Long Non-Coding RNAs in Thyroid Cancer: Implications for Pathogenesis, Diagnosis, and Therapy

Two identifiable cell types are observed in the normal parenchymal histology of the thyroid gland. The first type comprises follicular cells that surround the colloid follicles and uptake iodine to synthesize thyroid hormones; anaplastic thyroid and well-differentiated tumors such as papillary and follicular can develop in these cells. The second type comprises parafollicular cells whose main function is to secrete calcitonin; parafollicular cells have attracted interest primarily as the origin of medullary thyroid cancer (MTC) [1]. Thyroid cancer (TC) is the 9th most common malignancy and the most common endocrine cancer [2]. The most common treatment for TC without metastasis is surgery. Chemotherapy and radioactive iodine treatment are also medical treatment options in addition to surgery [3]. However, the prognosis of recurrent TC is not good with the above treatments [4]. Histologic grade and age at onset are 2 major prognostic factors [5]. Ultrasound and fine-needle aspiration (FNA) biopsy of suspicious nodules are routinely employed to detect TC. Identification of novel biomarkers is necessary to improve the management of papillary thyroid cancer (PTC) and the decision-making regarding the therapeutic approach. Long non-coding RNAs (lncRNA) are an important part of the newly discovered non-coding RNAs (ncRNAs) [6]. Increasing evidence suggests that lncRNA play a key regulatory role in the genome and important functional roles in different fundamental biologic processes in transcription and translation. The challenge in TC diagnosis and treatment is to identify patients that have a high risk of developing radioiodine-refractory disease on the one hand and to prevent overtreatment on the other hand. While the clinical risk stratifications proposed by the American Thyroid Association are very helpful in guiding decision-making, it would be beneficial to have additional biomarkers such as potentially lncRNAs. The novel TC biomarkers give insight into how the disease begins and progresses while improving the success rate of early diagnosis and treatment. In TC, gene mutations, proteins, and messenger RNA (mRNA) markers such as BRAF, RAS, RET/PTC, and PAX8/PPARγ have been used to increase the diagnostic accuracy of patients with thyroid nodules [4]. Several microRNAs (miRNAs) are associated with PTC [7]. According to American Thyroid Association guidelines, the use of molecular biomarkers for TC is recommended for FNA samples with an unclear cytological diagnosis [8]. Several lncRNAs, such as NAMA, BANCR, and PTCSC3, have been identified as contributing factors to PTC [9, 10, 11] and may therefore be used as biomarker in this setting (table 1).

LncRNAs are small classes of transcribed molecules that range from 200 nucleotides to 100 kb in length [11]. LncRNAs account for about 80% of ncRNAs and play an important role in genomic organization. In cancerous tissues, there is a clear relationship between the expression and epigenetic state of these molecules. These molecules can interact with chromatin-modifying enzymes and histones, including DNA methyl transferase 3 and PRC2 [12]. LncRNAs can act as tumor suppressors or oncogenes in cancer [13]. Research has shown that these molecules play an important role in tumorigenicity, angiogenesis, proliferation, migration, apoptosis, and differentiation [14]. LncRNAs induce epithelial-mesenchymal transition (EMT) through the PI3K-AKT and Wnt/β-catenin pathways, thereby promoting metastasis [15]. A number of studies have shown that lncRNAs can interact with miRNAs by preventing the binding of miRNAs to mRNA targets [16]. LncRNAs have different expression levels at different stages of differentiation and can be used as biomarkers to identify the disease according to their specific expression in the tissue [17].

Nuclease-based genome editing technology is a powerful tool for clarifying the function of lncRNA in both in vitro and in vivo conditions [18]. The CRISPR-CAS9 system rapidly and efficiently removes relative or complete lncRNA [19]. On the other hand, inhibition of lncRNA expression has also been accomplished via inter-promoter polyadenylation and lncRNA sequences [20]. CRISPR-CAS9 also has the ability to increase lncRNA expression by introducing a strong promoter to the upstream gene or through targeting the transcriptional activation complex [21]. These systems play a significant role in increasing and reducing the expression of lncRNA in cancer, leading to new therapeutic approaches.

BRAF-activated non-protein coding RNA (BANCR) is a 693 bp lncRNA primarily identified using vastly parallel cDNA sequencing in melanoma cells. It is highly expressed in melanoma and increases cell migration [22]. Studies have shown that BANCR functions as both an oncogene and a tumor suppressor. Wang et al. [23]showed that BANCR was overexpressed in 6 pairs of PTC and matching normal tissues and that it increased PTC cell proliferation by stimulating autophagy. In the study by Liao et al. [10]BANCR expression was downregulated in PTC tissues compared with normal tissues. The significance of BANCR in TC is still unclear. BANCR is ordinarily activated by BRAF. The BRAF V600E mutation is very frequent in PTCs but the association with poorer prognosis is very controversial. However, studies have shown that BRAFV600E in combination with TERT promoter mutations is linked to poorer prognosis in PTC patients [24]. BRAF-activated BANCR has been shown to increase cell migration and proliferation in melanoma [25] and lung carcinoma [26] via the MAPK signaling pathway. It was believed that BANCR can progress EMT in PTC. In TC, there are 2 classical cell signaling pathways: ERK/MAPK and phosphoinositide 3-kinase/protein kinase B. A number of studies have shown that the V600E mutation of BRAF stimulates the MAPK signaling pathway [27]. Overexpression of BANCR might upregulate the expression of p-c-Raf, p-MEK1/2, and p-ERK1/2. Therefore, BANCR might activate the Raf/MEK/ERK signaling pathway. This effect was reversed by U0126 treatment [28]. Downregulated BANCR expression was correlated with tumor size, multifocal lesions, and an advanced pathologic stage. This change might stem from tumor heterogeneity. Low BANCR levels are associated with poor results in PTC patients. Liao et al. [10]indicated that overexpression of BANCR elevates apoptosis in TPC1 and K1 cell lines.

H19 is a typical lncRNA that is associated with various cancer types. H19 was originally thought to possess tumor suppressive properties based on its capacity to prevent tumorigenesis, but current studies have shown that it has both tumor promoter and suppressive functions [29, 30, 31]. There is a high level of expression of H19 in extraembryonic tissues, the embryo itself, and most fetal tissues, but postnatal expression of H19 is highly reduced [32]. In bladder cancer, H19 is a tumor recurrence marker and its expression is correlated with tumor grade, and in hepatocellular carcinoma, ectopic expression of H19 stimulates tumorigenesis, suggesting that H19 acts as an oncogene [33]. A recent study indicated that H19 expression was higher in tumor tissues and TC cell lines than in normal thyroid cells. In addition, it was stated that overexpression of H19 elevated proliferation, migration, and invasion of TC cells, whereas low expression of H19 reduced cell viability, migration, and invasion and caused stimulated growth arrest in vitro and in vivo [34].

Metastasis-Associated Lung Adenocarcinoma Transcript 1 (MALAT1) is an lncRNA marker of metastasis and poor prognosis in lung and other cancers [35]. Prior studies have shown that MALAT1-deficient lung cancer cells showed inferior motility and decreased tumor formation in inoculated mice [36]. Moreover, the pro-oncogenic possibility of MALAT1 was supported by its regulation of cell cycle-related transcription factors B-MYB and p53 [37]. Since its characterization in lung cancer, MALAT1 upregulation has also been described in several tumor types including follicular cell-derived thyroid tumors [38], although MALAT1 has remained unknown in MTC. MALAT1, an lncRNA that is involved in the regulation of cell cycle and migration, has been identified to be deregulated in malignancies of multiple organs, including lung, cervix, uterus, colon, breast, stomach, pancreas, kidney, bone, and bladder [39]. The recent study by Huang et al. [40] on thyroid carcinoma cell lines suggested that the follicular thyroid carcinoma (FTC) cell line FTC133 expressed higher levels of MALAT1 than the SW1736 anaplastic thyroid carcinoma (ATC) cell line. Studies have indicated that PTC had higher levels of MALAT1 than FTC and ATC [38]. MALAT1 expression was higher in MTC compared to normal thyroid, and its inhibition produced in vitro anti-oncogenic effects including decreased tumor cell proliferation and invasion [41]. It is worth noting that formerly, in murine mammary and lung cancer models, MALAT1 knockdown by genetic knockout or systemic administration of antisense oligonucleotides was shown to repress tumor growth and metastasis [42], validating the role of MALAT1 as a pro-oncogenic lncRNA in several cancer types. In addition, MALAT1 was demonstrated to promote EMT by epigenetically silencing E-cadherin expression [43]. MALAT1 was remarkably upregulated through transforming growth factor beta (TGFβ)-induced EMT in TC cells [38]. MALAT1 upregulation was found in the primary stage and could affect cell invasion [41].

Homeobox transcript antisense RNA (HOTAIR) is a 2,000 bp lncRNA that encodes antisense to the HOXC locus [44]. HOTAIR acts as an oncogene in tumorigenesis and stimulates invasion and metastases in tumor progression [45]. Dysregulation of HOTAIR is correlated with metastasis and poor prognosis in patients with TC. Through loss-of-function analysis, the biologic function of HOTAIR has been validated in human TC cell lines. The expression of HOTAIR is upregulated in human TC cells compared with normal human thyroid cells. Moreover, knockdown of HOTAIR significantly inhibited cell growth and invasion in TPC-1 and SW579 human TC [46].

Papillary thyroid carcinoma susceptibility candidate 3 (PTCSC3) is a recently recognized ncRNA, which is highly thyroid-specific. PTCSC3 as a tumor suppressor was considered a competing endogenous RNA for miR-574-5p [47]. PTCSC3 expression is strictly thyroid-specific and is downregulated in thyroid tumor tissues and thyroid cell lines.

Maternally expressed gene 3 (MEG3) is a lncRNA gene expressed in several normal tissues. Loss of MEG3 expression has been found in several types of human tumors. The regulation of MEG3 and its exact mechanisms of action in TC are unknown. MEG3 is downregulated in PTC; downregulated MEG3 was significantly associated with lymph node metastasis. MEG3 is an important tumor suppressor in TC which suppresses migration and invasion by targeting Rac1 [48].

Antisense non-coding RNA in the INK4 locus (ANRIL) has been proven to bind to chromobox 7 (CBX7) and to SUZ12, and is involved in transcriptional suppression through these interactions. Silencing ANRIL, as a key lncRNA, inhibits the invasion and metastasis of TPC1 cells. Moreover, ANRIL may prevent the expression of the TGFβ/Smad signaling pathway to decrease p15INK4B expression and stimulate invasion and metastasis of TC cells, which provides a novel perception of the practical role of lncRNA ANRIL-driven tumorigenesis [49]. LncRNAs (ENST00000426615 and ENST00000537266) may be significant regulators of PTC cell proliferation and motility, which could provide new insights into PTC pathogenesis [50]. Downregulation of lncRNA PANDAR suppresses cell growth/cell cycle and produces apoptosis in TC [51].

Many studies have shown that lncRNAs have wide biologic activity, especially in cell growth, differentiation, and tumorigenicity. LncRNAs may be a worthy biomarker for the diagnosis and prognosis of cancer. These molecules are expressed naturally in different cell types and can act as oncogenes or tumor suppressors, and play an important role in controlling cell growth through regulation of the cell cycle and apoptosis. Increases or decreases in their expression can lead to various types of cancer, such as TC. These molecules can be used as novel biomarkers in diagnosis or as prognostic factors for various types of cancer, without the need for invasive methods. However, to elucidate the regulatory role of these molecules, more research is needed in the future. TC is a disease that develops due to genetic and epigenetic abnormalities. Several classical disorders that are involved in tumorigenesis and progression of TC include changes in point mutations, copy number variations, and DNA methylation. In addition, many studies have shown that lncRNAs also play a role in regulating all types of biochemical activity in TC cells, including cell growth and survival. This study gives a summary of disorders affecting regulation and of recent developments regarding lncRNAs in TC. Given that lncRNAs have advantages such as noninvasiveness and easy access, they have a high potential to be utilized in the diagnosis and treatment evaluation of TC. Moreover, there are other aspects of lncRNAs in TC that will be investigated in the near future. For example, FNA biopsy and cytological analysis are a primary diagnostic measure that is widely used in the evaluation of thyroid nodules. However, at least 20% of biopsies show unclear cytological findings that cannot distinguish malignant from benign nodules, leading to poor management for these patients. Detection of cancer-related lncRNAs in FNA biopsies can be a useful strategy for the detection of malignant thyroid tumors. The study of lncRNAs in TC is becoming an attractive and useful field. The role and mechanism of lncRNAs in TC should be investigated in the future, and some lncRNAs may serve as biomarkers for diagnosis, prognosis, and therapeutic purposes.

The authors declare that they have no competing interests.

This article does not contain any studies with human participants or animals performed by any of the authors.