MiR-19a Overexpression in FTC-133 Cell Line Induces a More De-Differentiated and Aggressive Phenotype

In recent years, microRNAs (miRNAs) have received increasing attention for their important role in tumor initiation and progression. MiRNAs are a class of endogenous small non-coding RNAs that negatively regulate the expression of several oncogenes or tumor suppressor genes. MiR-19a, a component of the oncogenic miR-17-92 cluster, has been reported to be highly expressed only in anaplastic thyroid cancer, the most undifferentiated, aggressive and lethal form of thyroid neoplasia. In this work, we evaluated the putative contribution of miR-19a in de-differentiation and aggressiveness of thyroid tumors. To this aim, we induced miR-19a expression in the well-differentiated follicular thyroid cancer cell line and evaluated proliferation, apoptosis and gene expression profile of cancer cells. Our results showed that miR-19a overexpression stimulates cell proliferation and alters the expression profile of genes related to thyroid cell differentiation and aggressiveness. These findings not only suggest that miR-19a has a possible involvement in de-differentiation and malignancy, but also that it could represent an important prognostic indicator and a good therapeutic target for the most aggressive thyroid cancer.


Introduction
Thyroid cancers are considered the most representative malignancies of the endocrine system, affecting about 1% of all newly diagnosed tumor cases with an incidence growing worldwide [1,2]. They comprise some histotypes that can be classified, according to definite histological features, into three different groups: (i) Well-differentiated thyroid carcinomas (WDTCs), (ii) poorly differentiated thyroid carcinomas (PDTCs) and (iii) undifferentiated thyroid carcinomas (UDTCs) [3]. WDTCs include papillary thyroid carcinoma (PTC), which represents about 80% of cases with better prognosis and follicular thyroid carcinoma (FTC) that represents approximately 15% of all thyroid carcinomas, with a less favorable survival rates [4][5][6][7].
PDTCs represents about 0.3% to 6.7% of all thyroid malignancies and display an aggressive clinical behavior, with a poor prognosis [8]. Anaplastic thyroid cancer (ATC) is the most aggressive and lethal form of UDTCs. It shows a quick invasive progression with an early metastatic diffusion, a considerably poorer prognosis and reduced overall survival [9].
MiRNAs, an evolutionally well-conserved class of noncoding RNAs, are capable of controlling gene expression through mRNA cleavage or translational repression [10,11] and their deregulation is involved in the development and progression of several human cancers [12][13][14][15][16]. They work Int. J. Mol. Sci. 2018, 19, 3944 2 of 14 in a complex functional system, in which a precise miRNA regulates hundreds of mRNAs and a single gene is targeted by several miRNAs [17]. MiRNAs regulate various human genes correlated with several cellular processes including proliferation, apoptosis, metastasis, cell immunity and differentiation [18][19][20]. Approximately 2000 miRNAs have been discovered in humans [21] and about 50% have been found in cancer-related genomic regions or fragile sites, confirming their role as drivers of neoplastic alterations [22]. In addition recently, it has been demonstrated that miRNAs expression is correlated with different clinical parameters and their application could be used for cancer diagnosis, prognosis and treatment response [23][24][25].
MiR-19a, one of the seven miRNAs belonging to the miR-17-92 cluster located on chromosome 13q31.3, has been demonstrated to be a tumor-related miRNA involved in cancerogenesis. Several findings have revealed that miR-19a expression is increased in many different types of solid tumors including ATC, as shown in vitro by cell growth induction. Takakura et al., demonstrated that miR-19a is more expressed in thyroid cancer cell lines than in normal primary thyreocytes [26].
The aim of the present study was to evaluate the involvement of miR-19a in promoting de-differentiation towards an aggressive phenotype. For this purpose, we induced miR-19a expression in follicular thyroid cancer cell line (FTC-133) and evaluated its effects on proliferation, cell death and gene expression profile. Our results indicate that miR-19a could contribute to promote a de-differentiation and a more aggressive phenotype of FTC-133 cells. Although further studies are needed, these data could represent a starting point for the development of new therapeutic targets for thyroid carcinomas.

Quantitation of MiR-19a Level in Basal and Overexpressing Conditions
In order to quantify the expression levels of miR-19a in FTC-133 cells, at both basal and overexpressing conditions, RT-PCR analysis has been performed on total RNA. MiR-19a levels were normalized to those of control U6 snRNA and miRNA levels of cells at 24 and 48 h after overexpression were compared to those of a control group.
Int. J. Mol. Sci. 2018, 20, 3944 2 is involved in the development and progression of several human cancers [12][13][14][15][16]. They work in a complex functional system, in which a precise miRNA regulates hundreds of mRNAs and a single gene is targeted by several miRNAs [17]. MiRNAs regulate various human genes correlated with several cellular processes including proliferation, apoptosis, metastasis, cell immunity and differentiation [18][19][20]. Approximately 2000 miRNAs have been discovered in humans [21] and about 50% have been found in cancer-related genomic regions or fragile sites, confirming their role as drivers of neoplastic alterations [22]. In addition recently, it has been demonstrated that miRNAs expression is correlated with different clinical parameters and their application could be used for cancer diagnosis, prognosis and treatment response [23][24][25].
MiR-19a, one of the seven miRNAs belonging to the miR-17-92 cluster located on chromosome 13q31.3, has been demonstrated to be a tumor-related miRNA involved in cancerogenesis. Several findings have revealed that miR-19a expression is increased in many different types of solid tumors including ATC, as shown in vitro by cell growth induction. Takakura et al., demonstrated that miR-19a is more expressed in thyroid cancer cell lines than in normal primary thyreocytes [26].
The aim of the present study was to evaluate the involvement of miR-19a in promoting dedifferentiation towards an aggressive phenotype. For this purpose, we induced miR-19a expression in follicular thyroid cancer cell line (FTC-133) and evaluated its effects on proliferation, cell death and gene expression profile. Our results indicate that miR-19a could contribute to promote a dedifferentiation and a more aggressive phenotype of FTC-133 cells. Although further studies are needed, these data could represent a starting point for the development of new therapeutic targets for thyroid carcinomas.

Quantitation of MiR-19a Level in Basal and Overexpressing Conditions
In order to quantify the expression levels of miR-19a in FTC-133 cells, at both basal and overexpressing conditions, RT-PCR analysis has been performed on total RNA. MiR-19a levels were normalized to those of control U6 snRNA and miRNA levels of cells at 24 and 48 h after overexpression were compared to those of a control group.
Specifically, to determine miR-19a basal level in the FTC-133 cell line, these cells were compared to anaplastic thyroid carcinoma cells (8505c cell line). Our analysis demonstrated that FTC-133 cells exhibit significantly more reduced expression of miR-19a compared to the 8505c (RQ = 0.0339; **p < 0.01; Figure 1). Furthermore, we quantified miR-19a expression level in FTC-133 cells after miR mimic overexpression and observed a significant increase at both time points (24 and 48 h)

MiR-19a Overexpression on FTC-133 Induces Phenotypic Changes on Cell Morphology
The morphological analysis of miR-19a mimic overexpressing FTC-133 cells showed phenotypic changes at both analyzed time points (

MiR-19a Overexpression on FTC-133 Promotes Proliferation and Cell Viability and Reduces Apoptosis
To evaluate the effects of miR-19a mimic overexpression on follicular thyroid carcinoma cells we analyzed proliferation, cell viability and apoptosis processes at 24 and 48 h after miR-19a overexpression ( Figure 3

MiR-19a Overexpression on FTC-133 Promotes Proliferation and Cell Viability and Reduces Apoptosis
To evaluate the effects of miR-19a mimic overexpression on follicular thyroid carcinoma cells we analyzed proliferation, cell viability and apoptosis processes at 24 and 48 h after miR-19a overexpression (Figures 3 and 4).
Both cell proliferation assays, Dapi (data not shown) and Trypan blue dye exclusion staining ( Figure 3B), showed that FTC-133 cells overexpressing the miR-19a mimic revealed a significant increase at both time points (24 and 48 h) compared to the control cells (** p < 0.01). Cell viability analysis showed that there was only a slight increase between control and miR-19a mimic overexpressing cells at the time point 48 h ( Figure 3C). Further, to detect the effect of miR-19a overexpression on cell growth, we analyzed two cell-cycle genes, CDC25a and Aurora kinase B (STK5) on both FTC-133 cells miR-mimic overexpressing and control cells. CDC25a on FTC-133 cells overexpressing miR-19a mimic displayed an evident up-regulation at both time points compared to the control, more marked at 48 h ( Figure 4A, ** p < 0.01). Similarly, STK5 showed an increased expression at both times, 24 and 48 h, statistically more significantly at 48h ( Figure 4A, 24 h ** p < 0.05 and 48 h ** p < 0.01).
On the other hand, we performed Caspase 3/7 and Caspase 9 analyses to assess the influence of miR-19a on apoptotic processes at 24 and 48 h after miR mimic overexpression. We showed that the overexpression of miR-19a mimic on FTC-133 cells, at both assessed time points, decreased Caspase 3/7 and Caspase 9 levels ( Figure 4B,C) even if the differences were significant only for Caspase 3/7 ( Figure 4B, 24 h ** p < 0.01 and 48 h * p < 0.05). Furthermore, we quantified the endogenous Caspase 3 level by Western blot and confirmed that miR-19a overexpression in FTC-133 reduced protein levels at both time points, especially at 24 h ( Figure 4D).
In addition, to confirm the possible role of miR-19a overexpression on proliferation and cell survival, we performed miR-19a inhibition on 8505c cells and evaluated its effects at 24  To evaluate the effects of miR-19a mimic overexpression on follicular thyroid carcinoma cells we analyzed proliferation, cell viability and apoptosis processes at 24 and 48 h after miR-19a overexpression (Figure 3 and 4).

MiR-19a Overexpression on FTC-133 Alters the Expression of Genes Related to Thyroid Differentiation and Poor Prognosis
To assess if miR-19a up-regulation in FTC-133 altered the expression profile of genes correlated with thyroid cells differentiation and poor prognosis, such as in 8505c cells, we performed qRT-PCR with thyrocyte-specific markers, including thyroid stimulating hormone receptor (TSHr), thyroglobulin (Tg), transcription termination factor 1 (TTF1), paired box gene 8 (Pax8), and poor prognosis genes, including Cadherin-1 (CDH1) and S100 calcium-binding protein A4 (S100A4).
The gene expression profile of FTC-133 cells, overexpressing miR-19a mimic, displayed a statistically significant reduction of genes related to thyroid cells differentiation at 24 and 48 h after miR mimic overexpression ( Figure 5A). Specifically, TSHr, Tg and Pax8 expression levels were significantly more down regulated at 24 h than at 48 h when compared to the control, while the levels of TTF1 were equally reduced at both time points. Moreover, FTC-133 cells overexpressing the miR-19a mimic exhibited an higher expression levels of Pax8 and Tg at both 24 and 48 h, compared to the 8505c group, although these were markedly down regulated compared to the FTC-133 control. On the other hand, TSHr and TTF1 were significantly lower than in 8505c, except for TSHr at 24 h ( Figure 5A). qRT-PCR has been performed for the mRNA of Tg, TSHr, Pax8, TTF1, CDH1 and S100A4. Glyceraldehyde-3-Phosphate Dehydrogenase (GAPDH) was used as endogenous controls. ANOVA test p value is reported (p < 0.0001) and * (p < 0.05), ** (p < 0.01) indicates significant differences between groups as reported by the post-hoc test.
The expression profile of CDH1 and S100A4, two poor prognosis genes involved in cell migration, proliferation and invasion, exhibited a striking decrease for CDH1 in FTC-133 transfected cells overtime, compared to the FTC-133 control ( Figure 5B; **p < 0.01) and a comparable expression with 8505c cells at 48 h; while S100A4 revealed an evident constant increased at 24 and 48 h, statistically significant compared to both FTC-133 control and 8505c cells ( Figure 5B; **p < 0.01).

Discussion
In the last years, it has been shown that miRNAs deregulation is involved in development and progression of several types of cancers including B cell chronic lymphocytic leukemia [27], breast tumor [28], lung cancer [29], gastric tumor [30], colorectal cancer [31], glioblastoma [32], prostate tumor [33] and thyroid carcinoma [34,35]. MiR-19a is the main component responsible for the oncogenic activity of miR-17-92, promoting the growth and migration of tumor cells [36].
The expression profile of CDH1 and S100A4, two poor prognosis genes involved in cell migration, proliferation and invasion, exhibited a striking decrease for CDH1 in FTC-133 transfected cells overtime, compared to the FTC-133 control ( Figure 5B; ** p < 0.01) and a comparable expression with 8505c cells at 48 h; while S100A4 revealed an evident constant increased at 24 and 48 h, statistically significant compared to both FTC-133 control and 8505c cells ( Figure 5B; ** p < 0.01).

Discussion
In the last years, it has been shown that miRNAs deregulation is involved in development and progression of several types of cancers including B cell chronic lymphocytic leukemia [27], breast tumor [28], lung cancer [29], gastric tumor [30], colorectal cancer [31], glioblastoma [32], prostate tumor [33] and thyroid carcinoma [34,35]. MiR-19a is the main component responsible for the oncogenic activity of miR-17-92, promoting the growth and migration of tumor cells [36].
MiR-19a is also involved in tumor development by promoting cell proliferation and migration, as well as by reducing apoptosis. For example, Liu et al. reported that miR-19a promotes cell proliferation and migration in colorectal cancer cells and accelerates tumor growth in xenograft mice [39]; Fu et al., demonstrated that miR-19a overexpression exerts its oncogenic activity by stimulating cell proliferation, cycle, growth and migration [50]. Finally, Paiva et al., suggested that the development of thyroid cancer derived from miR18a and miR19a altered expression depends primarily on cell-specific proteins [51].
Although miR-19a has also been found to be altered in ATC, its functional involvement is still poorly investigated.
The BRAF1799T>A mutation, causing the BRAFV600E oncoprotein, is very frequent in well differentiated thyroid cancer and in ATC derived from PTC [60]. Fuziwara et al., in their study demonstrated that BRAFV600E activates miR-17-92 via upregulation of Notch signaling [61].
However, given the severity of this neoplasm, because the scenario that characterizes ATC is still largely unknown, it becomes essential to evaluate the involved functional aspects to identify new and more promising therapeutic targets.
In this study, we firstly investigated miR-19a overexpression in FTC. Thus, to assess the association of miR-19a overexpression with de-differentiation and malignancy of thyroid cancer, we induced miR-19a expression in a well-differentiated thyroid tumor cell line, FTC, and analyzed the expression profile of specific markers of clinical relevance in the poor prognosis and thyroid cells differentiation. We found that miR-19a overexpression significantly increases the cell proliferation and reduces apoptosis consistently with the miR-19a role in other cancers [43,62]. Since it is known that PTEN [63] is one of the predicted targets of miR-19a, it is also possible that miR-19a overexpression might be associated with PTEN inactivation inducing cell proliferation in FTC-133, while its inhibition promotes PTEN activation inducing cell growth reduction. In a recent study, Frisk et al. have suggested that PTEN suppression is associated in extremely malignant or late-stage thyroid tumors, including ATC [64]. These results suggest that miR-19a leads to the poor prognosis in follicular thyroid cancer cells through stimulating cell proliferation.
Furthermore, we demonstrated that miR-19a overexpression correlates with the loss of the most differentiated phenotype, as suggested by the marked down regulation of TSHr, Tg, TTF1 and Pax8, specific genes and transcription factors associated to the thyroid cells differentiation [65][66][67][68]. TTF1 (also known as Nkx2-1) and Pax8 are two key genes for thyroid gland development and morphogenesis [69] and their expression is critical for the thyroid differentiated phenotype and for the transcriptional activation of the differentiation markers, such as Tg and TSHr [67,70,71]. Several studies reported that the neoplastic thyroid cells de-differentiation is correlated with the reduction or even the loss of specific thyroid genes, Tg and TSHr, and their transcription factors, TTF1 and Pax8 [72,73]. Our data support the idea that the decrease or loss of expression of thyroid cells differentiation markers is associated with the most de-differentiated phenotype of thyroid neoplasia.
Finally, we demonstrated that miR-19a is involved in malignant phenotype and aggressiveness as shown by the consistent increase S100A4, a main component of S100 family proteins, strictly associated to the malignant phenotype and metastatic behavior of various tumors [74][75][76] and decrease of CDH1, an E-cadherin, which down expression is found in several types of cancer and relates with infiltrative and metastatic behavior [77][78][79][80].
In conclusion, the present work supports the idea that the miR-19a may be involved in dedifferentiation and in the "poorer prognosis" of thyroid cancer, thus suggesting that this miRNA could represent a good prognostic indicator and a valid therapeutic target for highly malignant anaplastic tumors, although additional studies on human samples are also needed in order to prove the association of miR-19a expression with the relevant clinicopathological factors and prognosis.

MiRNA Mimic/Inhibitor Transfection on FTC-133 and 8505c Cells
FTC-133 and 8505c cells were transfected, respectively, with specific mirVana miRNA mimic and inhibitor (hsa-miR-19a-p3; Life Technologies, Camarillo, CA, USA) in accordance with the manufacturer's instructions. Briefly, both Lipofectamine RNAiMAX reagent (ThermoFischer Scientific, Italy) and miRNA (10 µM initial concentration) were dilute in Opti-MEM Medium (ThermoFischer Scientific, Italy). MiRNA solution was supplemented with an equivalent volume of Lipofectamine RNAiMAX reagent (1:1 ratio) and the resulting mix was incubated for 5 min at room temperature. After, the mix (miRNA mimic /inhibitor final concentration 0.1 µM) was added to 0.25 × 10 6 cells/6-well, as reported by the manufactures protocol. Lastly, transfected cells were incubated in a humidified environment at 37 • C and 5% CO 2 /95% air atmosphere for 24 and 48 h. Cells transfected only with the transfecting agent were used as control. Cell morphology was analyzed, and digital images were acquired using a Leica DMI4000B microscope.

Cell Proliferation Analysis Through DAPI and Trypan Blue Staining
For DAPI staining, FTC-133 cells were fixed in 4% PFA for 15 min and permeabilized in 0.3% Triton X-100 for 5 min. Successively, the cells were washed three times with PBS and the nuclei stained with DAPI (1:5000) in PBS for 5 min. Slides were mounted in fluorescent mounting medium Permafluor (Thermo Scientific) and digital images were acquired using a Leica DMI4000B fluorescence microscope. At least five images from each sample were taken for the count. Cell count analysis has been performed using Fiji image recognition software. For trypan blue dye exclusion staining we added 0.1 mL of trypan blue solution (0.4% in PBS) to 0.1 mL of cells, and after loaded on a hemacytometer, we counted immediately under a microscope at low magnification the number of blue staining cells and the number of total cells.

Measurement of Cell Viability Through MTT Assay
The cell viability of the two different cell samples, treated with only transfecting agent and treated with miR mimic or miR-inhibitor, was measured using an MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (Sigma-Aldrich, Italy) assay. Specifically, 24 and 48 h after cell transfection the medium from each well was removed and replaced with 200 µL of MTT solution (1 mg/mL in FBS-free medium). Following 2 h incubation at 37 • C and 5% CO 2 MTT solution was removed, each well was washed two times using cold PBS 0.01 M, and the formed crystals were melted using 200 µL of DMSO. Next, the absorbance at 570 nm was read using a synergy HT plate reader (BioTek Instruments, Inc., VT, United States).

Caspase-3/-7 and Caspase-9 Analyses
Briefly, the Caspase-Glo ® 3/7 or 9 Buffer and lyophilized Caspase-Glo ® 3/7 or 9 Substrate (Promega, Milan, Italy) as well as the 96-well plates containing cells were equilibrated at room temperature (10 min). Caspase-Glo ® 3/7 or 9 Buffer was mixed with Caspase-Glo ® 3/7 or 9 Substrate by inverting the contents. A total of 50 µL of the mixed solution was added to an identical volume of each sample, in agitation by using a plate shaker at 500 rpm for 1 min, and the plates left to incubate at room temperature for 1 h. A blank reaction, consisting of Caspase-Glo ® 3/7 or 9 Reagent and only vehicle was used as control. The luminescence of each sample was read using a synergy HT plate reader (BioTek Instruments, Inc., VT, United States).
For Western blot analysis, cell pellets were homogenized in lysis buffer (Tris-HCl pH 7.4, 1% Triton X100, NaCl 150 mmol/L and EDTA 1 mmol/L) supplemented with a cocktail of protease inhibitors (1:100, Sigma). For the quantification, 40 µg of protein were separated on a precast 4% to 20% tris-glycine gel (Thermo Scientific, Rockford, IL, USA) and transferred to a nitrocellulose membrane. After blocking the following primary antibodies: Mouse caspase 3 (Santa Cruz, 1:500) and mouse ß-actin (Santa Cruz, 1:500). After three washes in TBST, the membranes were incubated with anti-mouse HRP-conjugated secondary antibody (Thermo Scientific group; 1:6000) for 1 h at RT. Peroxidase activity was developed by enhanced chemiluminescent substrate (Pierce Biotechnology Inc., Thermo Scientific) and visualized by autoradiography. The density of each band was quantified using ImageJ analysis software and to ß-actin levels measured in the same membrane.

qRT-PCR
For qRT-PCR analyses total RNA from FTC-133 cells was isolated 24 and 48 hours after miR mimic transfection using RNeasy Mini Kit (Qiagen, Germantown, MD, USA) and quantified as previously described [81]. Three independently cultured samples were used for each of the two time points. cDNA was synthesized from 1 µg of total RNA using ImProm-II Reverse Transcription System (Promega, Milan, Italy). qRT-PCR was performed using SYBR Green method on a 7900HT Real Time PCR using manufactures protocol (Applied Biosystems). Specific primers for each of the investigated molecular endpoint were designed using primer blast and selecting exon-exon junctions on mRNA as target region for annealing. Each sample was tested in triplicate and gene expression was assessed using the 2 −∆∆Ct method. RNA from control cells was used as reference for relative expression quantitation. The following primers for qRT-PCR were used: CDH1 F: TGCCCAGAAAATGAAAAAGG, R: GTGTATGTGGCAATGCGTTC; S100A4 F: GTGAC GCCCTGTCTCTAAGC, R: ATAGCAACAGCGTGTGCAAG; Tg F: AACCCCATTGTGTTCTCAGC, R: CATTAGCCCAGGCTTCAGAG; Pax8 F: AAGGTGGTGGAGAAGATTGG, R: GCTGCTCTGT GAGTCAATGC; TSHr F: ATGGGGATGTACCTGCTCCT, R: AATGAGATTGGGGCCATGCA; TTF1 F: GCCTCAGTCAGAATCCCACC, R: GGTATGCACTCTCGAGGCTC; CDC25a F: GGC AAGCGTGTCATTGTTGT; R: GGCTCACAGTAAGACTGGCA; STK5 F: GTGTACTTGGC TCGGGAGAA; R: CAGCTCTTCTGCAGCTCCTT; GAPDH F: GGAAGGTGAAGGTCGGAGT, R: TGGGTGGAATCATATTGGAA.
Glyceraldehyde 3-Phosphate Dehydrogenase (GAPDH) was used as a reference gene. Expression Suite Software v1.1 was used for data analysis.

Statistical Analysis
Statistical analysis was performed by Student t-test and one-way ANOVA. Tukey HSD (Honestly Significant Difference) method was used as a post hoc test when the ANOVA reported statistically significant differences, to evaluate the differences between the individual time-points or treatment groups.
Author Contributions: All authors had substantial contribution to the present work. Conceptualization, G.C. and R.P.; Data curation, G.C. and A.Z.; Funding acquisition, R.G. and R.P.; Methodology, A.D. and F.T.; Resources, R.P.; Supervision, R.P.; Writing-review & editing, G.C. and R.P.; G.C. is also the corresponding author. All the authors gave the final approval of the version to be published.