miR-154-3p and miR-487-3p synergistically modulate RHOA signaling in the carcinogenesis of thyroid cancer

Abstract Background: miRs family members are often thought to have extensively overlapping targets and synergistically to modulate target gene expression via post-transcriptional repression. The present study was to determine whether miR-154-3p and miR-487-3p synergistically collaborated to regulate RHOA signaling in the carcinogenesis of thyroid cancer. Materials and methods: Candidate miRs were filtrated using miR microarray assays. Gene and protein expression levels were analyzed using RT-qPCR and Western blotting, respectively. Cell growth was evaluated using CCK8 assays and nude-mouse transplanted tumor experiments. Cell apoptosis was detected using Annexin V-FITC double staining. Results: miR-154-3p and miR-487-3p were significantly decreased in 63 thyroid cancer tissues and cell lines compared with those in paired non-tumor tissues and normal thyroid follicular epithelial cells. Low expression levels of miR-154-3p and miR-487-3p significantly correlated with tumor size, TNM stage, histological grade, lymph node metastasis and shorter overall survival in patients with thyroid cancer. Furthermore, the protein expression of RHOA was significantly inversely correlated with miR-154-3p (r = −0.404; P = 0.001) and miR-487-3p (r = −0.456; P < 0.001) expression in thyroid cancer tissues. We experimentally validated that miR-154-3p and miR-487-3p synergistically blocked thyroid cancer cell growth in vitro and in vivo. However, the anti-proliferative and pro-apoptotic activities of miR-154-3p/487-3p were neutralized by RHOA overexpressed vectors. Conclusions: Our present findings expounded a novel signal cascade employing miR-154-3p/487-3p and RHOA to fine-tune thyroid cancer cell proliferation and apoptosis. We corroborated that suppression of RHOA by miR-154-3p/487-3p may be a valuable therapeutic target for impeding thyroid cancer progression.


Introduction
Thyroid cancer as a common endocrine cancer ranks ninth for incidence worldwide [1]. Global cancer statistics in 2018 estimates that approximately 567,000 cases are diagnosed with thyroid cancer [1]. Among these cancer patients, about 15.9% cases originate from China, and the incidence rate in women is three times higher than in men [2,3]. Recently, the incidence of thyroid cancer is ascending, which may be attributed to the improvement of the early diagnosis of thyroid cancer [4]. Although the therapeutic strategies, including thyroidectomy, radioiodine therapy and thyroid-stimulating hormone inhibition therapy, have greatly improved the survival quality of early thyroid cancer patients, the prognosis of advanced and metastatic patients is dissatisfactory [5]. Therefore, it is very meaningful to investigate the underlying molecular mechanism for providing the meritorious therapeutic protocol of thyroid cancer patients. expressed miRs were selected out according to |Log2fold change| ≥ 1, P < 0.05 and false discovery rate < 0.05. The hierarchical clustering analysis was performed using MeV software (version 4.2.6).

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
RT-qPCR for miRs: total RNA was extracted using miRNeasy Mini Kit (Qiagen, Inc., Valencia, CA, U.S.A.), according to the manufacturer's protocol. TaqMan ® RT kit and TaqMan ® MicroRNA assay (Applied Biosystems) were used to detect miRs expression levels using Applied Biosystems 7300 Real-Time PCR System (Thermo Fisher Scientific, Inc.). miRs expression levels were calculated using 2 − C t method, as described previously [16], and U6 was used as an internal control.
RT-qPCR for mRNA: Moloney murine leukemia virus reverse transcriptase (Invitrogen) was used to synthesize cDNA with 2 μg of total RNA according to the manufacturer's protocol. Real-Time PCR was performed using Applied Biosystems 7300 System with the TaqMan Universal PCR Master Mix (Thermo Fisher Scientific). The relative expression levels of RHOA were calculated using the 2 − C t method [16], and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was served as the internal control. The primers were used as follows: RHOA: Forward 5 -AGCCTGTGGAAAGACATGCTT-3 and Reverse 5 -TCAAACACTGTGGGCACATAC-3 ; GAPDH: forward 5 -GCACCGTCAAGCTGAGAAC-3 and reverse 5 -TGGTGAAGACGCCAGTGGA-3 .

Immunohistochemical (IHC) staining
The paraffin-embedded tumor tissues and adjacent non-tumor tissues were cut into 3-μm sections and mounted on glass slides for staining with immunoperoxidase, and the procedures of immunohistochemical staining of RHOA (cat. no: ab86297; dilution: 1: 100; Abcam, Cambridge, U.K.) and ROCK1 (cat. no: ab45171; dilution: 1: 100) were performed as described previously [18]. The pictures were visual under a microscope (Leica DM 2500; Leica Microsystems GmbH, Wetzlar, Germany). Image Pro-Plus 6 software (Media Cybernetics, Inc., Rockville, MD, U.S.A.) was used for the analysis of the integrated optical density of positive-RHOA and -ROCK1.

Tumor formation assay in vivo
Human thyroid cancer K-1 cells stably transfected with pre-miR-Con, pre-miR-154-3p, pre-miR-487-3p or pre-miR-154-3p/487-3p, and K-1 were suspended in phosphate-buffered saline (PBS) and injected subcutaneously into the same side armpit of each nude mouse (1 × 10 7 cells per 0.1 ml). Four-week-old male BALB/c nude mice were purchased from (n = 24, Beijing HFK Bio-Technology. co., LTD., Beijing, China). Animal experiment was performed in the experimental center of the Gansu Provincial Cancer Hospital & Gansu Provincial Academic Institute for Medical Research, Lanzhou, China. Tumor weight was measured when mice were killed on week 4 by intraperitoneal injection of sodium pentobarbital (2%; 200 mg/kg; cat. no. P3761; Sigma-Aldrich; Merck Millipore, Germany). All protocols were approved by the Animal Care and Research Committee of the Gansu Provincial Cancer Hospital & Gansu Provincial Academic Institute for Medical Research, Lanzhou, China (Approval number: 2018030155).

Statistical analysis
Data were presented as mean + − SD. Statistical analysis was performed using GraphPad Prism Version 7.0 (GraphPad Software, Inc., La Jolla, CA, U.S.A.). Chi-Square (χ 2 ) tests were used to evaluate differences between the clinical characteristics and miRs expression. Student t-test was used to analyze two-group differences. Inter-group differences were analyzed by one-way analysis of variance, followed by Tukey's post hoc analysis. Survival analysis was performed using the Kaplan-Meier method with the log-rank test applied for comparison. Spearman's rank analysis was used to identify the correlation between the expression levels of RHOA and miR-154-3p or miR-487-3p in thyroid cancer tissues. P < 0.05 was considered to indicate a statistically significant difference.

miR-154-3p and miR-487-3p are down-regulated in thyroid cancer tissues
Based on |Log 2 fold change| ≥ 1, P < 0.001 and FDR ≤ 0.001, 220 differentially expressed miRs, including 98 down-regulated and 122 up-regulated miRs, were observed in three thyroid cancer tissues compared with corresponding non-tumor specimens using miR microarray ( Figure 1A). The top 2 differentially expressed miRs were miR-154-3p and miR-487-3p with the fold change (tumor/adjacent) −5.02 and −4.91, respectively. RT-qPCR assays were performed to validate the expression levels of miR-154-3p and miR-487-3p in 63 pairs of thyroid cancer tissues and adjacent non-tumor tissues, and the results corroborated that the expression levels of miR-154-3p and miR-487-3p were significantly lower in thyroid cancer tissues than those of in the adjacent non-tumor tissues ( Figure  1B). Furthermore, the expression levels of miR-154-3p and miR-487-3p in thyroid cancer cell lines were consistent with the results from thyroid cancer tissues ( Figure 1C).

miR-154-3p and miR-487-3p are associated with poor prognosis in thyroid cancer patients
Clinicopathological data indicated that low expression of miR-154-3p or miR-487-3p was significantly correlated with bigger tumor size, poor TNM stage and histological grade, and lymph node metastasis (Table 1). Interestingly, both miR-154-3p and miR-487-3p with low expression in patients with thyroid cancer had a shorter overall survival ( Figure  2A,B). Gender and age had no significant correlation with overall survival in patients with thyroid cancer ( Figure  2C,D). Bigger tumor size, poor TNM stage and histological grade, and lymph node metastasis were significantly associated with poor prognosis in patients with thyroid cancer ( Figure 2E-H).

RHOA and ROCK1 are up-regulated in thyroid cancer tissues
IHC staining suggested that a significant increase RHOA protein expression was observed in thyroid cancer compared with adjacent non-tumor tissues, and the protein expression of RHOA was up-regulated in 55 of 63 (87.3%) thyroid cancer tissues ( Figure 3A,B). In addition, the result of correlation analysis showed that the protein expression of RHOA was significantly inversely correlated with miR-154-3p (r = −0.404; P = 0.001) and miR-487-3p (r = −0.456; P < 0.001) expression in thyroid cancer tissues ( Figure 3C). IHC staining results also exhibited that ROCK1 protein expression was significantly elevated in thyroid cancer tissues compared with adjacent non-tumor tissues ( Figure  3D). Correlation analysis revealed that the protein expression of ROCK1 was significantly negatively correlated with miR-154-3p (r = −0.350; P = 0.005) and miR-487-3p (r = −0.338; P = 0.007) expression in thyroid cancer tissues ( Figure 3E).

RHOA is a direct target of miR-154-3p and miR-487-3p
Using on-line bioinformatics algorithm (TargetScan, http://www.targetscan.org/), miR-154-3p and miR-487-3p were identified as candidate miRNAs targeting to RHOA. The binding sites between miR-154-3p/-487-3p and RHOA were highly conserved across many species ( Figure 4A). As shown in Figure 4B, both miR-154-3p and miR-487-3p could bind with the 3 -UTR of RHOA and share sequence homology. The association between miR-154-3p/487-3p   . Spearman's rank analysis was used to identify the correlation between the expression levels of RHOA and miR-154-3p or miR-487-3p in thyroid cancer tissues (C). IHC staining is used to evaluate ROCK1 protein expression in thyroid cancer tissues and adjacent non-tumor tissues (D). Spearman's rank analysis was used to identify the correlation between the expression levels of ROCK1 and miR-154-3p or miR-487-3p in thyroid cancer tissues (E); * P < 0.05.

Discussion
Recently, RHOA is frequently reported as an oncogenic gene implicating in the initiation and progression of malignant tumors via exacerbating cell migration and invasion [6,8,19,20]. RHOA is widespreadly overexpressed in prostate cancer, cervical cancer and colorectal cancer, and associated with cancer metastasis [21][22][23]. Thus, we speculate that RHOA may serve as a particularly feasible molecular target for the treatment of malignant tumors. However, very little is known about the underlying molecular mechanism and the expression of RHOA in thyroid cancer. Thus, we aimed to investigate the malignant properties of RHOA in thyroid cancer, which might provide a novel therapeutic target in thyroid cancer.
In the present study, we found that RHOA protein expression was significantly elevated in 63 thyroid cancer tissues compared with adjacent non-tumor tissues. Bioinformatics algorithm revealed that RHOA is a direct target of miR-154-3p and miR-487-3p, which are belonged to a families of miRNAs. The expression of miR-154-3p is found to be down-regulated in breast cancer [24]. Interestingly, the protein expression of RHOA was significantly negatively correlated with miR-154-3p and miR-487-3p expression in 63 thyroid cancer tissues. miRs play crucial roles in modulating gene expression via post-transcriptional repression [25,26]. In fact, single miR can modulate multiple target genes, in contrast with that, one gene can be regulated by multiple miRs [25,26]. For example, miR-143 and miR-145 gain-of-function synergistically suppress cell proliferation and invasion in breast cancer through the repression of ERBB3 protein expression [25]. miR-148-3p and miR-152-3p realize synergistic effect to inhibit cell proliferation and induce apoptosis in prostate cancer cells [26]. In thyroid cancer, we discovered a collaborative mechanism of miRs that miR-154-3p and miR-487-3p tend to function as tumor suppressors in human thyroid cancer via post-transcriptional repression of RHOA protein expression, but not mRNA levels.
Our findings demonstrated that miR-154-3p and miR-487-3p were significantly decreased in 63 thyroid cancer tissues and cell lines compared with those in paired non-tumor tissues and normal thyroid follicular epithelial cells. Low expression levels of miR-154-3p and miR-487-3p significantly correlated with tumor size, TNM stage, histological grade, lymph node metastasis and shorter overall survival in patients with thyroid cancer. Furthermore, we experimentally validated that miR-154-3p and miR-487-3p synergistically blocked thyroid cancer cell growth in vitro and in vivo. However, the anti-proliferative and pro-apoptotic activities of miR-154-3p/487-3p were neutralized by RHOA overexpressed vectors. Our present findings expound a novel signal cascade employing miR-154-3p/487-3p and RHOA to fine-tune thyroid cancer cell proliferation and apoptosis. Therefore, we corroborate that the suppression of RHOA by miR-154-3p/487-3p may be a valuable therapeutic target for impeding thyroid cancer progression.
One primary goal of the present study was to explore whether miR-154-3p or miR-487-3p performs its function individually or synergistically. First, we found that miR-154-3p or miR-487-3p are grouped into families, which contribute to have sequence homology and possess the same seed sequence to bind with target gene [25]. Usually, miRs family members are often thought to have extensively overlapping targets [25]. Experimental measurements strongly supported the hypothesis that miR-154-3p and miR-487-3p exhibited a synergistic repression of luciferase activity and RHOA protein expression. In addition, miR-154-3p and miR-487-3p showed a cooperative effect to inhibit thyroid cancer cell growth in vitro and in vivo. Our results suggest that multiple miRs collaborate to suppress the same target gene that may be more efficient and potent.
However, some limitations existed in the present study. First, the number of clinical specimen is too small, which may influence the reliability of the present conclusions. Second, the roles of miR-154-3p, miR-487-3p and RHOA on cell migration and invasion had not been investigated in the present study. Figure 6. miR-154-3p and miR-487-3p repress thyroid cancer K-1 cell growth in vivo Thyroid cancer K-1 cells (1 × 10 7 cells per 0.1 ml) were established to steadily express miR-154-3p, miR-487-3p or miR-154-3p/487-3p. K-1 cells were implanted subcutaneously into 4-week-old BALB/c nude mice, and tumor growth was evaluated at week 4 after K-1 cells implantation (A and B). The mRNA and protein levels of RHOA are detected using RT-PCR and