Clonal culture of Hydra vulgaris Ind-Pune (Reddy et al., 2011) was maintained in Hydra medium by following standard methods at 18 ± 1°C (Horibata et al., 2004). Polyps were fed daily with freshly hatched Artemia nauplii larvae and washed 6–8 h after feeding. For the regeneration experiment, Hydra polyps starved for 24 h were decapitated just below the tentacle base and allowed to regenerate till 0, 1, 2, 4 and 8 h post-amputation (hpa). The polyps were then fixed and processed for immunofluorescence assay. For the budding experiment, Hydra polyps starved for 24 h were collected at different stages of bud development. These were then fixed and processed for immunofluorescence assay. The different stages of budding were identified and labelled as reported previously (Otto and Campbell, 1977).

Hydra magnipapillata genome draft comprising 82.5% of 1.05 Gbp sequenced genome available as Refseq was initially used for identifying Hippo Pathway core components (Chapman et al., 2010). This assembly turned out to be incomplete, and we were unable to fish out any homologs. Therefore, an in-house transcriptome assembly generated in the Galande laboratory (Reddy PC. et al., 2019) was used for the present study. The in-house transcriptome was merged with the NCBI RefSeq to improve the assembly further to generate a hybrid assembly. The hybrid assembly was found to be 99.6% complete as compared to 95.7% exhibited by NCBI RefSeq (Reddy PC. et al., 2019). Using the stand-alone NCBI BLAST program, hits of homologs of Hippo pathway core components were identified (Madden, 2013). To confirm the hits, Reverse BLAST was performed. Finding a hit of a homolog in different phyla or species would confirm the homolog status. To further confirm, the amino acid sequences of these homologs were searched in HMMER (Hmmer, RRID:SCR_005305) for affirmation based on the hits returned (Potter et al., 2018). Once the homologs were identified, they were further analyzed for domain organization by SMART (SMART, RRID:SCR_005026) (Letunic et al., 2002). After manual evaluation of the domain organization, the domain architecture was constructed to scale using DOG 2.0 software (Ren et al., 2009).

Sequences from different representative phyla were collected based on protein BLAST searches using the Human YAP sequence as a query. The collected sequences were aligned using MUSCLE (Edgar, 2004). The alignment was trimmed using an automated trimAl programme (Capella-Gutiérrez et al., 2009). This alignment was subjected to phylogenetic analysis using FastTree 2 to generate an approximately maximum likelihood (ML) tree (Price et al., 2010). This method was selected after testing PhyML and RaxML as FastTree 2 has given better confidence on branching points and this could be due to the highly divergent nature of the sequences. This pipeline was implemented in the online platform NGphylogeny.fr (Lemoine et al., 2018). Here, the LG substitution model was used with Felsenstein’s phylogenetic bootstrap with 1000 (Lemoine et al., 2019). The phylogenetic tree was visualized by using the iTOL web server (Letunic and Bork, 2019). The tree was rooted using Amphimedon queenslandica YAP-like sequence as an outgroup. The domain organization analysis and visualization were carried out using DoMosaics software (Moore et al., 2014) using embedded HMMER3 tools (Mistry et al., 2013) and Pfam data. The sequences details were provided in Supplementary Table S1.

To analyse the rest of the Hippo pathway components, alignments and molecular phylogenetic trees of the protein sequences were carried out using MEGA 6.0 software (Tamura et al., 2013). MUSCLE algorithm was used for amino acid sequence alignment (Edgar, 2004). The alignment was graphically represented using Jalview (Waterhouse et al., 2009).

Total RNA was extracted from Hydra polyps starved for 48 h and cDNA was synthesized from total RNA using Improm-II reverse transcriptase system (Promega™) according to the manufacturer’s instructions. Hippo pathway genes were amplified by polymerase chain reaction using Pfu DNA polymerase with the following primers:

Hvul_yap_forward:5′-ATGGATATGAATTCTACGCAACGGC-3′,

The PCR conditions: Initial denaturation at 94°C for 5 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at the respective annealing temperatures (Ta) for 45 s and extension at 72°C for 45 s with the final extension 72°C for 5 min. The PCR amplified products were gel eluted using Mini elute kit (Qiagen), followed by A-tailing reaction using KapaTaq enzyme and cloned in pGemT-Easy vector system (Promega™) or TOPO TA cloning vector as per the manufacturer’s instructions. The recombinant plasmids were sequenced using sequencing primers, and the nucleotide sequences of cloned genes were deposited at NCBI Genbank (Hvul_yap- MW650883; Hvul_tead- MW650884; Hvul_hpo- MW650879; Hvul_mob- MW650880; Hvul_lats- MW650881 and Hvul_sav- MW650882).

Digoxigenin-labelled sense and antisense RNA probes were prepared by in vitro transcriptions using recombinant plasmids of target genes made as mentioned above (Roche Life Science) and used for in situ hybridization. Whole-mount in situ hybridization was performed on the polyps as described by Martinez et al. (1997) with the following changes (Martinez et al., 1997). The animals were relaxed for 2 min in 2% urethane. Treatment with proteinase-K was performed for an optimum of 15 min, and heat-inactivation of the endogenous alkaline phosphatases was done at 70°C for 15 min in 1X SSC. Digoxigenin labelled RNA probes at a 200–600 ng/ml concentration of the probe was used for hybridization at 59°C. The post-hybridization washes were performed using 1X SSC-HS gradients. After staining with BM-purple AP substrate for 30 min–1 h at room temperature, the animals were mounted in 80% glycerol for imaging. Imaging was carried out using a ×10 DIC objective lens with Axio Imager Z1 (Zeiss).

The stained polyps were rehydrated to PBS gradually through PBS: methanol gradient (25, 50, 75, and 100% wash each for 10 min). These polyps were then shifted to a 30% sucrose solution by gradually taking it through 10 and 20% for 30 min each. The polyps were left in 30% sucrose overnight. These polyps were then embedded in 10% PVP (polyvinyl pyrrolidone) by making cubes of PVP (1 × 1 × 2 cm³) made from aluminium foil cast. The embedded polyps were then sectioned (25 µm thick) using Leica CM1950 – Cryostat. The sectioned ribbons were collected on a glass slide and covered and sealed under a coverslip. The sections were then photographed under ZEISS Axio Zoom V16 apotome microscope.

t-SNE plots and gene expression plots of Hippo pathway components were generated and extracted from the Single Cell Portal (https://portals.broadinstitute.org/single_cell/study/SCP260/stem-cell-differentiation-trajectories-in-Hydra-resolved-at-single-cell-resolution). In order to use the Single Cell Portal, gene IDs of Hippo pathway components were acquired through a BLAST search in the Juliano aepLRv2 nucleotide database via Hydra 2.0 Genome Project Portal (https://research.nhgri.nih.gov/Hydra/sequenceserver/). To determine the clusters of cells that express individual Hippo pathway components, differential gene expression was analyzed using edgeR, a tool to analyze RNA-seq data using the trimmed mean of M-values (TMM) method. Differential gene expression was calculated as fold change.

Cells from Hydra polyps were dissociated using a protocol described previously with minor modifications (Greber et al., 1992). The Pronase enzyme was replaced by more efficient, pure and negligible endotoxin-containing Liberase™ TL (Roche/Sigma Aldrich). The enzyme was used at a concentration of 0.2 U/ml. Dissociation was performed either in Hydra dissociation medium (Greber et al., 1992) with a slight modification of replacing TES with HEPES (15 mM) for buffering. Fifty Hydra polyps were washed twice in a sterile-filtered Hydra medium and transferred into a 1.5 ml tube. The medium was removed, and 1 ml Liberase in dissociation medium was added. Cells were dissociated for 90 min at room temperature (22–24°C) with gentle agitation on a nutator and followed by gentle pipetting.

A recently published paper reported the presence of Yorkie (YAP/Yki) in Clytia hemispherica and producing polyclonal antibody specific to CheYki in rabbit against the peptide FNRRTTWDDPRKAHS (Coste et al., 2016). This antibody along with the pre-immune serum was kindly gifted by Dr Michaël Manuel (Sorbonne Universités, Université Pierre et Marie Curie (UPMC), Institut de Biologie Paris-Seine (IBPS) CNRS). The antibody was validated by immunofluorescence analysis (the antibody doesn’t work for western blotting experiments). Proliferating cells in Hydra were identified using anti-Ki67 antibody [OTI5D7] (ab156956; Abcam).

Immunofluorescence assay was performed as per the protocol is given in Takaku et al., 2014 (Takaku et al., 2014). Hydra polyps were starved at least for 1 day before fixation. Animals were relaxed in 2% urethane for 1–2 min and fixed in 4% paraformaldehyde (in 1XPBS) overnight at 4°C or 1 h at RT. 1:100 concentration of primary antibody was used. 1:100 concentration of Invitrogen Alexa-conjugated secondary antibodies were used. Invitrogen Alexa 488 conjugated Phalloidin for staining. DAPI was used for nuclear staining. These samples were then imaged on ZEISS Axio Zoom V16 (for regeneration and whole animal images) or Andor Dragonfly Spinning Disc (for budding Hydra) microscopes.

The dissociated cells from Hydra were fixed and stained as described above, using an anti-CheYki antibody. DNA was counterstained using DAPI. Cells were imaged using multispectral imaging flow cytometry (ImageStreamX flow-cytometer; Amnis Corp., Seattle, WA, United States). For multispectral imaging flow cytometry, ∼1 × 10⁵ cells were collected from each sample and data were analyzed using image analysis software (IDEAS 4.0; Amnis Corp.). Images were compensated for fluorescent dye overlap by using single-stained controls. Cells were gated for single cells, using the “area” and “aspect ratio” features, and for focused cells, using the Gradient RMS feature, as previously described (George et al., 2006). Cells were gated for cell size- Large and Small based on the area feature. Cells with an area less than 200 μm² was labelled as “Small” and anything bigger as “Large.” The large cells were found to be non-significant when compared to the cells found in the negative control (due to autofluorescence). The “Small” cells were further gated for nuclear vs cytoplasmic localization. The “similarity” feature was used to achieve this by colocalizing the YAP (Yki) signal with the DAPI signal. Cells with a similarity value greater than two were considered nuclear-localized, and cells with a low similarity value of less than two were considered cytoplasmically localized (See Supplementary Figure S8).

Verteporfin (Vp) was procured from Sigma (SML0534) and dissolved in DMSO. A 2 mM stock was prepared and kept in 50 µl aliquots at −20 C for long-term storage in the dark. The vial was thawed for 30 min in the dark before use. For the assay, the polyps were treated with 5 μM Vp in Hydra medium and the same volume of DMSO in Hydra medium as vehicle control. For the budding assay, the non-budding polyps were collected and incubated in Vp. The Vp solution was replaced every 24 h, and readings for number buds were taken every 24 h. Every step of the Vp treatment assay was performed in the dark. For statistical analysis, the average number of buds per polyp was calculated by measuring the total number of buds from a set of 48 polyps and then divided by 48. For this experiment, three such biological replicates were done (N = 3) with 48 polyps per set (n = 48). Hence a total of 144 polyps were used for the experiment per treatment. For detachment assay, the total number of buds fallen till the specified day was calculated by adding the number of buds fallen on each previous day. This experiment was done using 48 polyps. Statistical significance of the data was calculated by measuring p-value using Heteroscedastic two-tailed t-test (Supplementary Table S2 budding kinetics tab for statistical data).

The 2.8 A° crystal structure of the human YAP-TEAD complex deposited on PDB (4RE1) was as a reference for modelling the TEAD binding domain and YAP binding domains of Hydra YAP-TEAD complex (Zhou et al., 2015). Modeller software (Webb and Sali, 2016) was used to build five optimum models based on 4RE1 in the multi-model mode. Among the five models, the model with minimum DOPE assessment score and maximum GA341 assessment score was chosen for the final analysis. The model was then visualized in CHIMERA for analysis, superpositioning and annotation (Pettersen et al., 2004). The non-covalent bond analysis was performed in Biovia Discovery Studio Visualizer (BIOVIA, 2017). The Binding energy calculations done using PRODIGY web server (Xue et al., 2016).

Core Hippo pathway homologs–hippo/mst, mob, lats, sav, yap and tead were identified from the in-house Hydra transcriptome using NCBI stand-alone BLAST (Reddy PC. et al., 2019). In mammals, hippo, mob, lats and yap have two paralogs each while tead has four paralogs. Sav, on the other hand, has no reported paralogs. The occurrence of these paralogs has been attributed to whole-genome duplication events correlated to certain fish species (Chen et al., 2019). Therefore, any species that evolved earlier than fishes do not contain paralogs reported for Hippo pathway genes. Conforming to these reports, Hydra consists of only one gene coding for each core Hippo pathway component. The Hydra Hippo pathway homologs were labelled as Hvul_hpo, Hvul_mob, Hvul_lats, Hvul_sav, Hvul_yap and Hvul_tead. These homologs in Hydra were confirmed by obtaining the corresponding amplicons from Hydra cDNA (Supplementary Figure S1). Upon determining the nucleotide percent identity with other reported model organisms used for studying the Hippo pathway, we find that Hydra had a higher percent identity with humans than Drosophila (Figure 1B). Hvul_hpo shows about 60% identity with humans and 56% identity with Drosophila. Hvul_sav is comparatively less conserved with a 24% identity with human and 22% identity with Drosophila. Hvul_mob is highly conserved across the animal phyla with about 84% identity with humans and 83% identity with Drosophila. Hvul_lats shares 41.6% identity with humans and 42% identity with Drosophila. Hvul_yap shows 34.6% identity with humans and 34% identity with Drosophila. Hvul_tead exhibits about 65% identity with humans and 59% identity with Drosophila.

An earlier study performed the phylogenetic analysis of YAP homologues found in selective phyla (Hilman and Gat, 2011). However, this analysis did not cover the majority of invertebrate phyla such as Annelida, Mollusca and Echinodermata. This could be due to a lack of reliable data for the identification of the YAP homologues. Here, we have combined the phylogenetic analysis with predicted domain architecture. We used a YAP-like sequence found in Amphimedon queenslandica as an outgroup for rooting the tree. Additionally, a protein sequence with BLAST similarity from a unicellular Eukaryote (Capsaspora owczarzaki) was used for domain organization comparison. In this analysis, we observed that the Hydra homologue of YAP exhibits a strong affinity to the chordate counterparts rather than non-chordate homologues (Figure 1C). An interesting observation after the inclusion of multiple invertebrate phyla in the analysis is that they are highly diverged compared to the Cnidarian and Chordata species. This can be interpreted based on the invalid branch support values (Figure 1C). Additionally, a molluscan homologue, Sepia pharaonic (SEPPH_YAP), showed more similarity with Drosophila Yki and Saccoglossus kowalevskii homologue (SACKO_YAP) showed more similarity with Echinodermata homologues (Figure 1C). Domain organization analysis has led to the identification of variability in the N-terminal homology domain (FAM181). This region contains TEAD binding domain (TBD). Surprisingly, in Capitella teleta (Annelida), Clytia hemisphaerica (Cnidaria), and Ciona intestinalis (Chordata), the FAM181 domain could not be detected (Figure 1C). This could be due to the higher sequence divergence in this region.

A detailed domain analysis using SMART website for the amino acid sequence of Hippo pathway homologs revealed a highly conserved domain organization of the proteins analyzed, indicating a fully functional pathway consisting of these core components (Figure 1A). Hvul_HPO domain analysis revealed a conserved N-terminal Protein kinase domain (PKinase Domain) and a C-terminal SARAH (Salvador-RASSF-Hippo) domain. The presence of these domains indicates the conserved regulation of activation of Hvul_HPO kinase activity (Glantschnig et al., 2002; Praskova et al., 2004; Boggiano et al., 2011). The Hvul_SAV also can be seen to have conserved the SARAH domain required for orchestrating the reported scaffolding activity (Yin et al., 2013). Hvul_LATS domain architecture indicates conservation of the hydrophobic motif (Motif: AFYEFTFRHFFDDGG) (a 40% hydrophobicity confirmed using web-based peptide analysis tool at www.peptide2.com/N_peptide_hydrophobicity_hydrophilicity.php) containing the Threonine residue (T993) required for the activation of LATS by HIPPO phosphorylation (T1079 in humans) (Supplementary Figure S2A) (Hergovich et al., 2006; Ni et al., 2015). The MOB binding motif is highly conserved in Hvul_LATS as compared to the human and mouse (Supplementary Figure S2B). The auto-activation T-loop (near S909 in humans) of Hvul_LATS is 100% conserved (Motif: AHSLVGTPNYIAPEVL) near S830 (Supplementary Figure S2B) (Ni et al., 2015). Hvul_MOB is highly conserved (84% identity with human MOB) as compared to any other components of Hippo pathway homologs in Hydra, indicating highly conserved function. The same site as reported for Human MOB is also highly conserved in Hvul_MOB at T35 (Motif: LLKHAEATLGSGNLR) (Supplementary Figure S2C). This site is crucial for the release of the LATS-MOB complex from the MST-SAV-LATS-MOB complex and further initiation of LATS auto-activation (Ni et al., 2015). Hvul_YAP domain analysis revealed that it had a conserved TEAD-Binding Domain (TBD) and two WW domains. A serine phosphorylation prediction for YAP primary sequence was performed using GPS 2.1 web-based tool (Xue et al., 2010). Based on GPS prediction and manual curation, Hvul_YAP is predicted to have an LATS phosphorylation site at S74 (motif: PIHTRARSLPSNIGQ) and S276 (motif: YTAYMNSSVLGRGSS) homologous to the S127 (motif: PQHVRAHSSPASLQL) and S381 (motif: SDPFLNSGTYHSRDES) (Supplementary Figure S2D). Similar to mammals, a phosphodegron motif (DSGLDG) was identified immediately downstream to the S276 site (S381 in humans), which could be phosphorylated by CK1-ɣ at S287 (S388 in humans) of Hvul_YAP (Supplementary Figure S2E) (Zhao et al., 2010). These analyses indicated that the Hippo pathway effector protein YAP is well equipped for regulation by the LATS and CK1-ɣ. With its defined TEAD binding domain and WW domain, it could interact with transcription factor TEAD and other reported PPXY domain-containing proteins.

The Hippo effector protein YAP is known to elicit its biological function as a transcription co-effector by interacting with transcription factors. Presently, YAP is known to interact with TEAD, β-catenin, SMAD, RUNX, p73 and ErbB4 for regulating their transcriptional responses as an activator or repressor (Strano et al., 2001; Komuro et al., 2003; Zhao et al., 2008; Szeto et al., 2016; Passaniti et al., 2017; Pan et al., 2018). Among these, YAP-TEAD interaction has been extensively studied and is known to be important for regulating cell growth and size and tissue architecture (Totaro et al., 2018). The interaction of YAP and TEAD was first shown to form through their specific interaction domains in 2001 (Vassilev et al., 2001). The structural features of this interaction in humans were first demonstrated in 2009, showing how the TEAD binding domain (TBD) in YAP (amino acids 53–99) interacted with the YAP binding domain (YBD) in the TEAD (position: amino acids 209–426) (Li et al., 2010). The YBD consists of 12 β strands that arrange into two β sheets in an opposing fashion to form a β-sandwich fold. The four α helices from the YBD are arranged at the two ends of the β-sandwich fold for stabilizing the structure. The study showed that TBD-YBD interaction occurs over three interfaces. Each interface consisted of one of the following secondary structure of the TBD-the β1 strand, α1 helix or α 2 helix responsible for interacting with the globular YBD of the TEAD at the C-terminal. It was shown that the β1 strand of TBD interacted with the β7 strand of the YBD (interface 1), The α1 helix from TBD interacted with α3 and α4 helices of the YBD (interface 2). The α2 of the TBD was bound to the YBD through its interaction with α1 and α2 helices (interface 3) (Li et al., 2010).

Amino acid sequence alignment of the predicted YBD (amino acids 240–251) and predicted TBD (position: 1–58) of Hvul_YAP and Hvul_TEAD respectively with Human YAP and TEAD revealed 71.2% sequence identity (82.9% sequence similarity) of YBD (Figure 2A) and a 37.9% sequence identity (56.9% sequence similarity) of TBD (Figure 2B) which indicates plausible structural conservation and hence interacting capability of TBD with YBD. To confirm the same, the 3D structure of the YBD and TBD of Hvul_YAP and Hvul_TEAD was modelled using MODELLER software (Webb and Sali, 2016). The modelling was done based on the 4RE1 X-ray diffraction structure deposited at Research Collaboratory for Structural Bioinformatics PDB (RCSB PDB-https://www.rcsb.org/) models the interaction of human homologs of TBD and YBD at a resolution of 2.20 Å. The model generated from the Hydra homologs was superimposed on the human YAP (hYAP) and hTEAD structure from 4RE1 and was found to be highly structurally similar (RMSD for Hvul_YAP:hYAP- 0.338 A° and for Hvul_TEAD:hTEAD- 0.310 A°) and indicated a conserved interaction capability of Hvul_YAP and Hvul_TEAD (Figure 2C). The modelled YBD-TBD complex of Hydra clearly shows how three different regions- Region 1, Region 2 and Region three of TBD (purple) interacts with the globular YBD (green) by non-covalent bond interactions (Supplementary Figure S3A). The Region 1 interface consisting of TBD β1 (amino acids 10–17) and YBD β7 (358–363) strands interact with seven hydrogen bonds in the human complex, forming an anti-parallel β sheet (Li et al., 2010). In Hydra counterpart, there are only six hydrogen bonds (green dotted lines) due to the presence of Gln18 in β1 instead of Gly59 found in humans (Li et al., 2010), introducing a rotation in the preceding Arg, which disables it from forming a hydrogen bond (Supplementary Figure S3B). The 2^nd interface (Region 2) has the α1 helix of the TBD (amino acids 20–32) fitting right into the binding groove of the YBD formed by the α3 and α4 helices of the YBD (amino acids 385–409) (Supplementary Figure S3C). Similar to humans, this region is mainly mediated by hydrophobic interactions with the α1 helix of the TBD having conserved LXXLF motif for hydrophobic groove binding (Li et al., 2010). This interaction mainly consists of Leu24, Leu27and Phe28 from TBD and Try386, Lys393 and Val406 of YBD (pink dotted lines). In Hydra, few hydrogen bonds (green dotted lines) not found in humans may lead to a more stable interface. The third region (third interface) consists of a twisted coil and α2 helix (amino acids 42–58) from the TBD interacting deeply with the pocket formed by the α1 helix, β4, β11 and β12 helices of the YBD. This region was found to be indispensable for the YAP-TEAD complex formation in humans (Li et al., 2010). Region three in Hydra contains the hydrophobic side chains of the TBD–Phe44 (Met86 in humans), Leu49, Pro50 and Phe53, forming extensive van der Waals interactions with the YBD of TEAD at Glu280, Ala281, Ile282, Gln286, Ile287, Leu312, Leu316, Val431, His444 and Phe446 (Supplementary Figure S3D). The interface is further strengthened by multiple hydrogen bonds (indicated in green dotted lines) – TBD_Arg47:YBD_Gln286, TBD_Lys48:YBD_Gln286, TBD_Ser52:YBD_Glu280 and YBD_Lys314:TBD_Phe53. The hydrophobic interactions in region three consist of Phe53, Pro50 and Phe44 from TBD and Lys314, Glu408 and Phe446 from the YBD. In comparison, the hydrophobic interactions involving Pro56 and Pro57 from TBD with Trp316 and His444 respectively help to push the proline residues out of the hydrophobic pocket. One of the unique aspects that can be predicted from the model is that the Hydra region three YAP-TEAD complex can form two salt bridges (orange dotted lines) - TBD_Arg47:YBD_Asp289:YBD_Asp289 and TBD_Lys48:YBD_Asp283:YBD_Asp451. The human complex only forms a salt bridge at TBD_Arg89:YBD_Asp249: YBD_Asp249. These observations indicate a more stable YAP-TEAD interaction in Hydra as compared to humans. To shed more light on the same, the computationally calculated binding energy of the YAP-TEAD complex between the two organisms were compared using the web-based server PRODIGY (PROtein binDIng enerGY prediction) in Protein-protein mode (Xue et al., 2016). The ΔG of the YAP-TEAD complex in humans is about −6.8 kcal mol^−1, while the complex in Hydra has a value of −14.7 kcal mol⁻¹. This large difference in the binding energy supports the possibility that the YAP-TEAD complex in Hydra is much more stable.

An RNA WISH study of Hvul_hpo showed expression throughout the gastric region (Figure 4A). No expression was observed at the differentiated zones of hypostome, tentacle or basal disk, which might indicate a role in stem-cell maintenance or differentiation but not in terminally differentiated cells. There is a slight reduction in expression at the budding zone and early buds, which might indicate the antagonistic role of HPO towards YAP activity in areas of high mechanical stress as reported in other organisms. Hvul_hpo expression can also be seen at mature bud-parent polyp boundary, indicating a fine-tuning of regulation of Hippo pathway-dependent during bud detachment. Hvul_mob expression showed a similar pattern to Hvul_yap with a distinct down-regulation at the basal disk region of both adult and budding Hydra (Figure 4B). Hvul_sav expression reflected the expression pattern of Hvul_hpo, indicating a similar role. It can also be noted that there is a marked reduction in expression at the budding region, early and late buds, unlike the Hvul_hpo (Figure 4C).

A recent study reported high-throughput sequencing of the transcriptome of 24,985 single Hydra cells using Drop-seq and identified the molecular signatures of various cell states and types (Siebert et al., 2019). The differential expression of the Hippo pathway components and their pattern were examined using the Single Cell Portal. The expression patterns of Hvul_yap, Hvul_tead, Hvul_hpo, Hvul_lats, Hvul_sav & Hvul_mob were queried. From the single-cell data, Hvul_yap expression was insignificantly dysregulated or differentially expressed between cell types (Supplementary Figure S5). Surprisingly, such a trend was commonly observed between all the other Hippo pathway components, namely, Hvul_tead, Hvul_hpo, Hvul_lats, Hvul_sav & Hvul_mob. This indicates a slight disparity with the WISH data. This could be due to the lack of enough resolution from the datasets used. The data shown here only represent a relative fold-change between the cells and may indicate that the expression levels are relatively the same. The WISH data also indicate that most of the cells express almost all types of Hippo pathway components, yet at the same time, we see that they are excluded from some regions. These may be highly stage-specific and hence difficult to be picked up in sc-RNAseq of whole polyps. Nevertheless, the findings from analyzing single-cell data argue in favor of the fact that the Hippo pathway components are essential for cells and need to be expressed in almost all cell types. Their activity might be regulated at the protein level, and hence a protein-based analysis is essential to better understand the regulation of Hippo pathway in Hydra.

Verteporfin (Vp) is a benzoporphyrin derivative small inhibitor routinely used as a potent inhibitor for YAP-TEAD interaction (Brodowska et al., 2014). Vp binds to YAP and prevents it from interacting with TEAD. A budding assay was performed to assess the effects of Vp on the budding of Hydra polyps throughout 10 days of treatment. Vp treatment led to an increase in the average number of buds per polyps (the total number of buds from a set of 48 polyps were calculated and then divided by 48 to obtain the value) (Figure 9A). To confirm that the increased buds per polyp were not due to delayed rate of bud detachment, an assay was performed to quantitate the cumulative number of fallen buds (detached). As shown in Figure 9B, the detachment rate did not change drastically in the presence of Vp. These observations indicate that Vp treatment causes the polyps to produce more new buds than the control. Since the budding rate is indicative of the rate of cell proliferation and maintenance of steady-state, Vp treatment is also indicative of increased cell proliferation.