The Hippo signaling pathway is an evolutionarily conserved pathway with a biological role in cell fate determination, organ size control, and tissue regeneration in most tissues and organs (1–4). Dysregulation of this pathway has been implicated in a variety of pathologies and has received extensive attention over the past two decades (5). In cancer research, the activated Hippo pathway is considered a tumor suppressor pathway due to its role in inhibiting cell proliferation and promoting apoptosis (6–8).

Growth factors, glucose, hypoxia, cell polarity, and mechanical cues from cell–cell or cell–extracellular matrix attachment regulate the Hippo pathway (9). The core components of the Hippo pathway in mammals consist of a kinase cascade, MST1/2 and LATS1/2, and the main effector YAP1/TAZ, a transcriptional coactivator without DNA-binding domains. The major binding partners of YAP1/TAZ are TEAD1–4. Other transcriptional factors, including AP1, PITX2, ZEB1, MYC, E2F, and SMADs, have also been reported to cooperate with the YAP1/TAZ–TEAD complex (10–15). In addition, RUNX2, TP73, and FOXO1 also directly bind to YAP1/TAZ (16). YAP1 overexpression has been reported to be oncogenic in bile duct, breast, colon, lung, and liver cancers (17–23). Inhibition of the Hippo pathway or overexpression of YAP1 may lead to the nuclear translocation of YAP1, which then binds to transcription factors to promote the expression of tumor-promoting genes. Of note, YAP1 also functions as a tumor corepressor to repress the expression of downstream genes, including the cell-cycle kinase inhibitor p27, by recruiting the NuRD (nucleosome remodeling deacetylase) complex, YY1, or EZH2, a polycomb repressive complex member (24, 25). Thus, the role of transcriptional cofactor or transcriptional corepressor of YAP1 is largely context-dependent. Therefore, better defining the role of YAP1 in each cancer type will be a key challenge for future studies about target identification and cancer therapy. We investigated the potential role of YAP1 in survival predication from a pan-cancer perspective.

Tumor-infiltrating immune cells are a major component of the ecosystem in tumor microenvironment (TME) and regulate tumor progression (26). Recently, the Hippo signaling pathway is emerging as an important pathway to affect immune function in cancer (27). YAP1 has been reported to affect the activity of B cells, Tregs, macrophages, and myeloid-derived suppressor cells (MDSCs) in several cancer types (28). However, the role of YAP1 in different cancer types and its mechanisms in immune regulation remain to be investigated. In this study, we analyzed the correlation between YAP1 expression and infiltration of various immune cells in 33 cancer types.

Strategies to inhibit YAP1 activity include the following (1): Disrupting the YAP1–TEAD binding or blocking the transcriptional activity of the YAP1–TEAD complex. Carbonic anhydrase 3 (CA3) and verteporfin act by disrupting the YAP1–TEAD binding (29–31). The TDU domain of vestigial like family member 4 (VGLL4), a natural antagonist of YAP1, competes with YAP1 to bind TEADs (32). Narciclasine and peptide17 competes with TEAD4 for binding to YAP1 (33, 34). In addition, K-975, a TEAD inhibitor, inhibits YAP1/TAZ–TEAD interaction (35) (2). Targeting downstream targets of YAP1/TAZ (BCL-xL, FOXM1, and TG2). A37, celecoxib, TP-0903, cyclic peptide RA-V (deoxybouvardin), navitoclax, thiostrepton, and NC-9 fall into this category (36–43). Of these drugs, only verteporfin has been approved by Food and Drug Administration (FDA). However, YAP1 may act in a TEAD-independent manner (44). The efficacy and selectivity of other drugs are not satisfactory. New drugs are needed to inhibit YAP1 activity. In this study, we calculated the correlation between YAP1 expression and IC₅₀s of drugs with FDA approval or in clinical trials, which is a simple way to roughly assess the drug sensitivity.

In this study, we analyzed YAP1 expression in 33 cancer types to reveal its role in predicating prognosis, modulating TME, and drug response to chemotherapeutic and targeted drugs that were FDA-approved or are in clinical trials.

YAP1 is a transcriptional co-activator and a major effector of the Hippo signaling pathway. Emerging work indicates that YAP1 is widely activated in human malignancies and is essential for cancer initiation, progression, and drug resistance in most solid tumors. High expression of YAP1 has been reported to promote the excessive cell proliferation in multiple tissues, including liver, gastrointestinal tissue, skin, and heart (67–70). TEAD-dependent YAP1 function has also been linked to invasive and metastatic behavior of tumor cells (71). Compelling evidence also showed that YAP1 can alter the TME by recruiting immunosuppressive cell types, suppressing cytotoxic T-cell function or promoting the polarization of tumor-associated macrophages towards the pro-tumor M2 phenotype (72–74). In this study, we analyzed YAP1 expression in pan-cancer and evaluated its role in prognostic value, immunomodulation, and drug response.

According to the TCGA database, YAP1 expression was elevated in CHOL, COAD, and THCA, but decreased in 11 of the 23 cancer types, namely, BLCA, BRCA, HNSC, KICH, KIRC, KIRP, LUAD, LUSC, PCPG, PRAD, and UCEC ( Figure 1 ). However, YAP1 has been reported to be oncogenic in BLCA (75, 76), BRCA (77), HNSC (78), KICH (79), KIRC (80), KIRP (81), LUAD (82), LUSC (83), PRAD (84), and UCEC (85). PCPG is a rare adrenal tumor. The role of YAP1 in PCPG remains to be investigated. The downregulation of YAP1 in those caners might be due to case number limitation, and in addition to YAP1 expression level, YAP1 activity also depends on its cellular location. High YAP1 expression is a risk factor for survival outcomes in ACC, BLCA, LGG, LUAD, PAAD, and COAD ( Figure 3 ). Moreover, in ACC, BLCA, COAD, and TGCT, YAP1 was associated with more severe pathologic stages ( Figure 3 ). Among them, the prognostic value of YAP1 on BLCA (84), LUAD (86), PAAD (87) and COAD (88) has been reported.

According to our result, high expression of YAP1 predicted poor survival outcomes in patients with LUAD, COAD, and PAAD, and was associated with more severe pathologic stages in patients with COAD, PAAD, and TGCT ( Figure 3 ). PAAD has been reported to be characterized by immunosuppressive TME. In this study, YAP1 in PAAD was positively correlated with a wide range of immune checkpoint regulators. YAP1 expression was positively correlated with levels of TMB and MSI in TGCT. In addition, important immune cells, including activated CD4+ memory T cells and T follicular helper cells, were positively correlated with YAP1 expression in LUAD. Therefore, we would like to discuss in more detail the role of YAP1 in these cancer types.

Elevated expression of the gene signature for YAP1/TAZ activity is associated with poor prognosis in patients with non-small-cell lung cancer (NSCLC) (89, 90), which is consistent with the results of LUAD in our study ( Figure 3 ). Our GSEA results suggested that high YAP1 expression was associated with reduced activities of DNA repair processes ( Figure 6 ), and defective DNA repair contributes to individual susceptibility to lung cancer development (91). We also showed that TGF-β signaling was upregulated in LUAD with high YAP1 expression ( Figure 6 ). TGF-β is the most potent inducer of EMT in NSCLC and is pivotal to the development of tumor-promoting microenvironment in the lung cancer tissues (92). In our findings, YAP1 may inhibit infiltration of CD8+ T cells, CD4+ Th1 cells, T follicular helper cells, and NKT cells, but increase infiltration of CAFs, which promote the development of LUAD. Zhang et al. have reported that YAP1 activation was not sufficient to trigger NSCLC formation, but promoted its progression to higher grades (89, 93). The role of YAP1 in immune cell infiltration in LUAD will be further experimentally confirmed in our future studies.

High level of YAP activity has been found to be prognostic for poor outcome in four datasets of CRC patients (94). In our study, YAP1 expression was associated with severe pathologic stage and poor survival outcome in COAD ( Figure 3 ). YAP1/TAZ is required for the formation and growth of intestinal tumors (95). Among the enriched hallmarks, higher YAP1 expression was associated with upregulated Wnt β-catenin signaling, TGF-β signaling via NFKB, and Hedgehog signaling ( Figure 6 ). Overactivation of the Wnt signaling pathway is the most oncogenic pathway in CRC. In addition, β-catenin-driven cancers require the YAP1 transcriptional complex for tumorigenesis (96), suggesting the network between the two signaling pathways. Barry et al. reported that forced YAP1 overexpression in the gut epithelium was not sufficient to form tumor (95). The lack of Wnt signaling pathway might be the reason. In COAD, YAP1 expression was positively correlated with the infiltration of MDSCs and CAFs, and was negatively correlated with the infiltration of CD8+ T cells, CD4+ Th1 cells, activated NK cells, and NKT cells ( Figure 4 ). Mechanically, YAP1 promotes MDSC induction by inhibiting PTEN expression, resulting in upregulation of COX-2, p-AKT, and p-p65 in CRC-derived cells, which leads to secretion of the cytokine granulocyte-macrophage colony-stimulating factor (97). YAP1-dependent matrix remodeling is required for the generation and maintenance of CAFs (98). Mechanisms of YAP1 in recruiting CAFs and suppressing CD4+ Th1 cells, T follicular helper cells, activated NK cells, and NKT cells are yet to be unraveled.

PAAD is a very aggressive neoplasia that seems to arise from pancreatic exocrine cells. Pancreas-specific YAP1 knockout halted tumor progression (99). Activation of YAP1 in acinar cells upregulated JAK-STAT3 signaling and promoted the development of pancreatic cancer (100). In this study, YAP1 expression was higher in PAAD patients with more severe pathological stages ( Figure 3 ). Mechanically, YAP1 and the ZEB1 complex activates ITGA3 to promote its metastasis (101). PAAD is characterized by immunosuppressive TME. In our results, YAP1 expression was positively correlated with MDSC infiltration and negatively correlated with NKT cells. Mielgo et al. reported that YAP1 recruited MDSCs to suppress T-cell function and generate an immunosuppressive microenvironment (102). The mechanism remains to be clarified.

Testicular cancer is a very common malignancy in young men. Although testicular cancer has a high cure rate, patients have a high long-term risk of secondary malignant tumors. Ji et al. have reported that different immune status in TME may be responsible for the different survival outcomes of TGCT patients (103). Moreover, YAP1 inhibition enhanced the chemosensitivity of cisplatin in TGCT (104). Studies about the YAP1 in immunomodulation in TGCT remain absent.

In this study, from a pan-cancer perspective, high expression of YAP1 was associated with upregulation of TGF-β signaling, KRAS signaling, Hedgehog signaling, EMT, and androgen response in most cancer types ( Figure 6 ). There was a close interplay between YAP1 and TGF-β signaling. YAP1 activation promoted TGF-β expression, which was involved in the biological processes of endothelial-to-mesenchymal transition (EMT) and fibrosis (105–107) in the liver, lung, and kidney. Their interactions in other organ tissues need to be investigated. We confirmed that YAP1 was able to upregulate smad2 in TGF-β signaling in CRC cell lines ( Figure 8 ). KRAS and YAP1 converged on the transcription factor FOS and activated a transcriptional program involved in regulating EMT (108). However, another study found limited overlap of gene expression between KRAS G12V and YAP1 S127A-driven tumors (109). The interplay between YAP1 and KRAS signaling remains largely unknown. In this study, YAP1 inhibitor reduced the KRAS pathway activity in CRC ( Figure 8 ). Hedgehog signaling activation could upregulate YAP1 expression and induce osteosarcoma development (110), or aid in generation of liver (111). The cooperation between Hedgehog and YAP1 signaling in tumor formation and progression remains to be uncovered. YAP1 mRNA was upregulated in androgen-insensitive prostate cancer cells (112), but the mechanism of their interaction is largely unknown. Furthermore, in our findings, the YAP1 inhibitor did reduce EMT markers in CRC.

Emerging evidence demonstrated the role of YAP1 in modulating TME. In this study, we found that YAP1 may suppress the infiltration of CD8+ T lymphocytes, γδ T cells, T follicular helper cells, NKT cells, and activated NK cells. Moreover, YAP1 may recruit the CAFs and MDSCs to tumor site to suppress the immune response. Among them, the relationship between YAP1 and infiltration of CD8+ T lymphocytes, MDSCs, and CAFs has been reported in several cancer types. However, the mechanisms are incomplete. The role of YAP1 in the infiltration and activities of T follicular helper cells, NKT cells, and activated NK cells remains to be investigated. Stampouloglou et al. reported that YAP1 overexpression in T cells reduced their activation, differentiation, and function, which translated in vivo into an impairment of T-cell infiltration and tumor repression (113). Another study confirmed that YAP1 attenuated CD8+ T cell-mediated anti-tumor response (114). Mechanically, YAP1 overexpression in cancer cells could upregulate PD-L1 expression and impede the activities of CD8+ T cells in melanoma (115). Another mechanism is that YAP1/TEAD directly upregulates CXCL5 in cancer cells to recruit CXCR2-expressing MDSCs, leading to decreased infiltration of CD8+ T cells (74). In addition, IL-6 and CSF1-3 induced by YAP1 in PAAD stimulated the accumulation of MDSCs (116), and upregulation of COX2 by YAP1 in human granulosa cells promoted the recruitment of MDSCs (117). The role of YAP1 in CD8+ T cells and MDSCs in other cancer types remains to be unraveled. In this study, we confirmed the results by experiments on CRC cells and found that the YAP1 inhibitor verteporfin decreased the expression of PD-L1, CXCL5, COX-2, IL-6, and CSF1-3 ( Figure 8 ). Mechanotransduction-mediated YAP1 activation establishes a feed-forward self-reinforcing loop that contributes to maintenance of the CAF phenotype and promotes breast cancer invasion (98). Mechanistically, active YAP1 promotes the expression of ANLN and DIAPH3 and stabilizes actomyosin proteins, which is required for the generation and maintenance of CAFs (98). Another mechanism is that high expression of YAP1 in the tumor stromal cells converts normal fibroblasts into CAFs in the TME of prostate cancer (118). The function of YAP1 in CAFs of other cancer types remains largely unknown. In this study, high expression of YAP1 was associated with TGF-β signaling in pan-cancer. TGF-β-associated extracellular matrix genes link CAFs to immune evasion and immunotherapy failure (57). Therefore, YAP1 may be involved in the function of CAF through TGF-β signaling. We found that YAP1 inhibitor verteporfin reduced smad2 expression in CRC cell lines, suggesting a downregulation of TGF-β signaling ( Figure 8 ). The role of YAP1 in the infiltration and activity of T follicular helper cells, NKT cells, and activated NK cells remains to be investigated. Another interesting result in this study is that YAP1 expression was negatively correlated with infiltration of CD8+ T cells, but positively correlated with infiltration of resting CD4+ memory T cells in most cancer types. This might not be due to the direct relationship between these two cells. YAP1 expression was also negatively correlated with infiltration of activated CD4+ memory T cells, suggesting that YAP1 may impede the activation of memory CD4+ T cells. CD4+ T cells are required for survival of CD8+ T cells during both primary and memory recall responses (119). Therefore, the possible relationship is that YAP1 may impair the activation of memory CD4+ T cells, and impede the survival of CD8 T cells. Further experimental studies on CD4+ T cells will be conducted in the near future to verify the role of YAP1 in memory CD4+ T-cell activation.

Further analyses showed that YAP1 expression was positively correlated to a wide range of immune checkpoints, especially in PAAD, PCPG, and PRAD, suggesting that YAP1 is a potential new drug target for anti-cancer immunotherapy in these cancer types. In PAAD, PRAD, and LGG, YAP1 expression was positively correlated with the expression of CTLA-4, TIM-3 (HAVCR2), and PD-1 (PDCD1), which are key checkpoint regulators that suppress immune reaction. In most cancer types, CD274 (PD-L1), NRP1, and TNFSF15 were positively correlated with YAP1 expression. YAP1 has been reported to induce PD-L1 expression (115). Apart from PD-L1, YAP1 might also regulate the expression of other immune checkpoints. The relationship between these immune checkpoints and YAP1 and the role of their interactions in TME regulation remain to be investigated. These checkpoint regulators may play key roles in YAP1-induced immunosuppressive TME. In PAAD, PRAD, and PCPG, YAP1 expression was highly correlated with the expression of most immune checkpoint genes in our result. Among them, PAAD has been reported to be characterized by the immunosuppressive microenvironment, suggesting a role for YAP1 in PAAD development (102).

Methylation of DNA cytosine bases leads to the inaccessibility of DNA regulatory elements to their transcription factors through a number of mechanisms, leading to the gene transcription shutdown. We found that YAP1 expression was negatively regulated by its DNA methylation in the 5’UTR. Cellular RNAs are naturally decorated with a variety of chemical modifications, which affect the mRNA stability and translation. In our study, extensive modification “effectors,” including enzymes of “writers” and “erasers” that alter the modification level and binding proteins of “readers” that recognize the chemical marks, were positively correlated with YAP1 mRNA level. Further research should be conducted to study whether aberrant DNA methylation and RNA modifications of YAP1 are involved in cancer development and how they work.

FDA-approved chemotherapeutic drugs that are capable of inhibiting DNA synthesis, including teniposide, dacarbazine, doxorubicin, and triethylenemelamine, and inhibitors of hypoxia-inducible factor, MCL-1, ribonucleotide reductase, and FASN in clinical trials are potential drugs to treat cancer patients with high YAP1 expression. Among them, the cytotoxicity effect of AZD2858 and varlitinib was partially attributed to YAP1 activity.

YAP1 was aberrantly expressed in various cancer types and regulated by its DNA methylation and post-transcriptional modifications. High expression of YAP1 was associated with poor survival outcomes in ACC, BLCA, LGG, LUAD, and PAAD. YAP1 may promote tumor progression through immunosuppression, particularly by suppressing the infiltration of CD8+ T lymphocytes, CD4+ Th1 cells, T follicular helper cells, NKT cells, and activated NK cells, as well as recruiting MDSCs and CAFs in pan-cancer. The YAP1-promoting tumor activity is probably attributed to the activation of TGF-β, Hedgehog, or KRAS signaling pathways. AZD2858 and varlitinib in clinical trials might be effective in cancer patients with high YAP1 expression.

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

XH contributed to designing the article structure, writing—original draft, editing the figures and funding acquisition. YRZ contributed to performing experiments, editing the figures, and writing—review. HY contributed to performing experiments. YYZ and XS contributed to formal analysis. QL contributed to funding acquisition. YW contributed to supervision, funding acquisition, and project administration. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (81904130, 82122075, and 82074232), the Project of Shanghai Science and Technology Committee (19YF1449900), the Shanghai Frontier Research Base of Disease and Syndrome Biology of Inflammatory Cancer Transformation (2021KJ03–12), the “Shu Gguang” project supported by Shanghai Municipal Education Commission and Shanghai Education Development Foundation (21SG43), the Clinical Research Plan of SHDC (SHDC2020CR4043), and the Shanghai Youth Talent Support Program.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.