KSRP-mediated Wnt/β-catenin activation promotes follicular thyroid cancer progression and stemness

Abstract

Background

K-homology-type splicing regulatory protein (KSRP) is an RNA-binding protein involved in mRNA decay and translational repression through recognition of adenine–uracil-rich elements. Although KSRP regulates approximately 16% of transcript expression, its role in cancer remains poorly defined.

Methods

KSRP expression was analysed using qPCR, Western blot, and immunohistochemistry. Its functional role in follicular thyroid cancer (FTC) was examined through in vitro and in vivo assays. Luciferase reporter and rescue experiments were performed to elucidate the underlying molecular mechanisms.

Results

KSRP was significantly upregulated in FTC tissues and metastatic cell lines. Functional studies demonstrated that KSRP enhances the invasiveness and stemness of FTC cells. Mechanistically, KSRP promotes nuclear accumulation and transcriptional activity of β-catenin by downregulating the Wnt inhibitors DACT2 and SFRP2.

Conclusion

This study identifies KSRP as an oncogenic factor in FTC that activates Wnt/β-catenin signalling, suggesting its potential as a therapeutic target for FTC patients.

Introduction

RNA-binding proteins (RBPs) are pivotal regulators of mRNA processing and translation [1]. By interacting with sequence-specific motifs or RNA secondary structures, RBPs primarily regulate posttranscriptional steps such as mRNA splicing, polyadenylation, stability, localisation, and degradation [2]. RBPs typically exert either a stabilising or a destabilising effect on mRNA [3]. Adenine–uracil-rich elements (AREs) located in the 3′-untranslated regions (3′UTRs) of mRNA are involved in mRNA degradation [4]. Approximately 16% of all transcripts contain AREs. The K-homology-type splicing regulatory protein (KSRP; also known as KHSRP), an RBP with four KH domains, can recognise AREs and regulate gene stability by degrading mRNA, thereby inhibiting translation and facilitating micro-RNA (miRNA) maturation [5–7]. KSRP has been implicated in several cellular functions, including cell differentiation, cell proliferation, innate immunity, and lipid metabolism [8, 9]. This protein is aberrantly expressed in various pathological conditions, including cancer. We previously deciphered the role of KSRP in tumour progression. However, reports regarding its role in various tumours are inconsistent. For example, KSRP has been demonstrated to inhibit non–small-cell lung cancer metastasis by promoting early growth response-3 (EGR3) degradation [10]. The instability of EGR3 mRNA is independent of ARE recognition and is primarily regulated by miRNA-23a. This miRNA directly binds to EGR3 3′UTR, downregulating EGR3 expression and inhibiting non–small cell lung cancer cell motility. However, KSRP has also been identified to be an oncoprotein that promotes clear cell renal cell carcinoma progression [11]. KSRP reduces neural precursor cell–expressed developmentally downregulated 4-like (NEDD4L) mRNA stability by inducing miR-629-5p expression, which targets the AREs in the 3′UTR of NEDD4L mRNA to downregulate its expression. The downregulation of NEDD4L induces the epithelial–mesenchymal transition and promotes the invasiveness of clear cell renal cell cancer. Therefore, identifying KSRP’s mRNA targets that might influence tumorigenesis may clarify the potential molecular mechanisms underlying key gene regulation and reveal novel targets for cancer treatment.

Thyroid cancer is the most common endocrine malignancy, accounting for approximately 1%–1.5% of all new cancer diagnoses worldwide in recent decades [12]. Various factors increase the risk of thyroid cancer, examples of which include exposure to ionising radiation, nodular disease of the thyroid, and a family history of thyroid cancer [13, 14]. Additional risk factors for thyroid cancer include mutations in RAS, TP53, and BRAF [15, 16]. Because >95% of all thyroid cancer cases affect thyroid follicular cells [17], the molecular mechanisms underlying thyroid cancer development must be investigated to identify effective diagnostic biomarkers and therapeutic targets. Follicular thyroid cancer (FTC) and papillary thyroid carcinoma (PTC) are two common types of differentiated thyroid cancer. FTC typically involves RAS mutations and PAX8/PPARγ rearrangements, whereas PTC commonly involves BRAF V600E mutations and RET/PTC rearrangements. Moreover, these two cancers differ in terms of metastasis patterns, with FTC primarily spreading through the bloodstream and PTC primarily spreading through the lymphatic system. A previous study had identified KSRP as a binding partner of the long noncoding RNA AB074169 in PTC [18]. This RNA functions as a tumour suppressor; it downregulates the expression of KSRP to suppress the proliferation of PTC cells. Mechanistically, the downregulation of KSRP expression upregulates p21 expression and downregulates CDK2 expression, thereby regulating cell cycle progression in PTC. Evidence indicates that KSRP is involved in multiple steps of cancer development; nevertheless, such evidence is primarily based on the function of KSRP as an mRNA/miRNA regulator. Few studies have comprehensively and directly investigated the role of KSRP in thyroid cancer, particularly in FTC.

To address the aforementioned gaps in the literature, the present study investigated the role of KSRP in the progression of FTC. We observed that KSRP expression was higher in FTC tissues than in normal or adenoma tissues. Our functional analyses revealed the key role of KSRP in the migration, invasiveness, and stemness of FTC cells. Furthermore, KSRP induced the transcriptional activity of β-catenin by negatively regulating two inhibitors of Wnt signalling, namely DACT2 and SFRP2. Thus, our findings clarify the potential role of KSRP in FTC progression.

Materials and methods

Chemicals, reagents, and cell culture

Fetal bovine serum, penicillin/streptomycin, and trypsin were purchased from Gibco (Life Technologies, USA), and 2× iTaq SYBR Green Supermix was purchased from BioRad (Hercules, CA, USA). The following antibodies were obtained for immunoblotting: KSRP (A302-022A; Thermo Fisher), DACT2 (AB79042; Abcam), SFRP2 (12189-1-AP; Proteintech), Flag (8146; Cell Signalling), GAPDH (10494-1-AP; Proteintech), and Lamin A/C (10298-1-AP; Proteintech). The human thyroid cancer cell lines FTC-133 and FTC-236 were kindly donated by Dr. Goretzki and Dr. Simon of the University of Dusseldorf, Germany. The cells were cultured in Dulbecco’s modified Eagle’s medium/nutrient mixture F-12 medium (catalog number: 23400-21; Gibco; Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum and 1% (v/v) penicillin/streptomycin and incubated at 37 °C in a humidified incubator containing 5% CO₂.

Lentiviral infection

Lentiviral particles were generated using calcium phosphate and used to infect the human embryonic kidney cell line HEK-293T. In brief, HEK-293T cells were seeded on 10-cm culture plates the day before transfection. After 24 h, the cells were transfected with 10 μg of DNA, 10 μg of a packaging vector (pCMVΔR8.91), and 1 μg of an envelope vector (pMDG). After 16 h of incubation, the medium was replenished with fresh culture medium. After 48 h of incubation, the lentivirus-containing culture medium was centrifuged at 380 × g for 5 min to precipitate cell debris. The resultant supernatant was filtered using a 0.45-μm filter membrane. The target cells were incubated in the lentivirus-containing culture medium (supplemented with 8 μg/mL polybrene) for 48 h.

Short hairpin RNA sequences

Short hairpin RNA (shRNA) constructs against human KSRP, DACT2, and SFRP2 were purchased from the National RNAi Core Facility (Academia Sinica, Taiwan). The target sequences were as follows:

shKSRP-1: 5′-CGC CTA CTA CTC ACA CTA CTA-3′

shKSRP-2: 5′-GAC TTC AAT GAC AGA AGA GTA-3′

shDACT2: 5′-ATG GAC GAG GCA ACA TCA TAT-3′

shSFRP2: 5′-CCA CCC GAA TCT TGT AGA AAT-3′

Patients and immunohistochemistry

Thyroid tumour tissues were collected from National Taiwan University Hospital (institutional approval number: 201402068RINA). The specimens included follicular adenoma (FA), FTC, and normal thyroid tissue samples. The samples were deparaffinized and rehydrated. This was followed by antigen retrieval through incubation in 0.1 M ethylenediaminetetraacetate (pH 8.0) at a subboiling temperature for 10 min. Subsequently, the samples were incubated with 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity. After 1 h of preincubation in 3% normal horse serum, the samples were incubated with anti-KSRP antibodies overnight at 4 °C. Next, the samples were washed thrice and incubated with secondary antibodies. Subsequently, the samples were incubated with diaminobenzidine–horseradish peroxidase and counterstained with hematoxylin. After immunostaining, the tissue slides were subjected to digital imaging for cell quantification. Two pathologists, who were blinded to patient outcomes, independently analysed the results of immunohistochemical staining. Both the intensity and percentage of immunoreactive cells were recorded. Staining intensity was graded on a 4-point scale: 0 (no staining), 1+ (weak staining), 2+ (moderate staining), and 3+ (strong staining). The extent of staining was evaluated in terms of the percentage of immunoreactive cells (0%–100%). The final immunohistochemical score was calculated by multiplying staining intensity by the percentage of immunoreactive cells.

Quantitative reverse transcription polymerase chain reaction

Total RNA was isolated using the NucleoZOL reagent (Macherey-Nagel). Complementary DNA was synthesised using a PrimerScript RT kit (Takara). Real-time polymerase chain reaction (PCR) was performed using a CFX Connect real-time PCR detection system (BioRad) in accordance with the iTaq SYBR Green Supermix (BioRad) protocol. Relative gene expression was normalised against actin by using the 2^−ΔΔCt method. The primers used in this experiment are listed in Table S1.

Transwell cell migration assay

Cell migration was assessed using a modified two-chamber migration assay kit (two chambers separated by a membrane; pore size: 8 μm; Merck Millipore) in accordance with the manufacturer’s instructions. Cells were seeded (density: approximately 5 × 10⁴) into the upper chamber of the transwell apparatus and allowed to migrate to the lower chamber over a period of 24 h. The cells in the upper chamber were then removed using a cotton swab. The cells at the bottom of the membrane were fixed and stained with 0.2% crystal violet plus 20% methanol. Finally, the stained cells were quantified.

Transwell cell invasion assay

Cell invasiveness was assessed using a modified two-chamber invasion assay kit (two chambers separated by a membrane; pore size: 8 μm; Merck Millipore) in accordance with the manufacturer’s instructions. Cells were seeded (density: approximately 1 × 10⁵) into the upper chamber (coated with 40 μL of Matrigel) of the transwell apparatus and were allowed to invade the lower chamber over a period of 24 h. The cells in the upper chamber were then removed using a cotton swab. The cells at the bottom of the membrane were fixed and stained with 0.2% crystal violet plus 20% methanol. Finally, the stained cells were quantified.

Spheroid formation assay

Cells were cultured in Dulbecco’s modified Eagle’s medium/nutrient mixture F-12 medium supplemented with 10 ng/mL recombinant human epidermal growth factor and 10 ng/mL basic fibroblast growth factor with B27. After 14 days of incubation (when the average diameter of the spheroids reached 100 μm), the spheroids were counted.

Luciferase reporter assay

Cells were transfected with a mixture of the TCF-luciferase reporter vector and a Renilla luciferase plasmid (ratio: 10:1). Luciferase activity was measured using a dual-luciferase reporter assay kit (Promega) in accordance with the manufacturer’s instructions. The relative level of luciferase activity was normalised against the activity level of Renilla luciferase.

Immunoblotting

Cells were lysed in RIPA buffer and heated at 95 °C for 10 min. The cell lysate was subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis. The resultant protein bands were transferred onto a polyvinylidene difluoride membrane, and the membrane was blocked, followed by incubation with relevant primary antibodies. After washing, the membrane was incubated with horseradish peroxidase-conjugated secondary antibodies. The resultant immune complexes were visualised using an enhanced chemiluminescence substrate.

In vivo metastasis assay

FTC-133 and FTC-236 cells (2 × 10⁵ per group) were incubated in 10 µL Matrigel on ice. Male NOD-SCID male mice (age: 6–8 weeks; 6 mice per group) were obtained for animal experiments. The mouse thyroid gland was injected with FTC-133 or FTC-236 cells. Briefly, the mice are anaesthetised and their skin and subcutaneous tissues are incised horizontally, and the salivary glands are reflected superiorly. The central component of the neck was exposed, and the overlaying strap muscles were dissected away from the right thyroid using forceps. Once the right thyroid was exposed, FTC-133 and FTC-236 cells resuspended in 10 μL of serum-free medium were injected into the thyroid [19, 20]. At 7 weeks after injection, all mice were euthanised. Their lungs, brains, and thyroid glands were excised for further assessment. Luciferase activity in the excised organs was measured using an IVIS spectral imaging system (Xenogen, National Taiwan University Animal Center).

All protocols for the animal experiments were approved by the Institutional Animal Care and Use Committee of National Taiwan University’s College of Medicine and College of Public Health (approval number: 20150359).

Xenograft model establishment

FTC-133 and FTC-236 cells were suspended in PBS with 50% Matrigel and incubated on ice before xenografting. Male NOD-SCID-γ mice were subcutaneously injected with FTC-133 or FTC-236 cells. After injection, the mice were examined every week for signs of tumorigenesis.

All protocols for the animal experiments were approved by the Institutional Animal Care and Use Committee of National Taiwan University’s College of Medicine and College of Public Health (approval number: 20150359).

Statistical analysis

Data were visualised using GraphPad Prism (version 8) and SigmaPlot (version 12.0). Between-group differences were analysed using a two-tailed t-test. Data are presented herein in terms of mean ± standard deviation values. Statistical significance was set at p < 0.05. Unless otherwise stated, statistical significance is indicated as *p < 0.05 and **p < 0.01.

Results

KSRP expression is upregulated in FTC tissues

To investigate the role of KSRP in FTC development, we first examined the KSRP mRNA expression levels in FA, FTC, and normal thyroid tissues. The KSRP mRNA expression levels were higher in FA and FTC tissues than in normal thyroid tissues (Fig. 1a). Furthermore, the KSRP protein expression levels were significantly higher in FA and FTC tissues than in normal thyroid tissues (Fig. 1b, c). Notably, compared to FA tissues, the expression of KSRP in FTC reached a significantly higher level. These findings suggest that the upregulation of KSRP expression is a key event in the tumorigenesis of FTC.

Fig. 1. Expression of KSRP during the development of FTC.

a Levels of KSRP mRNA in normal thyroid, FA, and FTC tissues; the levels were measured through quantitative reverse transcription polymerase chain reaction. b Representative images of normal thyroid, FA, and FTC tissues stained with anti-KSRP antibodies. Scale bar: 50 µm. c Plot depicting the immunohistochemical scores for the expression levels of KSRP in normal thyroid, FA, and FTC tissues. The scores were calculated by multiplying staining intensity by the percentage of immunoreactive cells. KSRP K-homology-type splicing regulatory protein, FA follicular adenoma, FTC follicular thyroid cancer.

KSRP promotes FTC metastasis

To evaluate the role of KSRP in FTC, we selected a metastatic series of FTC cell lines, FTC-133 and FTC-236, to examine the KSRP expression. These two cell lines were derived from the same patient [21]: The FTC-133 and FTC-236 cells were derived from the primary tumour and lymph node metastasis, respectively [22]. The KSRP expression level was significantly higher in FTC-236 cells than in FTC-133 cells (Fig. 2a). The extents of migration and invasion were higher for FTC-236 cells than for FTC-133 cells. Similar results were obtained in a mouse model of orthotopic metastasis. Mice bearing FTC-236 tumours developed larger primary tumours than did those bearing FTC-133 tumours. Moreover, the extents of lung and brain metastases were higher in mice bearing FTC-236 tumours than in those bearing FTC-133 tumours (Fig. 2b, c). Since cancer stem cell (CSC) populations are regarded as the seeds of metastases [23], we next assessed the stemness of the FTC-133 and FTC-236 cells through a spheroid formation assay. The FTC-236 cells formed significantly more and larger spheroids than did the FTC-133 cells (Fig. 2d). To investigate the tumorigenicity of the FTC-133 and FTC-236 cells, a subcutaneous tumorigenicity assay was conducted in vivo. The FTC-236 cells formed tumours more rapidly and grew into larger tumours than did the FTC-133 cells (Fig. 2e). Thus, we hypothesised that the enrichment of CSCs and the metastatic ability of the FTC-236 cells would be influenced by KSRP expression. To determine whether KSRP could regulate the malignant phenotype of FTC cells, we performed cell migration and invasion assays in KSRP-overexpressing FTC-133 cells and KSRP-knockdown FTC-236 cells. Successful manipulation of KSRP expression was confirmed through Western blotting (Fig. 3a). KSRP knockdown significantly reduced the extent of cell migration and invasion. Conversely, KSRP-overexpression promoted cell migration and invasion (Fig. 3b). To confirm the role of KSRP in FTC metastasis, we established a mouse model of orthotopic metastasis. The extents of lung and brain metastases were lower in FTC-236-Luc cells with stable KSRP knockdown than in shControl cells (Fig. 3c, d). These findings suggest that KSRP plays a vital role in regulating the migration, invasion, and metastasis of FTC cells. Regarding cancer stemness properties, KSRP overexpression in the FTC-133 cells significantly increased both spheroid size and number compared with the parameters in the vector control cells (Fig. 3e). In contrast, the KSRP-knockdown FTC-236 cells formed fewer and smaller spheroids than did the shScramble control cells (Fig. 3f). Thus, KSRP may contribute to FTC metastasis, potentially by influencing the enrichment of CSC populations.

Fig. 2. High expression of KSRP in FTC-236 cells enhances clonogenic and tumorigenic properties.

a Left panel: KSRP expression in FTC-133 and FTC-236 cells. Right panel: Migratory and invasive abilities of FTC-133 and FTC-236 cells. b Representative bioluminescence images of the thyroid, brain, and lung tissues of mice bearing FTC-133 or FTC-236 tumours. c Levels of bioluminescence (photons/s) in the thyroid, brain, and lung tissues of mice bearing FTC-133 or FTC-236 tumours. d Tumour spheroids formed by FTC-133 and FTC-236 cells. Scale bar: 100 µm. e Tumour volume in mice subcutaneously inoculated with FTC-133 or FTC-236 cells at a density of 1 × 10³ or 1 × 10⁴ cells/mouse. The tumour volume of the FTC-236 groups with different cell numbers was compared with the comparative cell number of the FTC-133 groups, respectively. Significance *p < 0.05 and **p < 0.01. KSRP K-homology-type splicing regulatory protein, FTC follicular thyroid cancer.

Fig. 3. KSRP enhances the metastatic and clonogenic properties of FTC cells.

a Left panel: KSRP expression in FTC-133 and FTC-236 cells with altered KSRP expression. b Migratory and invasive abilities of FTC-133 and FTC-236 cells with altered KSRP expression. c Representative bioluminescence images of the thyroid, brain, and lung tissues of mice orthotopically injected with KSRP-knockdown FTC-236 cells. d Levels of bioluminescence (photons/s) in the thyroid, brain, and lung tissues of mice injected with KSRP-knockdown FTC-236 cells. e Tumour spheroids formed by KSRP-overexpressing FTC-133 cells. Scale bar: 50 µm. f Tumour spheroids formed by KSRP-knockdown FTC-236 cells. Scale bar: 100 µm. Significance: *p < 0.05 and **p < 0.01. KSRP K-homology-type splicing regulatory protein, FTC follicular thyroid cancer.

KSRP promotes the self-renewal of FTC stem cells by activating β-catenin

To elucidate the potential mechanisms underlying the KSRP-mediated promotion of FTC metastasis, we used Correlation AnalyzeR, an online database, to further explore and analyse gene sets’ correlations with KSRP [24]. KSRP was strongly correlated with stem cell–related genes (Fig. S1). Furthermore, gene set enrichment analysis (GSEA) of TCGA thyroid cancer revealed that the expression of KSRP was positively correlated with that of genes regulated by the Wnt signalling pathway (normalised enrichment score: >2.4; Fig. 4a). The expression of KSRP was also correlated with β-catenin (Pearson correlation coefficient: 0.62; Fig. 4b). We further measured the transcriptional activity of β-catenin in the FTC-133 and FTC-236 cells; the level of transcriptional activity was higher in the FTC-236 cells than in the FTC-133 cells (Fig. 4c). Because the regulation of Wnt/β-catenin signalling is reliant on the nuclear accumulation of β-catenin, we examined the nuclear and cytoplasmic fractions of the tumour cells. The expression levels of β-catenin and KSRP were significantly increased in the nuclear fraction of FTC-236 cells compared with their levels in the nuclear fraction of FTC-133 cells (Fig. 4d). To investigate the correlation between KSRP expression and Wnt/β-catenin signalling, β-catenin transcriptional activity was measured in the KSRP-overexpressing and KSRP-knockdown cells. The level of β-catenin activity was higher in the KSRP-overexpressing cells than in the vector control cells (Fig. 4e). Stable KSRP knockdown led to a significant reduction in β-catenin activity (Fig. 4f). Accordingly, we noted a reduction in the nuclear accumulation of β-catenin in the KSRP-knockdown cells (Fig. 4g, Fig. S2a) while KSRP-overexpression induced nuclear translocation of β-catenin (Fig. S2b), suggesting that KSRP regulates the transcriptional activity and nuclear translocation of β-catenin. This prompted us to investigate whether KSRP-mediated enhancement of FTC stemness and motility is correlated with the activation of β-catenin. For this investigation, β-catenin was overexpressed in KSRP-knockdown cells; the cells were then subjected to spheroid formation and cell migration/invasion assays. As expected, β-catenin-overexpression successfully rescued the spheroid-forming, cell-migrating, and cell-invasive abilities of stable KSRP-knockdown cells (Fig. 4h, Fig. S2c). Overall, these findings suggest that β-catenin serves as a key effector in the KSRP-mediated enhancement of FTC stemness and cell motility.

Fig. 4. KSRP promotes the self-renewal of FTC cells by activating β-catenin.

a Results of gene set enrichment analysis of genes related to “GO positive regulation of canonical Wnt signalling pathway” and “reactome signalling by Wnt” in The Cancer Genome Atlas stratified by KSRP expression. b KSRP expression was correlated with β-catenin expression. c Transcriptional activity of β-catenin in FTC-133 and FTC-236 cells. d Levels of β-catenin expression in the nuclear and cytoplasmic fractions of FTC-133 and FTC-236 cells. GAPDH and Lamin A/C were used as cytoplasmic and nuclear loading controls, respectively. e Transcriptional activity of β-catenin in KSRP-overexpressing FTC-133 cells. f Transcriptional activity of β-catenin in KSRP-knockdown FTC-236 cells. g Levels of β-catenin expression in the nuclear and cytoplasmic fractions of KSRP-knockdown FTC-236 cells. GAPDH and Lamin A/C were used as cytoplasmic and nuclear loading controls, respectively. h Tumour spheroids formed by KSRP-knockdown FTC-236 cells. Scale bar: 100 µm. Significance: *p < 0.05 and **p < 0.01. KSRP K-homology-type splicing regulatory protein, FTC follicular thyroid cancer.

KSRP suppresses the inhibitors of Wnt signalling to activate β-catenin signalling and enhance FTC stemness

To elucidate the molecular mechanisms underlying the KSRP-mediated regulation of β-catenin activation, we evaluated the expression of genes involved in Wnt signalling. Although we observed changes in the levels of some ligands and receptors of Wnt signalling in the KSRP-knockdown FTC-236 cells, these changes were inconsistent with the activation of Wnt signalling. Quantitative reverse transcription PCR revealed that the DACT2 and SFRP2 expression levels were significantly upregulated in the KSRP-knockdown FTC-236 cells (Fig. 5a). Similarly, depletion of KSRP could induce the protein expressions of DACT2 and SFRP2 (Fig. 5b). Next, we investigated whether KSRP regulates FTC metastasis and stemness by altering the expression levels of DACT2 and SFRP2. For this investigation, shRNAs against DACT2 and SFRP2 were introduced into the KSRP-knockdown cells; cell migration, cell invasion, and spheroid formation assays were then performed. Successful suppression of either DACT2 or SFRP2 in the KSRP-knockdown cells was confirmed through Western blotting (Fig. 5c). The suppression of these inhibitors rescued the migratory, invasive, and stemness properties conferred by KSRP knockdown (Fig. 5d, e). Notably, the rescue effects were stronger when both DACT2 and SFRP2 were suppressed than when either of them was suppressed (Fig. 5d, e). These findings suggest that Wnt inhibitors mediate the role of Wnt/β-catenin signalling in the KSRP-mediated promotion of FTC metastasis and enhancement of FTC stemness. Overall, our findings suggest that KSRP regulates the transcriptional activation and nuclear translocation of β-catenin in FTC cells by negatively regulating DACT2 and SFRP2, thereby promoting cell migration, invasion, and stemness.

Fig. 5. KSRP negatively regulates the inhibitors of Wnt signalling to activate β-catenin signalling and promote FTC stemness.

a Heat map depicting the relative expression levels of genes involved in Wnt antagonists, those expressing ligands, and those expressing receptors in KSRP-knockdown FTC-236 cells. The levels were measured through quantitative reverse transcription polymerase chain reaction. Red indicates upregulation, whereas green indicates downregulation. Colour intensity represents the values of mRNA expression. b Immunoblotting results of DACT2 and SFRP2 expression in KSRP-knockdown FTC-236 cells. c Immunoblotting results of DACT2 and SFRP2 suppression in KSRP-knockdown FTC-236 cells. d Left panel: Migratory abilities of KSRP-knockdown FTC-236 cells with DACT2 and SFRP2 suppression. Right panel: Invasion abilities of KSRP-knockdown FTC-236 cells with DACT2 and SFRP2 suppression. e Tumour spheroids formed by KSRP-knockdown FTC-236 cells with DACT2 and SFRP2 suppression. Scale bar: 50 µm. Significance: **p < 0.01 when compared to shControl cells, ## p < 0.01 when compared to shKSRP cells, § p < 0.05, §§ p < 0.01 when compared to either shKSRP-shDACT2 or shKSRP-shSFRP2 cells by two-tailed Student’s t-test. KSRP K-homology-type splicing regulatory protein, FTC follicular thyroid cancer.

Discussion

In this study, we have unveiled the aberrant expression of KSRP and its profound association with Wnt/β-catenin signalling in the progression of FTC. Our findings not only identify KSRP as an oncoprotein that significantly enhances FTC’s clonogenic and tumorigenic ability but also shed light on its potential role in enhancing β-catenin transcriptional activity by negatively regulating Wnt inhibitors, DACT2, and SFRP2. The implications of the KSRP-DACT2/SFRP2-β-catenin axis are far-reaching, leading to the upregulation of cell migratory and invasive ability in FTC cells. This underscores the potential of KSRP as a promising therapeutic target for FTC treatment.

Changes in the expression and activity of RBPs could result in anomalous expression and function of oncogenes or tumour suppressors, leading to the development of tumours. Because RBPs regulate the fate and function of RNA, several studies have explored the RNA targets of RBPs and the role of these proteins in regulating cancer progression. Among the well-established oncogenic pathways, Wnt/β-catenin signalling has been associated with various RBPs. In colorectal cancer, an RBP called DDX5 interacts with NEAT1 to enhance its stability, activate Wnt/β-catenin signalling, and promote tumorigenesis [25]. Furthermore, in bladder cancer, another RBP called HnRNPA1 interacts with the long noncoding RNA BCYRN1 to upregulate Wnt5A expression and activate Wnt/β-catenin signalling [26]. In gastric cancer, the RBP Lin28B directly interacts with the 3′UTR of NRP1 mRNA to increase its stability and expression, thereby activating downstream Wnt/β-catenin signalling and enhancing cancer stemness [27]. By contrast, in lung adenocarcinoma, RNA-binding motif protein 10 (RBM10) inhibits cell proliferation, metastasis, and epithelial–mesenchymal transition by disrupting the association between β-catenin and TCF/LEF, thereby inactivating Wnt/β-catenin signalling [28]. Therefore, understanding how the RBPs function, dissecting their regulatory networks, and identifying robust biomarkers and therapeutic targets could be the main focus of RBP-related studies.

Alterations in Wnt/β-catenin signalling occur during the late stage of thyroid cell transformation and the early stage of thyroid cancer progression [16, 29, 30]. Mutations occur in several components of the Wnt pathway, such as the scaffold proteins APC, Axin, and β-catenin. In the FTC, β-catenin is primarily localised on the plasma membrane [31]. However, in poorly differentiated and undifferentiated thyroid cancer, β-catenin may be mislocalized because of mutations in CTNNB1 [32]. In addition to mutation-induced dysregulation of β-catenin, other posttranslational modifications may induce β-catenin activation in thyroid cancer [33]. The PI3K/AKT pathway mediates the activation of the Wnt/β-catenin pathway [34]. AKT regulates the cytoplasmic stabilisation of β-catenin; this phenomenon is dependent on the phosphorylation of GSK3β [35]. With the AKT/MAPK-dependent inhibition of GSK3β, the nuclear level of β-catenin could be increased, thus further promoting its transcriptional activity. The functional associations of β-catenin with MAPK and PI3K/AKT signalling could promote FTC cell proliferation and DNA synthesis [36]. In the present study, KSRP was determined to be a novel regulator of β-catenin localisation and transcriptional activity in FTC cells. Future studies should investigate how KSRP influences the subcellular localisation of β-catenin. Notably, the regulation of KSRP expression reduced β-catenin’s transcriptional activity and inhibited Wnt/β-catenin signalling; these findings can be probed further to develop a new therapeutic strategy for thyroid cancer.

The association of KSRP with Wnt/β-catenin signalling has been observed in other cancers with various molecular regulatory mechanisms. For example, a previous study demonstrated KSRP to interact with dishevelled protein 3 (Dvl3) and suppress Wnt/β-catenin signalling by destabilising CTNNB1 mRNA [37]. Another recent study identified a KSRP-binding compound that can suppress distant metastasis in colorectal cancer by degrading Dvl2 [38]. KSRP was also reported to promote the degradation of β-catenin mRNA [39]. These findings indicate that the association of KSRP dysregulation with β-catenin can sustainably regulate tumour development. Our study revealed a novel regulatory effect of KSRP on Wnt/β-catenin signalling. KSRP inhibited the expression of the Wnt inhibitors DACT2 and SFRP2, thus enhancing the nuclear translocation and transcriptional activity of β-catenin. These effects could subsequently activate Wnt/β-catenin signalling, promote FTC metastasis, and enhance FTC cell stemness.

KSRP interacts primarily with its target mRNA to regulate gene stability through the degradation of mRNA and thus the inhibition of translation. Considering this observation, we investigated whether KSRP-specific AREs are present in the 3′UTRs of DACT2 and SFRP2. Our investigation revealed several areas in the 3′UTRs of DACT2 and SFRP2 that could potentially interact with KSRP (data not shown). We therefore explored whether inhibition of KSRP could increase DACT2 and SFRP2 mRNA stability. The result in Fig. S3 supports the hypothesis that KSRP would regulate the degradation of DACT2 and SFRP2 mRNAs by binding the corresponding AREs in their 3′UTRs. We further investigated whether KSRP could inhibit the translation of these mRNAs by promoting the maturation of corresponding miRNAs. Using different algorithms to search miRNA databases, we identified candidate miRNAs that are commonly predicted to target DACT2 and SFRP2 mRNAs. Subsequently, we compared the candidate miRNAs with established KSRP-regulated miRNAs. The results revealed a well-known KSRP-regulated miRNA, let-7 (data not shown). KSRP promotes the biogenesis of let-7 by interacting with the terminal loop of let-7 precursors [40, 41]. Thus, we hypothesised that KSRP would regulate the maturation of let-7 to inhibit the expression of downstream Wnt inhibitors. However, these analyses only showed the possible molecular mechanisms by which KSRP might regulate DACT2 or SFRP2 expression. Further experiments should be conducted to validate these findings.

In conclusion, because of its oncogenic properties and its association with Wnt/β-catenin signalling, KSRP holds promise as a reliable prognostic factor for FTC. Future studies should further explore the posttranscriptional mechanisms through which KSRP suppresses the expression of DACT2 and SFRP2 in order to identify novel therapeutic strategies for FTC and other cancers.

Author contributions

Ke-Fan Pan: Study design and Writing - Original draft preparation. Han-Lin Chou: Methodology and Investigation. Wei-Li Wang: Investigation. Bo-Rong Chen: Investigation. Michael Hsiao: Resources. Kuo-Tai Hua: Writing – Review & editing, Supervision, and Project administration. Ming-Hsun Wu: Supervision and Project administration. All authors have read and approved the final version of the manuscript.

Funding

This study was supported by the National Science and Technology Council, Taiwan (grant number: 111-2628-B-002-023-MY3), Wan Fang Hospital (grant number: 112-wf-eva-04, 113-wf-swf-06), and Far Eastern Memorial Hospital (grant number: 114-FTN0014).

Data availability

All data supporting the findings of this study are available from the corresponding author upon reasonable request.

Competing interests

The authors declare no competing interests.

Ethical approval and consent to participate

This study was approved by the Institutional Review Board of National Taiwan University Hospital (institutional approval number: 201402068RINA), and written informed consent was obtained from all study participants enrolled, and all participants consented to participate in this work. Ethics approval for the animal studies was obtained from the Institutional Animal Care and Use Committee of National Taiwan University’s College of Medicine and College of Public Health (approval number: 20150359), and all procedures were performed in accordance with established guidelines.

Consent for publication

All authors were aware of the submission of the present study, and patients signed informed consent regarding publishing their data.

Supplementary information

The online version contains supplementary material available at 10.1038/s41416-025-03142-x.