In the last few years, studies have started to disentangle how mechanical signals are interpreted by cells during morphogenesis, and how this results in specific cell behaviors. In this section, we will focus on the role of YAP/TAZ in this process.

The transcriptional co-activators Yes-associated protein (YAP) and its paralog TAZ (transcriptional co-activator with a PDZ binding domain; encoded by the wwtr1 gene) are important regulators of tissue growth and regeneration (as reviewed in Hansen et al., 2015). YAP and TAZ regulation is best understood under the scope of the Hippo kinase cascade (Lei et al., 2008; Zhao et al., 2008). The Hippo pathway was initially identified through mosaic genetic screens for suppressors of tissue overgrowth in Drosophila melanogaster (Udan et al., 2003). Importantly, Hippo signaling cascade controls organ size and tissue homeostasis through the regulation of cell proliferation, apoptosis, and tissue regeneration (see review Zheng and Pan, 2019). Not surprisingly, deregulation of the pathway has been implicated in varieties of cancers and diseases (Plouffe et al., 2015). The core components of the Hippo pathway, the kinase Hippo (Hpo, or MST1 and MST2 in vertebrates), the kinase Warts (Wts, or LATS1 and LATS2 in vertebrates), and the effector Yorkie (Yki, or YAP and TAZ in vertebrates) are highly conserved from Drosophila to mammals (Hilman and Gat, 2011; Sebé-Pedrós et al., 2012). Despite the conservation of the core players, the upstream regulators of the pathway seem to be divergent (Hansen et al., 2015).

Activation of the Hippo kinase cascade results in the phosphorylation of Yki/YAP/TAZ, which inhibits their nuclear import. The upstream kinases of the cascade (Hippo or MST1/2) form a complex with the adaptor protein Salvador (SAV1 in vertebrates) that activates LATS1/2 kinases. LATS1/2 together with MATS/MOB1 phosphorylate and inactivate Yki (YAP and TAZ in vertebrates) by cytoplasmic retention and eventually ubiquitination and degradation (as reviewed in Panciera et al., 2017; Zheng and Pan, 2019). On the other hand, when Hippo is inactive, dephosphorylated YAP/TAZ can translocate into the nucleus and bind to the TEAD(1–4) transcription factors (Lei et al., 2008; Zhao et al., 2008). The YAP/TAZ–TEAD complex activates the expression of target genes that regulate cell proliferation, differentiation, and apoptosis (Dong et al., 2007; Lei et al., 2008; Zhao et al., 2008; Lian et al., 2010; Lee and Yonehara, 2012). Specifically, the YAP–TEAD complex controls gene transcription by mostly binding to distal enhancers (Galli et al., 2015). YAP/TAZ mainly act as co-activators but can also act as co-repressors together with TEAD factors (Beyer et al., 2013; Kim et al., 2015). Moreover, YAP/TAZ can bind to other transcription factors either with or without TEAD (as reviewed in Totaro et al., 2018b). Overall, although YAP/TAZ are the main known mediators of the Hippo pathway during development, YAP/TAZ-activity can be regulated by multiple microenvironmental cues beyond the Hippo pathway.

In the last years, YAP/TAZ have emerged as sensors of mechanical forces (Dupont et al., 2011; Panciera et al., 2017). YAP/TAZ mechanotransduction can be triggered by cell density –either by cell–cell adhesion (see below), or reduced cell area (Aragona et al., 2013; Benham-Pyle et al., 2015)–, ECM rigidity (Dupont et al., 2011; Aragona et al., 2013; Elosegui-Artola et al., 2016), and shear stress (Nakajima et al., 2017). This regulation can be dependent or independently of the Hippo pathway. Within the cell density context, cell junction proteins regulate YAP/TAZ activity. For instance, Neurofibromatosis type 2 (NF2) works as a scaffold of the Hippo cascade components in cell–cell junctions. NF2 recruits LATS1/2 to the plasma membrane, enabling LATS1/2 activation by MTS1/2, which leads to inhibition of YAP/TAZ activity (Yin et al., 2013). Focal adhesions components, which contact the cell at the adjacent ECM, also regulate YAP/TAZ activity by modulating YAP subcellular location (Kim and Gumbiner, 2015; Elosegui-Artola et al., 2016). ECM stiffness and cell geometry control YAP/TAZ activity through small RhoGTPases triggering the actomyosin cytoskeleton tension in a Hippo independent manner (Dupont et al., 2011). Accordingly, F-actin inhibitor proteins mediate the spatial distribution of YAP/TAZ activity by mechanical forces along the tissue (Aragona et al., 2013). Forces exerted from the ECM can drive YAP/TAZ activity through different mechanisms. The Ras-related GTPase RAP2 transduces ECM stiffness into YAP/TAZ cellular responses through the activation of the Hippo pathway (Meng et al., 2018). The nuclear SWI/SNF complex inhibits YAP/TAZ as a response to mechanical signaling, in such a manner that to trigger YAP/TAZ activity, both YAP/TAZ nuclear accumulation and SWI/SNF inhibition are required (Chang et al., 2018). ECM forces can regulate the transport through the nuclear pores driving YAP nuclear import in a Hippo independent manner (Elosegui-Artola et al., 2017). Moreover, Piezo1, a mechanosensitive ion channel, can mediate the effects of substrate stiffness on YAP nuclear location (Pathak et al., 2014). Remarkably, YAP/TAZ are not only modulated by mechanical forces but can contribute to changes in actomyosin-mediated mechanical forces in cell culture and in developing tissues by regulating the expression of cytoskeletal and ECM genes (Porazinski et al., 2015; Lin et al., 2017; Nardone et al., 2017). Furthermore, biochemical cues also control YAP/TAZ activity (for recent reviews see Pocaterra et al., 2020; Heng et al., 2021). Extracellular cues can activate or inhibit YAP/TAZ activity through G-protein–coupled receptors (GPCRs). RhoGTPases mediate YAP/TAZ activation by GPCRs, modulating actomyosin cytoskeleton tension (Yu et al., 2012). Altogether, YAP/TAZ act as core integrators of chemical and mechanical cues in different biological contexts.

YAP/TAZ function as gatekeepers of progenitor cell behavior in several contexts during embryonic development. Specifically, YAP/TAZ have mainly been described as regulators of cell proliferation and tissue growth (Camargo et al., 2007; Lei et al., 2008; Ota and Sasaki, 2008). YAP/TAZ trigger proliferation of gastrointestinal mesenchymal progenitors (Cotton et al., 2017) and cranial neural crest cells (Wang et al., 2016). YAP by itself also controls the proliferation of cardiomyocytes (Heallen et al., 2011) and lung epithelial cells (Lin et al., 2017). Moreover, YAP–TEAD maintains inner ear progenitors through the activation of cell cycle and stemness genes (Gnedeva et al., 2020). In the same line, YAP/TAZ regulate cell proliferation in neural progenitors in the chick spinal cord by controlling their stemness properties through the activation of the cell cycle regulator cyclinD1 and the inhibition of the neural differentiation marker NeuroM (Cao et al., 2008). YAP/TAZ–TEAD signaling also regulates the proliferation of neural progenitors in the mammalian embryonic brain (Han et al., 2015). YAP/TAZ drive the expansion of neural progenitors in the hippocampus downstream of NF2 (Lavado et al., 2013) and maintain neural progenitors in the developing cortex by activating the transcription of proliferation genes and preventing neural differentiation (Lavado et al., 2018). Along this, YAP also maintains the proliferative properties of basal progenitors in the developing ferret and human cortex (Kostic et al., 2019). Finally, YAP/TAZ–TEAD drives proliferation of neural progenitors in the zebrafish hindbrain boundaries downstream of actomyosin tension (Voltes et al., 2019). Although YAP/TAZ play essential roles in tissue growth during embryonic development, they are not required for normal physiology in most of adult tissues. Thus, while YAP/TAZ overexpression have a widely effect driving tissue hyper-proliferation and promoting tissue repair after injury, the deletion of YAP/TAZ in many adult contexts does not result in effects on tissue proliferation. Overall, YAP/TAZ play a crucial role in the self-renewal of progenitor cells during embryonic development.

The Notch signaling pathway is the main regulator of binary cell fate decisions during embryonic development (Artavanis-Tsakonas et al., 1999). Notch signaling operates through cell–cell communication, with one cell displaying the Notch receptor and its neighboring cell the Notch ligand. Nevertheless, Notch receptors and ligands can also form cis interactions that inhibit (de Celis and Bray, 1997; Micchelli et al., 1997) or activate (Nandagopal et al., 2019) Notch signaling in a cell-autonomous manner. The Notch pathway is highly conserved in metazoan species (Gazave et al., 2009). In mammals, there are four Notch receptors (Notch1–4) and five ligands: three Delta ligands (Dll1, Dll3, and Dll4) and two Jagged ligands (Jag1 and Jag2). Upon ligand binding (Delta or Jagged) in its extracellular domain, the Notch receptor undergoes several protease cleavages, thereby releasing the Notch Intracellular Domain (NICD), which then translocates into the nucleus. Once there, the NICD forms a complex with the transcription factor RBPJ (Recombination signal-Binding Protein for Ig Kappa J region) and recruits co-activators such as MAML (Mastermind-Like). The NICD–RBPJ complex activates transcription of the main effectors of the pathway, the transcriptional repressors genes Enhancer of Split (Espl) in Drosophila and Hes/Her in vertebrates. The Hes/Her transcription factors repress genes driving cell specification (e.g., proneural genes), cell differentiation, and cell cycle arrest (reviewed in Kageyama et al., 2007). Hes/Her can also repress – directly or indirectly – Notch ligand expression. Through this lateral inhibition mechanism, one cell is singled out from an equipotent field to acquire a specific fate, repressing this specific fate in the neighboring cells (reviewed in Henrique and Schweisguth, 2019).

However, the lateral inhibition paradigm should not only be viewed from a static perspective. Some of the Notch effectors, such as Hes1 and Hes7, show an oscillatory expression by a negative autoregulatory loop in different developing tissues (Hirata et al., 2002; Bessho et al., 2003; Lahmann et al., 2019; Seymour et al., 2020), resulting in the oscillation of their targets (Masamizu et al., 2006; Shimojo et al., 2008; Lahmann et al., 2019). The oscillation of Hes genes and their targets keep progenitors in an undifferentiated and proliferative state, whereas the sustained expression of one of the target genes drives cell specification (Shimojo et al., 2008; Lahmann et al., 2019). Overall, Hes oscillations constitute a crucial mechanism in the control of binary cell fate decisions during embryonic development.

Moreover, Notch not only controls cell fate through lateral inhibition, but also through lateral induction (de Celis and Bray, 1997). Lateral induction consists of a positive feedback loop in which Notch signaling activates the expression of the Notch ligand, thereby activating Notch signaling in the adjacent cell. Subsequently, both interacting cells acquire the same fate. For instance, Jag1–Notch signaling through lateral induction drives prosensory fate in the developing inner ear (Hartman et al., 2010) and vascular smooth muscle fate in neural crest cells (Manderfield et al., 2012). This highlights another level of complexity in the control of cell fate decisions by Notch signaling in the developing embryo.

The role of the Notch pathway during development is highly context dependent (Bray, 2016). In vertebrates, different combinations and spatiotemporal expression of Notch receptors and ligands account for part of this context dependency. For example, the control of neuronal fates in the zebrafish spinal cord relies on different combinations of Notch ligands and receptors (Okigawa et al., 2014). Different ligands can also trigger distinct responses through the same Notch receptor. For instance, Jagged and Delta ligands drive different outcomes in the control of cell fate decisions during inner ear development and angiogenesis (Benedito et al., 2009; Petrovic et al., 2014). Likewise, Dll1 and Dll4 induce different Notch activation dynamics, driving different gene programs and cell fates (Nandagopal et al., 2018). Additionally, Fringe glycotransferases modify the affinity between Notch receptors and ligands (Panin et al., 1997), providing an extra regulatory layer. Epigenetic mechanisms may explain part of the Notch context dependency. For example, epigenetic modifications in the regulatory regions of Notch targets regulate cell fate decisions in olfactory and cortical neurogenesis (Endo et al., 2011; Tiberi et al., 2012). Moreover, the NICD–RBPJ complex can interact with other transcription factors (reviewed in Bray, 2016). Hence, the interplay between Notch signaling and other pathways is relevant for the diversity of Notch responses. Cell geometry is also crucial for Notch regulation of cell fate during development (Shaya et al., 2017). All this complexity raises the question of how the interactions between Notch signaling, the microenvironment and other signaling pathways contribute to Notch pleiotropic effects during development.

The first binary cell fate decision in the mammalian embryo occurs during the transition from morula to blastocyst, with the decision made between becoming trophectoderm (TE) or inner cell mass (ICM). YAP/TAZ–TEAD drive the specification of the TE fate downstream of cell polarity through Hippo-dependent and independent mechanisms (Nishioka et al., 2009; Cockburn et al., 2013; Hirate et al., 2013; Leung and Zernicka-Goetz, 2013; Lorthongpanich et al., 2013). In the blastocyst, Notch is specifically activated in the outer cells, which will give rise to the TE. Notch and YAP–TEAD activate the expression of the TE specification gene, Cdx2, by binding to its TE-specific enhancer (Rayon et al., 2014). Moreover, the helicase-like protein Strawberry Notch1 (Sbno1) interacts with the YAP–TEAD and NICD–RBPJ complexes, operating as an integrator of these complexes in the activation of the TE-enhancer of Cdx2 (Watanabe et al., 2017). This activation is not redundant since Notch regulates the onset of Cdx2 expression whereas YAP–TEAD maintains Cdx2 expression. Thus, through this mechanism, YAP–TEAD and Notch cooperate in the specification of the TE in a parallel and independent manner (Menchero et al., 2019). In vitro stretching of embryonic stem cells results in the activation of the TE enhancer of Cdx2 in the presence of YAP, TEAD, and Notch (Watanabe et al., 2017). Therefore, mechanical forces are upstream of the Notch and YAP–TEAD synergic mechanism in the binary cell fate decision between TE or ICM (Figures 1A,A’).

The embryonic pancreas contains multipotent progenitors organized in tubular epithelial structures formed by a tip and a trunk domain. Bipotent pancreatic progenitors residing in the trunk domain give rise to the ductal and endocrine cells. The YAP–TEAD complex acts as a main regulator for the maintenance of human pancreatic progenitors by activating several targets, including Hes1 (Cebola et al., 2015). Further, YAP–TEAD forms a transcriptional complex with Hes1 that represses the expression of the endocrine specification gene, Neurog3. Importantly, mechanical signals regulate this cell fate decision both in vivo and in vitro: cell confinement drives endocrine specification whereas cell spreading triggers ductal specification. Accordingly, different ECM compositions define the lineage commitment of the mouse and human pancreatic progenitors. Pancreatic progenitors sensing fibronectin activate the YAP–TEAD–Hes1 complex through the α5-integrin–F-actin axis, thereby repressing the endocrine cell fate. On the other hand, progenitors sensing laminin reduce the activation of the α5-integrin–YAP–Hes1 axis, committing to the endocrine fate (Mamidi et al., 2018). In this scenario, YAP–TEAD is upstream of Notch signaling meanwhile cooperating with Hes1 in the repression of Neurog3 expression to maintain pancreatic progenitors, which leads to the default commitment to the ductal lineage (Figures 1B,B’).

YAP/TAZ forms a common transcriptional complex with the NICD in the control of smooth muscle differentiation from neural crest cells (Manderfield et al., 2015). Notch signaling plays a critical role in the differentiation of cardiac neural crest cells into smooth muscle cells through lateral induction (High et al., 2007). Firstly, the vascular endothelium displays a Jag1 ligand, which activates Notch signaling in the neighboring mesenchyme. Subsequently, Notch activates the smooth muscle differentiation program and Jag1 transcription, generating a positive feedback loop that controls smooth muscle differentiation (Manderfield et al., 2012). Specific deletion of YAP and TAZ in neural crest cells impairs smooth muscle differentiation. YAP/TAZ deletion decreases Jag1 and NICD expression in the mesenchyme, while NICD expression remains intact in endothelial cells. In this context, YAP physically interacts in a TEAD-independent manner with the NICD–RBPJ complex, activating Jag1 enhancer and the Hes1 promoter in vivo and in vitro (Manderfield et al., 2015). This common YAP–NICD–RBPJ transcriptional complex contrasts with the parallel cooperation of YAP and NICD in the trophectoderm (Rayon et al., 2014; Watanabe et al., 2017; Menchero et al., 2019). In brief, Notch and YAP cooperate to activate the transcription of Notch targets controlling smooth muscle fate. Noteworthy, the vascular tissue is highly exposed to mechanical forces during development. Therefore, mechanical forces could be controlling smooth muscle fate through the regulation of YAP/TAZ and Notch.

YAP/TAZ can act upstream of Notch signaling by activating Notch receptors. Notch participates in the binary decision between cholangiocytes and hepatocytes (Kodama et al., 2004; Zong et al., 2009). YAP controls the proliferation of hepatocytes and hepatic progenitors downstream of the Hippo pathway in the adult liver (Camargo et al., 2007; Zhou et al., 2009; Lee et al., 2010; Lu et al., 2010). In this context, YAP promotes the biliary cell fate downstream of NF2 regulation (Zhang et al., 2010). NF2 deficiency increases biliary precursors and cholangiocytes proliferation during development. Remarkably, Notch2 deficiency rescues this phenotype. In the absence of NF2, YAP activates Notch2 expression controlling biliary specification and cholangiocytes proliferation (Wu et al., 2017). Thus, YAP controls biliary cell fate and proliferation through the activation of the Notch2 receptor during intrahepatic bile duct development, as previously described in adult hepatocytes (Yimlamai et al., 2014). In the adult liver, YAP can also act upstream of Notch pathway by activating the expression of Jag1 (Tschaharganeh et al., 2013). In contrast, YAP/TAZ and Notch have been proposed to promote biliary cell fate through parallel mechanisms during development (Lee et al., 2016). Finally, YAP/TAZ and Notch control the binary decision between cholangiocytes and hepatocytes downstream of mechanical forces in the adult liver (Pocaterra et al., 2019, 2021). Whether this binary cell fate decision is controlled by YAP/TAZ–Notch downstream of mechanical forces during embryonic development is still unsolved.

On the other hand, YAP can inhibit Notch activity in a cell-autonomous manner, as it occurs during angiogenesis and somitogenesis. During embryonic angiogenesis, Notch controls the cell fate decision of tip vs. stalk cell. In tip cells, the Notch ligand Dll4 triggers the formation of new sprouts and activates Notch in neighboring stalk cells, leading to tip fate suppression. Notch in stalk cells can also activate Dll4 expression, thus triggering Notch activation in tip cells (Caolo et al., 2010). This mechanism maintains arterial identity while regulating the formation of new branches. YAP/TAZ are activated in tip cells through the activation of the GPCRs, LPA4 and LPA6, mediated by actomyosin tension. YAP/TAZ control sprouting angiogenesis through the blockage of βcatenin–NICD mediated endothelial Dll4 expression. Altogether, YAP/TAZ inhibits Dll4 expression in tip cells in a cell-autonomous and TEAD-independent manner to control sprouting angiogenesis (Yasuda et al., 2019). In other words, as in smooth muscle differentiation, YAP/TAZ acts in a TEAD-independent manner in the regulation of Notch ligand expression. The second case is observed in the presomitic mesoderm (PSM) for the genetic synchronous oscillations of the Notch target Hes7 that precede the formation of somites, known as the segmentation clock. The segmentation clock has been proposed to be an excitable system (Hubaud et al., 2017). In this model, YAP activation provides an excitability threshold and Notch acts as the stimulus triggering the oscillations once it exceeds the threshold. Therefore, the collaboration of YAP and Notch is required for triggering and maintaining PSM oscillations. Importantly, YAP activation in the PSM cells is controlled by mechanical cues, such as cell density. In this scenario, Notch controls the decision between the quiescent and oscillatory state in PSM cells downstream of YAP activation by mechanical cues (Hubaud et al., 2017). In both contexts, YAP inhibits Notch signaling to control cell behavior in a cell-autonomous manner downstream of mechanical cues.

YAP can also be upstream Notch in a non–cell-autonomous manner. During fetal myogenesis, Notch controls the binary cell fate decision between muscle progenitors and differentiated muscle cells (Vasyutina et al., 2007). In the chick embryonic limb, nuclear YAP is expressed in differentiated muscle cells and in a subpopulation of muscle progenitors. YAP controls the maintenance but not proliferation of muscle progenitors downstream of muscle contraction. In muscle fibers, YAP–TEAD binds to the Jag2 promoter and activates Jag2 expression. Consistently, Jag2 expression in muscle fibers as well as Notch in muscle progenitors decrease upon muscle immobilization (Esteves de Lima et al., 2016). Overall, YAP–TEAD drives Jag2 expression in muscle fibers upon muscle contraction, as a result, Jag2 activates Notch in neighboring cells, maintaining the muscle progenitor cell pool in a non–cell-autonomous manner (Figures 2A,A’).

The YAP/TAZ–TEAD and Notch pathways do not always have synergic functions. In the epidermis, basal progenitors specify from the basement membrane to the tissue surface. Notch signaling participates in the cell decision between basal progenitor and the epidermal fate. The NICD–RBPJ complex drives the expression of epidermal differentiation genes (Blanpain et al., 2006). Low cell density or high ECM rigidity trigger YAP/TAZ–TEAD activation in basal progenitors leading to progenitor maintenance through the inhibition of Notch signaling, as shown both in vitro and in vivo (Totaro et al., 2017). This process is cell–cell contact independent. In basal progenitors, the YAP/TAZ–TEAD complex activates the transcription of Dll1 and Dll3; thereafter, cis interactions of Dll1 and Dll3 with Notch receptors can block Notch activation, thus, preventing epidermal differentiation. Altogether, YAP/TAZ–TEAD controls epidermal fate decisions by inhibiting Notch signaling downstream of mechanical signals in a cell-autonomous manner (Figures 2B,B’).

Notch can be upstream of YAP/TAZ during embryonic brain development. Asymmetric divisions play a role in the balance between cell proliferation and differentiation. Radial glial progenitors (RGPs) divide asymmetrically to give rise to a neuron and another RGP. The polarity gene Pard3 is highly expressed in the apical cell surface and regulates asymmetric divisions in RGPs in the mammalian cortex (Costa et al., 2008). Pard3 removal by genetic depletion in mice leads to temporally distinct changes in RGP mitotic behavior: at the early neurogenic phase, it results in increased YAP expression and promotes symmetric proliferative divisions, while at late neurogenic phase, it results in decreased YAP expression and promotes RGP symmetric differentiation divisions. Notch expression decreases during cortical development, coinciding with the RGPs behavioral switch. Accordingly, Notch promotes high YAP levels in the nucleus upon Pard3 removal (Liu et al., 2018). This activation could be through the binding of the NICD–RBPJ complex to the YAP promoter, as described in neural stem cells (Li et al., 2012). Thus, the interplay of Pard3 with Notch and YAP/TAZ could explain the potential role that mechanical signals play in this process. Altogether, the Notch pathway upstream of YAP/TAZ controls the division cell mode and, therefore, cell fate in the developing cortex (Figures 3 A,A’).