The Hippo signal pathway was first discovered by mosaic genetic screens on Drosophila for mutants exhibiting tissue overgrowth in the 90s (1,2). This signal pathway plays pivotal roles in organ growth control, stem cell function, regeneration and tumor suppression (3,4). Wart (Wts) was identified as the first component of the Drosophila Hippo pathway, which encodes a protein kinase of the nuclear Dbf-2-related (NDR) family and plays an important role in controlling cell growth and shape as well as proliferation (5). Subsequent studies uncovered several other Hippo pathway components, including scaffold protein Salvador (Sav), Hippo (Hpo) and Mats in the following years (6). These four proteins formed the center part of the Drosophila Hippo pathway, in which Hpo, in association with Sav, phosphorylate and activate Wts and the adopter protein Mats. Therefore, the Hippo pathway is also known as the Salvador-Warts-Hippo pathway (7). Wts/Mats is a protein kinase complex, thus, it functions through its substrates. By protein interaction screens using Wts as bait, Huang et al identified Yorkie (Yki) as the major substrate of the Wts/Mats complex in the Drosophila (8).

Inactivation of Yki through phosphorylation or mutation caused tissue atrophy, while over-expression of Yki caused massive tissue overgrowth, phenocopying loss of Hpo, Sav or Wts function. Functionally, unphosphorylated Yki can translocate into the nucleus and drive growth-promoting related gene expression, such as cell cycle regulator cyclin E, cell-death inhibitor Diap1, and bantam microRNA (9,10). Thus the localization and phosphorylation of Yki are often taken as a measure of the activity of the Hippo pathway (11). The Hippo pathway is deregulated with a high frequency in many human cancers, which confirmed the Hippo signaling pathway is closely related to the development of cancers (12). Therefore, the discovery of Hippo pathway has generated excitement in both basic and clinical research because targeting the Hippo pathway provide us with new insight into development of anticancer drugs (13).

Herein, we provide an overview of the Hippo pathway, with an emphasis on the mammalian system, and summarize the current understanding of the biological functions of YAP and TAZ, and discuss the opportunities and challenges for therapeutic intervention.

The Hippo signaling pathway is an evolutionarily conserved pathway; hence, most components of the Drosophila Hippo pathway have mammalian homologues (Table I). For instance, Mst1/2, Sav1, Lats1/2, Mob1 in mammalian are homologues of Hpo, Sav, Wts and Mats, respectively; both YAP and TAZ are homologues of Yki. Similar to the Drosophila Hippo pathway, the center part of the mammalian Hippo pathway is also a protein kinase cascade. Once the upstream activation signals are received, the downstream Mst1/2 kinase phosphorylate and activate the Lats1/2 kinases and MOB1 cofactor with the help of SAV1. Then the activated Lats1/2 kinases phosphorylate and inactivate YAP/TAZ transcription co-activators by promoting their cytoplasmic localization and degradation. Conversely, unphosphorylated YAP/TAZ will translocated and accumulated in the nucleus and form complexes with transcription factors, consequently inducing the expression of targeted genes, such as connective tissue growth factor (CTGF), cysteine-rich 61 (Cyr61), and fibroblast growth factor (FGF1) (14) (Fig. 1).

The upstream signals that regulate the Hippo pathway have not been well delineated, but it consist of a large network of proteins: including a variety of kinases, such as Src, protein kinase A (PKA) (15); some soluble factors, such as amphiregulin (ARGE) (14); cell adhesion and cell junction proteins, such as Echinoid, cadherin-catenin complex (16,17); cell polarity proteins, such as the crumbs complex, scribble (18); other signaling pathways, such as the G-protein-coupled receptor (GPCR) signaling, Wnt and PI3K pathways (19–21); and the state of the actin cytoskeleton (22,23). However, how these proteins impact on YAP/TAZ is not entirely clear. According to the existing experimental conclusions, in addition to regulating YAP/TAZ phosphorylation; these upstream factors also regulate YAP/TAZ nuclear localization through physical interaction.

YAP was first identified as a proline-rich phosphoprotein bound to the SH3 domain of Yes and Src protein tyrosine kinases (24). The human YAP gene, located at 11q13, encodes a 65 kDa protein. Eight isoforms of YAP protein have been identified in humans to date, which are differed by either an alteration in the activation domain or a loss of WW domain (25,26). YAP mRNA is not only ubiquitously expressed in a broad range of tissues; but also expressed in the whole developmental process (27). YAP gene is amplified in various human cancers. TAZ, also known as WWTR1 (WW-domain containing transcriptional regulator 1), was first identified in 2000 as a 14-3-3 binding protein. The human TAZ gene, located at 3q23-q24, encodes a 43 kDa protein. Similarly, TAZ is expressed in various tissues, except thymus and peripheral blood leukocytes and amplified in various human cancers (28).

Structurally, YAP and TAZ share nearly half of the overall amino acid sequence and have very similar topology. YAP protein consists of 488 amino acids, has a TEAD-binding region (TB), two WW domains, which consist of two conserved tryptophan (W) residues separated by 20–23 amino acids, an SH3-binding motif, a coiled-coil domain, a transcription activation domain, an N-terminal proline-rich domain, and a C-terminal PDZ-binding motif. TAZ protein consists of 400 amino acids, has a similar domain organization with YAP, although it lacks the second WW domain, the SH3-binding motif, and the proline-rich domain (Fig. 2).

Biochemically, the WW domains of the YAP and TAZ have been shown to interact with PPXY motifs (proline/proline/any amino acid/tyrosine) of some transcriptional factors. The TB domain recognizes the TEAD family of transcription factors and activates target gene expression, whereas the 14–3–3 binding motif is crucial to the regulation of YAP and TAZ. The PDZ-binding motif is required for binding with PDZ domain, which are found in many transmembrane or cytoskeleton associated proteins. The PDZ-binding domains appear to direct YAP and TAZ localization (29,30).

At the functional level, both of the YAP and TAZ are serving as transcriptional co-activator and share some transcription factor partners, such as TEAD and Runx. Both of them promote cell proliferation, increases cell migration and invasion. However, both of them have their unique target transcription factors, such as ErbB4, p73 for YAP and PPARγ, Pax3, TBX5, TTF-1 for TAZ, which may contribute to the differential functions of YAP to TAZ. Moreover, gene knockout mouse models demonstrated the different physiological functions between YAP and TAZ: Knockout of the YAP gene in mice leads to early developmental arrest, suggesting an essential role in development. However, TAZ knockout mice are viable, although they have defects in the kidneys and lungs.

Although YAP and TAZ are the main downstream effectors of the Hippo pathway in the mammals, YAP and TAZ do not have DNA binding domains. Alternatively, once they localize to the nucleus, they interact with transcription factor partners and function as transcription co-activators. YAP and TAZ share some transcription factor partners, such as TEAD and Runx proteins.

The mammalian genome contains four highly homologous members of the TEAD protein family (TEAD1-4), which are the major targets of the YAP and TAZ in regulation of cell contact inhibition, epithelial-mesenchymal transition (EMT), oncogenic transformation and apoptosis inhibition (31). The interaction of the TEAD with the TB domains of YAP and TAZ can activate downstream gene transcripts, such as CTGF, and Cyr61. Disrupting the binding of YAP or TAZ to TEAD by genetic approach will abolish their ability to promote cell proliferation and tumorigenesis (32–34). For instance, a mutation of TEAD1 in Y421H leads to the human genetic disease Sveinsson’s chorioretinal atrophy, which is characterized by helicoid peripapillary chorioretinal degeneration or atrophia areata (35). Besides the TB domains, the WW domains of the YAP and TAZ have been shown to interact with PPXY motifs of some transcriptional factors. Runx transcription factors consist of 3 members (Runx1, Runx2 and Runx3), all have a PPXY motif and are involved in carcinogenesis and cancer metastasis. Both YAP and TAZ can bind to and potentiate Runx activity through WW-PPXY interaction (36,37).

In addition to the above binding modes, YAP and TAZ can regulate expression of some genes through other means. TAZ increases the transcriptional activity of thyroid transcription factor-1(TTF-1) on the surfactant protein C promoter by directly interacting with the N-terminal domain of TTF-1 (38). T-box 5 gene (TBX5) belongs to the T-box family of transcription factors that plays an essential role in cardiac and limb development. YAP and TAZ can stimulate TBX5 transcription by interaction with multiple domains including carboxyl-terminal structure of the TBX5 (39). Through the coiled-coil region, YAP and TAZ are able to interact with Smad2/3/4 and consequently activate transforming growth factor-β target gene expression and maintain human embryonic stem cell self-renewal (40).

However, both YAP and TAZ have their own specific binding partners, although they are very similar in their structure. For instance, TAZ inhibits the transcriptional activation of PPARγ, which is critical for adipocyte differentiation, through the interaction of WW domain with PPXY motif. This inhibition prevented mesenchymal stem cells from differentiating into adipocytes. However, although YAP has WW domains, it is unable to interact with PPARγ (41). TAZ, but not YAP, interacts with the PPXY motif of the Kruppel-like factor 5 (KLF5) through the WW domain and protects KLF5 from WWP1-mediated ubiquitination and degradation, thereby, promotes breast cell proliferation and tumorigenesis (42). On the other hand, YAP also has its specific binding partners, such as p73 and ErBb4. Strano et al demonstrated the WW domain of YAP interacts with PPXY motif of the p73, consequently enhancing p73 transcription activity and accelerating programmed cell death (43). Komuro et al reported YAP activates ErbB4-dependent transcription by forming the YAP-ErbB4 complex, which is mediated by the WW domain of YAP and the PPXY motif of ErbB-4 (44).

Phosphorylation of YAP S¹²⁷ and the corresponding site of TAZ by Lats1/2 promote 14-3-3 protein binding resulting in YAP/TAZ cytoplasm retention and functional inhibition, because they were separated from their nuclear binding partners. Mutation of YAP S¹²⁷ or equivalent residues in TAZ abrogated the 14-3-3 protein interaction thus enhancing their function as transcriptional co-activators (45).

In addition to promoting YAP/TAZ translocation, phosphorylation by Lats1/2 also suppress their activities by inducing proteasomal degradation. The Lats1/2 phosphorylate the C-terminal phosphodegrons in YAP/TAZ, primed them for subsequent phosphorylation by casein kinase 1, eventually resulting in the recruitment of the SCF^β-TrCP ligase, thereby leading to poly-ubiquitylation and degradation of YAP/TAZ (46). The N-terminal phosphodegron of TAZ can be phosphorylated by GSK3 via PI3K signaling, resulting in SCF^β-TrCP E3 ubiquitin ligase-mediated proteolysis of TAZ (47). Disruption of the interaction between YAP/TAZ and β-TrCP by mutation will stabilize TAZ/YAP and promote their function.

YAP/TAZ can also be inhibited in a protein-protein interaction manner, the PPXY motifs on Lats1/2, AMOT, and PTPN14 have been shown to directly interact with WW domains on YAP/TAZ and thereby convey inhibitory signals by sequestering YAP/TAZ in the cytosol or junctional compartments (48,49). Through its PDZ domain, junctional protein zona occludens-2 (ZO-2) can directly bind with YAP, consequently promoting YAP translocation to the nucleus (50,51).

Numerous studies show that overexpression of YAP/TAZ can induce EMT, inhibiting apoptosis and increase the number of cancer stem cells in vitro (52–54). Knockdown of YAP/TAZ expression reduced cancer cell migration, invasion and anchorage-independent growth in soft agar, as well as tumorigenesis in nude mice (54–56). The protein level of TAZ is correlated with the invasiveness of breast cancer cells (54). In transgenic mouse models, over-expression of YAP/TAZ or attenuation of Hippo signaling is sufficient to promote tumor formation (45,57,58).

Indeed, the levels and nuclear localization of YAP/TAZ are elevated in many human cancers, such as carcinomas of the lung, thyroid, ovarian, colorectal, prostate, pancreas, esophagus, liver, mammary gland (12,59,60), and positively correlated with poorly differentiated tumors (53). Higher YAP/TAZ protein levels are associated with shorter patient overall survival (61–63). Xie et al reported TAZ protein was expressed in 66.8% (121/181) of non-small cell lung cancer patients (NSCLC) cases and significantly associated with poorer differentiation and short survival (64). Yue et al reported TAZ protein expression was positive in 113 out of 146 (77.4%) gastric cancer samples and especially higher expressed in signet ring cell carcinoma (65). The studies of Yuen et al demonstrated that the mRNA level of TAZ is a prognostic marker in colon cancer progression (61). Wang et al reported YAP was overexpressed in 66.3% (61/92) specimens of NSCLC and significantly correlated with p-TNM stage and lymph node metastasis (62). Lam-Himlin et al found that YAP expression in the cytoplasm and nucleus is significantly increased in high-grade dysplasia and adenocarcinoma of the esophagus as well as gastric adenocarcinoma and metastatic gastric disease (66).

However, some researchers suggest YAP might function as a tumor suppressor. Knockdown of YAP in breast cell lines suppressed anoikis, increased migration and invasiveness, enhanced tumor growth in nude mice, and decreased Taxol responsiveness (67,68). Overexpression of YAP can induce p19 cell apoptosis by increase p73 alpha transcriptional activity (39). YAP is also a critical component of c-Jun-mediated induction of apoptosis (43,69). In contrast, TAZ have not been found to have tumor suppressor function. These research studies suggest the possibility that YAP may have dual roles, as tumor promoter or tumor suppressor, which may depend on the cell context.

Considerable number of studies show that the Hippo pathway plays a key role in tumorigenesis. Downregulation of the Hippo pathway activity not only promotes cancer cells growth, but also makes cancer cells resistant to some chemotherapy drugs (70). These founding raised the possibility that therapeutic intervention of Hippo signaling may improve current treatment strategies. We focus on small molecules that are designed to target Hippo signaling itself by manipulating core components of the Hippo pathway (Table II).

The main reasons why the Hippo signaling pathway gained so much attention since it was delineated are the crucial roles in organ size control, cell differentiation, as well as tumorigenesis. The Hippo pathway activity is associated with cancer cell’s proliferative potential, multi-drug resistance, EMT and metastasis. In the past dozen years, our understanding of the Hippo signaling pathway, in both Drosophila and mammalian, has largely increased; the kinase cascade and the main downstream effectors of this pathway were basically defined. However, there are still many questions to be answered: i) although a wide range of upstream signals have been identified, the details mechanism by which these upstream regulators integrated to impact on the core components is not yet fully understood. ii) Some components also interplay with other pathways; the in-depth mechanisms need to be uncovered. iii) YAP and TAZ have similar structure and both serve as downstream effectors in the Hippo pathway, but they also have distinct functions. Is there some internal relationship between YAP and TAZ? Are there any other transcription factors and target genes mediating YAP/TAZ function? iv) Why does the Hippo pathway vary in terms of cell context? Are there any predictive biomarkers that enable us to select appropriate patients for anti-YAP/TAZ drugs treatment? Answers to these questions would advance our understanding of the Hippo pathway, as well as provide us valuable information for new drug development.

There are numerous potential drug targets identified to modulate the Hippo pathway and some new drugs are under clinical trials. However, a batch of new problems have appeared: most of the targets lack of specificity, such as upstream regulators GPCR, which have broad physiological functions in normal cells. General interference of these targets may cause serious side effects. Directly targeting the core components, such as YAP and TAZ, would improve specificity, but YAP and TAZ also play important roles in organ size control and normal tissue regeneration. It means that long-term whole-body inhibition of either of these proteins may results in serious side effects. It would be ideal to develop effective means to manipulate the Hippo pathway and YAP/TAZ activity in both a tissue specific and transient manner. Undoubtedly, the Hippo pathway and YAP/TAZ may represent valuable targets for cancer therapy, although further research is necessary to definite the upstream regulators, downstream targets, as well as the regulation mechanisms of this pathway.