Effect of valproic acid on miRNAs affecting histone deacetylase in a model of anaplastic thyroid cancer

Thyroid cancer is the most common malignant tumor of the endocrine system seen in the thyroid gland. More than 90% of thyroid cancers comprise papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC). Although anaplastic thyroid carcinoma (ATC) accounts for less than 2% of thyroid cancer. But patients’ lifespan after diagnosis is about 6 months. Surgical interventions, radioactive iodine use, and chemotherapy are not sufficient in the treatment of ATC, so alternative therapies are needed. The WST-1 assay test was performed to evaluate the anti-proliferative effects of Valproic acid (VPA). Also, the effect of VPA on miRNAs affecting histone deacetylase was determined by Quantitative RT-PCR. In the SW1736 cell line, IC50 dose for VPA was found 1.6 mg/ml. In our study, the level of oncogenic genes expression in cells treated with VPA, including miR-184, miR-222-5p, miR-124-3p, and miR-328-3p, decreased. Also, the expression of tumor inhibitory genes including miR-323-5p, miR-182-5p, miR-138-5p, miR-217, miR-15a-5p, miR-29b-3p, miR-324-5p and miR-101-5p increased significantly. VPA can ad-just countless gene expression patterns, including microRNAs (miRNAs), by targeting histone deacetylase (HDAC). However, further studies are required for more accurate results.


Introduction
Thyroid cancer is the most common endocrine malignancy. Different types of cancer include papillary cancer (PTC), follicular cancer (FTC), medullary cancer (MTC), and anaplastic cancer (ATC), of which papillary cancer is the more common and the most dangerous type of anaplastic [1]. ATC has a poor prognosis because of the lack of early detection, its highly aggressive nature, as well as its failure with surgical interventions, radioactive iodine use, and chemotherapy [2,3]. So that the survival rate is less than 25 weeks. Therefore, success in therapeutic approaches for thyroid cancer is crucial for patient survival [4].
MicroRNAs (miRNAs) are a group of regulators of gene expression. Which are endogenously synthesized and can affect the expression of genes. These molecules are singlestranded and noncoding RNAs so that in functional mode the length of these miRNAs does not exceed approximately 23 nucleotides [5]. Important point is that a miRNA can target multiple mRNAs and thereby show a regulatory effect on different genes. Besides, studies have shown that one mRNA is regulated by several miRNAs [6]. MiRNA expression differs in tissues. This difference in expression is also seen in a particularly healthy cell and its cancer cell, which may indicate the importance of miRNAs involved in the pathogenesis process [7]. Some miRNAs act as tumor suppressors or oncogenes by targeting cancer-related genes. They also play important roles in apoptosis, proliferation, migration, and cell invasion [8,9]. MiRNAs decrease the expression of the target gene by reducing mRNA stability and/or preventing its translation [6,10]. By reducing the abundance of specific proteins in this way, miRNAs exert management effects on many physiological processes. It is not surprising that their dysregulated expression may affect neoplastic disease [11,12].

3
Cancer is associated with many genetic and epigenetic changes. Genetic changes are seen in the number and order of nucleotides of DNA, while epigenetic changes occur in the structural proteins of DNA (histones). Epigenetic modifications can alter the transcription and translation of target genes and increase or decrease their functional level [13]. Histone deacetylase (HDACs) enzymes remove acetyl groups from histone lysine units and reduce gene expression by altering DNA density [14]. These histone modifications can regulate the expression status of cancer-related genes [15]. The use of HDAC inhibitors induces overexpression of Let seven family miRNAs. This increase in expression by affecting HMGA2 leads to cell cycle blockage and cell death in cancer cells [16]. Therefore, HDACs inhibitors can be used as a treatment for cancer [17]. Valproic acid (VPA) is an anticonvulsant drug and inhibits HDAC [18]. It has been suggested that VPA binds to the catalytic site, inhibiting class I and II a HDACs [19]. Because it is well tolerated and induces differentiation and apoptosis of carcinoma cells, it has recently been tested in clinical trials as a potential chemotherapeutic agent for a variety of cancers [20,21]. VPA as an HDAC inhibitor changes the expression profile of miRNA in the ATC model. However, further studies are required for more accurate results.

Cell line and culture conditions
In our study, the effect of VPA as an HDAC inhibitor on the SW1736 cell line was evaluated by cell culture assays. These cells were commercially prepared as cancer cells (ATCC). For cell lines, RPMI 1640 (Biological industries) medium with l-Glutamine supported by %10 fetal bovine serum (FBS) (Biological industries) containing 2 mM l-Glutamine (Biological industries), 1% penicillin-streptomycin (Biological Industries) were used; incubated at 37 °C, 95% of humidity and 5% CO2 until adequate cell density obtained, and follow-ups were carried out using "inverted" microscope.

Cytotoxicity calculation
Individual effects of the active agents on cells were specified using Colorimetric WST-1 Assay Kits. This procedure lasted for 3 days and three replications were performed for the control group and different treated doses. For cytotoxicity experiments, (10 × 10 4 ) cell/ml was used for each well. Model cells were plated into 96 well plates; after the 24, 48, and 72-h (h) incubations following the addition of active agent solutions of appropriate concentrations, 10 μl/ well WST-1 solution was added and incubated for 1 h.
Quantitative measurement of the formazan dye produced after the incubation was performed on a microplate reader (Multiskan FC, Thermo) every 15 min at 450 nm absorbance 620 nm reference range. IC 50 doses which are defined as the inhibitor concentrations causing a 50% decrease in cell proliferation were determined for VPA on SW1736 cell lines.

Determination of gene expression changes
The expression changes produced by the IC 50 doses of active agents were determined with real-time PCR in 48 h. For expression experiments, (5 × 10 5 ) cell/ml were plated into 6-well plates the specified dose was applied with dilutions suitable for the cells. The Active agent was not applied to the cells that were accepted as the control group. MiRNA isolation was obtained from SW1736 cell line with miRNeasy mini Kit (Qiagen) and cDNA synthesis was performed using RT 2 First Strand Kit (miScript II RT kit). The expression changes of miRNAs were evaluated using the MIHS-108ZF plate (Qiagen).

Statistical analysis
The statistical analysis of the cytotoxic effect of VPA and change of gene expression in SW1736 cells was performed by using Graph pad Prism v8.0.2 and xCELLi-gence® RTCA (Roche) software, respectively. The ΔΔCT method was used for the miRNA fold change calculations.

Evaluation of the effects of valproic acid on cell viability on the SW1736 cell lines via WST-1
To specify the VPA IC 50 value, VPA dilutions of (10 -3 -10 mg/ml) were evaluated with WST-1 test in the 24 h, 48 h, and 72 h for SW1736 cell lines. VPA IC 50 values on SW1736 cell lines were found as 1.6 mg/ml in the 48 h (Fig. 1).

Discussion
The number of patients with ATC makes up a small fraction of thyroid cancer, but the survival rate is often less than 1 year. Therefore, appropriate treatments are desperately needed. Mutations in BRAF are commonly seen in thyroid carcinoma [22]. Understanding the role of epigenetic changes in thyroid cancer can provide valuable opportunities to identify molecular mechanisms to provide novel therapeutic approaches. Epigenetic changes can affect miRNA expression [23]. MiRNAs can have regulatory effects on tumorigenesis, cancer cell proliferation, and other physiological processes [24]. They can act as tumor suppressors or oncogenes in different types of cancer because they can target different genes [8,9]. Based on the findings, we examined the expression of mRNAs in the SW1736 cell line using microarrays.
In our study, increased expression of miR-217 in VPAtreated SW1736 cells showed that it could increase apoptosis of cells in anaplastic thyroid cancer. VPA increased the expression level of miR-217 by inhibiting HDAC. In a study by Zhu L and colleagues, it was found that MAPK1 is a target of miR-217. MiR-217 inhibits MAPK1 and reduces cell viability, migration, and invasion while enhancing cell apoptosis. They have introduced miR-217 as a tumor suppressor gene in cervical cancer [25]. Another study found that miR-217 expression was different in normal pancreatic cells compared with pancreatic cancer cells. And the increased expression of miR-217 inhibits cell growth and invasion of cancer cells, in contrast to the effect on E2F3 that induces apoptosis [26]. Other studies have shown that miR-217 binds to its target mRNA and acts as a tumor suppressor gene in gastric and liver cancers [27,28]. Studies have shown that miR-138 is less expression in thyroid cancer than in normal thyroid tissue [29]. Low expression of miR-138 is associated with increased cell migration. MiR-138-5p has been shown to directly target Vim (vimentin) in cancer cells. These changes allow cancer cells to migrate [30]. In our study, expression of this miRNA increased after treatment with VPA, indicating a positive effect of VPA on our studied cell line. And introduced miR-138 as a tumor suppressor gene.
The expression level of miR-15a in PTC tissue is lower than in normal tissue. Overexpression of miR-15a prevents the proliferation and invasion of PTC cells and induces apoptosis. Overexpression of miR-15a by targeting RET and AKT inhibits tumor growth [31]. In another study, increased expression of miR-15a-5p inhibited proliferation of HepG2 or SNU-182 cancer cells in vitro [32]. In our study, the expression of miR-15a-5p was also increased after the treatment of the studied cell line.
The results of our study showed that miR-323a-5p expression was induced by VPA on the SW1736 cell line which induced apoptosis of cells. And showed that VPA has antitumor activity. In other words, VPA by inhibiting HDAC could increase 323-miR expression. This increase in expression suppresses the tumor. One group showed that miR-323a-5p has a tumor suppressor function in neuroblastoma cells. And increased levels of miR-323a-5p decrease the proliferation of neuroblastoma cells [33]. A study by Men Y and colleagues found that miR-323a-3p expression was lower in esophageal cancer cell lines than in healthy esophageal epithelial cells and inhibits cell proliferation. Besides, miR-323a-3p prevented cell migration and invasion in some of these cell lines. These results further revealed the tumor suppressor status for miR-323a-3p [34].
In contrast, some miRNAs act as oncogenes. The expression of these miRNAs in cancer cells is higher than in healthy cells. These miRNAs induce cell proliferation and suppress apoptosis in the cell, leading the cells to become cancerous. MiR-184 has an oncogenic role in SCC squamous cell carcinoma. In a study by Wong TS and colleagues, plasma miR-184 levels were significantly higher in patients with SCC compared to normal subjects. They induced apoptosis of these cells by inhibiting miR-184 in cancer cells of the tongue [35]. Another study showed that miRNA-184 levels were higher in the EGFRm-BM group than in the control group [36]. A team of researchers showed that miR-184 enhances the biological behavior of cancer by targeting FIH-1 in the human glioma cell line A172 [36]. In microarray data analysis we found that expression of this miRNA in our SW1736 cell line was higher than in the control group. Expression of miR-184 decreased after treatment with VPA.
The cell cycle is controlled by a variety of factors. One of these factors is p27, which suppresses the cell cycle.
These factors involved in the cell cycle are regulated by miRNAs. The observed miR-222 targeting p27 enhances cell proliferation. And acts as an oncogene [37,38]. Some studies confirmed the effects of increased miR-222 expression on cancer cell metastasis [39,40]. Inhibition of miR-222 results in PTEN rearrangement. As a result, cell proliferation and invasion are significantly suppressed and apoptosis is increased [41]. In our study, this miRNA was also increased in the studied cell lines and was identified as an oncogene in line with previous studies.
The level of miR-328-3p in ovarian cancer stem cells (CSC) is very high and may play an important role in the metastasis of these cells. Inhibition of miR-328 in rat ovarian xenografts inhibited tumor growth and prevented tumor metastasis [42]. In our study, a decrease in miR-328 was observed with Valproic acid. Our data showed that our drug reduced the expression of this miRNA and had antitumor effects.
One study found that treatment with imatinib alone was not able to grow anaplastic cells in thyroid cancer cells, but that combination therapy (imatinib and VPA) had a cytostatic effect. This effect is mediated by stunted G1 growth and acts through p21 expression and impaired AKT phosphorylation [43]. So combination therapies can be very helpful.

Conclusion
Many miRNAs have shown different expression levels in response to treatment with HDAC inhibitors. Some of them can enhance or decrease the antitumor response. However, further studies are needed to identify the precise mechanisms of miRNAs for antitumor activity. It is difficult to accurately identify clinical miRNAs because of their low stability, which is why it needs accurate verification. We suggest that the effect of VPA on miRNA and its association with the cell cycle, apoptosis, be considered in future studies. However, VPA is well tolerated and can induce tumor suppressor miRNAs such as mir-323a, mir-15-a, mir-318, mir-217 by inhibiting HDAC in ATC cells. Data availability All data obtained in the study can be accessed if desired.