Abstract
Background/Aim: The Hippo signaling pathway comprises mammalian sterile 20-like kinase 1/2, large tumor suppressor 1/2, and Yes-associated protein 1 (YAP1). This study investigated phosphorylated YAP (pYAP, Ser 127) protein expression in oral squamous cell carcinoma (OSCC).
Patients and Methods: Tissues from patients with oral epithelial dysplasia (OED, n=7), carcinoma in situ (CIS, n=14), and OSCC (n=109) were analyzed.
Results: Cytoplasmic expression of pYAP was low in OED, CIS, and OSCC tissues. The expression of pYAP was correlated with differentiation, and the expression of low levels of YAP was significantly more common in well-differentiated to moderately differentiated OSCC than in poorly differentiated OSCC. High pYAP expression correlates with characteristics of epithelial-to-mesenchymal transition (EMT), e.g., loss of E-cadherin and increased expression of vimentin and laminin 5 (p<0.0001). Additionally, the protein arginine methyltransferase 1, a positive modulatory factor of YAP activity, was found to be correlated with elevated levels of pYAP expression (p<0.0007).
Conclusion: The presence of elevated pYAP expression may serve as a prognostic indicator of an aggressive OSCC with EMT during the invasive stage.
Introduction
The Hippo signaling pathway plays an important role in organ development, tissue regeneration, wound healing, maintenance of epithelial homeostasis and modulation of the immune system (1). The Hippo pathway is highly conserved from Drosophila to mammals (2).
The mammalian Hippo signaling cascade includes the proteins mammalian sterile 20-like kinase 1/2 (MST1/2), Salvador homolog-1 (SAV1), large tumor suppressor1/2 (LATS1/2), Mps One Binder kinase activator protein 1 (MOB1), yes associated protein/transcriptional coactivator with PDZ-binding motif (YAP/TAZ) and TEA domain family member 1/2/3/4 (3). When the Hippo pathway is activated, MST1/2-SAV1 heterodimers, which are phosphorylated, phosphorylate and activate the LATS1/2-MOB1 complex (2, 3). This complex then phosphorylates YAP/TAZ, preventing their nuclear translocation and the expression of target genes (1, 3). Phosphorylated YAP/TAZ is either degraded by the proteasome or retained in the cytoplasm by the 14-3-3 protein (1-3).
In contrast, when the Hippo signaling pathway is off, YAP/TAZ are dephosphorylated and translocated to the nucleus (1-3). YAP/TAZ do not bind to DNA. Instead, they primarily bind to a member of the TEA domain family and then initiate transcription of target genes (1, 2).
Recently, we conducted an immunohistochemical analysis to detect the expression of major Hippo signaling pathway proteins in oral epithelial dysplasia (OED), carcinoma in situ (CIS), and oral squamous cell carcinoma (OSCC). MST2 demonstrated a distinct and significant expression pattern in a considerable population of differentiated OSCC cells. This suggests a differential part for MST2 depending on OSCC progression. High YAP1 expression in OSCC may reflect epithelial-to-mesenchymal transition (EMT) (4).
Although expression of phosphorylated YAP (Ser 127) (pYAP) has been reported in breast (5) and colon (6) cancers, its role in OSCC is ill defined. We now show in this paper that pYAP expression in OSCC, CIS, and its precursor lesion OED and evaluate its clinicopathological importance in OSCC.
Patients and Methods
Patients and tumors. Tumor specimens were obtained from 130 patients who underwent surgery for OSCC without neoadjuvant chemoradiotherapy at the Jichi Medical University Hospital during the period 2010 to 2015. All the patients were selected from a previous study cohort (4). Detailed information is available in our previous reports (4). The study protocol was approved by the Ethics Committee of Jichi Medical University (Approval ID: A15-269).
Immunohistochemistry. The details of the antibodies used, antigen retrieval methods, and antibody dilutions are listed in Supplementary Table I. Detailed staining and evaluation protocols are shown in our previous report (4).
Evaluation of immunohistochemical findings. Immunohistochemical staining was independently analyzed by two pathologists (YA and DM) blinded to the clinical data obtained through microscopic examination. YAP1, Vimentin, PRMT1, PRMT5, E-cadherin, Slug, and laminin 5 were evaluated using previously described methods (4).
The intensity of pYAP (Ser 127) staining was quantified as negative (0), mild (1), moderate (2), or strong (3). The percentage of immune reactive cells was classified into four categories: 0, 0%; 1, <10%; 2, 11%-50%; and 3, >50%. The product of the intensity and percentage scores was used as the final staining score. Final scores were defined as low (<3) or high (≥4), as reported by Yeo et al. (7), with slight modifications.
Statistical analyses. Detailed information and protocols of the analysis of the prognosis are available in our previous report (4).
Ethics statement. The study protocol was approved by the ethics committee of Jichi Medical University (Approval ID: A15-269) and was conducted in accordance with the 1964 Helsinki Declaration and its amendments or comparable ethical standards. Informed consent was obtained in the form of opt-out on the web-site. Those who did not provide consent were excluded.
Results
Expression of phosphorylated YAP in the normal squamous epithelium, OED, CIS, and SCC. In the normal squamous epithelium, the cytoplasm of the basal cells was positive for pYAP (Figure 1A). In six cases, weak positive pYAP staining was found in the cytoplasm of OED cells (Figure 1B). In all CIS cases, uniformly positive pYAP staining was found in the cytoplasm of the CIS cells (Figure 1C). pYAP1 expression was low in differentiated SCC (67/101 cases) but high in poorly differentiated OSCC (4/8 cases) (Table I). Likewise, the ratio of pYAP-high cases was remarkably lower in SCC with a clearly invasive pattern (4/40 cases) than in SCC with an infiltrative invasive pattern (34/69 cases) (Figure 1D and E, Table I). High pYAP expressive cases were linked to nuclear and cytoplasmic localization, while low pYAP expressive cases were linked to cytoplasmic localization. The association between pYAP expression and other clinicopathological factors in OSCC is provided in Table I. The intensity of pYAP staining was not associated with sex, tumor location, pathological T (pT) category, pathological TNM stage, lymphovascular invasion, or lymph node metastasis (Table I). The pYAP staining patterns from normal epithelium to OSCC were similar to the YAP1 staining patterns (4) (Supplementary Figure 1).
Expression of phosphorylated Yes-associated protein (pYAP) in normal epithelium, oral epithelial dysplasia (OED), carcinoma in situ (CIS), and oral squamous cell carcinoma (OSCC). (A) Weak expression of pYAP, mainly in the cytoplasm of the basal and parabasal cells of the normal epithelium. (B) Weak expression of pYAP, mainly in the cytoplasm of OED cells of. (C) Weak expression of pYAP, mainly in the cytoplasm of CIS cells. (D) Weak expression of pYAP, mainly in the cytoplasm of OSCC cells, Yamamoto Kohama (YK) classification-2. (E) High expression of pYAP in both the nucleus and cytoplasm of OSCC cells, YK-4D. Bar: 50 μm.
Correlation between the localization of pYAP and clinicopathological factors.
Relationships among the expression of phosphorylated YAP and that of YAP1, EMT markers, and PRMT1 and 5. We showed that the expression of protein arginine methyltransferases (PRMT) 1 and PRMT5 is associated with EMT in OSCC (8, 9). High YAP1 expression may be an indicator of aggressive OSCC with PRMT-mediated EMT at the invasion front (4).
To determine whether the expression levels of these molecules were associated with pYAP expression, we compared sections for the expression levels of YAP1, each EMT marker, PRMT1 and PRMT5. Representative images of YAP1, each EMT marker, PRMT1 and 5, and a summary of the expression levels of these molecules are shown in Figure 2 and Table II.
Expression of phosphorylated Yes-associated protein 1 (YAP1), E-cadherin, vimentin, laminin 5, Slug, protein arginine methyltransferase (PRMT) 1, and PRMT5 in the invasive front of squamous cell carcinoma (SCC). (A) Low cytoplasmic YAP1 expression in Yamamoto Kohama (YK) classification-3 oral SCC (OSCC). (B) High phosphorylated Yes-associated protein (pYAP) expression in the both nucleus and cytoplasm of OSCC cells, YK-4C. (C) Membrane expression of E-cadherin was decreased in OSCC cells, YK-4C. (D) Membrane expression of E-cadherin in OSCC cells, YK-2. (E) Cytoplasmic expression of vimentin in stromal cells of OSCC, YK-3. (F) Cytoplasmic vimentin expression in both cancer and stromal cells in OSCC, YK-4C. (G) Low expression of laminin 5 in OSCC cells, YK-3. (H) High cytoplasmic expression of laminin 5 in OSCC cells, YK-4C. (I) Low nuclear expression of Slug in OSCC cells, YK-3. (J) High nuclear expression of Slug in OSCC cells, YK-4C. (K) Low nuclear expression of PRMT1 in OSCC cells, YK-2. (L) High nuclear and cytoplasmic expression of PRMT1 in OSCC cells, YK-4C. (M) Low expression of PRMT5 in the nuclei or cytoplasm of OSCC cells, YK-2. (N) High nuclear and cytoplasmic expression of PRMT5 in OSCC cells, YK-4C. Bar: 50 μm.
Correlation among pYAP, YAP1, EMT markers, and PRMT expression.
Statistical analysis revealed that decreased expression of E-cadherin and high expression of YAP1, vimentin, laminin 5, and PRMT1 correlated with high expression of pYAP (Table II).
Correlation between the expression of phosphorylated YAP1 and the prognosis of OSCC. To identify parameters that influence recurrence-free survival (RFS) and overall survival (OS) in patients with OSCC, we performed univariate and multivariate Cox proportional hazards analyses.
Univariate analysis revealed that only neural invasion (p =0.0013) correlated with RFS in patients with OSCC (Table III). Univariate analysis showed that pT category (p =0.048), neural invasion (p=0.023), and diffuse invasion patterns (YK-4C and 4D) (p=0.012) were correlated with OS in patients with OSCC. In the multivariate analysis, pT category (p=0.014), neural invasion (p=0.0021), and diffuse invasion pattern (YK-4C and 4D) (p=0.012) were correlated with RFS (Table IV). In patients with OSCC, pYAP expression was not significantly associated with OS or RFS on univariate analysis (Table III and Table IV).
Univariate and multivariate analyses of relapse-free survival in patients.
Univariate and multivariate analyses of overall survival in patients.
Kaplan-Meier analyses showed that high expression of YAP1 and/or pYAP tended to correlate with worse prognosis. However, the correlation was not significant (Figure 3A and B). The YK grade correlated with both RFS and OS (p=0.019 and p<0.0001, respectively) (Supplementary Figure 2A and B).
(A) Kaplan–Meier curves for relapse-free survival (RFS) and overall survival (OS) in patients expressing the phosphorylated Yes-associated protein (pYAP). (B) Kaplan–Meier curves for relapse-free and overall survival in patients with Yes-associated protein 1 (YAP1) and pYAP expression. (C) Kaplan–Meier curves for relapse-free survival in YK-4C and 4D patients expressing YAP1 and pYAP.
Subsequently, we classified the patients into groups according to the YK grade and expression of YAP1 and pYAP. OS was significantly shorter in YK-4C and 4D with YAP1 high and pYAP expression than in other cases (p =0.048) (Figure 3C).
Discussion
YAP is directly phosphorylated by Lats 1/2 in five HXRXXS consensus motifs. Three of these sites are conserved in the Yki protein and are phosphorylated. In the inactive state, Hippo signaling does not induce YAP phosphorylation, resulting in nuclear accumulation of YAP, which promotes growth-promoting transcription. The study of the Ser127 phosphorylation of human YAP demonstrated that its nuclear exclusion resulted in the established principle that Ser127-phosphorylated YAP is cytoplasmic and therefore transcriptionally inactive (1, 2). When stimulated, MST1/2 activates LATS1/2 via phosphorylation, initiating LATS1/2-dependent stimulation of YAP1 phosphorylation at Ser127, Ser397, and other residues, leading to cytoplasmic retention (via the interaction of 14-3-3 with phosphorylated Ser127) and/or degradation (triggered by Ser397 phosphorylation) (1, 2). YAP activators, SBP-3264 and XMU-MP-1, suppressed the growth of leukemia cells through induction of apoptosis and could be novel candidate molecular-targeted drugs for leukemia (10). In gastric cancer, YAP nuclear expression gradually increased from normal mucosa, low- and high-grade dysplasia to tubular or papillary adenocarcinoma (TPAD) but in signet ring cell carcinoma (SRCC) (11). These finding suggested that nuclear YAP plays important role in carcinogenesis, and differential patterns YAP expression between TPAD and SRCC imply the existence of different carcinogenic pathways (11). Several reports have demonstrated the expression and function of pYAP. Kim H et al. showed that pYAP expression in the nucleus of metaplastic carcinoma, exhibiting an association between stemness and EMT features, was higher than that in triple-negative breast cancer (12) and that the expression levels of YAP and pYAP in the nucleus differed according to the metastatic site in metastatic breast cancer (5). Nuclear expression of pYAP is high in brain metastases but low in lung metastases (5). Nuclear YAP and pYAP are associated with a poor prognosis (5). Kalita-de Croft P et al. revealed that nuclear pYAP was associated with a favorable prognosis in estrogen receptor (ER)-positive breast cancers but an unfavorable prognosis in ER-negative cancers (13). However, cytoplasmic pYAP levels are higher in esophageal SCC than in the normal epithelium (7). In colon cancer, nuclear pERK expression is positively correlated with nuclear YAP expression but negatively correlated with cytoplasmic pYAP expression (6). In OSCC, low cytoplasmic pYAP expression in fibroblasts is accompanied by an aggressive pattern of invasion; however, the role of cancer cells in OSCC remains to be elucidated (14).
In our analyses, the cytoplasmic expression of pYAP was low in basal cells of the normal epithelium, dysplastic cells of the OED, and cancer cells. In addition to its physiological role in basal cells, pYAP has been implicated in the early development of OSCC.
High nuclear pYAP expression was found in some cases, especially in those with advanced pT stage (pT4), infiltrative invasion pattern (YK-4C or YK-4D), loss of epithelial markers (E-cadherin), and increased expression of mesenchymal markers (vimentin, laminin 5, Slug, and PRMT1). A poor prognosis was also associated with high pYAP expression. Taken together, these findings implicate pYAP in OSCC malignancy with EMT, consistent with a previous report (12).
EMT is a critical process in embryonic development in which the epithelial cell phenotype is transformed into the motile mesenchymal cell phenotype (15). Acquiring the mesenchymal phenotype plays an important role in the progression of epithelial tumors (15). We previously showed that high PRMT1 and PRMT5 nuclear expressions were participated in EMT in OSCC (8, 9), and the decreased expression of E-cadherin and increased expression of vimentin, laminin 5, Slug, PRMT1, and PRMT5 were associated with high YAP1 expression (8). PRMT1 and 5 dimethylate their substrates at the protein arginine residues in an asymmetrical and symmetrical manner, respectively (16). The roles of PRMT have been implicated in signal transduction, mRNA splicing, transcriptional control, DNA repair, and protein translocation (16). Metaplastic carcinomas exhibit EMT features (17). Comparing the expression of YAP and pYAP in metaplastic carcinoma and triple-negative breast carcinoma, the expression of nuclear YAP and pYAP in tumor cells was significantly higher in metaplastic carcinoma (12). These findings are consistent with those of present study.
Study limitations. Although it has been observed in previous studies, the mechanisms underlying the presence of pYAP in the nucleus remain unclear. Kalita-de Croft P et al. hypothesized that serine-127 phosphorylation indicates negative feedback on YAP and that its presence in the nucleus is due to ‘overflow’, suggesting that inactivation of YAP is accompanied by reduced aggressiveness of ER-positive breast cancers (13). Nemo-like kinase (NLK)-mediated Yap1 phosphorylation at Ser128 (pS128-Yap1) inhibits 14-3-3 binding and increases Yap1 nuclear localization (18, 19). NLK is an evolutionarily conserved MAP kinase-related kinase (MAPK) involved in cell proliferation, differentiation and morphological changes during early vertebrate embryogenesis and nervous system development (20).
Conclusion
In this study, we immunohistochemically analyzed pYAP expression in OED, CIS, and OSCC. pYAP expression was low in OEDs, CISs and differentiated OSCCs, but high in the nucleus and cytoplasm of poorly differentiated OSCCs. High pYAP expression was correlated with a decrease in E-cadherin and high expression of YAP1, Slug, vimentin, and PRMT1. An infiltrative invasion pattern (YK-4C and 4D) with YAP1 high and pYAP expression associated with worse prognosis (OS, both p<0.05). These findings suggested that high pYAP expression represents aggressive OSCC with EMT.
However, the molecular pathways by which pYAP contributes to OSCC aggressiveness remain unclear. More research is needed to understand the exact molecular mechanisms by which pYAP1 promotes EMT in OSCC cells.
Acknowledgements
We thank Tomoko Tamura and Sachiko Oguni for assistant work. English language editing was provided by Editage.
Footnotes
Authors’ Contributions
Y.A.: Conceptualization, Investigation, Writing original draft. D.M.: Investigation and Validation. A.K.: Validation. T.Y.: Validation. N.F.: Resources. H.N.: Resources. Y.M.: Resources. K.I.: Supervision, Writing review & editing. T.N.: Resources, Supervision, Writing review & editing. All Authors critically reviewed and approved the manuscript.
Supplementary Material
Supplementary data are available at figshare (DOI: 10.6084/m9.figshare.29134160).
Conflicts of Interest
We declare that there are no conflicts of interest regarding this manuscript.
Funding
This work was supported in part by Japan Society for the Promotion of Science KAKENHI Grant Number 22K09947 (to YA).
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received March 2, 2025.
- Revision received May 21, 2025.
- Accepted May 23, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
