Abstract
Objective
Thyroid cancer (THCA) is the most common endocrine cancer in the world. Although most patients with THCA have a good prognosis, the prognosis of those with THCA who have an extra-glandular invasion, vascular invasion, and distant metastasis is poor. Therefore, it is very important to find potential biomarkers that can effectively predict the prognosis and progression of highly aggressive THCAs. It has been identified that forkhead box P4 (FOXP4) may be a new biomarker for the proliferation and prognosis for tumor diagnosis. However, the expression and function of FOXP4 in THCA remain to be determined.
Methods
In the present study, the function of FOXP4 in cells was investigated through the comprehensive analysis of data in The Cancer Genome Atlas and combined with experiments including immunohistochemistry (IHC), colony formation, Cell Counting Kit-8 assay, wound scratch healing, and transwell invasion assay.
Results
In the present study, relevant bioinformatic data showed that FOXP4 was highly expressed in THCA, which was consistent with the results of the IHC and cell experiments. Meanwhile, 10 FOXP4-related hub genes were identified as potential diagnostic genes for THCA. It was found in further experiments that FOXP4 was located in the nucleus of THCA cells, and the expression of FOXP4 in the nucleus was higher than that in the cytoplasm. FOXP4 knockdown inhibited in vitro proliferation of the THCA cells, whereas overexpression promoted the proliferation and migration of THCA cells. Furthermore, deficiency of FOXP4 induced cell-cycle arrest.
Conclusion
FOXP4 might be a potential target for diagnosing and treating THCA.
Introduction
The incidence of thyroid cancer (THCA) ranks first among head and neck cancers (1). Thyroid cancer can be divided into four histological types: papillary thyroid carcinoma (PTC), follicular carcinoma, anaplastic carcinoma, and medullary carcinoma. Among these types, PTC accounts for more than 90%, which makes it the most important type of THCA (2, 3). There are two subtypes of PTC: indolent and aggressive. Most PTCs have a slow progression, are highly indolent, and have a favorable prognosis. However, 5–20% of PTCs have an extrathyroidal invasion, vascular invasion, and distant metastasis with a relatively poor prognosis (4). The main treatment for PTC is surgical resection. The clinical effect is not significant for patients with aggressive PTC and distant metastasis, and they are not sensitive to radiotherapy and chemotherapy (5). For these invasive PTCs with a relatively high degree of malignancy, it is particularly important to find new and feasible molecular pathological markers and therapeutic targets and to identify them as soon as possible using the appropriate treatment. The role of forkhead box P4 (FOXP4) in THCA has not been reported.
The FOXP protein family is a super-family of transcriptional regulators that have been implicated in the onset, maintenance, progression, and drug resistance of cancer (6, 7). The FOXP family, consisting of FOXP1, FOXP2, FOXP3, and FOXP4, is involved in embryonic development, neural development, immune disorders, and progression of cancer. Growing evidence suggests that the dual functions of the FOXP family which act as tumor suppressors or oncogenes for multiple cancers may be closely correlated with the interactions of the microenvironmental factors of tumors (8, 9, 10, 11, 12, 13). Reports on the dual function of the FOXP family may be closely related to the tumor microenvironment factors and the interaction of the FOXP family members (14, 15).
FOXP4, a member of the FOXP subfamily, is located on human chromosome 6p21.1 and encodes a protein composed of 680 amino acids (7). It has been shown that FOXP4 may be significantly upregulated in breast cancer (16), hepatocellular carcinoma (17), lung cancer (18), and laryngeal cancer (19), and it has been observed that FOXP4 may be involved in tumorigenesis and progression through various molecular mechanisms. However, FOXP4 does not act as a tumor-promoting factor in all tumors. In renal tumors, FOXP4 acts as a tumor suppressor to inhibit the growth and invasion of tumor cells (20). Until now, the expression and function of FOXP4 in THCA have not yet been determined. In the present study, the expression and biological behavior of FOXP4 in THCA are investigated through bioinformatics analysis and related experimental analysis.
Materials and methods
Bioinformatic analysis of the Forkhead box P4 analysis of thyroid cancer
Data source
This study downloaded The Cancer Genome Atlas Thyroid Cancer (TCGA–THCA) dataset from the UCSC Xena official website (https://xena.ucsc.edu/), including the gene expression matrix (count form and fragments per kilobase million form) and sampled clinical information (presenting phenotypes). The TCGA–THCA dataset contained 510 tumor samples and 58 non-cancerous tissue samples.
Forkhead box P4 expression analysis of thyroid cancer
Raw expression gene data from TCGA were normalized, and the Ensembl IDs from TCGA were converted to gene symbols. The expressions of FOXP4 were extracted from the TCGA dataset and compared in tumor and normal groups (21).
Analysis of the potential relationship between long non-coding RNA forkhead box P4–antisense RNA 1 and forkhead box P4
The relationship between long non-coding RNA (lncRNA) FOXP4–antisense RNA 1 (FOXP4–AS1) and FOXP4 was predicted using a bioinformatics tool. DIANA tools (http://carolina.imis.athena-innovation.gr/diana_tools/web/index.php?r=site%2Ftools) and the starBase database (http://starbase.sysu.edu.cn/) were used to predict the micro-RNAs (miRNAs) with both lncRNA FOXP4–AS1 and FOXP4. Finally, the network was visualized and mapped using Cytoscape.
Gene set enrichment analysis
Gene set enrichment analysis (GSEA) was performed to explore the potential biological pathways. The samples were divided into high- and low-expression groups using the median expression levels of the FOXP4 gene in tumor samples. Then, the clusterProfiler (a universal enrichment tool for interpreting omics data) package was utilized to perform the GSEA of FOXP4. C2.cp.kegg.v7.0.symbols.gmt was utilized as the reference gene set for the GSEA.
Screening of differentially expressed genes
The differentially expressed genes (DEGs) were screened between the tumor samples and 58 normal samples with DESeq2 (differential gene expression analysis based on the negative binomial distribution) (22) under the criteria of the adjusted P-value < 0.05 and |log2-fold change | > 1. The heatmap and the volcano plot were constructed to display DEGs using the pheatmap and ggplot2 packages, respectively.
Kyoto Encyclopedia of Genes and Genomes enrichment analysis
Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the R package clusterProfiler (23).
Weighted gene co-expression network analysis
The weighted gene co-expression network analysis (WGCNA) (24) package in R was used to construct co-expression networks. First, a sample clustering tree map was constructed to remove abnormal samples. Then, the most suitable beta value was defined to ensure a standard scale-free network. Co-expression modules were then identified by a dynamic pruning method. Finally, modules with a high correlation with FOXP4 expressions were considered key modules. Therefore, the intersections between the DEGs and genes in key modules were identified using the VennDiagram R package (25) and were defined as FOXP4-related genes for subsequent analysis.
Protein–protein interaction network construction
We established a protein–protein interaction (PPI) network of FOXP4-related genes via the STRING database (http://www.string-db.org/) (26). The cytoHubba plugin of the Cytoscape software was used to identify hub genes. Cytoscape was used for visualization.
The receiver operating characteristic curve analysis and expression analysis
We used the pROC R package (27) to construct receiver operating characteristic (ROC) curves. The diagnostic values of hub genes were evaluated using the area under the curve (AUC). When the AUC value was > 0.7, the hub genes were considered to be capable of distinguishing THCA samples from normal samples with excellent specificity and sensitivity. Expression levels of hub genes between THCA samples and normal samples in TCGA were shown using box plots. The correlation analysis of diagnostic genes was performed using the Corrplot package (28).
Construction of diagnostic genes: miRNA/transcription factor regulatory network
In this study, the NetwokAnalyst database (https://www.networkanalyst.ca/) (21) was used to predict the diagnostic genes' transcription factors (TFs). The miTarbase database (https://mirtarbase.cuhk.edu.cn/) (29) was selected to predict the regulatory relationship between diagnostic genes and miRNAs. The miRNA/TF-diagnostic gene network was constructed using the Cytoscape. In the network, an orange node represented the diagnostic genes, a purple node represented the miRNA, and a blue node represented the TF.
Histopathological and cellular localization analysis of forkhead box P4 and in vitro cell phenotype experiments with forkhead box P4 knockdown and overexpression
Immunofluorescence staining
The KTC-1, TPC-1, and Nthy-ori 3-1 cell lines were purchased from the China Typical Culture Collection Center (Wuhan, China), and the cells were cultured for 24 h in the 24-well plates, fixed with 4% formaldehyde, blocked with 5% bovine serum albumin, and permeabilized with 0.5% Triton X-100 (WILBER Biological Company Limited, Guizhou, China). The washed cells were then incubated with anti-FOXP4 polyclonal antibody (1:150, Abeam, Dallas, TX, USA), followed by incubation with mouse/goat anti-rabbit antibody (1:800, Beyotime Institute of Biotechnology, Shanghai, China). Cells were then incubated with 4′,6-diamidino-2-phenylindole (1:1000, Bio-Sharp Biotechnology, Hefei, China) for 5 min and then were observed under a fluorescence microscope within 4 h of the incubation.
Sampling and immunohistochemistry
Thyroid cancer tissues and paired adjacent tissues were obtained from three patients with THCA in the Affiliated Hospital of Guizhou Medical University (Guizhou, China). The patients had not received any treatment before surgery. According to the microscopic examination results, the selected tissue paraffin sections should contain more tumor tissue and less interstitial components with avoidance of the necrotic and hemorrhagic areas. All specimens were formalin fixed and assayed according to the standard immunohistochemistry (IHC) protocols, and IHC staining was conducted following the manufacturer’s instructions. The adoptions of specific antibodies were as follows: FOXP4 (1:200, Abcam). FOXP4 expression was assessed based on the staining intensity (0, 1+, 2+, and 3+) and the percentage of positive cells. The scores of the expression were 0 (0%), 1 (1–25%), 2 (26–50%), 3 (51–75%), or 4 (76–100%). The calculation of the staining index (SI) was as follows: SI = (the intensity score in 1) × (the intensity score in 2). Those with an SI < 3 were assigned as low expression and those with an SI ≥ 4 were assigned as high expression. In addition, the present study was approved by the Human Ethics Committee of the Affiliated Hospital of Guizhou Medical University.
Cell culture and reagents
The KTC-1 (Wuhan, China) cell line was cultured in Dulbecco's modified Eagle medium. The TPC-1 and Nthy-ori 3-1 cell lines (Wuhan, China) were cultured in RPMI 1640 (HyClone, Logan, UT, USA). All cell lines were cultured in the medium supplemented with 10% fetal bovine serum (FBS) (Gibco). The primary antibody for FOXP4 protein was purchased from Abcam, and the secondary antibody, affinity-purified biotinylated rabbit anti-immunoglobulin G, was purchased from Sigma.
Lentiviral vectors, plasmids, small hairpin RNA, and transfection conditions
To temporarily silence the expression of FOXP4, small interfering RNA (Ribo-Bio, Guangzhou, China) was transfected into KTC-1 and TPC-1 cells using Lipofectamine (Guizhou Hejin Biotechnology Co, Ltd, China). Stable cell lines with FOXP4 knockdown were established using FOXP4 small hairpin RNA. A FOXP4 complementary DNA plasmid (Guizhou Hejin Biotechnology Co) was used to upregulate the expression of FOXP4. All transfection procedures were conducted according to the manufacturer’s instructions.
Real-time polymerase chain reaction and Western blot
Primers for real-time polymerase chain reaction were designed by the Laboratory of Pathophysiology, GuiZhou Medical University, and synthesized by Jiangsu Genecfe Biotechnology Co, Ltd (Zhejiang, China). The mRNA expression was quantified with the adoption of the 2-( ΔCt sample –- ΔCt control) method (30). Cells were lyzed in a radioimmunoprecipitation assay buffer containing phosphatase and protease inhibitors. The bicinchoninic acid protein assay kit (Pierce) was used to determine the protein concentration. The bands of interest in the Western blot were normalized with that of the glyceraldehyde 3-phosphate dehydrogenase protein.
Wound scratch healing and transwell assay
Cells (1 × 105) were cultured in the six-well plate. After 16 h, the complete medium was replaced with the fresh medium with a low concentration of serum (2%). With the cells reaching 90% confluency, a 10 µL pipette tip was used to draw a uniformly shaped wound across each well. Cells were gently washed twice with phosphate-buffered saline to remove the loose cells, and the serum-free medium was added. Multiple positioning marks were made in the center of the exposed surface to ensure comparability of wounds with the same area. An inverted microscope photographed scratched areas at 0 and 48 h. The migration of cells was analyzed using ImageJ 1.48 software (National Institutes of Health). The wound healing rate was calculated according to the following formula: wound healing rate =(((scratch width at 0 h) – (scratch width at 48 h))/(scratch width at 0 h)) × 100%. Each independent experiment was repeated at least three times.
Transwell invasion analysis was used to determine cell migration and invasion capacities. Briefly, Transwell chambers with 8 µm pore sizes (Guangzhou Jet Bio-Filtration Co, Ltd, Guangzhou, Guangdong, China) were coated with 80 µL 1:16-diluted Matrigel-coated (BD Biosciences, Franklin Lakes, NJ, USA) and incubated at 37°C for 2 h. After transfection for 24 h, approximately 2 × 104 cells were placed into the upper chamber with 300 µL serum-free RPMI 1640 medium, while 500 µL RPMI 1640 supplemented with 10% FBS was added to the lower chamber. Following incubation for 48 h, the cells in the upper chamber were removed by a cotton swab, and the cells in the lower chamber were then stained with 0.1% crystal violet (Merck KGaA, Sigma-Aldrich) for 10 min at room temperature. Finally, the invaded cells were counted and photographed under a light microscope.
Clone formation assay and Cell Counting Kit-8 viability assay
Briefly, 2 × 103 transfected cells were cultured in the six-well plates for 12 days, with the medium being changed every 3 days. After 12 days, colonies were fixed with methanol for 10 min at room temperature (25–27℃) and then stained with 0.1% crystal violet for 15 min at room temperature. Colonies were counted under a light microscope.
The Cell Counting Kit-8 (CCK-8) assay (Sevenbio, Beijing, China) was used to detect cell viability. Cells were seeded in the 96-well plates at a density of 4 × 103 cells per well in 200 μL of medium and cultured for 24, 48, and 72 h, respectively. The absorbance was measured at 450 nm after cells were treated with 10% CCK-8 for 2 h at 37°C. Cell viability was calculated as the ratio of the optical density values of drug-treated samples to that of the controls.
Cell-cycle detection by flow cytometry
For cell-cycle detection, KTC-1 and TPC-1 cells transfected with si-FOXP4 for 48 h were synchronized by starving in the G0/G1 phase and then fixed into cells in precooled 75% ethanol and stained with propidium iodide (Yeasen, China). The DNA content of G0/G1, S, and G2/M phases in the cell cycle was examined through flow cytometry.
Results
The essential role of forkhead box P4 in the thyroid cancer: a study with The Cancer Genome Atlas
Elevated expression of forkhead box P4 in thyroid cancer
This paper used TCGA to explore the expression levels of FOXP4 in normal and cancer tissues. The results showed that FOXP4 was more highly expressed in THCA compared with normal tissues, as shown in the box plot (Fig. 1A, P < 0.001). The AUC value for FOXP4 was 0.8157. The above result indicated that FOXP4 had a good diagnostic value for THCA (Fig. 1B). In addition, to further identify the possible functions of FOXP4 in THCA, a GSEA with TCGA was conducted. Based on the results of the GSEA, FOXP4 was mainly enriched in allograft rejection, chemokine signaling pathway, and cytokine receptor interaction. After considering the results of the GSEA, it was concluded that FOXP4 might be highly correlated with immune and inflammation responses, suggesting that FOXP4 may rely on these pathways to control THCA progression (Fig. 1C).
(A) The expression of FOXP4 in normal and tumor tissues detected by TCGA. Compared with that in the normal tissue, the expression of FOXP4 was higher in THCA (P < 0.001). (B) The AUC of FOXP4 was 0.8157, which indicated a better diagnostic value. (C) According to the investigation results of GSEA, FOXP4 was mainly enriched in allograft rejection, chemokine signaling pathway, and cytokine receptor interaction. (D) The lncRNAs FOXP4-as1 and FOXP4 might interact with hsa-miR-4525 to form a ceRNA network.
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
A competing endogenous RNA regulatory network of long non-coding RNA forkhead box P4–antisense RNA 1 and forkhead box P4 in thyroid cancer
Both bioinformatics analysis and mechanistic studies indicated that lncRNA FOXP4–AS1 and FOXP4 could interact with hsa-miR-4525 to form a ceRNA network (Fig. 1D). Mechanistically, it was identified that lncFOXP4–AS1 is a sponge for hsa-miR-4525 and that FOXP4 is a direct target of hsa-miR-4525.
Identification and function analysis of differentially expressed genes
There were 3164 DEGs screened, which consisted of 1873 upregulated genes and 1291 downregulated genes (tumor vs normal) (Fig. 2A). The heatmap for the DEGs is shown in Fig. 2B. As shown in Fig. 2C, the KEGG pathway analysis revealed that the upregulated DEGs were mainly enriched in the neuroactive ligand–receptor interaction, cytokine–cytokine receptor interaction, and cyclic adenosine 3′,5′-monophosphate signaling pathway. As shown in Fig. 2D, the KEGG pathway analysis revealed that the downregulated DEGs were mainly enriched in the neuroactive ligand–receptor interaction, cytokine–cytokine receptor interaction, and T cell receptor signaling pathway.
(A) The upregulated and downregulated genes screened by DEGs. (B) The heatmap for screening the upregulated and downregulated genes. (C) Main upregulated pathways in DEG gene enrichment by KEGG analysis. (D) Main downregulated pathways in DEG gene enrichment by KEGG analysis.
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
Identification of hub genes
The DEGs were intersected with the above three module genes, and a total of 80 FOXP4-related genes were selected (Fig. 3A). As shown in Fig. 3B, PPI networks of FOXP4-related genes were constructed. The hub genes, including arginine vasopressin (AVP), fibroblast growth factor receptor 3 (FGFR3), neurturin (NRTN), calcium voltage-gated channel auxiliary subunit beta 1 (CACNB1), corticotropin-releasing hormone receptor 2 (CRHR2), ephrin A3 (EFNA3), glutamic–pyruvic transaminase (GPT), potassium channel subfamily K member 5 (KCNK5), neuropeptide W (NPW), neurotrophic receptor tyrosine kinase 3 (NTRK3), phosphoinositide-3-kinase regulatory subunit 2 (PIK3R2), solute carrier family 12 member 5 (SLC12A5), and urocortin (UCN) were identified by the cytoHubba plug-in in Cytoscape (degree > 1) (Fig. 3C). The degree of hub genes is shown in Fig. 3D.
(A) Among DEGs and module–trait relationships, 80 FOXP4-related genes were screened by intersecting the most closely related genes with FOXP4. (B) PPI networks and hub genes for FOXP4-related genes. (C and D) Identification of hub gene expression by the cytoHubba plug-in in Cytoscape (degree > 1).
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
The expression analysis and ROC curve analysis of hub genes
As shown in Fig. 4A, the AUC values of AVP, FGFR3, NRTN, CACNB1, CRHR2, EFNA3, GPT, KCNK5, NPW, NTRK3, PIK3R2, SLC12A5, and UCN were 0.5669, 0.6917, 0.6999, 0.8921, 0.7577, 0.8686, 0.7163, 0.8831, 0.8562, 0.7008, 0.8362, 0.8923, and 0.8808, respectively, demonstrating that all these hub genes except for AVP, FGFR3, and NRTN had good diagnostic values. It was observed that the 10 diagnostic gene expression levels were significantly increased in the THCA group compared with those in the normal group (Fig. 4B). The diagnostic gene correlation analysis results found that the highest positive correlation was between KCNK5 and CACNB1 as well as between EFNA3 and PIK3R2 (correlation = 0.6, Fig. 4C).
(A) The AUC values of hub genes. (B) The expression levels of the 10 hub genes were significantly higher in the THCA group compared with the normal group. (C) The results of the diagnostic gene correlation analysis.
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
Prediction of potential micro-RNA/transcription factor-diagnostic gene regulatory network
The miRNA- and TF-diagnostic gene interaction networks were established by using the Cytoscape software. As illustrated in Fig. 5A, the interaction network consists of 7 diagnostic genes and 186 miRNAs. There were 186 potential miRNAs according to the miRNet database targeting and regulating the expressions of diagnostic genes. As illustrated in Fig. 5B, the interaction network consists of 10 diagnostic genes and 88 TFs. There were 88 potential TFs according to the miRNet database targeting and regulating the expressions of diagnostic genes.
(A) The miRNA-diagnostic gene and TF-diagnostic gene interaction network established with the adoption of the Cytoscape software. (B) There were 88 potential TFs according to the miRNet database targeting and regulating the expressions of diagnostic genes.
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
The essential role of forkhead box P4 in thyroid cancer: a study with tissues and stable cell lines
FOX4 expression was upregulated in thyroid cancer tissues and stable cell lines
Thyroid cancer tissue and corresponding adjacent tissue samples from three patients were tested by IHC. As shown in Fig. 6A, there was more positive staining for FOXP4 in THCA tissues than in adjacent tissues. In THCA cells, the expression of FOXP4 was higher in the nucleus than in the cytoplasm (Fig. 6B). Western blot verification found that the protein expression of FOXP4 in three THCA cell lines (TPC-1, KTC-1, and B-CPAP) and the abundance of FOXP4 proteins were increased in cancer cells compared with thyroid normal follicular Nthy-ori 3-1 epithelial cells (Fig. 6C).
(A) Expression of FOXP4 in thyroid cancer tissue and adjacent tissue detected by immunohistochemistry. (B) Localization of FOXP4 protein in normal thyroid epithelial cells and tumor cells. (C) Expression of FOXP4 protein in normal thyroid epithelial cells and different cancer cell lines. (D) Promotion of cancer cell growth by FOXP4 confirmed by colony formation experiment with FOXP4 knockdown and overexpression. (E) Promotion of cancer cell migration by FOXP4 confirmed by wound scratch healing with FOXP4 knockdown and overexpression. (F) Promotion of cancer cell proliferation by FOXP4 confirmed by cell-cycle experiments.
Citation: Endocrine Connections 12, 4; 10.1530/EC-22-0390
Forkhead box P4 knockdown suppressed the proliferation and migration of thyroid cancer cell lines and arrested cell cycles in thyroid cancer cells
The results of FOXP4 knockdown and overexpression revealed that FOXP4 overexpression might promote tumor proliferation and migration. In contrast, FOXP4 knockdown significantly inhibited tumor growth (Figs. 6D and E). It was found through flow cytometry that compared with the ov-NC group, the cells in the S phase of the ov-FOXP4 group were significantly increased, and the cells in the G0/G1 phase were significantly reduced. While compared with the sh-NC, the cells in the S phase of the sh-FOXP4 group were significantly reduced, and the cells in G0/G1 phase were significantly increased (Fig. 6F).
Discussion
Thyroid cancer is one of the most common malignancies of the head and neck region, with PTC accounting for more than 90% of the total cases, and it is the type with the highest proportion of THCA. Although most PTCs progress slowly, 5–20% of PTCs are accompanied by extrathyroidal invasion, vascular invasion, and distant metastasis and have a poor prognosis. According to TCGA, a correlation analysis was conducted on FOXP4 in THCA, and it was found that FOXP4 might have a certain value in the investigation of THCA. At the same time, clinical tissue samples and cell experiments were conducted to investigate PTC.
In the present study, it was found that FOXP4 was highly expressed in THCA, and through the ROC curve analysis, it was suggested that FOXP4 might be used as a biomarker for diagnosing THCA. Currently, investigation on cancer stem cells shows that they have the properties of stem cells, with strong self-renewal capacity, and may promote tumor cell invasion and growth in cancer (31) as well as the probability of differentiating into all types of cells. In addition, due to the heterogeneity, cancer stem cells are insensitive to chemoradiotherapy and play a very important role in the metastasis and differentiation of cancer (32).
It was found in the present study that there existed a significant negative correlation between FOXP4 and the mRNAsi, and the mRNAsi in the high-expression group was significantly lower than that in the low-expression group. The present study showed that mRNAsi and FOXP4 expressions are inversely correlated, so it is believed that FOXP4 is correlated to the stemness index, which is closely related to tumor self-renewal and drug resistance (the expression of mRNAsi is negatively correlated with the expression of FOXP4; thus, it was believed that FOXP4 might be correlated with the stemness index, together with a close correlation with tumor self-renewal and drug resistance). This might reveal that FOXP4 could be used as a potential gene therapeutic target. It was also suggested that high-risk groups might be screened through FOXP4 detection, and necessary close follow-ups and enhanced treatments might be conducted to delay the recurrence and distant metastasis and improve patients’ survival rate. Currently, the investigations of FOXP4 in other cancers have revealed that FOXP4 plays an important role as a key transcriptional hub molecule with duality (33). FOXP4 was overexpressed in A549 and H1703 non-small cell lung cancer (NSCLC) cells, and the depletion of FOXP4 significantly reduced the growth and invasion of these two NSCLC cells. FOXP4 was closely correlated with MIR4316 in breast cancer cells and could promote breast cancer proliferation (34). FOXP4 was significantly downregulated in patients with renal cancer, and FOXP4 overexpression inhibited the tumor growth (35). Therefore, FOXP4 protein might have dual functions, with the evidence of acting as either a proto-oncogene or an inhibitor of specific tumor types. Therefore, it might be necessary to clarify the role of FOXP4 in THCA and investigate its function.
DEGs were identified between THCA and normal samples, and in addition, WGCNA was conducted to construct a co-expression network associated with FOXP4. Among the nine gene modules identified, the blue module consisting of 850 genes had the strongest correlation with FOXP4 and was selected for further analysis. Finally, the FOXP4-related differential genes were screened, and the PPI network of these FOXP4-related differential genes was constructed. According to the results of the ROC analysis, 10 FOXP4-related differential genes (CCNB1, CRHR2, EFNA3, GPT, KCNK5, NPW, NTRK3, PIK3R2, SLC12A5, and UCN) were identified as potential biomarkers for the diagnosis of THCA. Among these 10 genes, it has been reported that adopting the Kaplan–Meier plotter to examine the expressions of KCNKs mRNA in THCA has important prognostic value. Patients with THCA and high expressions of KCNK5 and KCNK15 had better overall survival, which showed that KCNK5 and KCNK15 might be very important and valuable for prolonging overall patient survival. The signal transducer and activator of transcription, nuclear factor kappa-light-chain-enhancer of activated B cells, and early growth response 1 genes might be the key TFs regulating KCNK5 (36). For KCNK5, the most relevant miRNA target was miR-296 (37). Meanwhile, a rearrangement of NTRK3 is a well-known driver in a variety of human cancers (38), and its fusion with various partner genes can induce tumorigenesis by generating chimeric oncoproteins with constitutively activated kinase functions and leading to downstream stimulation of cell proliferation via the RAS/RAF/mitogen-activated protein kinase pathway (39). It has been reported that NTRK-rearranged THCA appeared to have a high metastatic rate (39); however, with the treatment of NTRK inhibitors, it is expected to be an effective treatment for highly aggressive tumors (40). Therefore, NTRK inhibitor therapy is expected to receive good results in patients with NTRK3 rearrangements. In targeting the PIK3R2 gene, miR-126 was found to act as a proliferation inhibitor-targeting gene that could reduce p85β (a regulatory subunit of PI3K kinase) protein translation and reduce the activity of AKT kinase (41). Currently, a relevant literature has reported that miR-126 expression has been shown to be associated with angiogenesis. The regulatory effect of miR-126 on vascular endothelial growth factor (VEGF)-A expression and its tumor-suppressive effect were first demonstrated in thyroid carcinoma and were closely associated with undifferentiated thyroid carcinoma, and therefore miR-126 may be an effective therapeutic tool for treating undifferentiated THCA (42). Therefore, we speculate miR-126 might be a potential therapeutic tool for the treatment of undifferentiated THCA.
In the three clinical specimens, FOXP4 was found to be highly expressed in cancer tissues by IHC. At the same time, it was found that the expression of FOXP4 in the nucleus was significantly higher than that in the cytoplasm. The expression of FOXP4 in the TPC-1, KTC-1, and B-CPAP cell lines was detected at the protein and mRNA levels, and the expression of FOXP4 was higher in the TPC-1, KTC-1, and B-CPAP THCA cell lines than that in the normal thyroid follicular Nthy-ori 3-1 epithelial cells. With FOXP4 overexpression or knockdown, it was found that high expressions of FOXP4 could promote the proliferation and migration of cancer cells, while FOXP4 knockdown significantly inhibited the proliferation of tumor cells. Compared with the control group, the cells in the S phase in the FOXP4 overexpression group were significantly increased, and the cells in the G0/G1 phase were significantly decreased. In addition, compared with the control group, the cells in the S phase in the FOXP4 knockdown group were significantly reduced, and the cells in the G0/G1 phase were significantly increased. The above results indicated that FOXP4 might participate in the regulation of the cell proliferation cycle. The effect of FOXP4 on the biological function of THCA cells was verified at the cellular level. The above results suggested that the FOXP4 gene might play an important role in THCA.
Based on the bioinformatics analysis of FOXP4 in THCA and related experiments, the expression and important value of FOXP4 in THCA were further confirmed. At the same time, in vivo functional experiments are still needed to further verify the accurate biological effects through investigating and discussing FOXP4 at different levels to explore its value in clinical diagnosis. In the future, further experiments will be carried out for verification based on these undeveloped investigations.
Conclusion
In conclusion, the role of FOXP4 in THCA was explored first in the present study, as well as the diagnostic value of FOXP4-related genes in THCA. The expressions and biological behaviors of FOXP4 in THCA were then investigated. FOXP4 was investigated in the present study, not only from the perspective of bioinformatics but also from the perspective of cell biology, with further exploration of FOXP4 and its biomarkers as new diagnostic and therapeutic targets for patients with THCA.
Declaration of interest
The authors declare that they have no competing interests.
Funding
This work was supported by National Natural Science Foundation of China (81860478), Guizhou Provincial Health Committee member Science and Technology Foundation (gzwkj2021-170), and Beijing Health Promotion Association (CJBSRAF-2022)
Ethics approval and consent to participate
I confirm that I have read the Editorial Policy pages. This study was conducted with approval from the Ethics Committee of Affiliated Hospital of Guizhou Medical University (2021 Ethical Review 356). This study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.
Data availability statement
The data used to support the findings of this study are available from the corresponding author upon request.
Acknowledgements
The authors would like to acknowledge the hard and dedicated work of all the staff who implemented the intervention and evaluation components of the study.
References
- 1↑
Bergdorf K, Ferguson DC, Mehrad M, Ely K, Stricker T, Weiss VL. Papillary thyroid carcinoma behavior: clues in the tumor microenvironment. Endocrine-Related Cancer 2019 26 601–614. (https://doi.org/10.1530/ERC-19-0074)
- 2↑
Filetti S, Durante C, Hartl D, Leboulleux S, Locati LD, Newbold K, Papotti MG, Berruti A & ESMO Guidelines Committee. Electronic address: clinicalguidelines@esmo.org. Thyroid cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Annals of Oncology 2019 30 1856–1883. (https://doi.org/10.1093/annonc/mdz400)
- 3↑
Asa SL The current histologic classification of thyroid cancer. Endocrinology and Metabolism Clinics of North America 2019 48 1–22. (https://doi.org/10.1016/j.ecl.2018.10.001)
- 4↑
Xing M Molecular pathogenesis and mechanisms of thyroid cancer. Nature Reviews. Cancer 2013 13 184–199. (https://doi.org/10.1038/nrc3431)
- 5↑
Omur O, Baran Y. An update on molecular biology of thyroid cancers. Critical Reviews in Oncology/Hematology 2014 90 233–252. (https://doi.org/10.1016/j.critrevonc.2013.12.007)
- 6↑
Kim JH, Hwang J, Jung JH, Lee HJ, Lee DY, Kim SH. Molecular networks of FOXP family: dual biologic functions, interplay with other molecules and clinical implications in cancer progression. Molecular Cancer 2019 18 180. (https://doi.org/10.1186/s12943-019-1110-3)
- 7↑
Castells-Nobau A, Eidhof I, Fenckova M, Brenman-Suttner DB, Scheffer-de Gooyert JM, Christine S, Schellevis RL, van der Laan K, Quentin C & van Ninhuijs L et al.Conserved regulation of neurodevelopmental processes and behavior by FoxP in Drosophila. PLoS One 2019 14 e0211652. (https://doi.org/10.1371/journal.pone.0211652)
- 8↑
Li H, Liu P, Xu S, Li Y, Dekker JD, Li B, Fan Y, Zhang Z, Hong Y & Yang G et al.FOXP1 controls mesenchymal stem cell commitment and senescence during skeletal aging. Journal of Clinical Investigation 2017 127 1241–1253. (https://doi.org/10.1172/JCI89511)
- 9↑
Koon HB, Ippolito GC, Banham AH, Tucker PW. FOXP1: a potential therapeutic target in cancer. Expert Opinion on Therapeutic Targets 2007 11 955–965. (https://doi.org/10.1517/14728222.11.7.955)
- 10↑
Neyroud D, Nosacka RL, Callaway CS, Trevino JG, Hu H, Judge SM, Judge AR. FoxP1 is a transcriptional repressor associated with cancer cachexia that induces skeletal muscle wasting and weakness. Journal of Cachexia, Sarcopenia and Muscle 2021 12 421–442. (https://doi.org/10.1002/jcsm.12666)
- 11↑
Liu Y, Chen T, Guo M, Li Y, Zhang Q, Tan G, Yu L, Tan Y. FOXA2-interacting FOXP2 prevents epithelial-mesenchymal transition of breast cancer cells by stimulating E-cadherin and PHF2 transcription. Frontiers in Oncology 2021 11 605025. (https://doi.org/10.3389/fonc.2021.605025)
- 12↑
Nudel R, Newbury DF. FOXP2. Wiley Interdisciplinary Reviews. Cognitive Science 2013 4 547–560. (https://doi.org/10.1002/wcs.1247)
- 13↑
Kim CH FOXP3 and its role in the immune system. Advances in Experimental Medicine and Biology 2009 665 17–29. (https://doi.org/10.1007/978-1-4419-1599-3_2)
- 14↑
Wang J, Gong R, Zhao C, Lei K, Sun X, Ren H. Human FOXP3 and tumour microenvironment. Immunology. 2023 168 248–255. (https://doi.org/10.1111/imm.13520)
- 15↑
Wu F, Ji A, Zhang Z, Li J, Li P. miR-491-5p inhibits the proliferation and migration of A549 cells by FOXP4. Experimental and Therapeutic Medicine 2021 21 622. (https://doi.org/10.3892/etm.2021.10054)
- 16↑
Ma T, Zhang J. Upregulation of FOXP4 in breast cancer promotes migration and invasion through facilitating EMT. Cancer Management and Research 2019 11 2783–2793. (https://doi.org/10.2147/CMAR.S191641)
- 17↑
Zhang G, Zhang G. Upregulation of FoxP4 in HCC promotes migration and invasion through regulation of EMT. Oncology Letters 2019 17 3944–3951. (https://doi.org/10.3892/ol.2019.10049)
- 18↑
Chen T, Liu Y, Chen J, Zheng H, Chen Q, Zhao J. Exosomal miR-3180-3p inhibits proliferation and metastasis of non-small cell lung cancer by downregulating FOXP4. Thoracic Cancer 2021 12 372–381. (https://doi.org/10.1111/1759-7714.13759)
- 19↑
Shi J, Wang J, Cheng H, Liu S, Hao X, Lan L, Wu G, Liu M, Zhao Y. FOXP4 promotes laryngeal squamous cell carcinoma progression through directly targeting LEF1. Molecular Medicine Reports 2021 24 831. (https://doi.org/10.3892/mmr.2021.12471)
- 20↑
Teufel A, Wong EA, Mukhopadhyay M, Malik N, Westphal H. FoxP4, a novel forkhead transcription factor. Biochimica et Biophysica Acta 2003 1627 147–152. (https://doi.org/10.1016/s0167-4781(0300074-5)
- 21↑
Zhou G, Soufan O, Ewald J, Hancock REW, Basu N, Xia J. NetworkAnalyst 3.0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis. Nucleic Acids Research 2019 47 W234–W241. (https://doi.org/10.1093/nar/gkz240)
- 22↑
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 2014 15 550. (https://doi.org/10.1186/s13059-014-0550-8)
- 23↑
Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics 2012 16 284–287. (https://doi.org/10.1089/omi.2011.0118)
- 24↑
Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008 9 559. (https://doi.org/10.1186/1471-2105-9-559)
- 25↑
Chen H, Boutros PC. VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics 2011 12 35. (https://doi.org/10.1186/1471-2105-12-35)
- 26↑
Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, Simonovic M, Roth A, Santos A & Tsafou KP et al.STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Research 2015 43 D447–D452. (https://doi.org/10.1093/nar/gku1003)
- 27↑
Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC, Müller M. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 2011 12 77. (https://doi.org/10.1186/1471-2105-12-77)
- 28↑
Newman AM, Steen CB, Liu CL, Gentles AJ, Chaudhuri AA, Scherer F, Khodadoust MS, Esfahani MS, Luca BA & Steiner D et al.Determining cell type abundance and expression from bulk tissues with digital cytometry. Nature Biotechnology 2019 37 773–782. (https://doi.org/10.1038/s41587-019-0114-2)
- 29↑
Huang HY, Lin YC, Li J, Huang KY, Shrestha S, Hong HC, Tang Y, Chen YG, Jin CN & Yu Y et al.miRTarBase 2020: updates to the experimentally validated microRNA-target interaction database. Nucleic Acids Research 2020 48 D148–D154. (https://doi.org/10.1093/nar/gkz896)
- 30↑
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001 25 402–408. (https://doi.org/10.1006/meth.2001.1262)
- 31↑
Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature 2001 414 105–111. (https://doi.org/10.1038/35102167)
- 32↑
Shibue T, Weinberg RA. EMT, CSCS, and drug resistance: the mechanistic link and clinical implications. Nature Reviews. Clinical Oncology 2017 14 611–629. (https://doi.org/10.1038/nrclinonc.2017.44)
- 33↑
Yang T, Li H, Thakur A, Chen T, Xue J, Li D, Chen M. FOXP4 modulates tumor growth and independently associates with miR-138 in non-small cell lung cancer cells. Tumour Biology: the Journal of the International Society for Oncodevelopmental Biology and Medicine 2015 36 8185–8191. (https://doi.org/10.1007/s13277-015-3498-8)
- 34↑
Wang N, Gu Y, Li L, Wang F, Lv P, Xiong Y, Qiu X. Circular RNA circMYO9B facilitates breast cancer cell proliferation and invasiveness via upregulating FOXP4 expression by sponging miR-4316. Archives of Biochemistry and Biophysics 2018 653 63–70. (https://doi.org/10.1016/j.abb.2018.04.017)
- 35↑
Xiong Y, Zhang J, Song C. CircRNA ZNF609 functions as a competitive endogenous RNA to regulate FOXP4 expression by sponging miR-138-5p in renal carcinoma. Journal of Cellular Physiology 2019 234 10646–10654. (https://doi.org/10.1002/jcp.27744)
- 36↑
Lin X, Wu JF, Wang DM, Zhang J, Zhang WJ, Xue G. The correlation and role analysis of KCNK2/4/5/15 in Human Papillary thyroid Carcinoma microenvironment. Journal of Cancer 2020 11 5162–5176. (https://doi.org/10.7150/jca.45604)
- 37↑
Lin X, Wu JF, Wang DM, Zhang J, Zhang WJ, Xue G. The correlation and role analysis of KCNK2/4/5/15 in Human Papillary thyroid Carcinoma microenvironment. Journal of Cancer 2020 11 5162–5176. (https://doi.org/10.7150/jca.45604)
- 38↑
Gatalica Z, Xiu J, Swensen J, Vranic S. Molecular characterization of cancers with NTRK gene fusions. Modern Pathology 2019 32 147–153. (https://doi.org/10.1038/s41379-018-0118-3)
- 39↑
Okamura R, Boichard A, Kato S, Sicklick JK, Bazhenova L, & Kurzrock R. Analysis of NTRK alterations in pan-cancer adult and pediatric malignancies: implications for NTRK-targeted therapeutics. JCO Precision Oncology 2018 2018 1–20. (https://doi.org/10.1200/PO.18.00183)
- 40↑
Amatu A, Sartore-Bianchi A, Siena S. NTRK gene fusions as novel targets of cancer therapy across multiple tumour types. ESMO Open 2016 1 e000023. (https://doi.org/10.1136/esmoopen-2015-000023)
- 41↑
Fu R, Tong JS. miR-126 reduces trastuzumab resistance by targeting PIK3R2 and regulating AKT/mTOR pathway in breast cancer cells. Journal of Cellular and Molecular Medicine 2020 24 7600–7608. (https://doi.org/10.1111/jcmm.15396)
- 42↑
Salajegheh A, Vosgha H, Rahman MA, Amin M, Smith RA, Lam AK. Interactive role of miR-126 on VEGF-A and progression of papillary and undifferentiated thyroid carcinoma. Human Pathology 2016 51 75–85. (https://doi.org/10.1016/j.humpath.2015.12.018)
