{"paper_id":"27e6251d-920b-47b5-8dca-65e8f585ce52","body_text":"1 \nMesoglea biogenesis reveals a cryptic aboral valve for \npressure regulation in cnidarian morphogenesis \n \nSoham Basu1,2, Petrus Steenbergen1, Florian Gabler1, Alexandre Paix1,3, Paolo Ronchi4, Gleb \nBourenkov5, Thomas Schneider 5, Jonas Hellgoth 6, Anna Kreshuk 6, Suat Özbek 7, Aissam \nIkmi1* \n \n Affiliations: \n1. Developmental Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany. \n2. Collaboration for Joint PhD Degree between EMBL and Heidelberg University, Faculty of \nBiosciences, 69117 Heidelberg, Germany \n3. Department of Algal Development and Evolution, Max Planck Institute for Biology Tübingen, Max-\nPlanck-Ring 5, 72076 Tübingen, Germany \n4. Electron Microscopy Core Facility, European Molecular Biology Laboratory, Heidelberg, Germany. \n5. European Molecular Biology Laboratory, EMBL Hamburg c/o DESY, Notkestrasse 85, 22603, \nHamburg, Germany \n6. Cell Biology and Biophysics Unit, European Molecular Biology Laboratory, Heidelberg, Germany. \n7. University of Heidelberg, Centre for Organismal Studies, Department of Evolutionary Neurobiology, \nIm Neuenheimer Feld 230, 69120 Heidelberg, Germany \n \n*Corresponding Author: Aissam Ikmi (aissam.ikmi@embl.de)  \n \n \nAbstract \n \nCnidarians are classically defined by a single oral opening, a hallmark of the “blind \ngut” model in early animal evolution. Here, we identify a pressure-sensitive aboral valve in \nNematostella vectensis that operates independently of digestion. This valve dissipates \nelevated hydraulic pressure during morphogenesis, by expelling fluid through transient \nepidermal ruptures triggered by muscular ring opening. This unexpected function emerged \nfrom a comprehensive analysis of mesogleal basement membrane biogenesis. We show that \nthe global dynamics of this extracellular matrix transduce muscular hydraulics to drive tissue \nrearrangement and stabilize shape, while localized FGFRb-dependent matrix remodeling \nestablishes the aboral valve. By positioning the mesoglea as an integrator of biomechanics, \ntissue remodeling, and aboral valve function, these findings expand non-bilaterian openings \nbeyond the digestive paradigm as a hydraulic regulator. \n \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n2 \nIntroduction \nThe emergence of multicellular animals required major innovations in tissue \narchitecture and mechanical coordination. Central to this evolutionary transition was the \nextracellular matrix (ECM), a structural scaffold essential for shaping body plans and \nmediating morphogenetic signaling ( 1–3). In cnidarians—such as jellyfish, corals, and sea \nanemones—which represent one of the early branching metazoan lineages ( 4), the ECM is \norganized into a distinctive compartment known as the mesoglea, positioned between the \nectoderm and endoderm within their diploblastic body plan ( 5–7). This fibrous interstitial \nmatrix is bordered by basement membranes and has diversified across cnidarian lineages, \nranging from thin sheets in sea anemones to voluminous matrices in jellyfish. The mesoglea \noffers critical insights into the primordial diversification of ECM-tissue functions.\n \nUnlike vertebrates with rigid internal skeletons or arthropods with external \nexoskeletons, cnidarians rely on a hydrostatic system. In this system, the mesoglea acts as a \ndynamic elastic antagonist to muscle contractions, counterbalancing the pressure within their \nfluid-filled body cavity ( 8). This hydraulic mechanism generates and distributes internal \npressure, enabling these seemingly simple animals to perform complex movements and \nmaintain their body shape without hard structural elements ( 9–12). This dependence on \ninternal pressure as a physiological driver underscores the need for precise regulatory \nmechanisms. Yet how this regulation is achieved and how it integrates with tissue \narchitecture remains poorly understood. The prevailing “blind gut” model, in which the oral \nopening serves both ingestion and excretion, has long shaped models of early animal \nevolution ( 13–15). This anatomical paradigm distinguishes cnidarians from bilaterians, in \nwhich a through-gut evolved to compartmentalize digestive and excretory functions. \nHowever, whether this architectural simplicity accommodates specialized mechanisms for \npressure regulation has not been explored. \nGiven the fundamental role of hydraulics in shaping cnidarian form and function (8, \n12, 16), understanding how these forces are integrated with tissue architecture—particularly \nthe mesoglea—is essential not only for uncovering the principles underlying their \ndevelopment and behavior, but also for informing bioinspired design in fields such as soft \nrobotics ( 13, 14). Despite its importance, the embryonic origin, assembly, and mechanical \nintegration of the mesoglea are still unresolved. To address this, we leveraged the genetically \ntractable sea anemone Nematostella vectensis  ( 17, 18) to dissect mesogleal basement \nmembrane biogenesis. During the larva-to-polyp transition, axial elongation is driven by \nmuscular hydraulics and arrested by experimental depressurization ( 16). This period of \nmorphogenesis coincides with extensive ECM remodeling, including the upregulation of \nmatrix-modifying enzymes and basement membrane components like Collagen IV and \nLaminin (19). Together, these features make this developmental window an ideal context to \ninvestigate how mesogleal assembly interacts with hydraulic forces to orchestrate \nmorphogenesis. \nHere, we uncover the developmental origins and biomechanical integration of the \nmesogleal basement membrane during Nematostella  morphogenesis. To achieve this, we \ncombined genetic knock-ins, quantitative imaging, 3D electron microscopy, and molecular \nand biophysical perturbations. Unexpectedly, our analysis revises the classical view of \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n3 \ncnidarian body architecture by revealing a pressure-sensitive aboral valve regulated by \nlocalized ECM remodeling and muscular control. During morphogenesis, this structure \nresponds to elevated hydrostatic pressure within the body cavity by facilitating controlled \nfluid expulsion, thereby maintaining hydraulic pressure homeostasis. \n \nResults  \nEmbryonic origin of mesogleal basement membrane \n         To identify which embryonic tissue generates mesogleal basement membrane, we \nperformed live imaging of the eGFP::ColIV knock-in (KI) line ( 18) during early  \ndevelopment (Fig. 1A; Movie S1). At the blastula stage (~12.5 hours post-fertilization, hpf), \nwe observed that eGFP::ColIV was initially expressed in a polarized pattern marking the pre-\nendodermal plate prior to gastrulation. This expression intensified as these cells underwent \ninvagination. To complement live imaging, we also examined fixed samples at multiple \ndevelopmental stages, from the onset of gastrulation to the primary polyp (Fig. 1B). During \ngastrulation, eGFP::ColIV was primarily localized intracellularly in punctate structures, \nconsistent with active synthesis. As development progressed, it transitioned to an \nextracellular localization upon completion of gastrulation, forming the basement membranes \nlining both germ layers (Fig. 1, B and C). Throughout this period, the mesoglea thickness \nremained relatively constant at approximately 2µm (Figure 1, C and D), while intracellular \neGFP::ColIV was consistently restricted to the endoderm/gastrodermis (Fig. 1C). Within this \ntissue, it displayed a distinct subcellular distribution, with large apical puncta and smaller \nbasal puncta. Given their proximity to the forming ECM, these basal puncta likely represent \nsites of Collagen IV secretion. \nTo track the temporal dynamics of Collagen IV deposition, we quantified both \nintracellular and extracellular eGFP::ColIV fluorescence intensities throughout development \n(Fig. 1, C and D; fig. S1A). These measurements were correlated with body aspect ratio \n(A/R; length divided by width), as a proxy for developmental progress during axial \nelongation. As eGFP::ColIV was uniformly distributed throughout the central body region \n(fig. S1A), we calculated the average local fluorescence intensities at each stage. We \nobserved that extracellular eGFP::ColIV levels increased from gastrula until mid-planula, \nthen plateaued during later development (Fig. 1D). This pattern was recapitulated by \nimmunostaining for Collagen IV ( 20) and laminin ( 19) (fig. S1B), although Collagen IV \nstabilized earlier than laminin, suggesting a sequential assembly of basement membrane \ncomponents. Collectively, these results demonstrate that the endoderm is the main and \ncontinuous source of developmentally regulated Collagen IV production. \n \nDevelopmental patterning of mesogleal basement membrane \nAfter characterizing Collagen IV production across development, we next \ninvestigated how it is spatially organized, along with Laminin, to form a functional basement \nmembrane. In gastrula, Collagen IV displayed a diffuse distribution, while Laminin was not \nyet detectable (fig. S2, A and B). In early larval stages, the diffuse pattern of Collagen IV \nbegan to resolve into a more structured arrangement (fig. S2B), while Laminin incorporated \nbroadly across the tissue and exhibited early signs of spatial organization.  As development \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n4 \nprogressed, localized enrichment of both Collagen IV and Laminin emerged at sites of \ndeveloping endodermal folds, marking the segment boundaries of future gastrodermal \nmesenteric folds (Fig. 1, E and F; fig. S2). These segment boundaries (inter-segments) \ncontinued to accumulate Collagen IV and Laminin, and new structures appeared—short \nlateral bridges spanning adjacent segments that were marked by Collagen IV, but not \nLaminin (Fig. 1F; fig. S2B). The spatial density of these bridges increased progressively \nthroughout development. Within the body wall segments (intra-segments), both Collagen IV \nand Laminin adopted an undulating arrangement that became increasingly pronounced in \npolyps (Fig. 1G; fig S2B). These spatial patterns were consistently observed in both the \neGFP::ColIV KI line and animals stained for Collagen IV and Laminin. \n \nTo examine how this structural organization emerges, we used photoconversion \nexperiments with the Dendra2::ColIV KI line ( 21) to track ECM remodeling over time. In \nlarvae, we photoconverted discrete lateral patches of Dendra2::ColIV (magenta) in intra-\nsegment regions (fig. S2, C and D). These patches realigned into an axial undulating \nconfiguration and incorporated newly synthesized Dendra2::ColIV (green) (Fig. 1I), \nindicating that pre-existing Collagen IV is continuously being remodeled as part of a \ndynamically evolving network. We also analyzed Collagen IV dynamics in inter-segment \nregions (fig. S2, E and F) to determine whether lateral bridges formed through sequential \naddition or by intercalation among existing ones. If formation occurred via sequential \naddition, photoconverted (magenta) and newly synthesized (green) bridges would remain \nspatially distinct. Instead, we observed extensive mixing between old and new bridges, \nsupporting an intercalation model in which new bridges integrate between pre-existing ones \n(Fig. 1K). These results reveal the remodeling of the developing mesogleal basement \nmembrane which progressively transitions from a diffuse, unstructured matrix to a spatially \norganized scaffold during morphogenesis. \n \nEndodermal morphogenesis drives basement membrane organization \nTo investigate the cellular processes underlying basement membrane organization, we \nexamined the relationship between endodermal morphogenesis and ECM architecture. During \ndevelopment, the endoderm undergoes both tissue folding and differentiation into a spatially \npatterned musculature composed of circular and longitudinal muscles ( 22). These \nmorphogenetic processes may influence basement membrane organization through \nmechanical interactions and/or localized ECM production. To simultaneously visualize \nCollagen IV organization and muscle development, we combined the eGFP::ColIV KI line \nwith F-actin staining (Fig. 1, F and G). In early larvae, Collagen IV intensity at endodermal \nfolds progressively increased in parallel with the differentiation of parietal longitudinal \nmuscles into thick bundles (Fig. 1F). At the same time, the emergence of short lateral bridges \ncoincided with the formation of circular muscles (Fig. 1F). As development proceeded, both \nthe lateral bridges and the intra-segmental undulating Collagen IV pattern became more \npronounced, paralleling the maturation of circular muscles (Fig. 1G). To test whether \nendodermal morphogenesis influences basement membrane architecture, we performed \ntargeted perturbations of key developmental regulators. BMP knockdown (KD), which \ndisrupts endodermal folding and abolishes longitudinal muscle formation (23, 24), resulted in \na complete loss of Collagen IV and Laminin at segment boundaries, including the lateral \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n5 \nbridges (fig. S2G). Tbx20 KD, which impairs muscle patterning ( 16), also resulted in a \ndramatic disorganization of both Collagen IV and Laminin (fig. S2G). These findings \nindicate that the architecture of the mesogleal basement membrane is actively shaped by \nendodermal morphogenetic processes. \n \nECM modulators alter basement membrane composition   \nHaving characterized the biogenesis of the mesogleal basement membrane, we next \ninvestigated its functional role during development. We initially performed Collagen IV KD \nexperiments using shRNA ( 25) in the eGFP::ColIV knock-in (KI) line. In Collagen IV KD \nembryos, eGFP fluorescence was completely abolished, confirming efficient KD (fig. S3A). \nThese embryos exhibited severe tissue disorganization in both the ectoderm and endoderm, \nincluding a failure of adhesion between the two germ layers (fig. S3B). Given the essential \nrole of Collagen IV in early embryogenesis, we shifted our focus to perturbing basement \nmembrane remodeling during post-embryonic development. To this end, we used two \npharmacological inhibitors targeting distinct steps in ECM regulation (fig. S3C): GM6001, a \nbroad-spectrum matrix metalloproteinase (MMP) inhibitor ( 21), and 2,2’-bipyridine (BPY), \nan inhibitor of prolyl-4-hydroxylase (P4HA) (26). GM6001 blocks ECM degradation, leading \nto accumulation of ECM components, whereas BPY impairs collagen hydroxylation, \npreventing proper polymerization and integration of Collagen IV into the basement \nmembrane. These treatments were expected to alter ECM composition and, in turn, modulate \nthe biomechanical properties of the mesoglea. \nPharmacological treatments were applied to 3-day-old planula larvae and maintained \nfor three days during the larva-to-polyp transition. GM6001 treatment led to Collagen IV \naccumulation and mesoglea thickening (Fig. 2, A and B), while Laminin levels remained \nlargely unchanged. Despite the increased Collagen IV deposition, the overall intra- and inter-\nsegmental organization of Collagen IV and Laminin was largely preserved (Fig. 2A). In \ncontrast, BPY-treated larvae showed a substantial reduction in Collagen IV level, whereas \nLaminin level was relatively unaffected (Fig. 2A). The basement membrane in BPY-treated \nanimals appeared loosely organized and highly irregular, with increased variability in \nmesoglea thickness compared to controls (Fig. 2B). Together, these results demonstrate that \nGM6001 and BPY exert distinct and opposing effects on Collagen IV levels, enabling us to \nexperimentally perturb the steady state of Collagen IV accumulation that normally stabilizes \nafter mid-planula stage. \n \nMesoglea integrity controls axial elongation and hydraulic homeostasis \nWe next used quantitative live imaging to determine how GM6001 and BPY \ntreatments influence larva-to-polyp morphogenesis—a process that requires coordinated \nchanges in body shape and size driven by muscle-generated fluid pressure (16) (Fig. 3A; \nMovie S2). Both treatments arrested axial elongation midway through the developmental \ntrajectory (Fig. 3A; fig. S4A; Movie S2), indicating that intact ECM remodeling is essential \nfor successful morphogenesis. However, their trajectories following arrest diverged \nmarkedly. GM6001-treated animals remained morphologically stable after arrest, whereas \nBPY-treated animals exhibited a progressive reversal of elongation, characterized by \nreductions in both shape and size (Fig. 3A; Movie S2). Strikingly, this regression in BPY-\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n6 \ntreated animals was accompanied by abrupt leakage from the aboral pole, during which \ninternal cavity contents were expelled into the surrounding medium (highlighted by dashed \nlines in Fig. 3B; Movie S3). This aboral leakage, presumably driven by elevated internal \ncavity pressure (16), led to a rapid loss of cavity volume and a corresponding decrease in \nbody size (Fig. 3, A and B). To determine whether this phenomenon was specific to BPY \ntreatment, we re-examined elongation dynamics in GM6001-treated and control animals. No \nleakage events were observed in GM6001-treated animals (Movie S2), while control animals \nexhibited occasional aboral leakage near the end of elongation (Fig. 3B; Movie S3), \nindicating that pressure release at the aboral pole may occur naturally but is normally \nrestricted to later developmental stages. These observations suggest the presence of a \npreviously unrecognized pressure-responsive zone at the aboral pole that facilitates fluid \nrelease during morphogenesis (Fig. 3C). Under normal conditions, this region resists leakage \nduring early development, but its structural integrity is compromised when ECM composition \nis disrupted.  \nMore broadly, these experiments indicate that a finely tuned mesoglea composition is \ncritical not only for sustaining body elongation and maintaining overall shape but also for \npreserving internal hydraulic homeostasis throughout morphogenesis. Inhibition of matrix \nmetalloproteinases by GM6001 reduces ECM turnover and promotes collagen accumulation, \nreinforcing the mesoglea and enabling tissues to withstand internal pressure— though at the \ncost of preventing tissue remodeling and arresting elongation . In contrast, BPY treatment \ndestabilizes and weakens the mesoglea, rendering the tissue unable to sustain hydraulic stress. \nThis effect results in both a failure to elongate and premature aboral leakage, culminating in a \nprogressive loss of shape stability and hydraulic homeostasis. In the following sections, we \ndissect the mechanisms by which the mesoglea ( i) supports body elongation and ( ii) \nmodulates aboral leakage during morphogenesis.\n \n \nGlobal ECM modulation impairs axial tissue rearrangement  \nSince body elongation during the larva-to-polyp transition depends primarily on tissue \nremodeling rather than cell proliferation ( 16), we investigated how GM6001 and BPY \ntreatments affect the underlying cellular mechanisms. In previous work, we showed that \nelongation proceeds through two distinct phases: an initial stage characterized by changes in \ncell shape that thin the body wall and enable surface expansion, followed by a phase of \noriented tissue rearrangement that drives directional elongation along the oral–aboral axis \n(16). In control animals, tissue thinning accounted for early body shape changes, whereas \ncontinued elongation was achieved through axial tissue reorganization (Fig. 3D; fig S4B).  \nBecause GM6001- and BPY-treated animals arrested elongation at an intermediate \nstage and phenocopied the effects of muscle anesthetics ( 16) (Fig. 3A; fig S4B), we \nhypothesized that ECM perturbation disrupts muscle-driven tissue rearrangement without \nimpairing body wall thinning. Consistent with this, drug-treated animals exhibited degrees of \nbody wall thinning comparable to controls (fig S4B), while photoconversion experiments \nusing Kaede-labeled larval tissue revealed striking defects in axial tissue rearrangement (Fig. \n3D). In GM6001-treated animals, photoconverted patches showed minimal elongation along \nthe body axis, indicating limited rearrangement (Fig. 3D). In contrast, BPY-treated animals \nexhibited lateral expansion of photoconverted patches rather than axial elongation, consistent \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n7 \nwith their reversal of body elongation and rounded morphology (Fig. 3D). These results \nsuggest that ECM modulation alters mesoglea organization in ways that specifically interfere \nwith axial tissue rearrangement driven by muscular hydraulics. \n \nLocalized mesoglea remodeling supports a pressure-sensitive aboral valve \nIn addition to its global role in tissue remodeling, the mesoglea may contribute to \nregion-specific functions during morphogenesis. Early aboral leakage in BPY-treated animals \nand its delayed onset in developing wild-type polyps suggest a localized difference in tissue \nbehavior at the aboral pole. These observations raise the possibility of spatial heterogeneity in \nmesoglea properties. To explore this possibility, we examined the structural organization of \nthe aboral pole in greater detail. We first analyzed the distribution of Collagen IV and \nLaminin in primary polyps (Fig. 4A). At the aboral pole, Collagen IV displayed a localized \nreduction, forming a discrete ~5\n/i1 µm-wide region with diminished signal. In contrast, \nLaminin remained broadly distributed but showed local thickening in the same area, \nconsistent with spatially restricted basement membrane remodeling. To further examine \ntissue architecture, we performed focused ion beam scanning electron microscopy (FIB-\nSEM) on a primary polyp, acquiring a volume of 3500 µm³ at nanometer-scale resolution, \nwhich we subsequently segmented to characterize tissue organization (Fig. 4B, Movie S4). \nDespite the occurrence of fluid leakage, we found no evidence of a persistent discontinuity in \nthe epidermal surface, ruling out the presence of a continuously open pore. Instead, we \nidentified a rosette-like arrangement at the basal epidermis, where neighboring cells had lost \ndirect contact, forming a structured extracellular space (Fig. 4B, Movie S4). Within this \nspace, protruding gastrodermal cells were observed, indicating a specialized aboral tissue \narchitecture. \nTo determine whether this aboral tissue architecture is consistently present across \nindividuals, we imaged FGFRb-eGFP transgenic polyps ( 27), in which a cluster of \ngastrodermal cells at the aboral pole is fluorescently labeled (21). These clusters adopted two \ndistinct configurations: a funnel-like structure with a contin uous surface or a flattened \nstructure featuring a central gap (Fig. 4C; Movie S5), suggesting that the gastrodermal \ninterface cycles between closed and open states. Additionally, FGFRb-eGFP–positive \nclusters formed concentric muscle rings characterized by enrichment of F-actin and \nphosphorylated myosin light chain (p-MLC) (Fig. 4C). This enrichment was particularly \nprominent at the tip of the funnel structure, suggesting that muscles actively control the \nconfiguration of this anatomical feature. To confirm that this structure functions as a \npressure-responsive valve, we artificially inflated the body cavity by introducing a dye-filled \ncapillary through the mouth, connected to a pressure pump (Fig. 4D, Movie S6). Upon \npressurization, dye was consistently expelled through the aboral pole rather than the mouth, \nshowing that this anatomical feature serves as a directional release valve activated by \nelevated internal pressure (Fig. 4D). Interestingly, the epidermis above the muscular rings \nalternated between continuous and disrupted states (Fig. 4C), suggesting that pressure release \ntransiently induces epithelial rupture.  \nTo test whether this rupture triggers a wound response, we examined ERK signaling, \na conserved marker of epithelial injury across metazoans (28 ). In control polyps, \nphosphorylated ERK (pERK) was variably activated in the aboral epidermis, with signal \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n8 \nintensity ranging from low to high across individuals (fig. S5). In contrast, physically injured \npolyps exhibited robust pERK activation in both the aboral epidermis and gastrodermis. \nThese results suggest that pERK activation in uninjured polyps reflects transient, pressure-\ninduced epithelial wound.  \nCollectively, the data support the existence of an aboral valve that mediates internal \npressure release through cycles of epithelial disruption and repair. Acting as a \"physiological \nwound,\" this dynamic structure enables rapid decompression when the oral opening is sealed, \nwith its activity regulated by muscle control and localized ECM remodeling (Fig. 4E). \n \nFGFRb-dependent aboral valve morphogenesis  \nTo investigate how the aboral valve forms during development, we examined the \norganization of Collagen IV and Laminin at the aboral pole across developmental stages. In \nembryos, Collagen IV was uniformly distributed throughout the mesoglea, with Laminin \nundetectable (fig. S2A). During the larval stage (fig. 6A), Collagen IV became progressively \ndepleted across a broad region of the aboral mesoglea, forming a distinct depletion zone that \ncontracted in size as development proceeded. Laminin was also reduced in this region during \nthe larval period. However, during the larva-to-polyp transition (fig. 6A), Laminin levels \nwere restored and became specifically enriched within the Collagen IV-depleted zone. These \nspatially divergent remodeling patterns were disrupted in GM6001- and BPY-treated animals \n(fig. 6B), suggesting that mesoglea weakening at the aboral pole is a temporally coordinated \nprocess. \nDespite these early signs of ECM remodeling, pressurized larvae consistently \nexpelled fluid through the mouth (Fig. 4D; Movie S6). In contrast, animals undergoing the \nlarva-to-polyp transition exhibited variable fluid release—either orally or aborally—until \nthey reached a more elongated polyp stage, when aboral release became predominant (Fig. \n4D; Movie S6). This shift likely reflects the progressive maturation of contractile muscular \nrings at the aboral pole. Supporting this, FGFRb-eGFP–positive concentric muscle rings \nemerged only during late transition stages and coincided with strong enrichment of p-MLC at \nthe aboral pole, a hallmark of an actively contractile structure (fig. 6A).  \nTo test whether FGFRb signaling is required for aboral valve morphogenesis, we \nanalyzed FGFRb mutant animals (27). Unlike their siblings, FGFRb mutants failed to exhibit \nlocalized Collagen IV depletion at the aboral pole (Fig. 5A), and their musculature was \nhighly disorganized, lacking the characteristic p-MLC enrichment observed in controls (Fig. \n5B). These structural defects suggest a failure to form a functional aboral valve. To test this, \nwe artificially inflated the body cavities of FGFRb mutants. While control animals \nconsistently expelled dye through the aboral pole, mutants released fluid exclusively through \nthe mouth, with only a single exception (Fig. 5C; Movie S7). Furthermore, mutants required \nsignificantly higher internal pressure to trigger fluid expulsion compared to controls, \nconfirming that aboral release is mechanically impaired in the absence of FGFRb signaling. \nTogether, these findings demonstrate that FGFRb signaling is essential for \ncoordinating the localized ECM remodeling and muscle ring assembly required to form a \nfunctional, pressure-sensitive aboral valve.  \n \nDiscussions  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n9 \nOur study reveals the dual roles of the mesogleal basement membrane in cnidarian \nmorphogenesis, operating at both global and local scales (Fig. 5D). At the global level, we \ndemonstrate that the dynamic mesoglea architecture orchestrates tissue remodeling and axial \nelongation. At the local level, we identify a specialized mesogleal domain at the aboral pole \nthat undergoes targeted remodeling to create a pressure-sensitive valve, a previously \nunrecognized function that revises the traditional view of cnidarian body architecture. \nWe demonstrate that Collagen IV is expressed precociously in presumptive endoderm \ncells prior to gastrulation, which likely stabilizes initial contacts between invaginating \nendodermal filopodia and the basal blastoderm surface (29, 30). This early expression \nfacilitates bilayer formation and enables subsequent morphogenetic events. Throughout \ndevelopment, endoderm-derived Collagen IV forms a regulated ECM that dynamically \ninteracts with muscular hydraulics (18), essential for directional tissue rearrangements during \naxial elongation. Excess Collagen IV prematurely restricts tissue rearrangement, halting \nelongation, whereas reduced collagen destabilizes morphology through misaligned \nremodeling. These results suggest that mesoglea mechanics must be finely tuned, allowing \nplastic deformation ( 31, 32) without compromising structural integrity. In parallel, the \nenrichment of Wnt/PCP signaling components in mesoglea proteomes ( 19) suggests \nintegration of mechano-chemical signaling in larva-polyp morphogenesis, an avenue for \nfuture investigation. \nAlthough scattered reports have noted possible secondary openings in cnidarians (33–\n35), such claims have historically been dismissed due to the fragility of cnidarian tissues. \nHere, we provide definitive structural and functional evidence for a dynamic aboral valve. \nThis site features localized ECM remodeling and contractile muscle rings, and opens \ntransiently under internal pressure through controlled epithelial rupture and wound signaling \nactivation. Crucially, this is not a through-gut. Unlike the permanent anal pore in ctenophores \n(15), which support unidirectional digestion, the aboral opening in Nematostella is transient, \nmuscle-controlled, and pressure-responsive. It functions not in waste expulsion, but as a \nbiomechanical safety valve to offload internal pressure. While pressure and fluid could also \nbe released through the oral opening, this aboral valve likely serves as a “backup exit” when \noral release is obstructed, such as during pharyngeal compression (fig. S7). This dual-exit \nsystem introduces mechanical redundancy, ensuring robust physiological control of internal \npressure. \nImportantly, this pressure-release mechanism may confer adaptive advantages in the \nbrackish coastal habitats of Nematostella, where rapid salinity fluctuations frequently occur \n(36). A stress-induced epithelial rupture also occurs at the oral pole in Hydra during feeding \n(37, 38), suggesting that stress-responsive epithelial discontinuities may be an ancient feature \nof cnidarian biology. Together, our findings redefine the concept of body openings in early \nbranching animals. Rather than viewing epithelial perforations solely through the lens of \nunidirectional gut evolution, we propose that transient, muscle-regulated rupture zones may \nhave served ancestral roles in pressure regulation. More broadly, these findings illuminate \nhow coordinated global and local ECM dynamics collectively shape organismal form and \nfunction in one of the earliest-diverging animal lineages with organized tissue layers. \n \nReferences and Notes  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n10 \n \n1.  T. Brunet, N. King, The Origin of Animal Multicellularity and Cell Differentiation. Developmental \nCell 43, 124–140 (2017). \n2.  R. O. Hynes, The evolution of metazoan extracellular matrix. J Cell Biol 196, 671–679 (2012). \n3.  A. Naba, Mechanisms of assembly and remodelling of the extracellular matrix. Nat Rev Mol Cell \nBiol 25, 865–885 (2024). \n4.  U. Technau, R. E. Steele, Evolutionary crossroads in developmental biology: Cnidaria. Development \n138, 1447–1458 (2011). \n5.  R. P. Tucker, B. Shibata, T. N. Blankenship, Ultrastructure of the mesoglea of the sea anemone \nNematostella vectensis (Edwardsiidae). Invertebrate Biology 130, 11–24 (2011). \n6.  B. G. Bergheim, S. Özbek, Extracellular matrix and morphogenesis in cnidarians: a tightly knit \nrelationship. Essays in Biochemistry 63, 407–416 (2019). \n7.  Sarras (Jr), M. P, Components, structure, biogenesis and function of the Hydra extracellular matrix \nin regeneration, pattern formation and cell differentiation. The International Journal of \nDevelopmental Biology 56, 567–576 (2012). \n8.  W. M. Kier, The diversity of hydrostatic skeletons. Journal of Experimental Biology 215, 1247–1257 \n(2012). \n9.  J. Zhu, X. Wang, C. He, H. Wang, Mechanical properties, anisotropic swelling behaviours and \nstructures of jellyfish mesogloea. Journal of the Mechanical Behavior of Biomedical Materials  6, \n63–73 (2012). \n10.  C. Gambini, B. Abou, A. Ponton, A. J. M. Corneli ssen, Micro- and macrorheology of jellyfish \nextracellular matrix. Biophys J 102, 1–9 (2012). \n11.  W. M. Megill, J. M. Gosline, R. W. Blake, The modulus of elasticity of fibrillin-containing elastic \nfibres in the mesoglea of the hydromedusa Polyorchis penicillatus. Journal of Experimental Biology \n208, 3819–3834 (2005). \n12.  M. A. R. Koehl, Mechanical Diversity of Connective Tissue of the Body Wall of Sea Anemones. \nJournal of Experimental Biology 69, 107–125 (1977). \n13.  A. Hejnol, M. Q. Martindale, “The mouth, the anus, and the blastopore—open questions about \nquestionable openings” in Animal Evolution: Genomes, Fossils, and Trees , M. J. Telford, D. T. J. \nLittlewood, Eds. (Oxford University Press, 2009; \nhttps://doi.org/10.1093/acprof:oso/9780199549429.003.0004), p. 0. \n14.  C. Nielsen, T. Brunet, D. Arendt, Evolution of the bilaterian mouth and anus. Nat Ecol Evol 2, 1358–\n1376 (2018). \n15.  J. S. Presnell, L. E. Vandepas, K. J. Warren, B. J. Swalla, C. T. Amemiya, W. E. Browne, The \nPresence of a Functionally Tripartite Through-Gut in Ctenophora Has Implications for Metazoan \nCharacter Trait Evolution. Current Biology 26, 2814–2820 (2016). \n16.  A. Stokkermans, A. Chakrabarti, K. Subramanian, L. Wang, S. Yin, P. Moghe, P. Steenbergen, G. \nMönke, T. Hiiragi, R. Prevedel, L. Mahadevan, A. Ikmi, Muscular hydraulics drive larva-polyp \nmorphogenesis. Curr Biol 32, 4707-4718.e8 (2022). \n17.  A. Ikmi, S. A. McKinney, K. M. Delventhal, M. C. Gibson, TALEN and CRISPR/Cas9-mediated \ngen\nome editing in the early-branching metazoan Nematostella vectensis. Nat Commun 5, 5486 \n(2014). \n18.  A. Paix, S. Basu, P. Steenbergen, R. Singh, R. Prevedel, A. Ikmi, Endogenous tagging of multiple \ncellular components in the sea anemone Nematostella vectensis. Proc Natl Acad Sci U S A 120, \ne2215958120 (2023). \n19.  B. G. Bergheim, A. G. Cole, M. Rettel, F. Stein, S. Redl, M. W. Hess, A. Ikmi, S. Özbek, Molecular \ndynamics of the matrisome across sea anemone life history. eLife 14 (2025). \n20.  H. Shimizu, R. Aufschnaiter, L. Li, M. P. Sarras, D.-B. Borza, D. R. Abrahamson, Y. Sado, X. \nZhang, The extracellular matrix of hydra is a porous sheet and contains type IV collagen. Zoology \n(Jena) 111, 410–418 (2008). \n21.  S. Cheung, D. Bredikhin, T. Gerber, P. J. Steenbergen, S. Basu, R. Bailleul, P. Hansen, A. Paix, M. \nA. Benton, H. C. Korswagen, D. Arendt, O. Stegle, A. Ikmi, Systemic coordination of whole-body \ntissue remodeling during local regeneration in sea anemones. Dev Cell 60, 780-793.e7 (2025). \n22.  S. M. Jahnel, M. Walzl, U. Technau, Development and epithelial organisation of muscle cells in the \nsea anemone Nematostella vectensis. Frontiers in Zoology 11, 44 (2014). \n23.  L. Leclère, F. Rentzsch, RGM regulates BMP-mediated secondary axis formation in the sea anemone \nNematostella vectensis. Cell Rep 9, 1921–1930 (2014). \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n11 \n24.  M. Saina, G. Genikhovich, E. Renfer, U. Technau, BMPs and Chordin regulate patterning of the \ndirective axis in a sea anemone. Proceedings of the National Academy of Sciences  106, 18592–\n18597 (2009). \n25.  S. He, F. Del Viso, C.-Y. Chen, A. Ikmi, A. E. Kroesen, M. C. Gibson, An axial Hox code controls \ntissue segmentation and body patterning in Nematostella vectensis. Science  361, 1377–1380 (2018). \n26.  J. C. Pastor-Pareja, T. Xu, Shaping Cells and Organs in Drosophila by Opposing Roles of Fat Body-\nSecreted Collagen IV and Perlecan. Developmental Cell 21, 245–256 (2011). \n27.  A. Ikmi, P. J. Steenbergen, M. Anzo, M. R. McMullen, A. Stokkermans, L. R. Ellington, M. C. \nGibson, Feeding-dependent tentacle development in the sea anemone Nematostella vectensis. Nat \nCommun 11, 4399 (2020). \n28.  T. Q. DuBuc, N. Traylor-Knowles, M. Q. Martindale, Initiating a regenerative response; cellular and \nmolecular features of wound healing in the cnidarian Nematostella vectensis. BMC Biology  12, 24 \n(2014). \n29.  Y. Kraus, U. Technau, Gastrulation in the sea anemone Nematostella vectensis occurs by \ninvagination and immigration: an ultrastructural study. Dev Genes Evol 216, 119–132 (2006). \n30.  U. Technau, Gastrulation and germ layer formation in the sea anemone Nematostella vectensis and \nother cnidarians. Mech Dev 163, 103628 (2020). \n31.  A. Elosegui-Artola, The extracellular matrix viscoelasticity as a regulator of cell and tissue \ndynamics. Current Opinion in Cell Biology 72, 10–18 (2021). \n32.  M.-C. Díaz-de-la-Loza, B. M. Stramer, The extracellular matrix in tissue morphogenesis: No longer \na backseat driver. Cells & Development 177, 203883 (2024). \n33.  H. Shimizu, Y. Takaku, X. Zhang, T. Fujisawa, The aboral pore of hydra: evidence that the digestive \ntract of hydra is a tube not a sac. Dev Genes Evol 217, 563–568 (2007). \n34.  L. H. Hyman, The invertebrates: protozoa through Ctenophora. (1940). \n35.  A. R. Amiel, H. T. Johnston, K. Nedoncelle, J. F. Warner, S. Ferreira, E. Röttinger, Characterization \nof Morphological and Cellular Events Underlying Oral Regeneration in the Sea Anemone, \nNematostella vectensis. Int J Mol Sci 16, 28449–28471 (2015). \n36.  C. Hand, K. R. Uhlinger, The Culture, Sexual and Asexual Reproduction, and Growth of the Sea \nAnemone Nematostella vectensis. Biol Bull 182, 169–176 (1992). \n37.  J. A. Carter, C. Hyland, R. E. Steele, E.-M. S. Collins, Dynamics of Mouth Opening in Hydra. \nBiophys J 110, 1191–1201 (2016). \n38.  R. D. Campbell, Structure of the mouth of Hydra spp. A breach in the epithelium that disappears \nwhen it closes. Cell Tissue Res. 249, 189–197 (1987). \n \nAcknowledgments: We thank Kresimir Crnokic for his support in animal husbandry. We \nthank the Ikmi lab members for their comments on the work. We also thank Muzamil Majid \nKhan for the discussion about the ECM and for sharing ECM-related reagents, as well as the \nfeedback on the manuscript. \n \nFunding: This work was supported by the German Science Foundation (DFG) (Collaborative \nResearch Center 1324 (B07) and OE 416/8-1 to S.Ö, and by the European Molecular Biology \nLaboratory (EMBL) to A.I. \nAuthor Contributions: S.B. and A.I. conceived the idea for this project and designed the \nexperiments. S.B. performed most of the experiments. A.P. designed the KI lines. P.S. \nperformed immunostaining and drug treatments. F.G. performed lightsheet imaging. G.B. \nT.H. and P.R. processed the sample and performed the experiments for FIB-SEM. J.H. and \nA.K. segmented the FIB-SEM data.  S.O. generated the Laminin antibody. S.B. and A.I. \nanalyzed the data. S.B. generated all figures. A.I. drafted the manuscript with inputs from all \nauthors.  \nCompeting interests: Authors declare no competing interests \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n12 \nData and materials availability: All data needed to evaluate the conclusions are present in \nthe paper and the supplementary materials. Transgenic lines are available upon request. \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n13 \nFigures \n \n \n \n \nFigure 1. Developmental dynamics of mesogleal basement membrane \n(A) Lightsheet imaging of a live eGFP::ColIV embryo. Left: Optical cross-section of a \ndeveloping blastula showing endogenous Collagen IV expression (green) before the onset of \ngastrulation. Right: Quantification of eGFP::ColIV intensity along the oral–aboral axis (0: \noral pole). Scale bar: 100 µm. \n(B) Confocal cross-sections of fixed eGFP::ColIV-positive animals from gastrulation to the \nprimary polyp stage. F-actin is stained in gray. Scale bar: 100 µm. \n(C) Magnified insets from B with corresponding average intensity profiles along the xy-axis. \nThe central plane represents the developing mesoglea.; upper and lower regions correspond \nto endoderm and ectoderm, respectively. The intensity of the extracellular eGFP::ColIV is \nhighlighted in gray. Apical and basal eGFP::ColIV puncta are indicated by arrowhead and \narrow, respectively. Scale bar: 100 µm. \n(D) Ratio of extracellular (mesoglea) to intracellular (inner layer) eGFP::ColIV intensity as a \nfunction of body aspect ratio, n = 50.  \n(E) Schematic illustrating the spatial localization of intra-segmental and inter-segmental \nregions during development. \n(F) Maximum intensity Z-projection of inter-segmental regions in eGFP::ColIV-positive \nanimals (green) with F-actin staining (gray).  Black arrowheads indicate short lateral \nCollagen IV bridges; black arrows mark Collagen IV accumulation at inter-segmental \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n14 \nboundaries. Red and blue arrowheads indicate circular and parietal muscles, respectively. \nScale bar: 25 µm. \n(G) Maximum intensity Z-projection of intra-segmental regions in eGFP::ColIV -positive \nanimals (green) with F-actin staining (gray). Dashed lines indicate the wavy pattern of \nCollagen IV network. Scale bar: 25 µm. \n \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n15 \n \n \n \nFigure 2. Perturbations of basement membrane composition \n(A) (Left) Maximum intensity projections showing Collagen IV and Laminin \nimmunostaining in untreated (t\n0) and treated (t 1, 3 days post-treatment) larvae under the \nindicated conditions. Scale bar: 20 µm. (Right) Quantification of average fluorescence \nintensity for Collagen IV and Laminin. n = 10 larvae; n = 8 DMSO-treated polyps; n = 8 \nGM6001-treated animals; n = 8 BPY-treated animals. \n(B) (Left) Cross-sectional views of the basement membrane stained for Collagen IV and \nLaminin under the indicated conditions. Scale bar: 20 µm. (Right) Quantification of mesoglea \nthickness across conditions. n = 12 larvae; n = 12 DMSO-treated polyps; n = 12 GM6001-\ntreated; n = 6 BPY-treated animals. \n \n \n \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n16 \n \n \n \nFigure 3. ECM modulation disrupts tissue rearrangement and hydraulics \n(A) Morphospace analysis showing elongation trajectories based on changes in body volume \n(size) and aspect ratio (shape) from t 0 to t 1 (3-day interval) under the indicated conditions. \nOnset of aboral leakage is indicated in the developmental trajectory. n = 20 DMSO-treated; n \n= 20 GM6001-treated; n = 20 BPY-treated animals. \n(B) (Left) Aboral leakage phenotypes in control and BPY-treated animals. Note the cloud of \ncellular debris expelled from the oral pole. Scale bar: 100µm. (Right) Quantification of the \nonset of aboral leakage as a function of body aspect ratio. n = 10 DMSO-treated; n  = 10 \nBPY-treated animals. \n(C) Schematic representation of aboral leakage observed across experimental conditions. \n(D) (Left) Maximum intensity projection images of photoconverted tissue patches in larvae \nand corresponding topological changes at t\n1. Green: non-photoconverted; magenta: \nphotoconverted. Scale bar: 100 µm. (Right) Quantification of patch elongation (length/width \naspect ratio) versus body column aspect ratio at each time point across experimental \nconditions. (Right) Schematic summarizing trends in patch shape dynamics across treatments. \n \n \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n17 \n \n \nFigure 4. Localized ECM remodeling defines an aboral valve \n(A) (Top) Schematic depicting the tissue and ECM remodeling events required for aboral \nleakage during the larva-to-polyp transition. (Bottom) Cross-sectional (side view) and \nmaximum intensity projection (aboral view) images showing Collagen IV and Laminin \nimmunostaining in polyps. Asterisk indicates the site of ECM remodeling. Scale bar: 20 µm. \n(B) (Top) Cross-sectional view of Segmented FIB-SEM data revealing the architecture of the \naboral pole. Scale bar: 5µm. (Bottom) Aboral view of the FIB-SEM dataset highlighting \nepidermal architecture. Asterisk marks the basal gap in the epidermis. Scale bar: 10 µm. \n(C) (Left) F-actin and phospho-Myosin Light Chain (pMLC) staining of muscular rings \nreveals its contractile organization (see arrow). Scale bar: 20 µm (top), 10µm (bottom). \n(Right) Cross-sectional view and 3D rendering showing the dual conformational states of \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n18 \nmuscular rings labeled with FGFRb-eGFP. Asterisk indicates epidermal discontinuity. Scale \nbar: 20 µm. \n(D) (Left) Cavity inflation assay in a primary polyp showing three representative time points: \nbefore inflation, during inflation, and at the point of leakage. Scale bar: 50 µm. (Right) \nQuantification of oral versus aboral leakage events as a function of body aspect ratio across \ndevelopmental stages. \n(E) Schematic model summarizing the mechanism of aboral valve function. \n \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n19 \n \n \n \n \nFigure 5. Developmental basis of aboral valve formation \n(A) (Top) Maximum intensity projections (aboral view) showing Collagen IV distribution in \nFGFRb knockout ( FGFRb-/-) mutants and representative siblings. (Middle) Cross-sectional \nviews of the aboral pole in the respective genotypes. Scale bar: 10µm. (Bottom) \nQuantification of Collagen IV spatial distribution in FGFRb-/- mutants versus siblings. \n(B) (Top) Maximum intensity projections (aboral view) showing F-actin and pMLC staining \nin FGFRb-/- mutants and representative siblings. (Middle) Cross-sectional views of the \naboral pole in each genotype. Scale bar: 20µm. (Bottom) Quantification of pMLC signal \ndistribution in FGFRb-/- mutants versus siblings. \n(C) Cavity inflation assay in FGFRb-/- mutants and siblings. Aboral leakage is observed in 1 \nout of 9 FGFRb-/- mutants, compared to 7 out of 9 siblings. Scale bar: 200 µm. \n(D) Schematic model illustrating the processes dependent on local and global mesoglea \nremodeling.\n \n  \n  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n20 \nMaterials and Methods \n \nAnimal husbandry and Spawning \nAdult Nematostella vectensis were cultured in 12 ppt artificial seawater (ASW; Instant Ocean \nsea salt) at 17°C under dark conditions. To induce spawning, animals were placed in a white \nlight incubator for 6–8 hours overnight, with the temperature increased to approximately \n28°C ( 1). Spawning typically occurred within 3–4 hours following a cold water change \n(17°C). Collected eggs were de-jellied by incubating them for 9 minutes in a 4% cysteine \nsolution (Sigma, 168149) prepared in ASW, then rinsed three times with fresh ASW prior to \nfertilization. \n \nTransgenic and mutant lines \nThe eGFP::ColIV and Dendra2::ColIV knock-in lines used in this study were previously \ngenerated (2, 3). The FGFRb-eGFP reporter and FGFRb KO lines were described earlier (4). \nTransgenic embryos were obtained by crossing transgenic males with wild-type females. \n \nLive imaging of early embryos \nLive confocal imaging was performed on a Leica SP8 CSU confocal microscope equipped \nwith a 20× objective, with embryos mounted non-confined in 0.22 µm-filtered 12 ppt \nartificial sea water on MatTek round glass-bottom dishes (#P35G-1.5-14-C) and imaged \novernight. \nLive light-sheet imaging was performed on a Luxendo MuVi-SPIM using a Nikon CFI Plan \nFluor 10×/0.30 NA illumination objective and an Olympus XLUMPLFLN 20×/1.00 NA \ndetection objective, both immersed in 0.22 µm-filtered 12 ppt sea water and maintained at 23 \n°C; fluorescence was excited at 488 nm (~1.6 mW) with a 3 µm beam width, 50 ms exposure, \nand captured on a Hamamatsu C11440-22C camera. Embryos were mounted non-confined on \na 1% agarose in filtered 12 ppt sea water bedding in 100 µL glass capillaries trimmed for the \nsample holder, enclosed in FEP tubing. Prior to mounting the embryos onto the bedding, the \nagarosesolidified in the glass capillaries at 4 °C for 10 min, to then fill the FEP tube with sea \nwater to eliminate bubbles. Time-lapse Z-stacks (1 µm steps) were recorded every 5 min for \n12.5–40 h, acquiring four orthogonal views per time point (two detection arms plus a 90° \nrotation of each). \n \nAcquifer microscope \nLive imaging of the larva-to-polyp transition was conducted using an Acquifer screening \nmicroscope (5). Larvae (3 dpf) were individually placed into wells of a 384-well plate \n(Corning, 3540), each containing 25\n/i1 µL of 12 ppt artificial seawater (ASW) with 0.1% \nDMSO or pharmacological inhibitors. Imaging was performed continuously every 5 minutes \nover three days at 27\n/i1 °C, using a 4× objective and brightfield illumination set to 20% \nintensity. Image analysis was performed as previously described (5). \n \nImmunostaining  \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n21 \nAnimals were first anesthetized in 7% magnesium chloride prior to fixation. Fixation was \nperformed at room temperature for 1 hour using either 4% paraformaldehyde in PBS \n(Electron Microscopy Sciences, E15710) for transgenic line samples or Lavdovsky’s fixative \n(3.7% formaldehyde, 50% ethanol, 4% acetic acid) for immunostaining of basement \nmembrane components. Fixed samples were treated for 20 minutes in 10% DMSO (Thermo \nFisher, 85190) in PBS, followed by rinses in PBS containing 0.2% Triton X-100 (Sigma, \nT8787) (PTx 0.2%). Blocking was carried out for 1 hour in PTx 0.1% supplemented with \n0.1% DMSO, 1% BSA (Sigma, A2153), and 5% goat serum (Sigma, 8182G9023) . Samples \nwere then incubated overnight at 4\n/i1 °C with primary antibodies diluted in blocking buffer \n(Table 1). After thorough washing in PTx 0.1%, Alexa Fluor–conjugated secondary \nantibodies (Thermo Fisher, 1:500) were applied in the same buffer and incubated overnight at \n4\n/i1 °C. For additional labeling, F-actin was stained with Alexa Fluor–conjugated phalloidin \n(Thermo Fisher, 1:100), and nuclei were counterstained with Hoechst 34580 (Sigma, 63493, \n1:1000), both in PTx 0.1% overnight at 4\n/i1 °C. Finally, all samples were washed in PTx 0.1% \nand mounted in Vectashield Plus (Vector Laboratories) for confocal microscopy. \n \nTable 1: Primary antibodies used for immunostaining and their working dilutions  \n \nAntibody Source Dilution \nanti-phospho-Myosin Light Chain Cell Signaling Technology #3671S 1:50 \nanti-eGFP MBL #598S 1:500 \nanti-eGFP Abcam #ab1218 1:500 \nanti-Laminin-γ  Bergheim et al., 2025 1:400 \nanti-ColIV - JK2 Gift from Haruko Tomono 1:400 \nanti-phospho-p44/42 MAPK \n(Erk1/2)(Thr202/Tyr204) \nCell Signaling Technology #4370 1:250 \n \nDirect visualisation of reporter lines \nAnimals expressing transgenes or KI constructs were anesthetized in 7% magnesium chloride \nbefore fixation. Fixation was carried out for one hour at room temperature using 4% \nparaformaldehyde (EMS, E15710) in PBS. Afterwards, the samples were washed four times \nfor 5 minutes each with 1X PBS. Additional staining for F-actin and nuclei was performed \nusing phalloidin Alexa Fluor (Thermo Fisher, 1:100) and Hoechst 34580 (Sigma, 63493, \n1:1000), respectively, in 1X PBS for 6 hours at 4°C. The samples were protected from light \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n22 \npost-fixation. Finally, the samples were directly mounted in Vectashield Plus for confocal \nimaging. \n \nConfocal imaging \nSamples were imaged using either a Zeiss LSM880 AiryFast or Zeiss LSM980 AiryFast \nmicroscope. For lower-resolution acquisition, a Plan-Apochromat 20×/0.8 M27 air objective \nor an LD-LCI Plan-Apochromat 25×/0.8 Imm autocorr FCS M27 objective was used. For \nhigh-resolution imaging, either a C-Apochromat 40×/1.2 W autocorr FCS M27 water-\nimmersion objective or a Plan-Apochromat 63×/1.4 Oil DIC M27 objective was employed. \nDepending on the fluorophores, laser lines at 405\n/i1 nm, 488/i1 nm, 561/i1 nm, or 633/i1 nm were \nused for excitation. \n \nKaede mRNA Synthesis and injection \nKaede mRNA was synthesized using the HiScribe™ T7 ARCA mRNA Kit with poly(A) \ntailing (New England Biolabs, E2060S) from a PCR-amplified template of the Kaede-H2B \nplasmid (Addgene, #57316). Following in vitro transcription, the mRNA was purified using \nSPRISelect magnetic beads (Beckman Coulter, B23319). The final injection mix contained \n200\n/i1 ng/μ L Kaede mRNA and fluorescein isothiocyanate (FITC; Thermo Fisher, 46425) as \nan injection tracer. The mixture was injected into fertilized Nematostella eggs. \n \nPhotoconversion \nLarvae (3 DPF) expressing Kaede mRNA or Dendra2::ColIV  were anesthetized in 7% \nmagnesium chloride and mounted in a round glass-bottom dish (MatTek, #P35G-1.5-14-C). \nPhotoconversion was performed using an Evident Rapp FV3000 confocal microscope with a \n375/i1 nm laser to induce conversion. After 3 days post-photoconversion, animals were \ntransferred to microscopy slides, screened for photo-converted patches, and imaged. Laser \nlines at 488 /i1 nm and 561 /i1 nm were used to detect the unconverted and converted forms of \nDendra2/Kaede, respectively. \n \nshRNA Synthesis and Injection \nshRNAs were designed using the siRNA Wizard tool ( 6) (Invivogen), and primers were \nsynthesized by IDT. Primer annealing was carried out at 98\n/i1 °C for 5 minutes, followed by \npassive cooling to room temperature. In vitro transcription was performed using the T7 \nMegaShortScript Kit (Invitrogen, AM1354) with a 6-hour incubation. RNA was purified \nusing SPRISelect magnetic beads (Beckman Coulter, B23319) in 46% isopropanol. Samples \nwere incubated at room temperature for 15 minutes, placed on a magnetic stand for 5 \nminutes, and washed twice with 80% ethanol. After brief drying, RNA was eluted in RNase-\nfree water, aliquoted, and stored at −80\n/i1 °C. Fertilized eggs were injected with 500–\n1500/i1 ng/μ L of each shRNA (see Table 2), together with Texas Red–labeled Dextran \n(ThermoFisher, D3328) as an injection tracer. Microinjections were performed using a \nFemtoJet Express system (Eppendorf). Injected embryos were maintained at room \ntemperature and transferred to 27\n/i1 °C on the following day for further development. \nTable 2: Primer sequences used for shRNA synthesis \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n23 \nshRNA Direction Sequence \nGFP \n(control) \nForward TAATACGACTCACTATAGGGGCACAAGCTGGAGTACAAT\nTCAAGAGATTGTACTCCAGCTTGTGCCCTT \nReverse AAGGGCACAAGCTGGAGTACAATCTCTTGAATTGTACTC\nCAGCTTGTGCCCCTATAGTGAGTCGTATTA \nTbx20  Forward TAATACGACTCACTATAGGGAACAGCTGCTTAAACATTC\nAAGAGATGTTTAAGCAGCTGTTCCCTT \nReverse AAGGGAACAGCTGCTTAAACATCTCTTGAATGTTTAAGC\nAGCTGTTCCCTATAGTGAGTCGTATTA \nBMP2/4  Forward TAATACGACTCACTATAGGACTGGATATTCAAGTGATTC\nAAGAGATCACTTGAATATCCAGTCCTT \nReverse AAGGACTGGATATTCAAGTGATCTCTTGAATCACTTGAA\nTATCCAGTCCTATAGTGAGTCGTATTA \nCol IV Forward TAATACGACTCACTATAGGGTGCAATGGTACTACAATTC\nAAGAGATTGTAGTACCATTGCACCCTT \n Reverse AAGGGTGCAATGGTACTACAATCTCTTGAATTGTAGTAC\nCATTGCACCCTATAGTGAGTCGTATTA  \n \nPharmacological inhibitor treatments \nLarvae at 3 dpf were incubated for 72 hours at 27 /i1 °C in 12 ppt ASW containing either 0.1% \nDMSO (vehicle control), 50 /i1 µM GM6001 (Abcam, ab120845), a broad-spectrum matrix \nmetalloproteinase inhibitor, or 50 /i1 µM 2,2 ′ -Bipyridine (BPY) (Sigma, 1030980005), an \ninhibitor of prolyl-4-hydroxylase or 0.5 mM Rocuronium bromide, a previously characterized \nmuscle relaxant (5 ). Drug solutions were prepared fresh and used without replacement over \nthe course of the incubation. \n \nQuantifications \nImage analysis was performed with Fiji ( 7) with its MorphoLibJ  package ( 8). PyBoat in \nPython was used for time series analysis ( 9). Data analysis and plotting was performed with \nR in RStudio environment (10), with the following packages: tidyverse (11), ggplot2 (12) and \nggsignif (13). \nMorphometric measurements: Confocal cross-sections at the oral plane were used, which is a \ngood approximation given the radial symmetry of the body plan. Only the body column was \nused for measurements, ignoring the tentacles. To obtain the shape metrics, a bounded box \nwas fitted on the body column, the dimensions of which provided body length l (parallel to \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n24 \noral-aboral axis) and w (perpendicular to oral-aboral axis). The aspect ratio was used as a \nmetric of shape, which was defined as l/w . The aspect ratio correlates strongly with the \ndevelopmental time, since the morphogenesis is characterized by axial elongation. A polyline \nROI was defined along the mesoglea of the body column denoting axial position (0 towards \naboral pole, 1 towards oral pole). Using the ROI the region was straightened, which defined a \nnew direction perpendicular to the body axis, where the mesoglea was denoted as 0, positive \nvalue (above) denotes the endoderm while negative value (below) denotes the ectoderm. The \ntissue thickness is reported to be the average thickness along this straightened axis. For \neGFP::ColIV+ animals, the eGFP intensity peaked close to 0, the full-width half-maximum \n(FWHM) value of which denoted the local mesoglea thickness while the signal within \ndenoted the local ColIV amount. The mesoglea thickness and intensity reported were the \nmean and standard deviation across the entire body column. The same was done for \nimmunostaining (Fig S1B). \nAspect ratio of the photo-converted patches: The photo-converted areas A were segmented \nand its skeleton was defined as l . The width was defined empirically as A/l , and the reported \nAR\npatch = l/w = l2/A. If the shape inverted during axial elongation, the AR was inverted. \nECM intensity measurements in perturbations: The intensity of the ECM was measured on \nmaximum intensity projection images by taking the average intensity normalized over equal \nsized square ROIs. \npErk intensity measurements: pErk intensity was measured at the aboral pore, 0µm denoting \noutside and 50µm denoting the body cavity. The profile plot was generated across a line of \nwidth 5µm. \nIntensity measurements at the aboral end:  A ROI of length 100µm was defined along the \naboral pore and rescaled, with 0 denoting the location of the pore. The intensity was \nmeasured as profile plots. \n \nCavity inflation assay \nTo artificially increase internal pressure, a glass needle containing fluorescein isothiocyanate \n(FITC; Thermo Fisher, #46425, diluted 1:100 in 12\n/i1 ppt ASW) was gently inserted into the \noral opening of the animal. Injection was carried out using a FemtoJet Express microinjector \n(Eppendorf), while monitoring the animal under a dissection microscope. Cavity inflation \nwas gradually induced by precisely controlling both the injection pressure and duration. \nTime-lapse recordings were acquired at 1-second intervals and continued until leakage was \nobserved through either the aboral or oral poles. \n \nFIB-SEM volume acquisition \nThe samples were high pressure frozen in a solution of sea water containing 20% Ficoll \n(molecular weight ~70,000, Sigma) using HPM010 (Abra Fluid). Following high-pressure \nfreezing, freeze-substitution was carried out using the EM-AFS2 system from Leica \nMicrosystems. The freeze-substitution medium consisted of 0.1% uranyl acetate in acetone. \nThe samples were subjected to freeze-substitution at a temperature of -90°C for a duration of \n48 hours. Subsequently, the temperature was gradually increased to -45°C at a rate of 3.5°C \nper hour, and the samples were further incubated for 5 hours. The samples were gradually \ninfiltrated in HM20 resin, then polymerized under UV light for a period of 48 hours at a \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n25 \ntemperature of -25°C. Following this, the temperature was gradually increased to 20°C at a \nrate of 5°C per hour, and the samples were further UV polymerized for an additional 9 hours. \nIn order to target the central aboral region of the animal with sufficient accuracy for FIB-\nSEM acquisition, we used an already established strategy ( 14). The samples, mounted on \nSPINE sample holders commonly used for crystallography, were imaged by phase-contrast \nX-ray on the EMBL beamline P14 on the PETRA III synchrotron (c/o DESY, Hamburg, \nGermany) using a previously characterized imaging setup (15) at an X-ray energy of 18 keV. \nX-ray images were recorded using an Optique Peter (Lyon, France) X-ray microscope \nconsisting of an LSO:Tb scintillator with 8 \nμ m active layer; an Olympus UPlanFL 20-fold \nobjective (Olympus, Tokio, Japan), numerical aperture 0.5; a 45° mirror; a 180 mm tube lens \nand a PCO.edge 4.2 sCMOS camera with 2048x2048 pixels, pixels (6.5 \nμ m pixel size). Thus, \nthe effective pixel size was 0.325 μ m with a field of view 666 x 666 μ m². This setup typically \ndelivers a resolution of about 0.5-0.7 μ m, as determined from the analysis of projection \nimages from a Siemens star (Ta on SiN; XRESO-50HC, NTT-AT, Japan). On the resin-\nembedded sample, projection images were acquired at four camera distances: 62.5, 67.5, 73.5 \nand 82.5 mm. At each distance, 3600 projections covering 360° of continuous rotation were \nrecorded with an exposure time of 10 ms per frame. Data collection (including robotic sample \ntransfer from a storage vessel to the rotation axis and sample centering via an on-axis optical \nmicroscope) was completed in 6 minutes. Flat-field corrections were applied by dividing each \nprojection image by the most similar flat-field image according to the SSIM criterion ( 16). \nFor lateral shift compensation at the four camera positions, images recorded at each \nprojection angle were registered using Fourier-space correlation with a sub-pixel \ninterpolation. Registered images were further processed by a multi-distance non-iterative \nholographic reconstruction (17, 18), using a complex refraction index decrement ratio \nβ /δ  = \n0.15 and a zero compensation of 0.1. Tomographic reconstructions were performed using the \nTOMOPY package (19), employing the built-in Gridrec algorithm and Shepp-Logan filtering \nwith default settings. All steps of the XIMG data processing were combined into a python-\nbased custom software pipeline, available at: \nhttps://git.embl.de/maxim.polikarpov/ximg_p14/-\n/blob/master/2019/Dec_2019_Platynereis/nematosella/ \nThe high contrast, high resolution X-ray data, were used to guide the trimming of the block to \nthe region of interest, which was done with an ultramicrotome (UC7, Leica Microsystems) \nand a diamond trimming knife (Cryotrim 90, Diatome). The trimmed sample was then glued \nto an SEM stub using conductive epoxy resin (Ted Pella) and imaged by FIB-SEM using a \nZeiss Crossbeam 550. The acquisition was performed using the Atlas 3D workflow. FIB \nslicing was obtained at 1.5 nA. Imaging was performed at an acceleration voltage of 1.5 kV \nand a current of 700pA, using a backscattered electron detector (ESB). The voxel size was \n16x16x16 nm, with a dwell time of 10 µs. \nFIB-SEM volume segmentation \nCell segmentation from FIB-SEM data was performed using a boundary-based semantic \nsegmentation pipeline adapted from ( 20). Initial boundary annotations were generated in \nIlastik (21), which was used to train a random forest classifier for preliminary segmentation. \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n26 \nThis segmentation was subsequently refined using a publicly available pretrained model \n(https://bioimage.io/#/?tags=enhancer&id=10.5281%2Fzenodo.6808325) applied as an \nenhancer. To further improve accuracy, a U-Net architecture ( 22) was trained from scratch \nusing the enhanced intermediate segmentation as training labels. Following the protocol in \n(20), the final boundary map was converted into an instance-level segmentation using a \nwatershed algorithm ( 23). Minor segmentation artifacts were manually corrected before \nrendering the target cells for visualization. \n \n \nReferences and Notes \n \n1.  G. Genikhovich, U. Technau, Induction of spawning in the starlet sea anemone Nematostella \nvectensis, in vitro fertilization of gametes, and dejellying of zygotes. Cold Spring Harb. Protoc. \n2009, pdb.prot5281 (2009). \n2.  A. Paix, S. Basu, P. Steenbergen, R. Singh, R. Prevedel, A. Ikmi, Endogenous tagging of multiple \ncellular components in the sea anemone Nematostella vectensis. Proc. Natl. Acad. Sci. U. S. A. \n120, e2215958120 (2023). \n3.  S. Cheung, D. Bredikhin, T. Gerber, P. J. Steenbergen, S. Basu, R. Bailleul, P. Hansen, A. Paix, \nM. A. Benton, H. C. Korswagen, D. Arendt, O. Stegle, A. Ikmi, Systemic coordination of whole-\nbody tissue remodeling during local regeneration in sea anemones. Dev. Cell 60, 780-793.e7 \n(2025). \n4.  A. Ikmi, P. J. Steenbergen, M. Anzo, M. R. McMullen, A. Stokkermans, L. R. Ellington, M. C. \nGibson, Feeding-dependent tentacle development in the sea anemone Nematostella vectensis. \nNat. Commun. 11, 4399 (2020). \n5.  A. Stokkermans, A. Chakrabarti, K. Subramanian, L. Wang, S. Yin, P. Moghe, P. Steenbergen, \nG. Mönke, T. Hiiragi, R. Prevedel, L. Mahadevan, A. Ikmi, Muscular hydraulics drive larva-\npolyp morphogenesis. Curr. Biol. CB 32, 4707-4718.e8 (2022). \n6.  S. He, F. del Viso, C.-Y. Chen, A. Ikmi, A. E. Kroesen, M. C. Gibson, An axial Hox code \ncontrols tissue segmentation and body patterning in Nematostella vectensis. Science 361, 1377–\n1380 (2018). \n7.  J. Schindelin, I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. \nRueden, S. Saalfeld, B. Schmid, J.-Y. Tinevez, D. J. White, V. Hartenstein, K. Eliceiri, P. \nTomancak, A. Cardona, Fiji: an open-source platform for biological-image analysis. Nat. \nMethods 9, 676–682 (2012). \n8.  MorphoLibJ: integrated library and plugins for mathematical morphology with ImageJ | \nBioinformatics | Oxford Academic. \nhttps://academic.oup.com/bioinformatics/article/32/22/3532/2525592. \n9.  G. Mönke, F. A. Sorgenfrei, C. Schmal, A. E. Granada, Optimal time frequency analysis for \nbiological data - pyBOAT. bioRxiv [Preprint] (2020). https://doi.org/10.1101/2020.04.29.067744. \n10.  RStudio: integrated development for R | CiNii Research. \nhttps://cir.nii.ac.jp/crid/1370004240607707919. \n11.  H. Wickham, M. Averick, J. Bryan, W. Chang, L. D. McGowan, R. François, G. Grolemund, A. \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint \n\n27 \nHayes, L. Henry, J. Hester, M. Kuhn, T. L. Pedersen, E. Miller, S. M. Bache, K. Müller, J. Ooms, \nD. Robinson, D. P. Seidel, V. Spinu, K. Takahashi, D. Vaughan, C. Wilke, K. Woo, H. Yutani, \nWelcome to the Tidyverse. J. Open Source Softw. 4, 1686 (2019). \n12.  H. Wickham, Ggplot2: Elegant Graphics for Data Analysis  (Springer-Verlag New York, 2016; \nhttps://ggplot2.tidyverse.org). \n13.  C. Ahlmann-Eltze, I. Patil, ggsignif: R Package for Displaying Significance Brackets for \n“ggplot2.” PsyArxiv, doi: 10.31234/osf.io/7awm6 (2021). \n14.  J. M. Musser, K. J. Schippers, M. Nickel, G. Mizzon, A. B. Kohn, C. Pape, P. Ronchi, N. \nPapadopoulos, A. J. Tarashansky, J. U. Hammel, F. Wolf, C. Liang, A. Hernández-Plaza, C. P. \nCantalapiedra, K. Achim, N. L. Schieber, L. Pan, F. Ruperti, W. R. Francis, S. Vargas, S. Kling, \nM. Renkert, M. Polikarpov, G. Bourenkov, R. Feuda, I. Gaspar, P. Burkhardt, B. Wang, P. Bork, \nM. Beck, T. R. Schneider, A. Kreshuk, G. Wörheide, J. Huerta-Cepas, Y. Schwab, L. L. Moroz, \nD. Arendt, Profiling cellular diversity in sponges informs animal cell type and nervous system \nevolution. Science 374, 717–723 (2021). \n15.  M. Polikarpov, G. Bourenkov, I. Snigireva, A. Snigirev, S. Zimmermann, K. Csanko, S. \nBrockhauser, T. R. Schneider, Visualization of protein crystals by high-energy phase-contrast X-\nray imaging. Acta Crystallogr. Sect. Struct. Biol. 75, 947–958 (2019). \n16.  Z. Wang, A. C. Bovik, H. R. Sheikh, E. P. Simoncelli, Image quality assessment: from error \nvisibility to structural similarity. IEEE Trans. Image Process. 13, 600–612 (2004). \n17.  S. Zabler, P. Cloetens, J.-P. Guigay, J. Baruchel, M. Schlenker, Optimization of phase contrast \nimaging using hard x rays. Rev. Sci. Instrum. 76, 073705 (2005). \n18.  P. Cloetens, W. Ludwig, J. Baruchel, D. Van Dyck, J. Van Landuyt, J. P. Guigay, M. Schlenker, \nHolotomography: Quantitative phase tomography with micrometer resolution using hard \nsynchrotron radiation x rays. Appl. Phys. Lett. 75, 2912–2914 (1999). \n19.  D. Gürsoy, F. De Carlo, X. Xiao, C. Jacobsen, TomoPy: a framework for the analysis of \nsynchrotron tomographic data. J. Synchrotron Radiat. 21, 1188–1193 (2014). \n20.  A. Matskevych, A. Wolny, C. Pape, A. Kreshuk, From Shallow to Deep: Exploiting Feature-\nBased Classifiers for Domain Adaptation in Semantic Segmentation. Front. Comput. Sci. 4 \n(2022). \n21.  ilastik: interactive machine learning for (bio)image analysis | Nature Methods. \nhttps://www.nature.com/articles/s41592-019-0582-9. \n22.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation | SpringerLink. \nhttps://link.springer.com/chapter/10.1007/978-3-319-46723-8_49. \n23.  S. Wolf, C. Pape, A. Bailoni, N. Rahaman, A. Kreshuk, U. Kothe, F. Hamprecht, “The Mutex \nWatershed: Efficient, Parameter-Free Image Partitioning” (2018; \nhttps://openaccess.thecvf.com/content_ECCV_2018/html/Steffen_Wolf_The_Mutex_Watershed_\nECCV_2018_paper.html), pp. 546–562. \n \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted May 27, 2025. ; https://doi.org/10.1101/2025.05.26.656114doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}