Thyroid cancer is the most common endocrine malignancy, and miR-574 is significantly upregulated in thyroid cancer. However, the role and underlying mechanism of miR-574 in thyroid cancer development are poorly understood. In this study, we showed that NF-κB/p65 signaling pathway was activated and miR-574 was upregulated in thyroid cancer cells. p65 directly bound to the promoter of miR-574 and activated miR-574 transcription. Functionally, miR-574 inhibited apoptosis, promoted proliferation and migration of thyroid cancer cells, and stimulated thyroid cancer–induced tube formation of endothelial cells. On the molecular level, miR-574 inhibited the expression of BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3) by binding to 3′-UTR of BNIP3. miR-574 also downregulated the expression of apoptosis-inducing factor (AIF), while elevated the levels of MMP2, MMP9, and VEGFA. In vivo, miR-574 promoted xenograft growth, which was associated with reduced apoptosis and enhanced angiogenesis. NF-κB/miR-574 signaling presents multiple oncogenic activities on thyroid cancer development by directly regulating the BNIP3/AIF pathway. Therefore, targeting NF-κB/miR-574 signaling may reduce the aggressiveness of thyroid cancer.

Implications:

miR-574, directly regulated by NF-κB/p65, promotes tumorigenesis of thyroid cancer via inhibiting BNIP3/AIF pathway.

Thyroid carcinoma is the most common malignancy in the endocrine system, presenting an increasing incidence worldwide and involving expanding populations of children, adolescents, and young adults (1, 2). In China, partially due to the development of screening techniques and to iodine supplementation, we witness a rapid increase in both the morbidity and mortality of thyroid cancer (3, 4). According to the histopathologic features, thyroid cancer is divided into differentiated thyroid cancer (DTC) that comprises papillary thyroid cancer (PTC) accounting for more than 80% and follicular thyroid cancer (FTC) accounting for approximately 9%, medullary thyroid cancer (MTC) responsible for less than 5%, and anaplastic thyroid cancer (ATC) contributing to less than 2% of all thyroid cancer cases (5). Although the majority of patients with DTC are associated with good prognosis following complete total thyroidectomy and radioactive iodine treatment, approximately 10% to 15% of patients with DTC presented recurrent and metastatic disease, associated with a higher chance of developing resistance to radioactive iodine therapy followed by a reduced survival (6). Therefore, understanding the molecular mechanisms of thyroid cancer will benefit the development of effective therapies improving the survival of patients with thyroid cancer.

Migration and angiogenesis are two interrelated biological processes essentially contributing to tumor metastasis, and are regulated by many factors such as VEGF and matrix metalloproteinases (MMP; ref. 7). Thyroid cancer are associated with consistent increase of proangiogenic factors, such as VEGF, angiopoietin-2, and their corresponding receptors (8). In addition, upregulated VEGFC or serum MMP2 level in thyroid cancer correlated with a higher risk of recurrence and worse prognosis of patients with PTC, suggesting that migration and angiogenesis critically regulates the development of recurrent and metastatic PTC after surgery (9–11). Therefore, therapeutic strategies targeting migration and/or angiogenesis are expected to improve the survival of patients with thyroid cancer.

p65 is a member in the nuclear factor-κB (NF-κB) family of transcription factors and by forming either heterodimers or homodimers, binds to the consensus sequence within the promoter and regulates the transcription of various target genes (12). Studies over the past few decades have revealed elevated NF-κB activity and its pleotropic roles during cancer development, including but not limited to establishing a microenvironment of chronic inflammation to favor tumor initiation and development, stimulating the proliferation and inhibiting the apoptosis of cancer cells, enhancing VEGF production and angiogenesis, promoting metastasis by upregulating MMPs or factors involved in epithelial–mesenchymal transition, and boosting tumor metabolism (13). For thyroid cancer, earlier studies reported the upregulation as well as the significance of p65 (14, 15), yet the oncogenic signaling cascades downstream of p65 are not completely understood.

miRNAs are small noncoding RNA molecules of approximately 22 nucleotides. Distinct miRNA profiles are associated with different phenotypes of thyroid cancer (16–18). Of various miRNAs, the downregulated endogenous miR-574 was associated with inhibited proliferation and migration of PTC cells (19, 20), supporting that miR-574 may present oncogenic activities during thyroid cancer development, which, however, remains to be characterized. In addition, miR-574 presents oncogenic activities in other human cancers, including breast cancer (21) and lung cancer (22). A recent study showed that p65 could regulate miR-574 expression (23), which prompted us to explore the potential cross-talk between p65 and miR-574 in thyroid cancer development.

In this study, we examined the biological functions of miR-574 in thyroid cancer, characterized the potential control of miR-574 by p65, and explored the downstream targets of miR-574 that may mediate its effects in thyroid cancer. Our findings showed that NF-κB directly activated the transcription of miR-574. In turn, miR-574 targeted the expressions of BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3), but upregulated MMP2, MMP9, and VEGFA. Inhibition of the NF-κB signaling and knockdown of miR-574 increased apoptosis, decreased proliferation and migration of thyroid cancer cells, and inhibited thyroid cancer–associated angiogenesis by activating BNIP3/AIF signaling pathway. Taken together, NF-κB–mediated miR-574 signaling presented multiple oncogenic activities in thyroid cancer development and thus may become an important therapeutic target for thyroid cancer treatment.

Cell culture

The normal human thyroid follicular epithelial cell line Nthy-ori3–1, the human thyroid cancer cell line FTC-133, and the human umbilical vein endothelial cells (HUVEC) were purchased from Sigma-Aldrich. The human thyroid cancer cell lines TPC-1, BCPAP, and K1 cells were purchased from ATCC. The Nthy-ori3-1, TPC-1, and BCPAP cell lines were cultured in RPMI1640 containing 2 mmol/L l-glutamate, 10% FBS, and 1% penicillin/streptomycin; the FTC-133 and K1 cell lines were cultured in DMEM/F-12 medium containing 2 mmol/L l-glutamate, 10% FBS, and 2% penicillin/streptomycin. HUVECs were cultured in endothelial cell growth medium (Sigma-Aldrich). All culture reagents were purchased from Invitrogen unless otherwise stated. Cell lines were authenticated by short tandem repeat (STR) analyses, and Mycoplasma tests were performed each month by qRT-PCR. Experiments were performed on cells under 15 passages.

Cell transfection

The full-length human p65 cDNA was PCR amplified from total RNA and cloned into pcDNA3.1 vector (pcDNA3.1-p65). p65 and BNIP3 specific short hairpin RNAs (shRNA) were cloned into pSicoR vector (shp65 and shBNIP3, respectively). Human miR-574 mimics and the corresponding control mimics (mimics-NC), and human miR-574 inhibitor and the corresponding control inhibitor (inhibitor-NC) were synthesized by GenePharma. Transfection was performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol.

qRT-PCR

Total RNA was extracted from cultured cells or isolated tumor tissues using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. cDNA was synthesized using TaqMan MicroRNA Reverse Transcription Kit (Invitrogen) following the manufacturer's instructions. qRT-PCR was performed on an ABI 7500 thermocycler (Applied Biosystems) using SYBR-Green Universal qPCR Master Mix (Bio-Rad) following the manufacturer's instructions. The thermal cycling condition was set as 95°C for 3 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 30 seconds. The following primer sequences were used: human p65 forward 5′-GCGAGAGGAGCACAGATACC-3′ and reverse 5′-AGGGGTTGTTGTTGGTCTGG-3′; human miR-574 forward 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACACAC-3′, and reverse 5′-GCCTGAGTGTGTGTGTGTGA-3′; human BNIP3 forward 5′-CTGGACGGAGTAGCTCCAAG-3′, and reverse 5′-CCGACTTGACCAATCCCATA-3′; U6 (internal control for miRNAs) forward 5′-CTCGCTTCGGCAGCACA-3′, and reverse 5′-AACGCTTCACGAATTTGCGT-3′ and GAPDH (internal control for mRNAs) forward, 5′-CCAGGTGGTCTCCTCTGA-3′ and reverse 5′- GCTGTAGCCAAATCGTTGT-3′. Relative gene expression was calculated using 2−ΔΔCt method.

Western blot analysis

Total proteins were extracted from cultured cells or xenograft tumors using RIPA buffer (Thermo Fisher Scientific) supplemented with protease inhibitor cocktail (Cell Signaling Technology), and measured for concentration using BCA Kit (Thermo Fisher Scientific). Proteins (30 μg) were separated by 10% SDS-PAGE and transferred onto a PVDF membrane. After blocking with 10% BSA for 1 hour, the membrane was incubated with one of the following primary antibodies (all from Cell Signaling Technology, unless otherwise noted) at 4°C overnight: p65, BNIP3, AIF, MMP2, MMP9, VEGFA, and GAPDH (internal control), followed by horseradish peroxidase–conjugated secondary antibodies (Cell Signaling Technology) at room temperature for 2 hours. The signal was developed using the ECL system (Beyotime) according to the manufacturer's instructions. The relative protein expression was analyzed by NIH Image J software and presented as the density ratio to that of the internal control.

Bioinformatics analysis and dual luciferase reporter assay

The promoter sequence of human miR-574 gene was cloned into the psiCHECK-2 vector (Promega), and then transfected into shNC versus shp65 and pcDNA3.1 versus pcDNA3.1-p65 (TPC-1 or BCPAP) cells using Lipofectamine 2000 according to the manufacturer's instructions.

Bioinformatics analysis using Targetscan (http://www.targetscan.org/vert_71/) was used to identify potential binding targets of miR-574 and to locate the binding sequences within the 3′ untranslated region (3′-UTR) of BNIP3. Mutations were then introduced into the potential binding sequences using the Site-Directed Mutagenesis Kit (Stratagene). The wild-type (WT) or mutant (MUT) 3′-UTR sequence of BNIP3 was cloned into the psiCHECK-2 vector, downstream of the reporter luciferase gene, and transfected into TPC-1 or BCPAP cells expressing miR-574 mimics versus mimics-NC or miR-574 inhibitor versus inhibitor-NC using Lipofectamine 2000. After 48 hours, luciferase activity was detected using the Dual Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) assay was performed using SimpleChIP Kit (Cell Signaling Technology) according to the protocols provided by the manufacturer. Briefly, cells were crosslinked with 1% formaldehyde and lysed to prepare nuclei. Then chromatin was partially digested in micrococcal nuclease followed by sonication to generated DNA/protein fragments of 150–900 base pairs (bps) in length. Upon incubating the digested chromatin with anti-p65 antibody, anti-Histone 3 antibody (positive control), or normal rabbit IgG (negative control; all from Cell Signaling Technology) at 4°C overnight, the immune complexes were pulled down using ChIP-grade protein G magnetic beads. After eluting chromatin from the antibody/protein G magnetic beads, DNA was purified using the spin column provided with the kit and examined with PCR analysis.

Colony formation assay

The cell proliferation of TPC-1 or BCPAP cells stably expressing miR-574 mimics versus mimics-NC or miR-574 inhibitor versus inhibitor-NC was analyzed using colony formation assay as described previously (24). Briefly, cells were trypsinized, resuspended in fresh medium at 200 cells/mL, and seeded into 6-well plate. After 7 days of culture at 37°C, cell colonies were stained with nitroblue tetrazolium chloride solution (Sigma-Aldrich), imaged under a microscope, and quantified using NIH Image J software (https://imagej.nih.gov/ij/).

Detection of apoptosis by staining cells with Annexin V and PI

Apoptotic cells were labeled using Annexin-V-FITC and PI Kit (Thermo Fisher Scientific) following the manufacturer's instructions. Briefly, cells were trypsinized and prepared into single-cell suspension in 1× Annexin-binding buffer containing 5 μL of FITC-Annexin V and 1 μL of 100 μg/mL propidium iodide (PI). After a 15-minute incubation at room temperature, cells were analyzed on FACSCalibur (BD Biosciences), and apoptotic cells were presented as the percentage (%) of Annexin-V+ PI+ cells.

Wound healing migration assay

The wound healing migration assay was performed as described previously (25). Briefly, a 1-mm wide scratch was made across the confluent cell layer using a sterile pipette tip in a 24-well plate (0 hours). At 0 and 24 hours, respectively, each well was photographed at an identical location and the width (W) of the scratch was measured. The migration rate was calculated as (W0hW24h)/W0h × 100%. Each experiment was performed in triplicates and repeated three times.

Transwell migration assay

Transwell insert (8.0 μm; Corning) was used for assessing cell migration. Briefly, the cells (1 × 105 cells/well) were seeded into the top well and cultured in serum-free medium at 37°C. Serum-containing complete medium (600 μL/well) was added to the lower chamber. After 24 hours, noninvaded cells were removed with cotton swabs from the upper side of the membrane, and the invaded cells were fixed in 4% paraformaldehyde, stained with 1% crystal violet, imaged, and counted under an inverted microscope.

Tube formation assay

TPC-1 or BCPAP cells (mimics-NC vs. miR-574 mimics, inhibitor-NC vs. miR-574 inhibitor) were cultured in serum-free growth medium overnight and the conditioned medium was collected. Then 96-well plate was coated with Matrigel (BD Biosciences). Upon gel solidification, HUVECs were added at 1 × 104/well in triplicate and cultured in conditioned medium at 37°C for 16 hours. The cells were then washed with PBS twice, fixed with 4% paraformaldehyde. Images were captured using an inverted live cell microscope (IX-71, Olympus).

In vivo xenograft model

All animal-related protocols were reviewed and approved by the Institutional Animal Care and Use Committee of Xiangya Hospital of Central South University (Changsha, P.R. China). Six-week-old nude mice were acquired from SJA Laboratory Animal Co., Ltd. All mice were housed in a specific-pathogen-free facility with a 12-hour light/dark cycle at a temperature of 21°C ± 2°C and a relative air humidity of 55% ± 10%. To establish the xenograft model, TPC-1 or BCPAP cells were injected subcutaneously into the dorsal flank region of each mouse (5 × 106 cells/injection). Mice were divided randomly into four groups (i.e., inhibitor NC group, miR-574 inhibitor group, mimics NC group, miR-574 mimics group, 8 mice/group). Tumor volume (V) was measured every 5 days for 30 days and calculated according to the formula: V (mm3) = 0.5 × (W)2 × (L), where L represents the length and W represents the width. Thirty days later, the xenograft tumors were isolated, imaged, and weighed, with partial tissues snap frozen in liquid nitrogen and partial fixed in 10% neutral formalin for further analysis.

Histologic analysis

The mouse xenograft tissues were fixed in 10% neutral formalin and prepared into paraffin blocks. Serial sections of 4 μm in thickness were prepared. Hematoxylin and eosin (HE) staining was performed using HE Staining Kit (Vector Labs) according to the manufacturer's instructions.

For IHC detection of CD34, after deparaffinization, antigen retrieval, and blocking, the slides were incubated with the primary antibodies of anti-CD34, anti-BNIP3, anti-AIF, anti-MMP2, anti-MMP9, and anti-VEGFA (Abcam) overnight at 4°C. The sections were counterstained with Mayer's hematoxylin.

Statistical analysis

All data were presented as the means ± SD from at least three independent experiments or 8 mice/group. Statistical analysis was performed using GraphPad Prism 5 software (GraphPad Software). Comparisons between two groups were performed using two-tailed Student t test and multiple comparisons using one-way ANOVA followed by Tukey post hoc test. A P value of <0.05 was considered statistically significant.

NF-κB signaling was activated in thyroid cancer cells, correlating with elevated expression of miR-574 and reduced expression of BNIP3

To explore the clinical significance of and potential interactions among NF-κB signaling, miR-574, and BNIP3, we first examined the association of miR-574 with the overall survival of patients with thyroid cancer as well as the expression of BNIP3 between normal (n = 59) and thyroid cancer tissues (n = 505) using The Cancer Genome Atlas database. The correlation analysis of miR-574 level and patient survival was performed using LinkedOmics portal (http://www.linkedomics.org/login.php) and the comparison of BNIP3 level between normal and cancer tissues using UALCAN portal (http://ualcan.path.uab.edu/analysis.html). We found that higher miR-574 level was associated with shorter overall survival of patients with thyroid cancer (Fig. 1A) and BNIP3 level was significantly lower in thyroid cancer than in normal tissues (Fig. 1B), supporting the clinical significance of both molecules. Next, we measured their expressions in four distinct thyroid cancer cell lines (TPC-1, K1, FTC-133, and BCPAP) versus normal thyroid follicular epithelial cells Nthy-ori3-1. As shown in Fig. 1C, p65 mRNA level was significantly upregulated in TPC-1, K1, and BCPAP thyroid cancer cells (P < 0.05, when compared to Nthy-ori3-1 cells). Although the p65 mRNA level in FTC-133 cells was higher than in Nthy-ori3-1 cells, the difference was not statistically significant (P > 0.05). Correlated with the increase of p65 mRNA in thyroid cancer cells, miR-574 was upregulated, yet BNIP3 downregulated in all thyroid cancer cells (P < 0.05, when compared with Nthy-ori3-1 cells, Fig. 1D and E) except that the increase of miR-574 in K1 cells over Nthy-ori3-1 cells was not statistically significant (P > 0.05). On the protein level, p65 was significantly higher in TPC-1 and BCPCP cells, while that of BNIP3 markedly reduced in all four thyroid cancer cells than in normal thyroid epithelial cells (all P < 0.05; Fig. 1F and G). In K1 and FTC-133 cells, p65 level was higher but presented no statistical difference from that in Nthy-ori-3-1 cells (P > 0.05, Fig. 1F and G). Therefore, TPC-1 and BCPAP cells were selected for the follow-up experiments in this study. Overall, the data support a positive regulation between NF-κB signaling and miR-574, and a negative one between these two molecules and BNIP3.

Figure 1.

NF-κB and miR-574 were upregulated while BNIP3 was downregulated in thyroid cancer cells. A, Analysis of TCGA database on the association of miR-574 level and the overall survival of patients with thyroid cancer. Red and blue represent miR-574 high and low expression, respectively. B, Analysis of TCGA database on the expression of BNIP3 mRNA between normal (n = 59) and thyroid cancer tissues (n = 505). The relative expressions of p65 (C), miR-574 (D), and BNIP3 (E) were examined by qRT-PCR and compared between the normal thyroid follicular epithelial cell line Nthy-ori3-1 and thyroid cancer cell lines, TPC-1, K1, FTC-133, and BCPAP. F, The expressions of BNIP3 and p65 were examined in indicated cell lines by Western blotting. GAPDH was examined as the internal control. G, The quantification of the band density of a target protein to that of GAPDH. *, P < 0.05 and **, P < 0.01.

Figure 1.

NF-κB and miR-574 were upregulated while BNIP3 was downregulated in thyroid cancer cells. A, Analysis of TCGA database on the association of miR-574 level and the overall survival of patients with thyroid cancer. Red and blue represent miR-574 high and low expression, respectively. B, Analysis of TCGA database on the expression of BNIP3 mRNA between normal (n = 59) and thyroid cancer tissues (n = 505). The relative expressions of p65 (C), miR-574 (D), and BNIP3 (E) were examined by qRT-PCR and compared between the normal thyroid follicular epithelial cell line Nthy-ori3-1 and thyroid cancer cell lines, TPC-1, K1, FTC-133, and BCPAP. F, The expressions of BNIP3 and p65 were examined in indicated cell lines by Western blotting. GAPDH was examined as the internal control. G, The quantification of the band density of a target protein to that of GAPDH. *, P < 0.05 and **, P < 0.01.

Close modal

p65 bound to the promoter and directly activated the transcription of miR-574

Promoter analysis revealed the potential binding sites to p65 within the promoter region of miR-574. To assess whether NF-κB directly activates the transcription of miR-574, we focused on two thyroid cancer cells presenting relatively higher p65 and miR-574 levels, TPC-1 and BCPAP, and decreased and increased the expression of p65 by transfecting these cells with shp65 and pcDNA3.1-p65, respectively. qRT-PCR (Fig. 2A) and Western blotting (Fig. 2B and C) showed that shp65 significantly reduced the endogenous p65 (P < 0.05, when compared with shNC cells), while pcDNA3.1-p65 robustly elevated p65 level in both TPC-1 and BCPAP cells (P < 0.05, when compared with pcDNA3.1 cells). Corresponding to alterations of p65 level in TPC-1 or BCPAP cells, miR-574 was downregulated in shp65 cells but upregulated in pcDNA3.1-p65 cells (P < 0.05, when compared with the relative control cells; Fig. 2D). To further validate the regulation of p65 on miR-574, we treated both TPC-1 and BCPAP cells with pyrrolidine dithiocarbamate (PDTC), an inhibitor of NF-κB activation. As shown in Fig. 2D, PDTC significantly reduced endogenous miR-574 level, supporting that NF-κB is an upstream regulator and its activation is essential for the upregulation of miR-574. When a reporter construct where the luciferase gene was driven by the miR-574 promoter was transfected into shp65 versus shNC cells or pcDNA3.1-p65 versus pcDNA3.1 cells, we found that the luciferase activity was significantly suppressed in shp65 cells (P < 0.05, when compared with shNC cells) but dramatically boosted in pcDNA3.1-p65 cells (P < 0.05, when compared with pcDNA3.1 cells; Fig. 2E). Furthermore, ChIP assay revealed the direct binding of p65 to the promoter sequence of miR-574 (Fig. 2F). Taken together, these data suggest that p65 directly acts on the promoter of miR-574 and not only necessarily but also sufficiently stimulates the transcription of miR-574.

Figure 2.

p65 directly activated the transcription of miR-574. A, The mRNA level of p65 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with shNC, shp65, pcDNA3.1, or pcDNA3.1-p65. B and C, The protein level of p65 was examined in indicated cells by Western blotting. GAPDH was examined as the internal control. Representative Western image shown in B and the band density ratio of a target gene to GAPDH shown in C. D, The relative expression of endogenous miR-574 in indicated cells was examined by qRT-PCR. PDTC, an inhibitor of NF-κB activation. E, The promoter sequence of miR-574 containing the p65-binding site was cloned upstream of the luciferase reporter gene and transfected into indicated cells. The luciferase activity was measured and compared between shNC versus shp65 or between pcDNA3.1 and pcDNA3.1-p65 cells. F, The binding of p65 to the promoter of miR-574 was examined by ChIP. No template and normal rabbit IgG (IgG) were used as the negative control. Anti-histone 3 (H3) antibody was used as the positive control. *, P < 0.05 and **, P < 0.01.

Figure 2.

p65 directly activated the transcription of miR-574. A, The mRNA level of p65 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with shNC, shp65, pcDNA3.1, or pcDNA3.1-p65. B and C, The protein level of p65 was examined in indicated cells by Western blotting. GAPDH was examined as the internal control. Representative Western image shown in B and the band density ratio of a target gene to GAPDH shown in C. D, The relative expression of endogenous miR-574 in indicated cells was examined by qRT-PCR. PDTC, an inhibitor of NF-κB activation. E, The promoter sequence of miR-574 containing the p65-binding site was cloned upstream of the luciferase reporter gene and transfected into indicated cells. The luciferase activity was measured and compared between shNC versus shp65 or between pcDNA3.1 and pcDNA3.1-p65 cells. F, The binding of p65 to the promoter of miR-574 was examined by ChIP. No template and normal rabbit IgG (IgG) were used as the negative control. Anti-histone 3 (H3) antibody was used as the positive control. *, P < 0.05 and **, P < 0.01.

Close modal

NF-κB signaling, via upregulating miR-574, necessarily and essentially promoted multiple malignant phenotypes of thyroid cancer cells

To understand the functional significance of NF-κB signaling and NF-κB–induced miR-574 in thyroid cancer cells, we detected cell proliferation, apoptosis, migration, and endothelial cell tube formation. The colony formation assay revealed that shp65 significantly suppressed, while p65 markedly stimulated cell proliferation of both TPC-1 and BCPAP cells (P < 0.05, Fig. 3A and B). In contrast, the apoptosis assay showed that shp65 promoted, while p65 inhibited the apoptosis of both TPC-1 and BCPAP cells (P < 0.05; Fig. 3C and D). Similar to the effect of p65 on cell proliferation, shp65 significantly inhibited, while p65 stimulated cell migration (as detected by wound healing assay, Fig. 3E and F, and Transwell migration assay, Fig. 3G and H) of both TPC-1 and BCPAP cells (P < 0.05). In addition, to investigate whether altering p65 level in cancer cells may have any paracrine effect, we examined the apoptosis of HUVECs in response to conditioned medium collected from thyroid cancer cells. As shown in Supplementary Fig. S1A and S1B, the conditioned medium from shp65 thyroid cancer cells significantly promoted, while that from p65-overexpressing thyroid cancers reduced the apoptosis of HUVECs (P < 0.05, when comparing shp65+ mimics NC or p65+mimics NC cells with Vector +mimics NC cells). Correspondingly, the tube formation of HUVECs was significantly dampened by the conditioned medium collected from TPC-1 or BCPAP cells expressing shp65, but was greatly stimulated by the conditioned medium collected from the thyroid cancer cells expressing p65 (P < 0.05; Supplementary Fig. S1C and S1D). More importantly, miR-574 mimics was sufficient to rescue all shp65-induced phenotypes as examined above (Fig. 3; Supplementary Fig. S1). Collectively, the data suggest that p65, via upregulating miR-574, was not only necessary, but also sufficient to promote multiple malignant phenotypes thyroid cancer cells.

Figure 3.

NF-κB regulated cell proliferation, apoptosis, and migration by targeting miR-574 in thyroid cancer cells. A and B, The proliferation capacity of indicated cells was examined by colony formation assay. Representative images of colonies from indicated cells shown in A and the quantification of colony numbers in B. C and D, The apoptosis of indicated cells was examined by Annexin V/PI dual staining followed by flow cytometry. The representative flow images shown in C and the quantification of % of Annexin V+PI+ apoptotic cells in D. E and F, The migration of indicated cells was examined by wound healing assays and the representative images of wound on cell monolayer shown at 0 and 24 hours, respectively, with the relative migration rate (%) calculated. G and H, The migration of indicated cells was examined by Transwell migration assay. The representative images of migrated cells after 24 hours shown in G and quantification of migrated cells shown in H. *, P < 0.05 and **, P < 0.01.

Figure 3.

NF-κB regulated cell proliferation, apoptosis, and migration by targeting miR-574 in thyroid cancer cells. A and B, The proliferation capacity of indicated cells was examined by colony formation assay. Representative images of colonies from indicated cells shown in A and the quantification of colony numbers in B. C and D, The apoptosis of indicated cells was examined by Annexin V/PI dual staining followed by flow cytometry. The representative flow images shown in C and the quantification of % of Annexin V+PI+ apoptotic cells in D. E and F, The migration of indicated cells was examined by wound healing assays and the representative images of wound on cell monolayer shown at 0 and 24 hours, respectively, with the relative migration rate (%) calculated. G and H, The migration of indicated cells was examined by Transwell migration assay. The representative images of migrated cells after 24 hours shown in G and quantification of migrated cells shown in H. *, P < 0.05 and **, P < 0.01.

Close modal

miR-574 targeted BNIP3/AIF signaling and elevated the expressions of MMP2, MMP9, and VEGFA in thyroid cancer cells

To understand the molecular mechanisms harnessed by miR-574 to carry out its biological functions, we explored the regulatory relationship between miR-574 and BNIP3. As shown in Fig. 4A, miR-574 inhibitor significantly reduced the endogenous miR-574 level (P < 0.05, when compared with inhibitor-NC), while miR-574 mimics potently increased miR-574 level (P < 0.05, when compared with mimics-NC) in both TPC-1 and BCPAP cells. Furthermore, treating TPC-1 or BCPAP cells with miR-574 inhibitor significantly increased, while miR-574 mimics reduced the expression level of BNIP3 (Fig. 4B). Bioinformatics analysis was used to screen for miR-574 target genes and identified a potential binding sequence within the 3′-UTR of human BNIP3 gene (Fig. 4C). The wild-type (WT) or mutant (MUT) 3′-UTR sequence of human BNIP3 was then cloned downstream of the luciferase reporter gene and transfected into either TPC-1 or BCPAP cells. By dual luciferase reporter assay, we found that miR-574 inhibitor specifically upregulated, while miR-574 mimics downregulated the luciferase activity regulated by WT-BNIP3 sequence, but not that by MUT-BNIP3 sequence (Fig. 4D). These results demonstrate that miR-574 not only directly binds to the 3′-UTR of BNIP3, but also inhibits its expression on the posttranscriptional level. Consistently, we detected the upregulation and downregulation of BNIP3 protein in TPC-1 or BCPAP cells expressing miR-574 inhibitor and mimics, respectively (P < 0.05, when compared with inhibitor-NC and mimics-NC, respectively; Fig. 4E and F). Concomitant with the alterations of BNIP3, miR-574 inhibitor significantly elevated the protein level of AIF, while reduced that of MMP2, MMP9, and VEGFA in both TPC-1 and BCPAP cells; in contrast, miR-574 mimics markedly reduced the expression of AIF and increased that of MMP2, MMP9, and VEGFA, suggesting that miR-574, by essentially inhibiting BNIP3/AIF signaling and upregulating MMP2, MMP9, and VEGFA, mediates multiple malignant phenotypes of thyroid cancer cells.

Figure 4.

miR-574 inhibited BNIP3/AIF signaling and upregulated MMP2, MMP9, and VEGFA in thyroid cancer cells. A, The relative expression of miR-574 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with inhibitor NC, miR-574 inhibitor, mimics NC or miR-574 mimics. B, The mRNA level of BNIP3 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with inhibitor NC, miR-574 inhibitor, mimics NC or miR-574 mimics. C, Bioinformatic analysis identified the potential binding sites between miR-574 and 3′-UTR of BNIP3. D, The luciferase activity was measured and compared between inhibitor-NC versus miR-574 inhibitor or between mimics-NC and miR-574-mimics cells. E and F, The expressions of BNIP3, AIF, MMP2, MMP9, and VEGFA were examined in indicated cells by Western blotting. GAPDH was examined as the internal control. The representative Western blot image presented in D and the quantification of the band density of a target gene to that of GAPDH shown in E. *, P < 0.05 and **, P < 0.01.

Figure 4.

miR-574 inhibited BNIP3/AIF signaling and upregulated MMP2, MMP9, and VEGFA in thyroid cancer cells. A, The relative expression of miR-574 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with inhibitor NC, miR-574 inhibitor, mimics NC or miR-574 mimics. B, The mRNA level of BNIP3 was examined by qRT-PCR in TPC-1 and BCPAP cells transfected with inhibitor NC, miR-574 inhibitor, mimics NC or miR-574 mimics. C, Bioinformatic analysis identified the potential binding sites between miR-574 and 3′-UTR of BNIP3. D, The luciferase activity was measured and compared between inhibitor-NC versus miR-574 inhibitor or between mimics-NC and miR-574-mimics cells. E and F, The expressions of BNIP3, AIF, MMP2, MMP9, and VEGFA were examined in indicated cells by Western blotting. GAPDH was examined as the internal control. The representative Western blot image presented in D and the quantification of the band density of a target gene to that of GAPDH shown in E. *, P < 0.05 and **, P < 0.01.

Close modal

BNIP3 played a central role in antagonizing the protumor activities of miR-574

To further examine the significance of BNIP3 in protumor activities of miR-574, we transfected TPC-1 or BCPAP cells with either control vector (Vector) or shBNIP3-expressing vector (shBNIP3), which significantly reduced endogenous BNIP3 level on both the steady-state mRNA (Fig. 5A) and protein levels (Fig. 5B and C). Then, we compared multiple phenotypes among miR-NC+Vector, miR-574 inhibitor+Vector, miR-574 mimics+Vector, and miR-574 inhibitor +shBNIP3 cells. miR-574 mimics significantly boosted cell proliferation (Fig. 5D and E), inhibited apoptosis (Fig. 5F and G), stimulated migration (Fig. 5HK), suppressed the apoptosis of HUVECs (Supplementary Fig. S2A and S2B), and tumor-induced angiogenesis (Supplementary Fig. S2C and S2D) of HUVECs, while miR-574 inhibitor played the opposite roles (all P < 0.05, when compared to inhibitor NC+Vector cells), suggestive of the malignancy-promoting and -suppressive effects of the former and the latter, respectively. Interestingly, the malignancy-suppressive effects of miR-574 inhibitor were completely abolished when endogenous BNIP3 was knocked down, as demonstrated by the enhanced colony formation, reduced apoptosis, increased migration and tumor-induced angiogenesis (Fig. 5; Supplementary Fig. S2), indicating that BNIP3 is a critical negative regulator of miR-574 signaling in thyroid cancer cells.

Figure 5.

BNIP3 played a critical role in antagonizing the protumor activities of miR-574-5p. AC, The relative expression of BNIP3 was examined by qRT-PCR (A) and Western blot (B and C) in TPC-1 and BCPAP cells transfected with empty plasmid vector or shBNIP3. The representative Western blot image presented in B and the quantification of the band density of a target gene to that of GAPDH shown in C. D and E, The proliferation capacity of indicated cells was examined by colony formation assay. Representative images of colonies from indicated cells shown in D and the quantification of colony numbers in E. F and G, The apoptosis of indicated cells was examined by Annexin V/PI dual staining followed by flow cytometry. The representative flow images shown in F and the quantification of % of Annexin V+PI+ apoptotic cells in G. H, The migration of indicated cells was examined by wound healing assays and the representative images of wound on cell monolayer shown at 0 and 24 hours, respectively. I, The quantification of migration rate (%) from indicated cells. J and K, The migration of indicated cells was examined by Transwell migration assay. The representative images of migrated cells after 24 hours shown in J and quantification of migrated cells shown in K. *, P < 0.05 and **, P < 0.01.

Figure 5.

BNIP3 played a critical role in antagonizing the protumor activities of miR-574-5p. AC, The relative expression of BNIP3 was examined by qRT-PCR (A) and Western blot (B and C) in TPC-1 and BCPAP cells transfected with empty plasmid vector or shBNIP3. The representative Western blot image presented in B and the quantification of the band density of a target gene to that of GAPDH shown in C. D and E, The proliferation capacity of indicated cells was examined by colony formation assay. Representative images of colonies from indicated cells shown in D and the quantification of colony numbers in E. F and G, The apoptosis of indicated cells was examined by Annexin V/PI dual staining followed by flow cytometry. The representative flow images shown in F and the quantification of % of Annexin V+PI+ apoptotic cells in G. H, The migration of indicated cells was examined by wound healing assays and the representative images of wound on cell monolayer shown at 0 and 24 hours, respectively. I, The quantification of migration rate (%) from indicated cells. J and K, The migration of indicated cells was examined by Transwell migration assay. The representative images of migrated cells after 24 hours shown in J and quantification of migrated cells shown in K. *, P < 0.05 and **, P < 0.01.

Close modal

miR-574 stimulated xenograft growth and tumor-associated angiogenesis in vivo

The in vitro assays performed above support the oncogenic activities of miR-574 in thyroid cancer cells. To explore the in vivo activities of miR-574 during thyroid cancer development, we established xenografts from TPC-1 or BCPAP cells. As shown in Fig. 6A, xenograft tumors from thyroid cancer cells expressing miR-574 inhibitor were much smaller, while those from thyroid cancer cells expressing miR-574 mimics much larger than the other groups of tumors (P < 0.05; Fig. 6A). Consistently, miR-574 inhibitor significantly suppressed the xenograft growth (P < 0.05, when compared with inhibitor NC group), while miR-574 mimics potently stimulated the xenograft growth (P < 0.05, when compared with mimics NC group; Fig. 6B). By day 30, the differences in xenograft growth were also reflected by weights of xenograft tumors (Fig. 6C). Consistent with in vitro functional phenotypes of miR-574, immunostaining for CD34, a biomarker for angiogenesis (26), revealed significantly reduced angiogenesis in xenografts from miR-574 inhibitor cells (P < 0.05, when compared with inhibitor NC group), yet enhanced angiogenesis in those from miR-574 mimics cells (P < 0.05, when compared with mimics NC group; Fig. 6D). In addition to CD34, we also observed the changes of BNIP3, AIF, MMP2, MMP9, and VEGFA in xenografts, where miR-574 inhibitor xenografts presented increased expressions of BNIP3 and AIF, and reduced expressions of MMP2, MMP9 and VEGFA (when compared with inhibitor NC xenografts), while miR-574 mimics xenografts demonstrated the opposite regulations of these proteins (Fig. 6D).

Figure 6.

miR-574 promoted xenograft growth in vivo. A, Representative pictures of three isolated xenografts from indicated groups. B, Tumor sizes measured every 5 days with electronic caliper. C, Weights of tumors harvested from mice at the end of experiments. D, H&E staining for tissue morphology and IHC staining analysis of vasculature (CD34), BNIP3, AIF, MMP2, MMP9, and VEGFA. Scale bar, 50 μm. *, P < 0.05 and **, P < 0.01.

Figure 6.

miR-574 promoted xenograft growth in vivo. A, Representative pictures of three isolated xenografts from indicated groups. B, Tumor sizes measured every 5 days with electronic caliper. C, Weights of tumors harvested from mice at the end of experiments. D, H&E staining for tissue morphology and IHC staining analysis of vasculature (CD34), BNIP3, AIF, MMP2, MMP9, and VEGFA. Scale bar, 50 μm. *, P < 0.05 and **, P < 0.01.

Close modal

When analyzing the xenografts on day 30, we found that miR-574 level remained significantly downregulated in miR-574 inhibitor group (P < 0.05, when compared with inhibitor NC group) and upregulated in miR-574 mimics group (P < 0.05, when compared to mimics NC group; Fig. 7A), verifying the in vivo efficiency of miR-574 inhibitor and mimics in downregulating and upregulating miR-574 level, respectively. On the molecular level, we detected opposite regulations on BNIP3, AIF, MMP2, MMP9, and VEGFA by miR-574 inhibitor versus by miR-574 mimics. The expressions of BNIP3 and AIF were reduced while those of MMP2, MMP9, and VEGFA increased in xenografts from miR-574 mimics cells (Fig. 7B and C), suggesting that by targeting BNIP3/AIF signaling and upregulating MMP2, MMP9, and VEGFA, miR-574 stimulates angiogenesis, and promotes xenograft growth in vivo.

Figure 7.

miR-574 essentially and sufficiently promoted tumor-associated angiogenesis in vivo. A, The relative expression of miR-574 in xenografts from indicated groups was examined by qRT-PCR. B and C, The expressions level of BNIP3, AIF, MMP2, MMP9, and VEGFA were examined in indicated xenografts by Western blotting. GAPDH was examined as the internal control. The representative Western blot image presented in B and the quantification of the band density of a target gene to that of GAPDH in C. *, P < 0.05 and **, P < 0.01.

Figure 7.

miR-574 essentially and sufficiently promoted tumor-associated angiogenesis in vivo. A, The relative expression of miR-574 in xenografts from indicated groups was examined by qRT-PCR. B and C, The expressions level of BNIP3, AIF, MMP2, MMP9, and VEGFA were examined in indicated xenografts by Western blotting. GAPDH was examined as the internal control. The representative Western blot image presented in B and the quantification of the band density of a target gene to that of GAPDH in C. *, P < 0.05 and **, P < 0.01.

Close modal

Retrospective analyses showed that macroscopic metastasis strongly predicts the persistence and recurrence of thyroid cancer (27, 28), supporting the therapeutic value of targeting metastasis in PTC treatment. The constitutive activation of NF-κB, through the canonical p50/p65 or the noncanonical p52/RelB heterodimer formation, presents multiple oncogenic roles during the initiation, development, and progression of different human malignancies (13). When it comes to specific NF-κB subunits, p65 overexpression is dominantly associated with solid cancers, such as ovarian cancer and adrenal gland carcinoma (29). In thyroid cancer, the overexpression and constitutive activation of p65 subunit was detected in thyroid cancer tissues (when compared with in normal tissues), in the highly aggressive ATC tissues (30), in thyroid cancer cell lines (31), as well as in the invasive front of PTC tissues (32). Instead of being an epiphenomenon during thyroid cancer development, NF-κB is functionally critical for regulating multiple “hallmarks of cancer,” including inhibiting cell apoptosis and promoting cell transformation, growth, migration, invasion, and tumor-associated angiogenesis (33, 34). Moreover, p65 activation in thyroid cancer functions as not only a diagnostic but also a prognostic biomarker for poor prognosis and metastases (35, 36). Therefore, targeting NF-kB becomes a promising strategy for thyroid cancer treatment (37). However, the detailed molecular mechanisms mediating NF-κB signaling are poorly understood in thyroid cancer cells. In this study, we identified a novel downstream mediator of NF-κB, miR-574, and presented evidence that miR-574 was a direct transcription target activated by NF-κB in thyroid cancer cells. Recently, Ku and colleagues showed that NF-κB, by regulating miR-574, induced synaptic and cognitive impairment in mice receiving oropharyngeal aspiration of particulate matter ≤2.5 μm (PM2.5; ref. 23). However, the regulatory relationship between NF-κB and miR-574 in cancer has not been reported. This is therefore the first study showing that NF-κB could directly activate the expression of miR-574, thereby promoting cell proliferation, migration, and angiogenesis in TP.

Similar to the well-characterized protumor activities of NF-κB, the oncogenic phenotypes of miR-574 have been reported in different human cancers through quite distinct molecular mechanisms. In non–small cell lung cancer (NSCLC), miR-574 was overexpressed in both serum and cancer tissues of patients with metastatic NSCLC, and miR-574 promoted the migration and invasion of NSCLC cells by inhibiting receptor-type tyrosine-protein phosphatase PCP-2 and subsequently activating β-catenin (22). In breast cancer cells, miR-574 contributed to drug resistance by down-regulating Smad4 (21). A recent study by Zhang and colleagues showed that miR-574 targeted Quacking proteins and thus activated Wnt/β-catenin signaling in thyroid cancer cell lines, BCPAP and FTC-133, essentially inhibiting apoptosis and promoting cell-cycle progression of these thyroid cancer cells (15). In the current study, we found the essential roles of miR-574 in suppressing apoptosis, and stimulating proliferation and migration of thyroid cancer cells, and thyroid cancer–induced angiogenesis, demonstrating that miR-574 is an oncogenic miRNA in thyroid cancer. To our knowledge, this is the first study showing that miR-574 critically promotes tumor angiogenesis, an essential step for the metastatic spread of cancer cells. Mechanistically, we revealed at least a partial list of novel targets regulated by and mediated the oncogenic activities of miR-574 in thyroid cancer cells, that is, BNIP3, AIF, MMP2, MMP9, and VEGFA.

BNIP3 is a BH3-only proapoptotic protein induced by hypoxia and can function to contribute to apoptosis, necrosis, and autophagy (38). In solid tumors, BNIP3 was reported to be either up-regulated (due to the presence of hypoxic regions) or downregulated (as a mechanism for cancer cells evading apoptosis; refs. 39, 40). Studies also showed that BNIP3 and AIF mutually and positively regulated the expression of each other, and act together to induce apoptosis (39). By analyzing the expression of autophagy-related proteins in different types of thyroid cancers, Kim and colleagues reported the absence of BNIP3 expression in MTC and ATC, and also association between the negative expression of BNIP3 in FTC and capsular invasion, suggesting the importance of BNIP3 in controlling malignant development of thyroid cancer (41). However, no further information is available on how BNIP3 is regulated in thyroid cancer. Here, we showed that miR-574 was upregulated while BNIP3 was downregulated in four different thyroid cancer cell lines. In addition, miR-574 bound to the 3′-UTR of BNIP3 gene, repressed its expression, and concomitantly downregulated AIF level, which, when acting together with BNIP3, may mediate the effects of miR-574 on inhibiting the apoptosis of thyroid cancer cells. In addition to regulating apoptosis, we showed that targeting BNIP3 with shRNA in thyroid cancer cells treated with miR-574 inhibitor completely abolished antitumor activities of the latter, including cell proliferation, apoptosis, migration, and tumor-induced angiogenesis. Therefore, BNIP3 plays a critical role in antagonizing multiple malignant and metastatic phenotypes induced by p65/miR-574.

In addition to targeting BNIP3/AIF signaling, we showed that miR-574 elevated the expressions of two matrix metalloproteinases, MMP2 and MMP9, as well as a proangiogenic factor VEGFA. MMPs are zinc-dependent endopeptidases that can not only degrade basement membrane and extracellular matrix, promoting cancer invasion, but also facilitate the establishment of metastatic niche in distant organs (42). Therefore, the upregulation of MMP2 and MMP9 may account for increased migration induced by miR-574. VEGFA is a major driver for tumor angiogenesis and an important anticancer therapeutic target (43). Our findings suggest that miR-574 enhanced the expression of VEGFA in thyroid cancer cells, correlated with the stimulation of miR-574 on angiogenesis, both in vitro and in vivo. Therefore, it is critical to further characterize the effects of miR-574 on angiogenesis in different cancer paradigms and comprehensively explore the underlying mechanisms.

In this study, we not only revealed the signaling cascade NF-κB/miR-574/BNIP3 and its regulation on several downstream targets, but characterized the significance of this cascade on multiple levels. First, the association between miR-574 and overall survival of patients with thyroid cancer as well as the upregulation of BNIP3 in thyroid cancer versus normal tissues supported the clinical significance of these two molecules. Second, we applied both gain-of-function and loss-of-function strategies and used two distinct thyroid cancer cell lines to comprehensively explore the biological significance of p65 and miR-574. Third, we focused on individual malignant phenotypes using the in vitro cultured cells as well as overall thyroid cancer development using the in vivo xenograft development. However, because the in vitro analyses suggest that NF-κB/miR-574/BNIP3 signaling impacted multiple metastasis-related phenotypes, it would be interesting for future studies to examine the effect of targeting this signaling pathway on thyroid cancer metastasis using an in vivo model.

Conclusions

In summary, the current study reveals the diverse oncogenic activities of NF-κB–induced miR-574 signaling in thyroid cancer cells, identifies downstream targets that may mediate these activities, and thus advances our understanding on the pathogenesis of thyroid cancer. To characterize the significance of these findings in clinical settings, we need to examine the expressional correlation between miR-574 and NF-κB in thyroid cancer tissues, and more importantly, whether specifically targeting miR-574 will benefit patient outcome.

No potential conflicts of interest were disclosed.

Conception and design: Z.-J. Zhang, X.-Y. Li

Development of methodology: Q. Xiao

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Z.-J. Zhang, Q. Xiao

Writing, review, and/or revision of the manuscript: Z.-J. Zhang, X.-Y. Li

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Q. Xiao

Study supervision: X.-Y. Li

This work was supported by the National Natural Science Foundation of China (No.81672885).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Pellegriti
G
,
Frasca
F
,
Regalbuto
C
,
Squatrito
S
,
Vigneri
R
. 
Worldwide increasing incidence of thyroid cancer: update on epidemiology and risk factors
.
J Cancer Epidemiol
2013
;
2013
:
965212
.
2.
Vergamini
LB
,
Frazier
AL
,
Abrantes
FL
,
Ribeiro
KB
,
Rodriguez-Galindo
C
. 
Increase in the incidence of differentiated thyroid carcinoma in children, adolescents, and young adults: a population-based study
.
J Pediatr
2014
;
164
:
1481
5
.
3.
Du
L
,
Wang
Y
,
Sun
X
,
Li
H
,
Geng
X
,
Ge
M
, et al
Thyroid cancer: trends in incidence, mortality and clinical-pathological patterns in Zhejiang Province, Southeast China
.
BMC Cancer
2018
;
18
:
291
.
4.
Wang
Y
,
Wang
W
. 
Increasing incidence of thyroid cancer in Shanghai, China, 1983–2007
.
Asia Pac J Public Health
2015
;
27
:
NP223
9
.
5.
Ferrari
SM
,
Fallahi
P
,
Politti
U
,
Materazzi
G
,
Baldini
E
,
Ulisse
S
, et al
Molecular targeted therapies of aggressive thyroid cancer
.
Front Endocrinol
2015
;
6
:
176
.
6.
Worden
F
. 
Treatment strategies for radioactive iodine-refractory differentiated thyroid cancer
.
Ther Adv Med Oncol
2014
;
6
:
267
79
.
7.
Ghobrial
IM
. 
Myeloma as a model for the process of metastasis: implications for therapy
.
Blood
2012
;
120
:
20
30
.
8.
Bunone
G
,
Vigneri
P
,
Mariani
L
,
Buto
S
,
Collini
P
,
Pilotti
S
, et al
Expression of angiogenesis stimulators and inhibitors in human thyroid tumors and correlation with clinical pathological features
.
Am J Pathol
1999
;
155
:
1967
76
.
9.
Kilicarslan
AB
,
Ogus
M
,
Arici
C
,
Pestereli
HE
,
Cakir
M
,
Karpuzoglu
G
. 
Clinical importance of vascular endothelial growth factor (VEGF) for papillary thyroid carcinomas
.
APMIS
2003
;
111
:
439
43
.
10.
Lai
CW
,
Duh
QY
,
Chen
CW
,
Chuang
FJ
,
Chang
YJ
,
Lin
MT
, et al
VEGF-D and a preoperative serum levels predict nodal and distant metastases in differentiated thyroid cancer patients
.
World J Surg
2015
;
39
:
1742
9
.
11.
Shi
Y
,
Su
C
,
Hu
H
,
Yan
H
,
Li
W
,
Chen
G
, et al
Serum MMP-2 as a potential predictive marker for papillary thyroid carcinoma
.
PLoS One
2018
;
13
:
e0198896
.
12.
Hayden
MS
,
Ghosh
S
. 
NF-kappaB, the first quarter-century: remarkable progress and outstanding questions
.
Genes Dev
2012
;
26
:
203
34
.
13.
Xia
Y
,
Shen
S
,
Verma
IM
. 
NF-kappaB, an active player in human cancers
.
Cancer Immunol Res
2014
;
2
:
823
30
.
14.
Meng
Z
,
Lou
S
,
Tan
J
,
Xu
K
,
Jia
Q
,
Zheng
W
. 
Nuclear factor-kappa B inhibition can enhance apoptosis of differentiated thyroid cancer cells induced by 131I
.
PLoS One
2012
;
7
:
e33597
.
15.
Zhang
Z
,
Li
X
,
Xiao
Q
,
Wang
Z
. 
MiR-574–5p mediates the cell cycle and apoptosis in thyroid cancer cells via Wnt/beta-catenin signaling by repressing the expression of quaking proteins
.
Oncol Lett
2018
;
15
:
5841
8
.
16.
Graham
ME
,
Hart
RD
,
Douglas
S
,
Makki
FM
,
Pinto
D
,
Butler
AL
, et al
Serum microRNA profiling to distinguish papillary thyroid cancer from benign thyroid masses
.
J Otolaryngol Head Neck Surg
2015
;
44
:
33
.
17.
Min
XS
,
Huang
P
,
Liu
X
,
Dong
C
,
Jiang
XL
,
Yuan
ZT
, et al
Bioinformatics analyses of significant prognostic risk markers for thyroid papillary carcinoma
.
Tumour Biol
2015
;
36
:
7457
63
.
18.
Zhang
X
,
Mao
H
,
Lv
Z
. 
MicroRNA role in thyroid cancer pathogenesis
.
Front Biosci
2013
;
18
:
734
9
.
19.
Fan
M
,
Li
X
,
Jiang
W
,
Huang
Y
,
Li
J
,
Wang
Z
. 
A long non-coding RNA, PTCSC3, as a tumor suppressor and a target of miRNAs in thyroid cancer cells
.
Exp Ther Med
2013
;
5
:
1143
6
.
20.
Wang
X
,
Lu
X
,
Geng
Z
,
Yang
G
,
Shi
Y
. 
LncRNA PTCSC3/miR-574–5p governs cell proliferation and migration of papillary thyroid carcinoma via Wnt/beta-catenin signaling
.
J Cell Biochem
2017
;
118
:
4745
52
.
21.
Sun
FD
,
Wang
PC
,
Luan
RL
,
Zou
SH
,
Du
X
. 
MicroRNA-574 enhances doxorubicin resistance through down-regulating SMAD4 in breast cancer cells
.
Eur Rev Med Pharmacol Sci
2018
;
22
:
1342
50
.
22.
Zhou
R
,
Zhou
X
,
Yin
Z
,
Guo
J
,
Hu
T
,
Jiang
S
, et al
MicroRNA-574–5p promotes metastasis of non-small cell lung cancer by targeting PTPRU
.
Sci Rep
2016
;
6
:
35714
.
23.
Ku
T
,
Li
B
,
Gao
R
,
Zhang
Y
,
Yan
W
,
Ji
X
, et al
NF-kappaB-regulated microRNA-574-5p underlies synaptic and cognitive impairment in response to atmospheric PM2.5 aspiration
.
Part Fibre Toxicol
2017
;
14
:
34
.
24.
Crowley
LC
,
Christensen
ME
,
Waterhouse
NJ
. 
Measuring survival of adherent cells with the colony-forming assay
.
Cold Spring Harb Protoc
2016
;
8
:
721
4
.
25.
Saxena
NK
,
Sharma
D
,
Ding
X
,
Lin
S
,
Marra
F
,
Merlin
D
, et al
Concomitant activation of the JAK/STAT, PI3K/AKT, and ERK signaling is involved in leptin-mediated promotion of invasion and migration of hepatocellular carcinoma cells
.
Cancer Res
2007
;
67
:
2497
507
.
26.
Vieira
SC
,
Silva
BB
,
Pinto
GA
,
Vassallo
J
,
Moraes
NG
,
Santana
JO
, et al
CD34 as a marker for evaluating angiogenesis in cervical cancer
.
Pathol Res Pract
2005
;
201
:
313
8
.
27.
de Castro
TP
,
Waissmann
W
,
Simoes
TC
,
de Mello
RC
,
Carvalho
DP
. 
Predictors for papillary thyroid cancer persistence and recurrence: a retrospective analysis with a 10-year follow-up cohort study
.
Clin Endocrinol
2016
;
85
:
466
74
.
28.
Ito
Y
,
Kudo
T
,
Kobayashi
K
,
Miya
A
,
Ichihara
K
,
Miyauchi
A
. 
Prognostic factors for recurrence of papillary thyroid carcinoma in the lymph nodes, lung, and bone: analysis of 5,768 patients with average 10-year follow-up
.
World J Surg
2012
;
36
:
1274
8
.
29.
Kaltschmidt
B
,
Greiner
JFW
,
Kadhim
HM
,
Kaltschmidt
C
. 
Subunit-specific role of NF-kappaB in cancer
.
Biomedicines
2018
;
6:pii:E44
.
30.
Pacifico
F
,
Mauro
C
,
Barone
C
,
Crescenzi
E
,
Mellone
S
,
Monaco
M
, et al
Oncogenic and anti-apoptotic activity of NF-kappa B in human thyroid carcinomas
.
J Biol Chem
2004
;
279
:
54610
9
.
31.
Visconti
R
,
Cerutti
J
,
Battista
S
,
Fedele
M
,
Trapasso
F
,
Zeki
K
, et al
Expression of the neoplastic phenotype by human thyroid carcinoma cell lines requires NFkappaB p65 protein expression
.
Oncogene
1997
;
15
:
1987
94
.
32.
Vasko
V
,
Espinosa
AV
,
Scouten
W
,
He
H
,
Auer
H
,
Liyanarachchi
S
, et al
Gene expression and functional evidence of epithelial-to-mesenchymal transition in papillary thyroid carcinoma invasion
.
Proc Natl Acad Sci U S A
2007
;
104
:
2803
8
.
33.
Bauerle
KT
,
Schweppe
RE
,
Haugen
BR
. 
Inhibition of nuclear factor-kappa B differentially affects thyroid cancer cell growth, apoptosis, and invasion
.
Mol Cancer
2010
;
9
:
117
.
34.
Pacifico
F
,
Leonardi
A
. 
Role of NF-kappaB in thyroid cancer
.
Mol Cell Endocrinol
2010
;
321
:
29
35
.
35.
Sethi
K
,
Sarkar
S
,
Das
S
,
Rajput
S
,
Mazumder
A
,
Roy
B
, et al
Expressions of CK-19, NF-kappaB, E-cadherin, beta-catenin and EGFR as diagnostic and prognostic markers by immunohistochemical analysis in thyroid carcinoma
.
J Exp Ther Oncol
2011
;
9
:
187
99
.
36.
Zhang
Y
,
Meng
Z
,
Zhang
M
,
Tan
J
,
Tian
W
,
He
X
, et al
Immunohistochemical evaluation of midkine and nuclear factor-kappa B as diagnostic biomarkers for papillary thyroid cancer and synchronous metastasis
.
Life Sci
2014
;
118
:
39
45
.
37.
Namba
H
,
Saenko
V
,
Yamashita
S
. 
Nuclear factor-kB in thyroid carcinogenesis and progression: a novel therapeutic target for advanced thyroid cancer
.
Arq Bras Endocrinol Metabol
2007
;
51
:
843
51
.
38.
Burton
TR
,
Gibson
SB
. 
The role of Bcl-2 family member BNIP3 in cell death and disease: NIPping at the heels of cell death
.
Cell Death Differ
2009
;
16
:
515
23
.
39.
Qi
Y
,
Tian
X
,
Liu
J
,
Han
Y
,
Graham
AM
,
Simon
MC
, et al
Bnip3 and AIF cooperate to induce apoptosis and cavitation during epithelial morphogenesis
.
J Cell Biol
2012
;
198
:
103
14
.
40.
Ye
F
,
Wang
H
,
Zhang
L
,
Zou
Y
,
Han
H
,
Huang
J
. 
Baicalein induces human osteosarcoma cell line MG-63 apoptosis via ROS-induced BNIP3 expression
.
Tumour Biol
2015
;
36
:
4731
40
.
41.
Kim
HM
,
Kim
ES
,
Koo
JS
. 
Expression of autophagy-related proteins in different types of thyroid cancer
.
Int J Mol Sci
2017
;
18
:
pii:E540
.
42.
Kessenbrock
K
,
Plaks
V
,
Werb
Z
. 
Matrix metalloproteinases: regulators of the tumor microenvironment
.
Cell
2010
;
141
:
52
67
.
43.
Rapisarda
A
,
Melillo
G
. 
Role of the VEGF/VEGFR axis in cancer biology and therapy
.
Adv Cancer Res
2012
;
114
:
237
67
.