotting assays. Expression of listed proteins was quantified, normalized to the loading control Erk1/2 (C). Data were presented as mean ± standard deviation (SD).
*P < 0.05 vs. "Ca" tissues. The experiments in this figure were repeated five times with similar results obtained.