The Hippo pathway in clear cell renal cell carcinoma (ccRCC): a nexus with the VHL disruption?

Highlights

• •First comprehensive Hippo-VHL nexus analysis in ccRCC tumorigenesis.
• •3p loss creates unique 'double hit' disrupting VHL and RASSF1 pathways.
• •LATS1/2 convergence creates self-reinforcing HIF-Hippo oncogenic loop.
• •Hippo components serve as prognostic biomarkers for patient stratification.
• •Complements VHL-centric view, positions Hippo as emerging therapeutic target.

Abstract

Background

Clear cell renal cell carcinoma (ccRCC) represents 70% of kidney cancers, with 20–50% recurrence risk after surgery. Despite therapeutic advances, no reliable biomarkers have been identified for patient stratification or treatment response prediction. While VHL gene alterations are well-established in ccRCC pathogenesis, the role of the Hippo pathway remains underexplored despite ample evidence of its involvement.

Objective

This review synthesizes current knowledge on Hippo pathway alterations in ccRCC and examines its crosstalk with the VHL/HIF axis, identifying potential biomarkers and therapeutic targets.

Strategy

We comprehensively analyzed literature on Hippo pathway components in ccRCC, focusing on molecular mechanisms, clinical correlations, and interactions with VHL signaling.

Results

Multiple Hippo pathway alterations characterize ccRCC: RASSF1A hypermethylation, NF2 mutations (particularly in aggressive variants), SAV1 downregulation associated with 14q loss, and LATS1/2 methylation-mediated inactivation. These changes result in YAP/TAZ nuclear accumulation and oncogenic transcription. Importantly, chromosome 3p loss simultaneously disrupts both VHL and RASSF1, creating a unique double-hit scenario. The VHL-Hippo crosstalk operates through multiple mechanisms: HIF-induced GPRC5A and VEGFR signaling inhibit LATS1/2 phosphorylation, promoting YAP/TAZ activation, while active YAP/TAZ enhances pro-angiogenic gene transcription, amplifying hypoxic responses. Low expression of RASSF1A, SAV1, and LATS1/2, coupled with high YAP/TAZ activity, correlates with advanced tumor stage, higher grade, and poorer survival.

Conclusions

The Hippo pathway represents a critical yet underappreciated dimension of ccRCC biology, offering promising biomarkers for risk stratification and novel therapeutic targets. The Hippo-VHL nexus presents multiple intervention points that could enhance current treatment.

Graphical abstract

Introduction

With 434,840 new cases worldwide in 2022 [1] and 155,953 deaths, the incidence of kidney cancer is on the rise [2]. Renal cell carcinoma (RCC) accounts for the vast majority (90 %) of kidney cancers while 70 % of all RCC are of the clear-cell type (ccRCC) [3].

In the absence of metastases, the preferred treatment is surgery with partial or total nephrectomy [4,5]. However, there is a 20 to 50 % risk of recurrence at 5 years depending on the pTNM stage (pathological Tumor Node Metastasis) [[6], [7], [8]]. As such, Pembrolizumab immunotherapy (IO) is indicated for patients with a pT3, pT4 or pT2 stage associated with an ISUP 4 histopronostic score (International Society of Urological Pathology) in order to reduce the relative risk of recurrence to around 30 % [9]. In the event of recurrence, or when there is synchronous metastasis detected (around 20 % of cases), treatment is based on a combination of dual immunotherapy (IO – IO) or immunotherapy combined with a tyrosine kinase inhibitor (IO – TKI) [4,5]. While therapeutic innovations and the development of predictive algorithms using artificial intelligence, such as UROPREDICT [10] looks promising, there are currently no predictive biomarkers for recurrence or therapeutic response in adjuvant or metastatic situations in ccRCC. In contrast to other cancers, it is therefore very difficult to personalize follow-up and adjuvant or metastatic treatment. Furthermore, despite improved response rates with IO-TKI combinations, acquired resistance remains inevitable. The development of novel combined therapies could help overcome these resistances. However, due to extensive crosstalk between signaling pathways, VEGF blockade may paradoxically exacerbate tumor hypoxia, which in turn could activate YAP through the HIF-GPRC5A axis, thereby maintaining tumor survival and angiogenesis independently of VEGF and contributing to therapeutic resistance [11].

The Hippo pathway, a pathway altered in most human cancers [12] is a promising candidate for such biomarkers as over 90 % of ccRCC cases are associated with the loss of the short arm of chromosome 3 (3p) which contains the locus of RASSF1 (Ras Associated Domain Family 1), one of the key upstream regulators of the Hippo signaling pathway [13]. Mutations in the Von Hippel-Lindau (VHL) gene are also associated with 3p loss [14]. Moreover, several pieces of evidence support a pathophysiological link between the VHL and the Hippo pathways in ccRCC.

The aim of this review is to synthesize current knowledge on the involvement of the Hippo pathway and its connection with VHL alterations in ccRCC, providing a comprehensive summary of the available literature and identifying key mechanisms that promote carcinogenesis and therapeutic resistance.

VHL in ccRCC

TheVHL gene, located at 3p25–26 is classified as a tumor suppressor gene. It was first identified following studies of patients with Von Hippel Lindau disease, a hereditary cancer syndrome that predisposes particularly to ccRCC. Tumorigenesis of ccRCC follows a double-hit sequence involving allele deletion with concomitant contralateral alteration of the VHL gene [19]. Somatic mutations of the VHL gene are identified in 30 to 60 % of ccRCCs. Loss of the second allele is most often due to promoter hypermethylation or loss of heterozygosity (3p LOH) [20]. Inactivation of VHL leads to activation of the HIF pathway, which is a driver of oncogenesis in ccRCC [21]. Indeed, loss of function of the VHL protein initiates tumor development and contributes to disease progression [14]. However, additional genetic alterations in clonal tumor populations further enhance their capacities for progression and metastasis. HIF (hypoxia inducible factors protein) is a heterodimeric transcription factor of the basic Helix-Loop-Helix (bHLH) family regulating over a hundred genes [22]. There are three isoforms: HIF-1, HIF-2 and HIF-3. Each isoform consists of an α subunit (HIF-1α, HIF-2α, or HIF-3α) paired with a common β-subunit, encoded by respectively the ARNT1 and ARNT2 genes. HIF-1 is ubiquitous while HIF-2 is tissue specific [23]. HIF is particularly active in the angiogenesis pathways (VEGF and PDGF pathway), on glucose metabolism and transport (GLUT1 - Glucose Transporter Protein Type 1), on glycolysis (6-phosphofructose-2-kinase), pH control (CA9 - Carbonic Anhydrase 9), on cell proliferation (and TGF pathway) and on erythropoiesis (EPO).

In normoxic conditions, the HIF-1α or HIF-2 α subunits are hydroxylated by prolyl-hydroxylases (PHD 1 to 4), which requires dioxygen. HIF-1α and HIF-2α are then polyubiquitinated by the VHL protein and its complex, before being directed to and degraded by the 26S proteasome. Thus, the functional loss of VHL and/or hypoxia allows HIF α to no longer be degraded by the proteasome. It can then bind to HIF β and act as a pro-tumor and pro-angiogenic transcription factor.

The respective roles of HIF-1α and HIF-2α in renal cell carcinogenesis remain unclear. Loss of chromosome 14q (harboring HIF-1α) in patients with ccRCC appears to be correlated with lower survival [24] and faster metastatic progression [25]. However, a recent study showed that HIF-1α was essential for cell proliferation and development [21], whereas HIF-2α appeared to play a more limited role in this context. Nevertheless, the Food and Drug Administration (FDA) approved the marketing of Belzutifan (MK-6482), a selective HIF-2α inhibitor, for the treatment of VHL-mutated ccRCC in 2021 [26]. This underscores the importance of the hypoxic environment as a key feature in the development of ccRCC. Currently, these signaling pathways are targeted by TKI treatments. However, such drugs exert systemic effects that can lead to adverse events, including cytopenias, colitis and cutaneous xerosis [26]. Thus, identifying agents that directly target factors overexpressed in hypoxic conditions could be a new avenue of therapeutic research. Among these actors linked to hypoxia, the Hippo signaling pathway and its effectors YAP1 and TAZ could be good candidates.

The Hippo pathway and ccRCC

The Hippo pathway and kidney

The Hippo signaling pathway is a highly conserved signaling pathway throughout evolution and is involved in organogenesis and the control of cell proliferation and death [27].

This pathway is composed of three groups of proteins: upstream regulators, a core of kinases whose canonical functions is to control the activation of two transcription cofactors: Yes-Associated Protein-1 (YAP1) and transcriptional co-activator with PDZ-binding motif (TAZ) through phosphorylation. When active (dephosphorylated) in the nucleus, YAP1/TAZ associate with other transcription factors (TEAD, RUNX) to promote the transcription of genes leading to cellular development [27] (Fig. 1).

Fig. 1.

The Hippo pathway.

The Hippo pathway can be broadly divided into three main components: upstream regulators (which includes NF2 and RASSF members), a core kinase cascade (MST1/5, NDR) and downstream effectors, primarily the transcriptional co-activators YAP1 and TAZ.

The active (phosphorylated) Hippo kinases phosphorylate YAP1 and TAZ [15,16]. As a result, YAP1/TAZ is sequestered in the cytoplasm and subsequently degraded by the proteasome. When Hippo kinases are inactive (either physiologically or due to an epigenetic or genetic abnormality), the YAP1/TAZ are translocated to nucleus [17], interact with transcription factors such as TEAD [18] and lead the expression of many genes controlling cell proliferation, survival, differentiation and migration/invasion.NF2: merlin; Rassf: Ras associated domain family, MAP4Ks: Mitogen-activated protein kinase; MST1/5: mammalian sterile 20‑like kinase; SAV1: Salvador Family WW Domain Containing Protein 1; NDR1/2: nuclear dbf2‑related kinase; LATS1/2: large tumour suppressor 1/2; MOB1: Mps One Binder Kinase Activator-Like 1; YAP, Yes‑Associated Protein; TAZ: Transcriptional coactivator with a PDZ‑binding domain; TEAD: TEA DNA‑binding protein.

In kidney, the Hippo pathway is involved in the glomerular and lower urinary tract embryonic development as well as the maintenance of podocyte homeostasis, the integrity of the glomerular filtration barrier, regulation of renal tubular cyst growth, renal epithelial injury in diabetes, and renal fibrogenesis [28].

While alterations of the Hippo pathway are reported in most human cancers [12], this signaling pathway has been under-studied in ccRCC. A recent study based on TCGA (The Cancer Genome Atlas) database demonstrated that contrary to papillary and chromophobe subtype cancers, low Hippo pathway activity is associated with poor clinical outcomes in ccRCC [29].

Hippo regulators’s alterations in the ccRCC

RASSF

Ras associated domain family (RASSF) consists of ten proteins presenting RA domains [25]. RASSF are protein scaffolds and can be divided into two groups: C-terminal (C-) RASSFs 1 to 6 and N-terminal (N-) RASSFs 7 to 10 [30]. In addition to the RA domain, C-RASSFs also contain helical Salvador-Rassf-Hippo (SARAH) motifs that can directly bind with the kinases MST1 and MST2 (mammalian sterile 20-like kinases 1 and 2) [[31], [32], [33]]. N-RASSFs lack the SARAH motif, and it is unclear whether they play a role in the Hippo pathway [30].

Epigenetic hypermethylation of the promoter of RASSF genes, leads to silenced RASSF protein expression. In a cohort of 179 patients, Kawai et al. have shown that high levels of RASSF1 gene methylation are independently associated with decreased survival in ccRCC patients (p = 0.0053) [34].

Potential involvement of RASSF1A in ccRCC was highlighted through the comparison of healthy and malignant tissue from 90 patients, showing that the partial loss of RASSF1A expression due to hypermethylation of its gene promoter could be a premalignant event in early ccRCC tumorigenesis (average methylation: 11 % in healthy vs 20 % in malignant; p < 0.001) [35]. These results were confirmed in a cohort of 318 patients where higher immunohistochemical expression of RASSF1A (using a cut-off value of 25 % for the relative number of cells positively) was positively correlated with pT stage (p = 0.001), high-grade (III - IV) ISUP (p = 0.029) but not with the presence of metastasis at diagnosis (p = 0.410) [36]. Moreover, no significant correlation was found with overall survival (OS) when immunostaining was interpreted as a continuous variable (p = 0.239, hazard ratio (HR) = 1.005, 95 % CI 0.997–1.013) or as a quantitative variable, using a cut-off value of 25 % for the proportion of positively stained cells (p = 0.054).

Another study in 86 patients demonstrated that decreased RASSF1A expression, due to loss of the 3p region, which is commonly lost at the earliest stage of ccRCC developmentand RASSF1 hypermethylation were associated with poorer progression free survival (PFS) and OS. Thus, a decrease in RASSF1A mRNA levels impacted PFS and OS with respective HR of 2.93 and 2.25 (p = 0.02) [37]. While RASSF1C mRNA levels were higher in ccRCC, no correlation with clinicopathological variables was found. Consequently, the RASSF1 gene does not appear to be involved in ccRCC progression.

NF2

The NF2 (Neurofibromatosis type 2) protein, also referred as Merlin, is encoded by the NF2 tumor suppressor gene [38,39] and is an important mediator of cell contact inhibition. This contributes to the arrest of cell proliferation as it stabilizes the intercellular junctions [40,41]. Accordingly, NF2 plays a role in establishing and maintaining the apical-basal polarity of cells, and this polarity is lost early during the malignant transformation process [42]. NF2 inactivation disrupts the Hippo pathway, leading to reduced LATS1/2 activity. Subsequently, YAP1 is translocated to the nucleus which leads to the expression of its transcriptional targets [43].

Deleterious mutations of NF2 are found in 1 % to 3 % of RCC [44,45]. Most of those cases are paired with loss of heterozygosity involving chromosome 22q, according to a haploinsufficiency model, resulting in loss of expression of a functional NF2 protein [46]. Such loss of NF2 then induces type II neurofibromatosis. Genome-wide sequencing of primary ccRCC tumors revealed that around 33 % of wild-type VHL clear-cell ccRCC had NF2 mutations [47].

NF2 dysfunction seems associated with greater aggressivity in RCC. Indeed, mice were generated with a targeted deletion of NF2 in the proximal convoluted epithelium using Villin-Cre transgene and then suffered intratubular neoplasia by 3 months, which progressed to invasive carcinoma by 6–10 months [48]. Furthermore, 11 cases with a somatic NF2 mutation were found amongst 62 unclassified aggressive histology RCC (uRCC) [49]. Molecular analysis on 65 sarcomatoid renal tumors of different histological subtypes revealed an increase of NF2 mutations in RCC papillary subtypes [50]. Targeting sequencing of samples from 49 ccRCC that had been previously microdissected showed that all mutations of NF2 were deleterious and affected functional domains of the protein [51].

Moreover, Hippo pathway alterations, especially those targeting NF2, were more frequent in ccRCC than in a control cohort of 268 non RCC (p = 0.001). In the NF2-mutant ccRCC cell line JHRCC12, NF2 reconstitution inhibited cell proliferation, transformation, and invasion. These results were then confirmed in a model of immunocompromised male NOD-SCID IL2Rg-/- (NSG) mice using xenografts of the JHRCC12 line. Additionally, YAP1 knockdown restored cellular homeostasis despite NF2 mutation. This finding was supported by immunohistochemical analysis of NF2-mutant cases (n = 3), which showed significantly stronger nuclear YAP/TAZ staining compared to NF2-wildtype ccRCC (p = 0.019).

Using a cohort of 24 nephrectomy patients and RCC cell lines (A498, 786-O, ACHN), miRNA-572 was shown to inhibit NF2 and to be upregulated in ccRCC [52].

Adaptors and Hippo kinases cascade’s alterations in the ccRCC

SAV1

Protein Salvador (SAV1), a core member of the hippo pathway, is usually localized at the apical-lateral cell junction’s regions [53,54]. Activation of the Hippo pathway leads to phosphorylation of SAV1, which then forms a complex with MST1/5 to in turn phosphorylate the NDR kinase [55].

Array CGH (comparative genomic hybridization) and gene expression analysis of 8 RCC cell lines (786-O, 769-P, KMRC-1, KMRC2, KMRC-3, KMRC-20, TUHR4TKB, and Caki-2) revealed that SAV1 downregulation correlates with gene copy number loss at 14q in these cell lines [56]. qRT-PCR quantification of SAV1 mRNA levels showed an increase in cells with 14q loss (n = 10) and a significant decrease in cells retaining 14q (n = 12) (p < 0.005), with levels comparable between 14q loss and normal kidney tissue. Moreover, in high grade ccRCC (n = 7), levels of SAV1 mRNA were significantly lower than those in low-grade (n = 8) and normal kidney (n = 5) (p < 0.005). Consequently, high grade ccRCC was also correlated with a decreased expression of SAV1 (p = 0.0469). SAV1-siRNA transfection demonstrated that SAV1 downregulation inhibits apoptosis. Additionally, in 786-O cells, the phosphorylation of YAP1 was suppressed in SAV1-deficient ccRCC cell lines and enhanced when SAV1 was re-expressed. Immunohistochemistry analysis further showed that SAV1 downregulation promotes YAP1 nuclear translocation.

The involvement of SAV1 in ccRCC oncogenesis was confirmed in vivo in Cdh16-Cre;Sav1^fl/fl mice, demonstrating that SAV1 is necessary for the sustenance of growth, nuclear size and structure and that its deficiency leads to enhanced proliferation [57].

In a cohort of 43 nephrectomy patients, combined qRT-PCR and RNA FISH analyses [58] demonstrated that the long noncoding RNA HOTAIR, known to activate the Wnt/β-catenin pathway by promoting histone H3K27 methylation at the WIF-1 promoter [59], was overexpressed in patients with advanced TNM stage and/or metastases (p < 0.001). In RCC cell lines (786-O and OSRC2), HOTAIR was shown to regulate Enhancer of Zeste Homolog 2 (EZH2), a methyltransferase responsible for methylation activity [60].

In a cohort of 174 ccRCC patients treated with the TKI Sunitinib, low SAV1 expression was significantly associated with poorer OS (HR = 0.19 [0.06–0.56]; p < 0.001) [61]. SAV1 was also associated with metachronous metastatic disease (p = 0.013).

MST1/5

Proteasome Activator (REGγ), also known as PSME3 or PA28γ, is a member of the 11S proteasome activator family that regulates the degradation of many proteins [62]. REGγ thus interacts with the casein kinase 1 ε (CK1ε) [63] which in turn interacts with Mammalian sterile 20-like kinase 1 (MST1) [64]. In a cohort of 81 ccRCC, upregulation of REGγ was significantly correlated with pTNM (p = 0.008), ISUP grade (p = 0.032), and presence of metastasis (p = 0.033). REGγ was also an independent risk factor of poorer OS (HR = 1.649, 95 % IC [1.327 – 3.287]; p = 0.0035) [63]. These results were confirmed in vivo with a xenograft model using 786-O and Caki-1 cell lines: REGγ expression is negatively correlated with the presence of CK1ε, MST1, and phosphorylated YAP1 (p-YAP1, sequestrated in the cytoplasm), while positively correlated with active YAP1 presence in the nucleus.

Analysis of the TCGA database revealed that MST1 gene expression in ccRCC patients correlates with TNM tumor stage, ISUP grade, and metastasis occurrence [65]. These findings were validated by immunohistochemistry, western blot, and qPCR analyses. High MST1 expression was significantly associated with improved survival (p < 0.001) and negatively correlated with both plasmacytoid dendritic cells (pDCs) (p < 0.001) and CD56-bright natural killer (NK) cells (p = 0.017).

NDR kinases: LATS1/2, NDR1/2

LATS1 and LATS2 are highly similar in structure and function and are at the core of the Hippo pathway. When activated by phosphorylation, LATS1/2 then phosphorylates YAP1/TAZ, leading to their sequestration in the cytoplasm and degradation by the proteasome [55]. In a cohort of 86 ccRCC patients, decreased LATS1 mRNA expression was significantly correlated with increased metastasis (p < 0.001) and advanced TNM stage (p < 0.001) [66]. In a cohort of 54 ccRCC patients, LATS1 immunoreactivity was significantly reduced in tumor cells compared to normal kidney tissue (p < 0.001). Additionally, cytoplasmic LATS1 immunoreactivity positively correlated with nuclear YAP1 expression (r = 0.34, p = 0.0127) and was associated with poorer OS (p = 0.0181) [67]. The last result was confirmed in multivariate analysis (HR = 0.90).

In a cohort of 30 ccRCC samples and 786-O cells, decreased LATS1 expression was linked to promoter hypermethylation [68]. Pharmacological demethylation (5-Aza-2′-deoxycytidine) restored the expression of LATS1 mRNA and protein in the in vitro model and downregulated the expression of YAP1 which inhibited cell proliferation and induced cell apoptosis [68]. Consequently, LATS1 expression was also related to a poorer TNM stage, and deleterious mutations of LATS2 with a splicing mutation and with missense mutations were reported [51]. Ablation of LATS1/2 in Cre line -Slc34a1CreERT2 mice is sufficient to induce sarcomatoid ccRCC with lung metastasis [69]. Analysis of the TCGA database confirmed decreased LATS1 expression in ccRCC while in vitro studies using CAKI-1 cells demonstrated that EZH2 is highly enriched at the LATS1 promoter. Furthermore, EZH2 inactivation led to hypomethylation of the LATS1 promoter region, underestimating the role of LATS1 promoter methylation in ccRCC carcinogenesis [70].

A high expression of SH3-Binding Glutamate-Rich Protein-like 2 (SH3BGRL2) was correlated with better parameters in TCGA database: low ISUP, low pTNM and better PFS [71]. These results were confirmed for OS in multivariate analysis: HR = 0.658 95 % IC [0.482 – 0.900]; p = 0.009. These findings were further validated in a cohort of 112 ccRCC patients and in a xenograft model using 786-O and A498 cell lines [71]. They showed that SH3BGRL2 interacts with LATS1/2 thus preventing the YAP1 phosphorylation.

Except for the data extracted from The Human Protein Atlas presented in Table 1, there is no data available for NDR1/2 (STK38/STK38L) in ccRCC.

Table.

Expression analysis of Hippo pathway players according to The Human Protein Atlas in ccRCC [72].

Proteins	Difference in expression between normal and tumor tissues (N)	Correlation between High Expression and OS (N)
MST1 (STK4)	Overexpression p < 0.001 (194)	Positive p = 0.037 (521)
MST2 (STK3)	Subexpression p < 0.001(194)	Negative p = 0.038 (282)
MST3 (STK24)	Subexpression p < 0.001 (194)	Negative p < 0.001(100)
MST4 (STK26)	None p = 0.3 (194)	Negative p = 0.021 (521)
MST5 (STK25 - YSK1)	Subexpression p < 0.001 (194)	Positive p < 0.001 (521)
SAV1	None p > 0.999 (194)	Positive p < 0.001 (282)
LATS1	None p > 0.999 (194)	Positive p < 0.001 (521)
LATS2	None p > 0.999 (194)	Positive p < 0.001 (521)
NDR1 (STK38)	Overexpression p < 0.001(194)	Positive p = 0.003 (521)
NDR2 (STK38L)	None p > 0.999 (194)	Positive p = 0.001 (521)
YAP1	None p > 0.999 (194)	None p = 0.47 (282)
TAZ (WWTR1)	Overexpression p < 0.001	Positive < 0.001 (521)

Hippo effectors’s alterations in the ccRCC

YAP1

YAP1 is the principal effector of the Hippo pathway. A study on a cohort of 31 ccRCC patients highlighted a predominance of YAP1 expression at the tumoral front (p < 0.001) [73]. Interestingly, there was no correlation between YAP1 presence and high ISUP grade. However, knockdown of YAP1 significantly inhibited proliferation, migration, and reduced tumor size in an in vivo model using xenograft with ACHN cell lines. Moreover, the overexpression of YAP1 observed in immunohistochemistry is associated with an overexpression of YAP1 mRNA in tumor tissue compared to healthy tissue and was correlated with tumor differentiation in 786-O and ACHN cell lines but also in a cohort of 30 ccRCC patients [74]. Higher expression rate of YAP1 mRNA is correlated with more aggressive tumors as they are less differentiated (p = 0.018). This was also associated with higher pTNM stages (p = 0.034).

Quantitative PCR analysis in 86 ccRCC cases revealed that high YAP1 expression significantly correlates with the presence of metastases (p < 0.0001) [66]. Additionally, in a cohort of 54 ccRCC patients, cytoplasmic YAP1 presence in tumor cells was associated with significantly shorter OS (median = 26.8 months) compared to patients lacking cytoplasmic YAP1 [67]. Multivariate regression analysis confirmed that the expression levels, based on immunoreactivity, of YAP1 (HR =4.53) acts as independent prognostic factor in ccRCC.

TAZ

The transcriptional co-activator with PDZ-binding motif (TAZ), also known as WW domain-containing transcription regulator-1 (WWTR1), is a paralog protein of YAP1 [12]. TAZ is abundantly expressed in ccRCC, as demonstrated in vivo using RCC4 and 786-O cell lines [75]. Analysis of the TCGA database further confirmed TAZ overexpression, which was positively associated with higher ISUP grade (p = 0.003), advanced pTNM stage (p = 0.001), distant metastases (p = 0.001), and shorter overall survival (OS) (p < 0.001). TAZ expression also emerged as an independent prognostic factor (HR = 1.661; 95 % CI: 1.208–2.286; p = 0.002), although no significant association was found with PFS. TAZ levels were consistently higher in tumor tissues compared to adjacent normal kidney tissues. In vitro experiments using 786-O and A498 cell lines, as well as xenograft models, showed that TAZ knockdown via shRNA reduced proliferation, clonogenicity, tumor growth, and induced cell cycle arrest [76]. These findings suggest that dysregulation of the Hippo pathway may contribute to ccRCC progression. In a TAZ knockout model (A498 and 786-O), high TAZ expression was shown to inhibit mitophagy, suggesting that impairment of the Hippo pathway may facilitate tumor development through mitophagy suppression [77].

TAZ regulates the expression of programmed cell death ligand 1 (PD-L1) by interacting directly with proline-rich/Ca-activated tyrosine kinase 2 (PYK2) [[78], [79], [80]]. This protein is targeted by anti PD-1 - Programmed cell Death Protein 1 - (Nivolumab) and (Pembrolizumab) which can be used in ccRCC treatment [4,5]. YAP1 can also induce PD-L1 expression and while this mechanism is well described in lung cancer [12] and even in papillary kidney cancer [81], it has not been clearly identified to date in ccRCC.

TAZ plays also a critical role in regulating ferroptosis susceptibility in ccRCC through the EMP1-NOX4 axis [75]. Ferroptosis sensitivity is highly influenced by cell density: ccRCC cells grown at low density are highly susceptible to erastin-induced ferroptosis, whereas resistance develops at high confluency. This density-dependent regulation is mediated by TAZ, which undergoes nuclear/cytosolic translocation according to cell confluency. TAZ knockdown confers ferroptosis resistance, while overexpression of constitutively active TAZ sensitizes cells to ferroptosis. Mechanistically, nuclear TAZ transcriptionally activates EMP1 (Epithelial Membrane Protein 1), which in turn induces the expression of NOX4 (NADPH Oxidase 4), the hallmark of ferroptosis. Thus, pharmacological inhibition of NOX4 by GKT136901 protects ccRCC cells from erastin-induced death.

The dual nature of YAP1/TAZ in ccRCC: oncogenic driver or tumor suppressor?

Overall, the literature highlights the complex and context-dependent roles of YAP1/TAZ in ccRCC. Most studies report that YAP1 and/or TAZ are overexpressed or hyperactivated, promoting proliferation, epithelial–mesenchymal transition (EMT), invasion, metastasis, and metabolic reprogramming [[66], [73], [74], [81], [82], [83], [84]]. Their transcriptional partners TEADs, particularly TEAD4, are associated with poor prognosis [82,85], and YAP1/TAZ activation contributes to therapeutic resistance, notably to TKI [86]. Inactivation of several upstream regulators such as SAV1 [57,56], MST1/STK4 [65], KIBRA [87], and ARRDC1/3 [88] further enhances YAP1 deregulation, while factors like EFHD1 [89] or CENPK [90] can amplify YAP1/TAZ activity.

Conversely, several reports have revealed an unexpected tumor-suppressive function of YAP1, which can inhibit key ccRCC pathways such as HIF-2α [82] and NF-κB signaling [83]. In these contexts, YAP1 loss promotes pro-tumoral inflammation or HIF-2α-dependent activation. Regulators such as SH3BGRL2 [71], CCDC25 [91], ENO2 silencing [83], and Leukemia Inhibitory Factor Receptor (LIFR) [92] can reactivate the Hippo pathway and exert antitumor effects.

Beyond the YAP1/HIF-2α competition for TEAD binding, recent evidence has uncovered an additional tumor-suppressive mechanism of YAP in ccRCC involving the NF-κB (Nuclear Factor kappa-light-chain-enhancer of activated B cells) pathway [93] - transcription factors that control DNA transcription, cytokine production, and cell survival [94]. YAP1 in nuclear situation inhibits NF-κB signaling by disrupting the cooperativity between the p65 subunit and ZHX2, a VHL substrate that accumulates in VHL-deficient cells and functions as a critical NF-κB cofactor. Specifically, YAP/TEAD complexes compete with ZHX2 (Zinc fingers and homeoboxes protein 2) for p65 binding while simultaneously repressing ZHX2 expression, resulting in diminished NF-κB target gene transcription (IL6, IL8, CCL2) and growth inhibition.

Together, these findings underline that the Hippo–YAP1/TAZ status in ccRCC should be interpreted contextually, as it may display both oncogenic and suppressive functions depending on the molecular landscape. Therapeutic strategies targeting this pathway—such as TEAD inhibitors, stabilization of upstream components (SAV1, MST1, KIBRA), or pharmacological modulation of YAP1 activity-hold promising potential [95]. Ultimately, personalized therapeutic approaches based on reliable biomarkers (TEAD4, ZHX2, HIF2α, NF-κB, SAV1, among others) will be essential to optimize treatment response and improve outcomes for ccRCC patients.

VHL and Hippo pathway nexus in ccRCC

Chromosome 3p carries both VHL and RASSF1A. The simple loss of this unique chromosomal locus therefore simultaneously disrupts the VHL/HIF axis and the Hippo pathway through RASSF1 hemizygosity, creating a “double whammy” scenario that could explain the aggressiveness of ccRCC. Recent work has demonstrated that combined VHL and RASSF1A deletion in mouse models causes chromosomal segregation defects and increased micronuclei formation. This indicates that the two proteins cooperate to maintain genomic integrity [96]. This genomic instability resulting from the simultaneous loss of VHL and RASSF1A may accelerate tumor evolution and contribute to the intratumoral heterogeneity characteristic of ccRCC. Chromosome 3p also carries PBRM1 (Polybromo 1), SETD2 (SET Domain containing 2), BAP1 (BRCA1-Associated Protein 1) who are known to be involved in the ccRCC through their role in chromatin remodeling [97]. The co-localization of these functionally interconnected genes on 3p suggests that this chromosomal region represents an "Achilles' heel" in renal epithelium, where a single genetic event can simultaneously compromise hypoxia sensing (VHL), chromatin remodeling (PBRM1, SETD2, BAP1), and growth control pathways (RASSF1/Hippo).

Some recent studies highlight interactions between the VHL and Hippo signaling pathways. Notably, the VHL pathway seems to impact on the phosphorylation of LATS1/2 and thus allows an amplification of cellular responses to hypoxia. Among the HIF-1 α target, GPRC5A (G Protein-coupled Receptor class C group 5 member A) belongs to the family of G proteins coupled to receptors (GPCRs) located on the cell surface [11]. These receptors play a role in hepatic lipid metabolism [98], in the development of the nervous system [99], in the inflammatory response [100] as well as in tumor development [101,102]. In hypoxic situations, ferropt prevents the phosphorylation of LATS via RhoA GTPase (guanosine triphosphatase) [16] and facilitates the passage of YAP1 into the nucleus and the transcription of its target genes. Thus, activation of the HIF-GPRC5A-YAP axis promotes cell growth and survival in a hypoxic environment. In colorectal cancer, a high level of GPRC5A mRNA is associated with lower PFS [11]. Other GPRCs act on the cytoskeleton leading to remodeling activating Rho GTPases and preventing the MST1/2-LATS1/2 phosphorylation cascade. Finally, some GPRCs act directly with HIF-1α, promoting cell proliferation and mobility in particular [103,104].

VEGF (Vascular Endothelial Growth Factor), another target gene of HIF, also appears to interact with the Hippo pathway [15,105]. Once bound to its receptor VEGFR, VEGF will activate PI3K (phosphoinositide 3 kinase), a MAP-Kinase. This kinase prevents the phosphorylation of MST1/2 and therefore that of LATS1/2. Thus, YAP1 is not retained in the cytoplasm and can translocate into the nucleus in order to activate its target genes. In addition, VEGFR2, once activated, can modulate the cytoskeleton and activate Rho GTPases, leading to the translocation of YAP into the nucleus [105].

There also appear to be several interactions between the Hippo and VHL signaling pathways at the transcription factor level. Indeed, the TEADs transcription factors, stimulated by YAP/TAZ, appear to promote the transcription of VEGF [106] and create a positive stimulation loop of the VHL pathway. Similarly, the VHL and Hippo pathways can potentiate each other in the activation of ferroptosis via their action on EMP1 and NOX4 [[75], [107], [108], [109]].

HIF-1α also appears to be a co-activator of TAZ and thus promote lymph node and metastatic dissemination in breast cancer [110]. The HIF-1α/TAZ complex regulates in particular the CTGF (Connective Tissue Growth Factor), PAI1 (Plasminogen 1 Activator Inhibitor) and BIRC5 (Baculoviral IAP Repeat Containing 5) genes which play a role in tumorigenesis [111]. This complex also transcribes SIAH1 (Siah E3 Ubiquitin Protein Ligase 1) which induces the nuclear localization of TAZ by degrading LATS2 with the help of TGFβ (Transforming Growth Factor Beta) (a target gene of HIF) [110]. When the VHL pathway is activated, SIAH2 also inhibits the phosphorylating action of PHD2 (prolyl hydroxylase 2) on YAP1 [112]. HIF-1α can also be the co-activator of YAP with TEAD and transcribe CYR61 (cysteine-rich angiogenic protein 61) and CTGF. Note that YAP1 can also prevent HIF hydroxylation by pVHL.

The Hippo and VHL pathways are therefore two pathways that activate each other in hypoxia situations. LATS1/2 appears to be one of the links between these two signaling pathways. These interactions are summarized in Fig. 2.

Fig. 2.

Interaction between Hippo and VHL pathway.

The Hippo pathway (elements in green) is frequently dysregulated in ccRCC. The proteins RASSF, whose inactivation by promoter methylation is associated with poor prognosis [34,37], and NF2, frequently mutated in certain subtypes of renal carcinoma [[46], [48]], positively regulate the MST1/5-SAV1 complex. Decreased expression of SAV1 is associated with high-grade carcinomas [56]. The MST1/5-SAV1 complex phosphorylates LATS1/2, whose methylation and inactivation are documented in ccRCC [[67], [68]]. Loss of LATS1/2 in adult renal epithelium leads to the development of renal carcinoma [69]. LATS1/2 phosphorylates the effectors YAP-1 and TAZ, whose overexpression is correlated with poor prognosis [[66], [74],76]. The 14–3–3 protein sequesters phosphorylated YAP/TAZ in the cytoplasm, leading to their degradation. The non-phosphorylated forms translocate to the nucleus where they act as transcriptional co-activators, regulating cell proliferation, survival, metastasis, and immune evasion [55,73].

The VHL pathway (elements in pink) interacts with the Hippo pathway. Under normoxic conditions, pVHL targets HIF-1α for degradation [19]. Under hypoxic conditions or in the presence of VHL mutations, HIF-1α accumulates and regulates the expression of pro-angiogenic genes [21]. The ubiquitin ligase SIAH2 regulates the activation of the Hippo pathway under hypoxic conditions [111]. YAP-1 and HIF-1α cooperate to regulate VEGF expression and angiogenesis [[105], [112]]. The hypoxia-regulated GPRC5A receptor activates YAP [11], while VEGFR also regulates the Hippo pathway [15].

TAZ regulates ferroptosis in ccRCC [75], a process dependent on lipid metabolism [107]. SH3BGRL2 inhibits growth and metastasis via activation of the Hippo pathway [71]. TAZ also regulates PD-L1 expression, influencing immune evasion [78].

ccRCC: clear cell renal cell carcinoma EMP1: Epithelial membrane protein 1 EPO: Erythropoiesis GPRC5A: G Protein-coupled Receptor class C group 5 member A GTPase: guanosine triphosphatase HIF: hypoxia inducible factor LATS1/2: Large Tumor Suppressor Kinase 1/2 MAP4Ks: Mitogen-activated protein kinase MOB1A et 1B: Mps One Binder Kinase Activator-Like 1A et 1B MST 1/5: Mammalian STE20-Like Protein Kinase 1/5 NDR1/2: nuclear dbf2-related kinase NF2: Merlin protein NOX4: NADPH Oxidase 4 PD-1: Programmed cell Death Protein 1 PDGF: Platelet Derived Growth Factor RASSF1: Ras Associated Domain Family 1 SAV1: Salvador Family WW Domain Containing Protein 1 SH3BGRL2: SH3-Binding Glutamate-Rich Protein-like 2 SIAH1 et 2: Siah E3 Ubiquitin Protein Ligase 1 et 2 TAZ: Transcriptionnal co activator with pdZ binding motif VEGF: Vascular Endothelial Growth Factor VHL: Von Hippel-Lindau YAP1: Yes Associated Protein 1.

Conclusions /perspectives

Summary of key findings

This review synthesizes the current knowledge of the Hippo pathway alterations in ccRCC and reveals its complex interplay with the VHL/HIF axis. We have shown that multiple Hippo pathway components are dysregulated in ccRCC through various mechanisms: (i) epigenetic silencing of RASSF1A via promoter hypermethylation, (ii) mutations and loss of NF2, particularly in aggressive and sarcomatoïd variants, (iii) downregulation of SAV1 associated with chromosome 14q loss, (iv) methylation-mediated inactivation of LATS1/2, and (v) consequent nuclear accumulation of YAP/TAZ driving oncogenic transcription. Importantly, these alterations correlate with advanced TNM stage, higher ISUP grade, increased metastatic potential, and reduced overall survival, establishing the Hippo pathway as a key player in ccRCC progression.

Future perspectives and applications

Several research priorities emerge from this review. The Human Protein Atlas data presented in Table 1 reveals that several Hippo pathway members remain largely unexplored in ccRCC. MST3 (STK24) and MST5 (STK25) show striking correlations with patient outcomes but lack mechanistic studies. Similarly, NDR1/2 kinases, which can partially compensate for LATS1/2 function, warrant investigation as potential therapeutic targets or resistance mechanisms.

Clinically, developing a "YAP1 signature" could improve risk stratification for adjuvant pembrolizumab selection both through oncological mechanisms and the regulation of PD-L1 by the Hippo pathway. In addition, in metastatic situations the Hippo pathway status might predict immunotherapy response, warranting prospective validation.

Therapeutically, the Hippo-VHL nexus offers multiple intervention points. Combining LATS1/2 activators or YAP/TAZ inhibitors with existing VEGFR-TKIs or belzutifan could overcome resistance mechanisms. Additionally, targeting the GPRC5A-Rho GTPase axis might disrupt the crosstalk between pathways. The ferroptosis connection through NOX4 and EMP1 opens another therapeutic avenue, particularly relevant given ccRCC's unique metabolic vulnerabilities.

In conclusion, this review establishes the Hippo pathway as a critical, actionable component of ccRCC (cellular?) biology that has been overshadowed by a focus on VHL/HIF signaling. Future studies should prioritize the understanding of the Hippo components identified here, validate biomarker combinations in clinical cohorts, and develop therapeutic strategies targeting the Hippo-VHL nexus to improve outcomes for ccRCC patients.

Funding

Not applicable.

Availability of data and materials

Not applicable.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

CRediT authorship contribution statement

T Waeckel: Writing – review & editing, Writing – original draft, Validation, Project administration, Funding acquisition, Conceptualization. R Lefranc: Writing – review & editing, Writing – original draft, Validation. M Waeckel: Writing – review & editing, Writing – original draft. M Riffet: Writing – review & editing. X Tillou: Writing – review & editing, Supervision, Project administration, Conceptualization. G Levallet: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Funding acquisition, Conceptualization. C Bazille: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: WAECKEL Thibaut reports financial support, administrative support, and article publishing charges were provided by Centre Hospitalier Universitaire de Caen. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.