Sage Journals: Discover world-class research

Abstract

Thyroid carcinoma is one of the most frequent malignant tumors of the endocrine system, which accounts for nearly 1% population in newly diagnosed carcinoma worldwide and the incidence has an increasing tendency in recent years. To explore whether miR-135a-5p could affect the proliferation, invasion and migration of thyroid carcinoma cells by targeting VCAN. The expression levels of miR-135a-5p and VCAN were detected in human thyroid carcinoma tissues and cells, para-carcinoma tissues, as well as human normal thyroid cells using RT-qPCR and Western blot. In addition, dual-luciferase reporter gene assay, MTT assay, colony formation assay, wound healing assay, Transwell assay, flow cytometry analysis and in vivo tumorigenesis assay were also conducted. The results demonstrated that miR-135a-5p was down-regulated while VCAN was up-regulated both in thyroid carcinoma tissues and cells. Furthermore, the up-regulation of miR-135a-5p inhibited cell proliferation, invasion and migration of thyroid carcinoma. Dual-luciferase reporter gene assay provided evidence indicating that miR-135a-5p targeted VCAN in thyroid carcinoma. MiR-135a-5p could inhibit cell proliferation, invasion and migration of thyroid carcinoma by targeting VCAN. MiR-135a-5p and VCAN might emerge as a target for the treatment of thyroid carcinoma.

Keywords

Thyroid carcinoma miR-135a-5p VCAN

1. Introduction

Thyroid carcinoma is one of the most frequent malignant tumors of the endocrine system, which accounts for nearly 1% population in newly diagnosed carcinoma worldwide and the incidence has an increasing tendency in recent years [1]. Thyroid carcinoma can be mainly classified as papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC), Hürthle cell carcinoma (HCC), anaplastic thyroid carcinoma (ATC), and medullary thyroid carcinoma (MTC) [2]. Among the subtypes of thyroid carcinoma, PTC accounts for about 80% of the total incidence, while FTC, HCC and poorly differentiated thyroid carcinoma (PDTC) have a high risk in metastasizing to lung or bone [3]. Nevertheless, the mortality remains high as many patients have no response to these conventional thyroid carcinoma therapies including radioiodine, radiotherapy, surgery, and chemotherapy [4]. Thus, new treatments like gene therapy are needed urgently. Although significant progress has been made these years in the understanding of molecular and gene basis of the formation of thyroid carcinoma, there is still a great need of understanding the undiscovered mechanisms in tumorigenesis and metastasis to prevent, diagnose, and treat thyroid carcinoma.

MicroRNAs (miRNAs, miR) are non-coding single-stranded RNAs (around 21 nucleotides) that are involved in regulating the gene expression by binding to the 3 ${}^{\prime}$ -untranslated regions (3 ${}^{\prime}$ -UTR) of mRNA transcripts of target genes and repressing their translation or regulating their degradation [5]. It has been validated that miRNAs play an important role in tumorigenesis by regulating cell processes in tumor cells. For thyroid carcinoma, a number of miRNAs are reported to contribute to the development and cellular progress of the tumor [6]. For instance, miRNA-145 has been reported as a suppressor of thyroid carcinoma growth and metastasis by targeting AKT3 [7]. Many researches have shown that miR-135a-5p can act as a regulator of the differentiation, regeneration, and carcinogenesis in human cells, as well as a tumor suppressor in gastric carcinoma and gallbladder carcinoma by targeting correlating genes [8, 9]. The target genes of miR-135a in tumor cells had been found as focal adhesion kinase, FOXO1, ERRalpha, and KIFC1, most of which are connected with cell invasion and sensitivity to chemotherapy [10, 11, 12, 13]. However, the biological role and molecular mechanism of miR-135a in thyroid carcinoma have not been studied yet.

Versican (also known as CSPG2 or VCAN), a chondroitin sulfate proteoglycan as one of the main component of the extracellular matrix (ECM), presents a significantly higher level in many diseases, interacting with hyaluronan (HA), than that in normal tissues [14]. The N-terminal globular (G1) domain and C-terminal globular (G3) domain of VCAN play a role in regulating migration, adhesion and proliferation in normal tissues, as well as an interactive role in varieties of intracellular and extracellular molecules [15]. Its overexpression associated with the development and relapse of tumor has been reported in many types of carcinoma cells, such as colorectal carcinoma, breast carcinoma, gastric carcinoma etc. [16, 17, 18]. It is suggested that VCAN is involved in promoting proliferation, motility, and invasion of malignant tumor cells, therefore the inhibition of its synthesis or process may be a potential therapeutic target in carcinoma [19]. But up to now, the expression and regulation of VCAN gene in thyroid carcinoma cells remain unclear.

Several miRNAs, including miR-144, miR-133a and miR-136, have been reported as potential regulator of structural ECM proteins, collagen and VCAN in the gathering, adhesion, and organization of the ECM [19]. Previous studies also proved that the level of VCAN in carcinoma cells can be controlled by the miRNA-mRNA regulation network. For example, a study indicated that the migration of malignant melanoma cell could be inhibited by targeting VCAN gene with a miRNA, which revealed the possibility of study on the targeting relationship between miRNA and VCAN [20]. In the present study, we aimed to verify the expression and function of both miRNA-135a-5p and VCAN in thyroid carcinoma cells and find the certain relationship between them.

2. Materials and methods

2.1 Clinical specimens

Fifty-three pairs of human thyroid carcinoma and para-carcinoma tissues (distance of 3 cm from the lesion) were collected from patients (19 males and 34 females) diagnosed with thyroid carcinoma who had undergone tumor resection at People’s Hospital of Weifang during Sep 2014 to Sep 2016. Tissues were snap-frozen in liquid nitrogen after surgery and reserved at $-$ 80 ${{}^{\circ}}$ C. This research was approved by Ethics Committee of People’s Hospital of Weifang, and all patients signed the informed consent.

2.2 Cell culture

The human thyroid carcinoma cell line FTC-133, human PTC cell lines TPC1 and K1, human squamous thyroid carcinoma (STC) cell line SW579, and human normal thyroid cell line HT-ori3 (ATCC, Maryland, USA) were used in this research. FTC-133 cells were cultured in dulbecco’s modified eagle medium (DMEM) with 10% fetal bovine serum (FBS) and Ham’s F12 (1:1). K1 and TPC1 cells were cultured in 1640-RPMI medium with 10% FBS. SW579 cells were also cultured in DMEM with 10% FBS. HT-ori3 cells were cultured in Ham’s F12 with 5% FBS, supplemented with 1 mU/mL thyrotrophic hormone, 10 ng/mL hydrocortisone, 5 $\mu$ g/mL transferrin and 10 $\mu$ g/mL insulin (Sigma-Aldrich, Missouri, USA). All cells were incubated in 5% CO ${}_{2}$ at 37 ${{}^{\circ}}$ C.

2.3 Cell transfection

Cells (5 $\times$ 10 ${}^{4}$ /well) were inoculated in 6-well plates. Cell transfection was done using Lipofecta- mine™ 2000 (Invitrogen, Carlsbad, CA) following the manufacturer’s protocols. MiR-135a-5p mimics, negative controls (NCs) and Pgv208-GFP vectors (Gene- Pharma, Shanghai, China) were used in this research. We divided cells into 6 groups: Control, NC-1 (cells transfected with miR-135a-5p NCs), NC-2 (cells transfected with Pgv208-GFP vectors), Mimics (cells transfected with miR-135a-5p mimics), cDNA VCAN (cells transfected with cDNA VCAN), Mix (cells transfected with miR-135a-5p mimics and cDNA VCAN).

Table 1
RT-qPCR primer sequences

cDNA		Primer sequences
miR-135a-5p	F	5 ${}^{\prime}$ -CTCCTAGGTATGGCTTTTTATTC-3 ${}^{\prime}$
	R	5 ${}^{\prime}$ -TCAACTGGTGTCGTGGAGTC-3 ${}^{\prime}$
U6	F	5 ${}^{\prime}$ -CTCGCTTCGGCAGCACA-3 ${}^{\prime}$
	R	5 ${}^{\prime}$ -AACGCTTCACGAATTTGCGT-3 ${}^{\prime}$
VCAN	F	5 ${}^{\prime}$ -GAGATAAGATGGGAAAGGCAGG-3 ${}^{\prime}$
	R	5 ${}^{\prime}$ -GGGGACAGTGAGGTGGAACA-3 ${}^{\prime}$
GAPDH	F	5 ${}^{\prime}$ -CCTGCCTCTACTGGCGCTGC-3 ${}^{\prime}$
	R	5 ${}^{\prime}$ -GCAGTGGGGACACGGAAGGC-3 ${}^{\prime}$

F: Forward primer; R: Reverse primer.

2.4 RNA extraction and RT-qPCR

All cellular RNA was extracted from tissues and cells using QIAzol (Qiagen, Hilden, Germany). Then they were transcribed into cDNAs, which were amplified following the manufacturer’s specification of PrimeSeript TM Reverse Transeriptase kit (TaKaRa, Japan) and the ABI 7900 real time PCR instrument (ABI, California, USA). U6 and GAPDH were used as standard for calculating relative miRNA and mRNA levels respectively by the 2 ${}^{\rm-\Delta\Delta Ct}$ method. The primer sequences for RT-PCR were listed in Table 1.

2.5 Western blot

Total proteins were extracted from cells after 24 h transfection. The concentration of protein was determined using Bradford method. The relative proteins were separated by SDS-PAGE under 220 v and transferred to the PVDF membranes. The membranes were incubated overnight with primary antibodies of VCAN and $\alpha$ -Tubulin, followed by horseradish-peroxidase secondary antibodies for 1–2 hours. Blots were visualized by electrogenerated chemilumines (ECL) and $\alpha$ -Tubulin acted as the internal control

2.6 MTT assay

K1 cells were plated onto 96-well plates at a density of 1 $\times$ 10 ${}^{4}$ cells/well. MTT solution (20 $\mu$ L) was added to every well at each indicating time point (24, 48, 72, 96 h). After 4 hours’ incubation at 37 ${{}^{\circ}}$ C, 150 $\mu$ L DMSO was added to each well. The optical absorbance of cells at 490 nm wavelength was read under a microplate reader, each group was done for 3 times and growth curves were drawn with the mean value.

2.7 Colony formation assay

K1 cells of 1 $\times$ 10 ${}^{6}$ /L were incubated in 5% CO ${}_{2}$ atmosphere at 37 ${{}^{\circ}}$ C for 2–3 weeks until cell colonies were seen by naked eyes. Cells were immobilized with 4% paraformaldehyde and dyed with GIEMSA. Stained colonies were counted and the colony number was recorded.

2.8 Wound healing assay

The cells were inoculated into 6-well plates. When the cell confluence achieved about 90%, scratches were created by scraping the cell layer across each well. Detached cells were rinsed out using serum-free medium. The remaining cells were cultured for another 12 or 24 hours in 5% CO ${}_{2}$ atmosphere at 37 ${{}^{\circ}}$ C. The cell culture surfaces were photographed at each time point (12, 24 h).

2.9 Transwell assay

The invasion of transfected cells was determined using Transwell assay. The membrane of upper chambers was coated with 50 mg/L Matrigel (1:8). A total of 100 $\mu$ L serum-free medium were accreted to the upper chambers, meanwhile, medium with 10% FBS were added to lower chambers. After 48-hour’ transfection, the cells were digested using trypsin, and then washed with PBS. Cells were adjusted to a density of 5 $\times$ 10 ${}^{4}$ /mL and inoculated in 6-well plates for 24 h at 37 ${{}^{\circ}}$ C. Cells that failed to invade through the membranes were wiped off, on the other hand, those invaded through successfully were fixed with 95% alcohol and stained with haematoxylin, the numbers were counted under a microscope as well.

2.10 Dual-luciferase reporter gene assay

The amplification of VCAN 3 ${}^{\prime}$ -UTR was cloned into the pmirGLO vectors to construct the recombinant plasmid pmirGLO-3 ${}^{\prime}$ -VCAN-wt. Mutated VCAN 3 ${}^{\prime}$ -UTR were also amplified and cloned into plasmid to construct recombinant plasmid pmirGLO-3 ${}^{\prime}$ -VCAN-mut. MiR-135a-5p mimics or NCs were transfected to K1 cells along with pmirGLO-3 ${}^{\prime}$ -VCAN-wt, pmirGLO-3 ${}^{\prime}$ -VCAN-mut or pmirGLO vectors. The luciferase activity was measured by Dual-Luciferase ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ Reporter Assay System (Promega, Madison, USA) and presented as the ratio of Firefly to Renilla luciferase activity.

Figure 1.

The expressions of miR-135a-5p and VCAN in thyroid tissues and cells. (A) RT-qPCR was used to detected miR-135-3p expression in thyroid tissues ( $n=$ 53) and adjacent tissues ( $n=$ 53), * $P<$ 0.05 compared with adjacent tissues; (B) RT-qPCR was used to detected miR-135a-3p expression in normal thyroid cell line (HT-ori3) and thyroid carcinoma cell lines (FTC-133, K1, TPC1 and SW579), ** $P<$ 0.01 compared with HT-ori3 cells; (C) Western blot was used to detected VCAN protein expression in normal thyroid cell line (HT-ori3) and thyroid carcinoma cell lines (FTC-133, K1, TPC1 and SW579), $\alpha$ -Tubulin was used as the internal control.

Figure 2.

MiR-135a-5p directly targets VCAN. (A) The predicted miR-135a-5p binding site on wild-type VCAN 3 ${}^{\prime}$ -UTR and the corresponding mutations in mutant VCAN 3 ${}^{\prime}$ -UTR; (B) Luciferase activities of wild-type pmirGLC-VCAN-3 ${}^{\prime}$ UTR, mutant pmirGLO-KLF4-3 ${}^{\prime}$ UTR, and empty vector in K1 cells were detected by dual-luciferase reporter gene assay after cells were transfected with miR-135a-5p mimics or negative control, * $P<$ 0.05 compared with NC group; (C) the expression levels of VCAN protein in K1 cells transfected with miR-135a-5p mimics and negative control were determined by Western blot, $\alpha$ -Tubulin was used as the internal control.

Figure 3.

MiR-135a-5p inhibits cell proliferation, migration and invasion of thyroid carcinoma by targeting VCAN. (A) Cell viability was determined by MTT assay on K1 cells from different groups (Control, cells without transfection; NC-1, cells transfected with miR-135a-5p Negative control; NC-2, cells transfected with Pgv208-GFP vectors; Mimics, cells transfected with miR-135a-5p mimics; cDNA VCAN, cells transfected with VCAN cDNA; Mix, cells transfected with miR-135a-5p mimics and VCAN cDNA); (B) colony formation assay was used to detect the proliferation of cells in different groups (control, NC-1, NC-2, mimics, cDNA VCAN, Mix), the results were shown as the percentage colony number of control group; (C) cell migration was detected by wound healing assay, the relative wound closure was quantified as the ratio of the migration distance to control distance after 12 h and 24 h; (D) cell invasion was detected by Transwell assay, invaded cells were stained and counted in three randomly selected visual fields. * $P<$ 0.05, ** $P<$ 0.01 compared with control group.

Figure 4.

MiR-135a-5p affects cell cycle and apoptosis of thyroid carcinoma by targeting VCAN. (A) The cell cycle of K1 cells from different groups (Control, cells without transfection; NC-1, cells transfected with miR-135a-5p Negative control; NC-2, cells transfected with Pgv208-GFP vectors; Mimics, cells transfected with miR-135a-5p mimics; cDNA VCAN, cells transfected with VCAN cDNA; Mix, cells transfected with miR-135a-5p mimics and VCAN cDNA) was detected by flow cytometry with PI staining; (B) the cell apoptosis of cell from different groups (Control, NC-1, NC-2, mimics, cDNA VCAN, Mix) was analysed by flow cytometry with Annexin V-FITC/PI staining. * $P<$ 0.05 compared with control group.

Figure 5.

MiR-135a-5p suppresses tumor growth of thyroid carcinoma in vivo. (A) The tumors from the nude mice injected with K1 cells from different groups (Control, cells without transfection; NC-1, cells transfected with miR-135a-5p Negative control; NC-2, cells transfected with Pgv208-GFP vectors; Mimics, cells transfected with miR-135a-5p mimics; cDNA VCAN, cells transfected with VCAN cDNA; Mix, cells transfected with miR-135a-5p mimics and VCAN cDNA) were separated five weeks after the injection; (B) The volume of tumor from different groups (Control, NC-1, NC-2, mimics, cDNA VCAN, Mix) was counted each 5 days and tumor growth curve was calculated at 30 days after injection, * $P<$ 0.05 compared with control group; (C) The VCAN expression of tissue samples from different groups (Control, NC-1, NC-2, mimics, cDNA VCAN, Mix) was detected by immunohistochemistry assay, VCAN was stained with brown.

2.11 Flow cytometry analysis

Approximately 2 $\times$ 10 ${}^{6}$ K1 cells were collected from each group and they were fixed with 75% alcohol at 4 ${{}^{\circ}}$ C overnight. The cells were then washed with PBS at a centrifugalization of 1000 rpm for 5 min. After that, the cells were incubated in dark for 30 min at 37 ${}^{\circ}$ C with 1% BSA, 100 $\mu$ L PI and 100 $\mu$ L RNA enzyme (10 mg/mL) added sequentially. Thereafter, the cell cycle was analysed using the flow cytometer.

As for apoptosis, the cells (2 $\times$ 10 ${}^{6}$ ) were digested by 0.25% trypsin and washed by PBS (1000 rpm, 5 min), followed by the addition of 100 $\mu$ L AnnexiV-FITC and 10 $\mu$ L PI to the cell suspension (1 $\times$ 10 ${}^{6}$ cells). The cell apoptosis rate was detected in 15 minutes using the flow cytometry instrument.

2.12 In vivo tumorigenesis assay

0.2 mL K1 single-cell suspension (2 $\times$ 10 ${}^{6}$ cells) was injected subcutaneously into the back of the male nude mice (4–5 weeks old). The mice were divided into six groups (4 mice per group): Control, NC-1, NC-2, Mimics, cDNA VCAN and the Mix. Tumor volumes were measured using the formula: carcinoma volume $=$ 0.5 $\times$ (length $\times$ width ${}^{2}$ ). Five weeks after the injection, mice were euthanized and tumors were harvested. The formaldehyde-fixed and paraffin-embedding carcinoma tissues were sectioned and analyzed by immunohistochemistry.

2.13 Statistical analysis

SPSS 20.0 (Chicago, Illinois, America) was employed for statistical analysis and GraphPad Prism 6.0 (GraphPad Software, Inc., CA, USA) was used for plotting. The data were presented as mean $\pm$ standard deviation (SD). The differences between two groups were analyzed using t test. One-way analysis of variance (ANOVA) was utilized to determine the significance of differences among groups. $P<$ 0.05 was seen as significantly different.

3. Results

3.1 MiR-135a-5p and VCAN expression in thyroid tissues and cells

RT-qPCR was applied to detect the expression levels of miR-135a-5p in 53 pairs of thyroid carcinoma and para-carcinoma tissues. As presented in Fig. 1A, miR-135a-5p expression was dramatically down-regulated in thyroid carcinoma tissues compared with para-carcinoma tissues ( $P<$ 0.05). In addition, the expression levels of miR-135a-5p were also compared in normal thyroid cell line (HT-ori3) and thyroid carcinoma cell lines (FTC-133, TPC1, K1 and SW579). As shown in Fig. 1B, miR-135a-5p expression was relatively lower in thyroid carcinoma cells than in normal thyroid cells ( $P<$ 0.01). VCAN, on the other hand, demonstrated quite the opposite trends in cells (Fig. 1C). Since K1 showed the highest expression level of miR-135-5p, it was then chosen for further experiments.

3.2 MiR-135a-5p directly targets VCAN

To confirm that whether miR-135a-5p could target VCAN, plasmids carrying miR-135-5p or VCAN 3 ${}^{\prime}$ -UTR gene sequences were constructed. The wild-type and mutated VCAN 3 ${}^{\prime}$ -UTR as well as miR-135a-5p sequences were all illustrated in Fig. 2A. The relative luciferase activity of the cells transfected with pmirGLO-3 ${}^{\prime}$ -VCAN-wt and miR-135a-5p mimics was significantly lower than that of cells transfected with NCs ( $P<$ 0.05, Fig. 2B). However, the cotransfection of pmirGLO-3 ${}^{\prime}$ -VCAN-mut and miR-135a-5p mimics did not influence the luciferase activity significantly. We further detected the expression levels of VCAN in K1 cells transfected with miR-135a-5p mimics by Western blot. We could see from Fig. 2C that the expression of VCAN was significantly down-regulated in Mimics group comparing with NC-1 group. Taken together, we could conclude that miR-135a-5p directly targeted VCAN.

3.3 MiR-135a-5p suppresses cell proliferation and invasion of thyroid carcinoma through targeting VCAN 3 ${}^{\prime}$ -UTR

3.3.1 Cell proliferation

To investigate the effects of miR-135a-5p on K1 cell proliferation, we carried out MTT and colony formation assays. As shown in the MTT assay results (Fig. 3A), there was no obvious difference in cell numbers among groups 48 h after transfection, whereas, 72 h after transfection, substantial decreased cell number was observed in Mimics group and significant increased cell number was seen in cDNA VCAN group ( $P<$ 0.05). Similarly, 96 h after transfection, cells in Mimics group demonstrated weaker cell growth and those in cDNA VCAN group showed stronger cell growth ( $P<$ 0.01). On the other hand, cells in Mix group had almost the same cell growth condition with the control group at every indicating time point, suggesting that overexpression of miR-135a-5p could inhibit VCAN-induced cell proliferation. The colony number in Mimics group was less than that in NC-1 group ( $P<$ 0.05), while the colony number in cDNA VCAN group was more than that in NC-2 group ( $P<$ 0.05, Fig. 3B). Besides, Mix group was seen no difference from the control group, which revealed that the up-regulation of miR-135a-5p has an inhibitory effect on cell proliferation induced by VCAN.

3.3.2 Cell migration and invasivion

To define the role of miR-135a-5p in cell migration and invasion, we conducted wound healing and Transwell assays. As expected, migration rate of cells in the Mimics group markedly decreased, while a significantly risen migration rate was observed in cells transfected with cDNA VCAN ( $P<$ 0.05, Fig. 3C), and the Mix group showed no prominent difference from the control group, indicating that miR-135a-5p suppressed cell migration in thyroid carcinoma cells by repressing VCAN.

We also conducted Transwell assay to explore the effect of miR-135a-5p on cell invasion. Similarly, the number of invaded K1 cells was significantly smaller in Mimics group ( $P<$ 0.05), while that was substantially larger in cDNA VCAN group ( $P<$ 0.05, Fig. 3D). Mix group was almost the same as Control group in terms of the invasiveness of cells, suggesting that the overexpressed miR-135a-5p inhibited thyroid carcinoma cell invasion.

3.4 MiR-135a-5p affects cell cycle and apoptosis of thyroid carcinoma by targeting VCAN

3.4.1 Cell cycle

Flow cytometry was applied to analyze cell cycle, results were presented in Fig. 4A. Apparently, significant cell number reduction in S and G2/M, and accumulation in G0/G1 were observed in Mimics group ( $P<$ 0.05). Nevertheless, in cDNA VCAN group, cells were mostly entering S and G2/M phase ( $P<$ 0.05). There was no remarkable difference among NCs, Mix and Control groups in this case.

3.4.2 Cell apoptosis

As shown in Fig. 4B, comparing with NC groups, apparent increase in cell apoptosis rate was noticed in the Mimics group while significant reduction was observed in the cDNA VCAN group ( $P<$ 0.05). Moreover, there were also no differences among NCs, Control and Mix groups, suggesting that the up-regulation of miR-135a-5p could promote cell apoptosis.

3.5 MiR-135a-5p suppresses tumor growth of thyroid carcinoma in vivo

To investigate whether miR-135a-5p overexpression repressed tumor growth in vivo, we conducted the nude mice tumorigenesis assay. Photographs of tumors obtained from mice 5 weeks after the injection were presented in Fig. 5A, the tumor volume was remarkably smaller in the Mimics group, in contrast with the cDNA VCAN group. Obviously, lower tumor growth rate and apparent reduction in tumor volume were noticed in the Mimics group, while higher tumor growth rate and tumor volume increase were seen in cDNA VCAN group ( $P<$ 0.05, Fig. 5B) compared with NCs. We also conducted immunohistochemistry assay to explore the distribution of VCAN in K1 cells. As shown in Fig. 5C, VCAN was distributed mainly in extracellular interstitium. Besides, sparse and dense VCAN were seen in the Mimics group and cDNA VCAN group respectively. There was no difference in carcinoma volume, growth rate and VCAN distribution among NCs, Mix and Control groups, illustrating that miR-135a-5p overexpression counteracted the promoting effect of VCAN on tumorigenesis. These findings suggested that miR-135a-5p suppressed tumor growth of thyroid carcinoma by targeting VCAN.

4. Discussion

As one of the most frequent malignant carcinomas of the endocrine system with an ever-increasing incidence, thyroid carcinoma has been studied a lot on its pathogenesis and therapies. However, the molecular and gene basis of its development is still unclear, which is proved to be a difficulty to surmount through. In recent years, researchers began focusing on the regulating role of miRNAs in tumors, and several miRNAs have been proved to function by overexpressing or suppressing the carcinoma progress. For instance, Qiu et al. found that the overexpression of miRNA-613 down-regulated cell proliferation, migration, and invasive ability in PTC by targeting SphK2 [21]. Minna et al. identified that miR-451a, targeting AKT/mTOR pathway, was suppressed in thyroid carcinoma cells and inhibited the neoplasia progress [22]. Similarly, miR-191 was also reported by Colamaio et al. as a suppressor in FTC and PTC, playing a negative role in pathogenesis by targeting CDK6 [23]. In the present study, we demonstrated that miR-135a-5p was remarkably suppressed in human thyroid carcinoma tissues. The expression of miR135a-5p was inversely correlated with the expression of VCAN gene. Moreover, functional assays proved that the overexpression of miR-135a-5p inhibited cell proliferation, migration and invasion of thyroid carcinoma cells, and slowed the growth of thyroid carcinoma in vivo by targeting VCAN 3 ${}^{\prime}$ -UTR, which suggests that miR-135a-5p is a potential respect for the treatment of thyroid carcinoma.

Recently, miR-135a-5p has been identified as a biomarker in tumor tissues as its ectopic expression. Golubovskaya et al. confirmed that the overexpression of miR-135 inhibited the invasion of cancer cells and increased the sensitivity to chemotherapy by targeting focal adhesion kinase, which could be a key for tumor therapy [10]. MiR-135a-5p has also been proved as a tumor suppressor in gastric carcinoma and gallbladder carcinoma by targeting very low density lipoprotein receptor (VLDLR) and AP-2 $\alpha$ /BCL-2 pathway, respectively [8, 9]. In our study, we proved that miR-135a-5p was significantly suppressed in thyroid carcinoma compared with normal tissues, which is in line with other tumors in previous studies. Meanwhile, its inhibitory effect on cell migration and invasion was supported by the results of wound healing and Transwell assays. These results suggested that the expression level of miR-135a-5p in thyroid carcinoma cells may contribute to tumor genesis and development.

Several target genes of miR-135a-5p have been discovered in previous studies. For further understanding the mechanism of miR-135a-5p in thyroid carcinoma cells, we identified versican (VCAN) as a potential target based on bioinformatics analysis. As a main component of ECM, the overexpression of VCAN was indicated to be associated with improvement of cell proliferation, migration and invasion in malignant tumor tissue [19]. We observed in the present study that the expression level of VCAN in thyroid carcinoma cells was remarkably higher than other tissues, accelerating cell proliferation and migration as well as increasing the invasion of thyroid carcinoma cells, which supported that VCAN could be a potential therapeutic target in thyroid carcinoma cells.

Up to now, although the mechanism that miR-135a-5p could target VCAN is agreed by bioinformatics analysis, there is no experimental evidence published yet. Our experiment of dual-luciferase reporter gene assay proved that VCAN 3 ${}^{\prime}$ -UTR could be directly targeted by miR-135a-5p and there was an inverse correlation existed in their expressions. The targeted relationship was also confirmed by our further functional studies: cells with the cotransfection of miR-135a-5p mimics and cDNA VCAN performed similarly to NCs groups in cell progress in vitro, as well as tumor growth in vivo, while cells transfected with miR-135a-5p mimics or cDNA VCAN respectively indicated significant differences.

For the first time, our study investigated the functional mechanism of miR-135a-5p as a suppressor in thyroid carcinoma cells as well as targeting and regulating VCAN, which would provide evidence for further studies on the miR135a-5p/VCAN pathway and its functional roles in thyroid carcinoma cells. However, there are still several limitations in the study. Firstly, since thyroid carcinoma is classified as several types with different characteristics, the result of this study may not be appropriate for all types of thyroid carcinoma. Secondly, we cannot exclude that other genes may be targeted by miR-135a-5p, which can result in the regulation of thyroid carcinoma cells in this experiment. Lastly, VCAN may be regulated by other miRNAs besides miR-135a-5p, thus further study on the functional mechanism of VCAN is also needed.

In summary, our data put forward a new idea that miR-135a-5p can suppress cell progress including cell proliferation, migration, and invasion in thyroid carcinoma by targeting VCAN 3 ${}^{\prime}$ -UTR, and updated the current knowledge about the functional roles of miRNAs in thyroid carcinoma. In addition, we provided evidence for using the miR-135a-5p/VCAN pathway as a novel gene target for treating thyroid carcinoma.

Conflict of interest

None.

References

Nikiforova

M.N.

and Nikiforov

Y.E.

, Molecular genetics of thyroid cancer: Implications for diagnosis, treatment and prognosis, Expert Rev Mol Diagn 8(1) (2008), 83–95.

Brown

R.L.

et al., Thyroid cancer: Burden of illness and management of disease, J Cancer 2 (2011), 193–199.

Cabanillas

M.E.

et al., Thyroid cancer, Lancet 388(10061) (2016), 2783–2795.

Antonelli

et al., New targeted therapies for anaplastic thyroid cancer, Anticancer Agents Med Chem 12(1) (2012), 87–93.

Griffiths-Jones

et al., miRBase: microRNA sequences, targets and gene nomenclature, Nucleic Acids Res 34(Database issue) (2006), D140–144.

Deng

et al., MiR-146b-5p promotes metastasis and induces epithelial-mesenchymal transition in thyroid cancer by targeting ZNRF3, Cell Physiol Biochem 35(1) (2015), 71–82.

Boufraqech

et al., miR-145 suppresses thyroid cancer growth and metastasis and targets AKT3, Endocr Relat Cancer 21(4) (2014), 517–531.

Zhou

et al., MicroRNA-135a acts as a putative tumor suppressor by directly targeting very low density lipoprotein receptor in human gallbladder cancer, Cancer Sci 105(8) (2014), 956–965.

Pan

et al., Regulation of BGC-823 cell sensitivity to adriamycin via miRNA-135a-5p, Oncol Rep 32(6) (2014), 2549–2556.

10.

Golubovskaya

V.M.

et al. cells

MiR-138andMiR-135directlytargetfocaladhesionkinase,inhibitcellinvasion,andincreasesensitivitytochemotherapyincancer

, Anticancer Agents Med Chem 14(1) (2014), 18–28.

11.

Ren

J.W.

et al., MiR-135 post-transcriptionally regulates FOXO1 expression and promotes cell proliferation in human malignant melanoma cells, Int J Clin Exp Pathol 8(6) (2015), 6356–6366.

12.

Tribollet

et al., miR-135a inhibits the invasion of cancer cells via suppression of ERRalpha, PLoS One 11(5) (2016), e0156445.

13.

Zhang

et al., miR-135a acts as a tumor suppressor in gastric cancer in part by targeting KIFC1, Onco Targets Ther 9 (2016), 3555–3563.

14.

Keire

P.A.

et al., A role for versican in the development of leiomyosarcoma, J Biol Chem 289(49) (2014), 34089–34103.

15.

W.W.

et al., Roles of versican in cancer biology-tumorigenesis, progression and metastasis, Histol Histopathol 28(6) (2013), 701–713.

16.

Chida

et al., Stromal VCAN expression as a potential prognostic biomarker for disease recurrence in stage II–III colon cancer, Carcinogenesis 37(9) (2016), 878–887.

17.

W.W.

et al., The role of versican in modulating breast cancer cell self-renewal, Mol Cancer Res 11(5) (2013), 443–455.

18.

Kim

N.S.

et al., Versican expression in tumor epithelial cells is correlated with a good prognosis in gastric cancer, Anticancer Res 34(10) (2014), 5613–5619.

19.

Ricciardelli

et al., The biological role and regulation of versican levels in cancer, Cancer Metastasis Rev 28(1–2) (2009), 233–245.

20.

and Yang

, MicroRNA-203 inhibits malignant melanoma cell migration by targeting versican, Exp Ther Med 8(1) (2014), 309–315.

21.

Qiu

et al., MicroRNA-613 inhibits cell growth, migration and invasion of papillary thyroid carcinoma by regulating SphK2, Oncotarget 7(26) (2016), 39907–39915.

22.

Minna

et al., miR-451a is underexpressed and targets AKT/ mTOR pathway in papillary thyroid carcinoma, Oncotarget 7(11) (2016), 12731–12747.

23.

Colamaio

et al., miR-142-3p down-regulation contributes to thyroid follicular tumorigenesis by targeting ASH1L and MLL1, J Clin Endocrinol Metab 100(1) (2015), E59–69.

MiR-135a-5p modulates biological functions of thyroid carcinoma cells via targeting VCAN 3 ′ -UTR