TWF2 Drives Tumor Progression and Sunitinib Resistance in Renal Cell Carcinoma through Hippo Signaling Suppression

Abstract

Renal cell carcinoma (RCC) remains a formidable clinical challenge, characterized by a high propensity for metastasis and the frequent emergence of intrinsic or acquired resistance to targeted therapies. However, the molecular mechanisms underlying sunitinib resistance and tumor progression in RCC are not fully understood. This study aims to identify Twinfilin actin-binding protein (TWF2) as a key mediator of tumor aggressiveness and therapeutic resistance. TWF2 expression is markedly upregulated in RCC cells, particularly in sunitinib-resistant subtypes, and significantly associated with poor prognosis and therapeutic nonresponsiveness. Functional analyses demonstrate that TWF2 promotes RCC cell invasion, migration, metastasis, and sunitinib resistance by inhibiting the Hippo signaling. Mechanistically, TWF2 interacts with Yes-associated protein (YAP) via the binding residues: TWF2 M99 and YAP M225. By competitively displacing large tumor suppressor kinase 1, TWF2 prevents YAP ubiquitination and degradation, leading to its stabilization and subsequent nuclear translocation. Mutation of the M99 residue abolishes the tumor-promoting activity of TWF2. Furthermore, salvianolic acid E is identified as a small-molecule inhibitor of the TWF2-YAP interaction, and synergistically enhances sunitinib efficacy in RCC cell lines and patient-derived xenograft models. These findings highlight TWF2 as a promising therapeutic target for overcoming drug resistance in RCC.

Keywords: Hippo signaling; TWF2; renal cell carcinoma; sunitinib resistance; tumor progression.

Figure 1

TWF2 is identified as a master regulator for RCC drug resistance and tumor progression. A) Schematic diagram illustrating the establishment of sunitinib‐resistant (786‐O‐R, 769‐P‐R) and sunitinib‐sensitive (786‐O‐S, 769‐P‐S) RCC cell lines. B) Flow chart of screening key regulators involved in RCC drug resistance and tumor progression. C) Venn diagram displaying the intersection of five datasets: MS Top100, RNA‐seq Top100, TCGA–KIRC OS (genes upregulated in tumors vs normal tissues), TCGA–KIRC Metastasis (genes upregulated in metastatic vs primary tumors), and TCGA–KIRC Prognosis (genes associated with poor prognosis). This analysis identified six overlapping genes: CTHRC1, TWF2, COL6A3, SLC38A5, IFI44, and OASL. D) Volcano plot of differentially expressed genes in sunitinib‐resistant versus sunitinib‐sensitive 786‐O cells based on transcriptomic analysis. Genes upregulated in resistant cells are shown in red, and downregulated genes in blue. E) Relative TWF2 mRNA expression in paired tumor and adjacent normal tissues from the SYSU ccRCC cohort. F) Representative western blot (left) and the corresponding statistical analysis (right) of TWF2 protein expression levels in twelve paired ccRCC tumors (T) and adjacent normal tissues (N). G) Representative western blot (top) and mRNA expression analysis (bottom) of TWF2 in sunitinib‐sensitive (S) and ‐resistant (R) 786‐O cell. H) Representative immunohistochemical (IHC) staining showing TWF2 expression in ccRCC tumors and adjacent normal tissues. I) Representative IHC images of TWF2 expression in ccRCC tissues from patients classified as responders or nonresponders to sunitinib treatment. J) Overall survival (OS) in ccRCC patients with low (n = 60) or high (n = 60) TWF2 expression. K) Disease‐free survival (DFS) in ccRCC patients with low (n = 60) or high (n = 60) TWF2. Data are presented as means ± SD and are analyzed by Student's t‐test (E–G) or log‐rank test (J, K). **p < 0.01; ***p < 0.001.

Figure 2

TWF2 promotes RCC progression and contributes to sunitinib resistance. A,B) Colony formation assay of TWF2‐knockdown 769‐P cells (A) and TWF2‐overexpressing 786‐O cells (B). C,D) Transwell assays evaluating migration and invasion of TWF2‐knockdown 769‐P cells (C) and TWF2‐overexpressing 786‐O cells (D). E) Representative bioluminescence images (left) and the corresponding statistical analysis (right) of lung metastases in mice injected with TWF2‐overexpressing 786‐O cells. F) Representative gross of lung images and hematoxylin–eosin (H&E) staining of metastatic lesions (left), with statistical analysis (right), from the metastasis model shown in (E). G,H) Flow‐cytometry‐based apoptosis analysis of TWF2‐knockdown (G) and TWF2‐overexpressing (H) cells and their respective controls following treatment with sunitinib (2 µm) or DMSO for 60 h. I) Relative cell viability (left) and resistance index (right) in TWF2‐knockdown and control 786‐O‐R cells treated with sunitinib, based on Cell Counting Kit‐8 (CCK‐8) assays. J) Representative bioluminescence images (left) of orthotopic tumors formed by TWF2‐knockdown or control 786‐O‐R cells treated with sunitinib, with the corresponding statistical analysis (right). K) Gross images of orthotopic renal tumors (left) and the corresponding tumor weights (right) from (J). 786‐O‐R: sunitinib‐resistant 786‐O cells. Data are presented as means ± SD and are analyzed by Student's t‐test (A–K) or one‐way ANOVA (J). ns, no significance; **p < 0.01; ***p < 0.001.

Figure 3

TWF2 inhibits Hippo signaling through enhancing dephosphorylation and nuclear translocation of YAP. A) KEGG pathway enrichment analysis of differentially expressed genes following TWF2 knockdown in 769‐P cells, indicating altered Hippo signaling. B) Western blot analysis of Hippo signaling pathway components in TWF2‐knockdown 769‐P cells and TWF2‐overexpressing 786‐O cells. C) Western blot analysis of Hippo‐pathway‐associated proteins in sunitinib‐sensitive (S), sunitinib‐resistant (R), and TWF2‐knockdown sunitinib‐resistant (R + shTWF2) 786‐O cells. D) Representative IHC images (left) and correlation analysis (right) between TWF2 and YAP based on their expression in human ccRCC tissues. E) Representative immunohistochemical images of TWF2 and YAP in human ccRCC tissues from sunitinib‐responsive and nonresponsive patients. F) Representative immunofluorescence images (left) and quantification (right) showing YAP subcellular localization following TWF2 knockdown. G) Representative immunofluorescence images (left) and quantification (right) showing YAP subcellular localization following TWF2 overexpression. H,I) Western blot analysis of nuclear and cytoplasmic YAP distribution in TWF2‐knockdown 769‐P cells (H) and TWF2‐overexpressing 786‐O cells (I). J) Western blot analysis of nuclear and cytoplasmic YAP levels in the indicated cell lines. Data are presented as means ± SD and are analyzed by Student's t‐test (F, G) or Pearson correlation test (D). ***p < 0.001.

Figure 4

Identification of YAP as a binding partner of TWF2 in RCC cells. A) Representative silver‐stained gel showing proteins specifically immunoprecipitated by the TWF2 protein. B) LC–MS/MS analysis identified YAP as a TWF2‐interacting protein in Flag‐TWF2 immunoprecipitated samples. C) Co‐IP of overexpressed Flag‐tagged TWF2 and endogenous YAP in 786‐O cells. D) Co‐IP of endogenous TWF2 and YAP in 769‐P cells. E) BLI assay measuring the binding affinity between purified TWF2 and YAP. F) Representative immunofluorescent images showing colocalization of endogenous TWF2 and YAP in 786‐O (top) and 769‐P (bottom) cells. G) Schematic diagrams of wild‐type YAP (full‐length, 1–504) and its truncation mutants. H) Co‐IP analysis demonstrating interactions between TWF2 constructs and YAP. 786‐O cells transfected with Flag‐TWF2 and the indicated YAP constructs with HA‐tag were subjected to immunoprecipitation using an anti‐Flag antibody, followed by immunoblotting with anti‐HA and anti‐Flag antibodies.

Figure 5

TWF2 binding to YAP enhances YAP stability by competing with LATS1. A,B) Quantitative RT‐PCR analysis of YAP mRNA levels in TWF2 knockdown 769‐P cells (A) and TWF2‐overexpressing 786‐O cells (B). C–E) Western blot analysis of YAP protein stability using CHX chase assays in TWF2 knockdown 769‐P cells (C), TWF2‐overexpressing 786‐O cells (D), and sunitinib‐sensitive versus ‐resistant 786‐O cells (E). Cells were treated with 20 µg mL⁻¹ CHX for the indicated time points. Statistical analyses are shown in the right panels. F) Western blot showing YAP protein levels in TWF2 knockdown and control 769‐P cells following treatment with chloroquine (CQ, 10 µm) or MG132 (10 µm) for 12 h. G,H) Co‐IP assays showing increased YAP ubiquitination in TWF2‐deficient 769‐P cells (G) and decreased YAP ubiquitination in TWF2‐overexpressing 786‐O cells (H). I,J) Co‐IP assays with anti‐YAP antibody showing the precipitated LATS1 levels in TWF2‐knockdown 769‐P cells (I) and TWF2‐overexpressing 786‐O cells (J). K,L) Co‐IP assays with anti‐YAP antibody showing p‐YAP levels in TWF2 knockdown 769‐P cells (K) and TWF2‐overexpressing 786‐O cells (L). M) Western blot showing the effect of increasing LATS1 and TWF2 expression on exogenous YAP levels in 293T cells transfected with the indicated plasmids. Data are presented as means ± SD and are analyzed by Student's t‐test (A, B) or one‐way ANOVA (C–E). ns, no significance; ***p < 0.001.

Figure 6

TWF2 binds to YAP in a Met99‐dependent manner. A) Structure model of the TWF2–YAP complex generated using the HDOCK server, indicating predicted binding sites. B) Four amino acids in TWF2 predicted to mediate binding were individually mutated to alanine. The interaction between mutant TWF2 and YAP was assessed using Co‐IP in 293T cells using an anti‐Flag antibody. C) Co‐IP assay showing the effect of the TWF2 M99A mutation on its interaction with YAP in 786‐O cells. D) Four YAP residues predicted as TWF2‐binding sites were individually mutated to alanine. Their interaction with TWF2 was evaluated using Co‐IP in 293T cells using an anti‐HA antibody. E) Western blotting analysis of p‐YAP and Hippo signaling components in 786‐O cells overexpressing either Flag‐tagged wild‐type TWF2 or the M99A mutant. F) Co‐IP assays with anti‐YAP antibody showing the level of coprecipitated LATS1 in cells overexpressing wild‐type or M99A mutant TWF2. G) Co‐IP assays with anti‐YAP antibody detecting the level of coprecipitated p‐YAP in 786‐O cells overexpressing wild‐type or M99A mutant TWF2. H) YAP ubiquitination levels in 786‐O cells cotransfected with His‐ubiquitin and either Flag‐tagged wild‐type TWF2 or the M99A mutant. Cell lysates were subjected to IP with anti‐YAP antibody. oeWT, overexpression of wild‐type TWF2; oeM99A, overexpression of TWF2 M99A mutant.

Figure 7

M99A mutation inhibits RCC progression and sunitinib resistance in vitro and in vivo. A) CCK‐8 proliferation assays measuring cell proliferation in 786‐O cells overexpressing wild‐type TWF2, the M99A mutant, or control vector. B) Colony formation assay comparing the clonogenic potential of 786‐O cells overexpressing wild‐type TWF2, the M99A mutant, or control vector. C) Wound healing assay evaluating the migratory capacity of 786‐O cells overexpressing wild‐type TWF2, the M99A mutant, or control vector. D) Transwell assays assessing migration and invasion of 786‐O cells overexpressing wild‐type TWF2, the M99A mutant, or control vector. E) Representative bioluminescence images (left) and statistical analysis (right) of lung metastases in a nude mouse model established using the indicated 786‐O cells (n = 5 per group). F) Representative images of lungs with metastatic nodules and H&E‐stained sections of metastatic lesions (left), with the corresponding quantification (right) from the metastasis model in (E). G) Relative cell viability (left) and resistance index (right) of 786‐O cells overexpressing control vector, wild‐type TWF2, or M99A mutant following sunitinib treatment, measured using CCK‐8 assay. H) Representative images of tumors (top) and tumor weights (bottom) from nude mice bearing sunitinib‐sensitive 786‐O tumors overexpressing control vector, wild‐type TWF2, or M99A mutant (n = 5 per group) following sunitinib treatment. Data are presented as means ± SD and are analyzed by one‐way ANOVA (A) or Student's t‐test (B–H). ns, no significance; **p < 0.01; ***p < 0.001.

Figure 8

Salvianolic acid E disrupts the TWF2–YAP interaction and enhances sunitinib efficacy in RCC PDX models. A) Flowchart of virtual screening for small‐molecule inhibitors targeting TWF2–YAP interaction. B) Chemical structure of Sal E. C) 3D structural model of the TWF2–Sal E binding interface. The docking score of Sal E for TWF2 is −9.619. D) 2D interaction map depicting the binding between Sal E and TWF2. E) ITC analysis quantifying the binding affinity between TWF2 and Sal E. The fitted curve yields a dissociation constant (K _d) of 16.7 µm. F) Co‐IP analysis demonstrating disruption of the TWF2–YAP interaction in 786‐O‐R cells treated with increasing concentrations of Sal E. G) Western blot analysis of Hippo signaling components in 786‐O‐R cells treated with Sal E. H) Relative cell viability (left) and resistance index (right) of 786‐O‐R cells treated with Sal E or vehicle in combination with sunitinib, as determined using CCK‐8 assay. I–K) Colony formation (I), migration (J), and invasion (K) assays in 769‐P cells treated with Sal E or vehicle. L) Experimental timeline for RCC PDX mouse model treatment with Sal E, sunitinib, or both. Day −21 indicates the day of PDX tumor implantation. M,N) Tumor growth curves (M) and tumor weights (N) in PDX mice treated with vehicle, Sal E, sunitinib, or combination therapy (n = 5 per group). O) TUNEL staining showing apoptotic cells in tumors from the indicated treatment groups. P,Q) Representative IHC staining for YAP and p‐YAP (P) and the corresponding quantification of IHC scores (Q) in tumors from treated PDX mice. Sal E, Salvianolic acid E; 786‐O‐R, sunitinib‐resistant 786‐O cells. Data are presented as means ± SD and are analyzed by Student's t‐test (H–K, N, Q) or one‐way ANOVA (M). *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 9

Schematic model illustrating the role of TWF2 in sunitinib resistance and RCC progression. In sunitinib‐resistant and high grade malignant RCC cells, TWF2 is upregulated and interacts with YAP, competing with LATS1 and thereby preventing YAP ubiquitination and degradation. Stabilized YAP translocates into the nucleus and activates transcription of target genes, promoting RCC proliferation, metastasis, and drug resistance. In sunitinib‐sensitive and low grade malignant RCC cells, reduced TWF2 expression permits YAP phosphorylation and subsequent degradation, limiting transcriptional activity. The small‐molecule compound Sal E disrupts the TWF2–YAP interaction and enhances the therapeutic efficacy of sunitinib in RCC.