Overexpression of miR-100-5p inhibits papillary thyroid cancer progression via targeting FZD8

Abstract Papillary thyroid cancer (PTC) is the most prevalent type of TC worldwide; however, its pathological process remains unclear at the molecular level. In the current study, we analyzed the microarray data of PTC tissues and non-neoplastic thyroid tissues, and confirmed miR-100-5p as a downregulated miRNA in PTC. Via bioinformatic approach, western blotting, and TOP/FOP-flash assay, miR-100-5p was observed to be involved in the inactivation of Wnt/β-catenin signaling in TPC-1 and KTC-1. Frizzled Class Receptor 8 (FZD8), the coupled receptor for canonical Wnt/β-catenin signaling, was verified to be targeted and inhibited by miR-100-5p in TPC-1 and KTC-1. In the function assay, miR-100-5p mimic repressed PTC cell proliferation and induced cell apoptosis of TPC-1 and KTC-1; meanwhile, transfection of full-length FZD8 attenuated the effect of miR-100-5p mimic. Moreover, in the collected samples, miR-100-5p was lowly expressed in PTC tissues compared with normal tissues, especially in those of advanced stage (Stage III/IV vs Stage I/II), while FZD8 was highly expressed in PTC tissues, which in PTC tissues was inversely correlated to miR-100-5p. Thus, we suggest that overexpression of miR-100-5p inhibits the development of PTC by targeting FZD8.


Introduction
Thyroid cancer (TC) was one of the top ten cancer types for estimated new cases globally for both sex in 2020 [1].
Statistics indicate that TC is relatively more prevalent in Eastern Asia, North America, and Micronesia/Polynesia compared to other regions in the world [1]. Most TC (85%) are originated from thyroid follicular cells, and TC can be further divided into papillary thyroid cancer (PTC, 75%), follicular thyroid cancer (FTC, 15%), and Hürthle cell cancer (HCC, 10%) [2]. Generally, the 5-year survival rate of patients with PTC is high when diagnosed early [3]; however, the recurrence happens to a small portion of them and some of them are under the threat of death upon cancer cell metastasis [4]. Surgery is the most commonly used treatment for patients with PTC. Owing to decades of basic research, several novel targets have been discovered for PTC and data from clinical research suggest their effectiveness in the treatment of PTC [5,6]. It is urgent to further explore the molecular mechanism of PTC.
In the current study, miR-100-5p was screened out as a downregulated miRNA in PTC. Our assays further showed that, miR-100-5p inhibited cell proliferation and induced cell apoptosis of PTC cells by targeting Frizzled Class Receptor 8 (FZD8) and followed by inactivation of Wnt/β-catenin signaling in PTC.

CCK8 and flow cytometry assay
Cell proliferation was measured by a Cell Counting Kit (CCK-8, Dojindo, Tokyo, Japan). TPC-1 and KTC-1 (5 × 10 3 ) were seeded in wells in 96-well plates. On day 1, 2, and 3 following plating, 10 μL of CCK-8 solution was mixed with the fresh medium and replaced the medium in the wells. After 4 h, the medium was scanned by a Microplate Reader and OD 450.
Flow cytometry was utilized to measure the percent of apoptotic TPC-1 and KTC-1 cells. Cells were stained by Annexin V-FITC and PI (Annexin V-FITC cell apoptosis detection kit, Beyotime, Shanghai, China), and analyzed by the flow cytometry (FACSCanto II, BD Biosciences, Franklin Lakes, NJ, USA). The data were analyzed by FlowJo. The cells were divided into live, early apoptosis, late apoptosis, and necrosis groups.

RT-PCR
Total RNA from cells and tissue samples was extracted by TRIzol (Invitrogen). Nanodrop 2000 was used to determine the RNA concentration and quality. PrimeScript ® miRNA cDNA Synthesis kit (TaKaRa, Tokyo, Japan) and PrimeScript ® RT reagent kit (TaKaRa) were utilized to conduct reverse transcription of miRNA and general genes, respectively. TB Green ® Fast qPCR Mix was used to perform RT-PCR. The condition was: Step 1, 95°C (30 s); Step 2, 95°C (5 s), 60°C (10 s), 35 cycles. U6 and β-actin served as references for miRNA and mRNA, respectively. 2 −ΔΔCt was used for data analysis.

Dual luciferase reporter assay
The cells were transfected with pmirGLO in combination with RNA oligonucleotide. At 48 h following transfection, cells were collected and subjected to lysate with reagent in Dual Luciferase Reporter Assay System kit (Promega Corp.). Luciferase activity was measured on a GloMax luciferase detector (Promega Corp.).

TOPFlash/FOPFlash assay
Cells were transfected with TOPFlash and FOPFlash vector in combination with RNA oligonucleotide. On day 2 following transfection, they were treated with reagent from Dual Luciferase Reporter Assay System kit (Promega Corp.) and measured on the GloMax luciferase detector (Promega Corp.).

Statistical analyses
The data were analyzed by GraphPad Prism 6.0 and expressed as mean ± SD. Student's t-test and one-way ANOVA were performed in two and three groups, respectively. Student-Newman-Keuls (S-N-K) method was selected as post-analysis for one-way ANOVA. p < 0.05 was statistical significance. Each experiment was repeated three independent times.

miR-100-5p inactivated Wnt/β-catenin signaling in PTC cells
To study the downstream molecules of miR-100-5p, we used TargetScan7.2 to search them (Table A1). By KEGG pathway analysis, these targets were mainly involved in several signaling pathways including Wnt/β-catenin pathway, MAPK pathway, and Hippo pathway (Figure 2a). In comparison with MAPK pathway and Hippo pathway, there were relatively much more genes enriched in Wnt/β-catenin pathway, which was selected as the research subject in the current study. More importantly, Wnt/β-catenin signaling remains one of the most well-characterized oncogenic pathways in PTC [14][15][16].
Among the 30 targets, FZD8 which is the receptor for Wnt proteins [18] attracted our attention. Upon binding between Wnt and FZD8, FZD8 can activate β-catenin dependent/independent signals [17]. However, the relationship between miR-100-5p and FZD8 in PTC has not been studied yet.

miR-100-5p was negatively correlated with FZD8 in PTC tissues
In the collected clinical samples in the current study, RT-PCR displayed a significant decrease of miR-100-5p in PTC tissues compared to the matched non-tumor tissues (Figure 5a). In addition, lower miR-100-5p levels were found from tumors of advanced stage compared with those of early stage (Figure 5b). On the contrary, FZD8 mRNA levels were elevated in PTC tissues compared to the matched non-tumor tissues (Figure 5c). The Pearson correlation analysis indicated that miR-100-5p was inversely (r = −0.414) correlated to FZD8 in these PTC tissues (Figure 5d).

Discussion
Two decades of studies have revealed multiple oncogenic pathways in PTC; however, the potential roles of miRNAs in regulating these pathways and their contribution to PTC remains poorly defined. The pro-cancer and anticancer functions of miR-100-5p have been shown from cancers of numerous origins, for example, miR-100-5p is lowly expressed in prostate cancer, which impairs cancer cell proliferation, migration, and invasion via downregulating mTOR [19]; in contrast, overexpression of miR-100-5p is observed in ovary cancer, which promotes cell invasion via directly targeting SMARCD1 [20]. In the present study, by analyzing previously published microarray data, miR-100-5p was identified as one of the most significantly downregulated miRNAs in PTC. The experiments confirmed Figure 2: Wnt/β-catenin signaling was repressed by miR-100-5p in PTC: (a) KEGG pathway analysis was utilized for analyzing targets of miR-100-5p predicted by TargetScan7.2, (b) RT-PCR presented that miR-100-5p mimic elevated miR-100-5p in TPC-1 and KTC-1, (c and d) the TOPFlash/FOPFlash assay presented that miR-100-5p mimic reduced the activity of Wnt/β-catenin pathway in TPC-1 and KTC-1 in a dosedependent manner, (e) western blotting revealed that miR-100-5p decreased Myc and Cyclin D1 protein levels in TPC-1 and KTC-1, and (f) RT-PCR revealed that miR-100-5p decreased Myc and Cyclin D1 mRNA levels in TPC-1 and KTC-1. **, p < 0.01 and ***, p < 0.001. Figure 3: FZD8 was a target for miR-100-5p: (a) the targets of miR-100-5p were predicted using PITA, TargetScan7.2, and miRanda. The overlapping of the targets is presented in a vein map, (b) RT-PCR showed that FZD8 mRNA levels were increased in TPC-1 and KTC-1 compared to Nthy-ori3-1, (c) RT-PCR presented that miR-100-5p mimic decreased FZD8 mRNA levels in TPC-1 and KTC-1, (d) western blotting revealed that miR-100-5p mimic decreased FZD8 protein levels in TPC-1 and KTC-1, (e) the conserved binding sites for miR-100-5p on FZD8 mRNA are presented, (f) FZD8 3′UTR and its mutant form was inserted into pmirGLO plasmid, respectively, and (g and h) luciferase assay revealed that miR-100-5p reduced luciferase activity of FZD8 3′UTR but not FZD8 3′UTR Mutant in TPC-1 and KTC-1. **, p < 0.01 and ***, p < 0.001. that miR-100-5p was decreased in PTC cells and PTC tissues. miR-100-5p mimic repressed cell proliferation and induced cell apoptosis in PTC cells. Thus, the data implied that miR-100-5p functioned as a tumor suppressor in PTC for the first time. However, the potential molecules responsible for the function of miR-100-5p remain undiscovered. By KEGG pathway analysis and TOPFlash/FOPFlash assay, miR-100-5p was discovered to inactivate Wnt/βcatenin signaling in PTC cells. Regarding Wnt/β-catenin signaling, its overactivation results in PTC cell proliferation, resistance to cell apoptosis, cell invasion ,and drug resistance [14][15][16]. However, the potential molecules which link miR-100-5p and Wnt/β-catenin signaling remain undiscovered.
Studies have revealed that many key positive and negative modulators for Wnt/β-catenin signaling are targets of miRNAs, and dysregulation of miRNAs is one of the most crucial mechanisms for Wnt/β-catenin signaling activation, e.g., Bai et al. report that miR-150 targets RAB11A to inactivate Wnt/β-catenin signaling in PTC [21]. Li et al. demonstrate that upregulation of miR-320a elevates ANRIL to inactivate Wnt/β-catenin signaling in PTC [22]. Accordingly, current study aimed to explore the potential mRNA targets for miR-100-5p.
Among the 30 targets of miR-100-5p predicted by PITA, TargetScan7.2, and miRanda, FZD8, which can activate β-catenin dependent/independent signals by serving as receptor for Wnt proteins [17,18], is selected as the research subject. Afterwards, overexpression of FZD8 was observed in TPC-1 and KTC-1 compared to Nthy-ori3-1, indicating the oncogenic role of FZD8 in PTC, which was in consistent with the function of FZD8 in other cancer types, including prostate cancer [23], non-small cell lung cancer [24], and head and neck squamous carcinomas [25]. Moreover, Chen et al. reported that FZD8 was regulated by circRNA_NEK6/ miR-370-3p, and contributed to the progression of TC [26]. Very recently, Mao et al. reported that FZD8 was regulated by circRPS28 (hsa_circ_0049055)/miR-345-5p, contributed to promotion of cell growth and blockage of cell apoptosis of PTC [27].
Thereafter, FZD8 was shown to harbor conserved binding sites for miR-100-5p, and miR-100-5p acted as a novel regulator of FZD8. Additionally, in PTC cells, the biological function of miR-100-5p mimic was attenuated by overexpression of FZD8. Moreover, in the PTC samples, the negative correlation between miR-100-5p and FZD8 was discovered. The data added novel insights for understanding the complex signaling transduction in PTC.
Collectively, data from clinical samples and cellbased assays indicated that miR-100-5p was functionally significant for PTC. Targeting miR-100-5p may be a promising treatment for patients with PTC. Figure 5: miR-100-5p was inversely associated to FZD8 in PTC: (a) RT-PCR showed that miR-100-5p was lower in PTC tissues than non-tumors tissues from 50 patients with PTC, (b) miR-100-5p levels were relatively lower in tumors at Stage III/IV than those at Stage I/II, (c) RT-PCR showed that FZD8 mRNA expression was higher in PTC tissues than non-tumors tissues from 50 patients with PTC, and (d) the Pearson correlation analysis indicated an inverse correlation between miR-100-5p and FZD8 in PTC tissues. ***, p < 0.001.
However, there was a limitation in the current study, among the three signaling pathways, the present study only studied the relation between miR-100-5p and Wnt/βcatenin signaling, but not MAPK or Hippo, which will be investigated in our future work.
Funding information: Not received.

Conflict of interest:
The authors declare that there are not any types of conflict of interest in the current study.
Data availability statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.