The anti-neural role of BMP signaling is a consequence of its ancestral function in dorsoventral patterning

Figures Abstract In Bilateria with centralized nervous systems (e.g., in vertebrates or arthropods), the minimum of the BMP signaling activity gradient defines the position of the central nervous system. BMP-dependent patterning of the secondary body axis is ancestral for Bilateria and possibly also for the bilaterian sister clade Cnidaria. However, the variety of levels of centralization of the nervous systems in Bilateria—from diffuse to fully centralized—as well as the lack of centralization of the nervous system in Cnidaria, suggest that BMP signaling cannot be perceived as a universally “anti-neural” signal. Here we use transgenic reporter lines in the anthozoan cnidarian Nematostella to show that BMP signaling is active in distinct neuronal populations. Moreover, attenuation of BMP signaling followed by RNA-Seq shows that BMP signaling is a positive regulator of many neuronal genes, including the top-tier neural progenitor marker soxB(2). Furthermore, we analyze BMP signaling activity in the cubozoan jellyfish Tripedalia and the scyphozoan jellyfish Stomolophus and prove that BMP signaling in parts of the diffuse nervous system is not an anthozoan but an ancestral cnidarian feature, shared by anthozoans and medusozoans. Finally, we show that the highly centralized ventral nervous system of the nonmodel spiralian, the chaetognath Spadella, forms out of paired BMP signaling-positive domains on the lateral sides of the embryo. Together, our data suggest that one of the ancestral roles of BMP signaling may have been in promoting neurogenesis. We propose that the “anti-neural” function of BMP signaling in vertebrates and arthropods is a consequence of its global role in the dorsoventral patterning of the ectoderm. Citation: Knabl P, Ordoñez JF, Montenegro Cabrera JD, Abed-Navandi D, Halbauer R, Link O, et al. (2026) The anti-neural role of BMP signaling is a consequence of its ancestral function in dorsoventral patterning. PLoS Biol 24(5): e3003803. https://doi.org/10.1371/journal.pbio.3003803 Academic Editor: Marianne E. Bronner, California Institute of Technology, UNITED STATES OF AMERICA Received: June 9, 2025; Accepted: May 4, 2026; Published: May 21, 2026 Copyright: © 2026 Knabl et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: Newly generated sequencing data is deposited as bioprojects at https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1269470 and http://www.ncbi.nlm.nih.gov/bioproject/1441074. All code required to reproduce the computational analyses of this study can be found on https://github.com/Genikhovich-Lab/ and, additionally, at 10.5281/zenodo.19686897 and 10.5281/zenodo.19686882. All other relevant data are within the paper and its Supporting information files. Funding: This research was funded by the Austrian Science Fund (FWF) grants P32705 and PAT2395824 to G.G. and the Austrian Science Fund (FWF) grant P34665 to T.W. During his time in the Genikhovich lab, P.K.‘s salary was paid from the grants to G.G. P.K. is a recipient of the Dimitrov Fellowship of the Austrian Academy of Sciences (OeAW) and of the Writing-up Fellowship of the Konrad Lorenz Institute for Evolution and Cognition Research (KLI). During her time in the Wollesen lab, J.O.’s salary was paid from the grant to T.W. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: bHLH, basic helix-loop-helix; CNS, centralized nervous system; DMH1, Dorsomorphin Homolog 1; DV, dorsoventral; hpf, hours post fertilization; NM, Nematostella Medium; NGS, normal goat serum; PMR, parietal muscle region; SD, Standard deviation; SEM, standard error of the mean Introduction The evolutionary origin of BMP signaling-dependent regulation of animal nervous systems is uncertain. Both in vertebrates and arthropods, the establishment of a BMP signaling gradient is required for patterning the dorsoventral (DV) body axis and the formation of the centralized nervous system (CNS). In these models, the position of the neuroectoderm is defined by the inhibition of BMP signaling on one side of the DV axis—dorsally in vertebrates and ventrally in arthropods [1]. Later during development, similar genetic programs regulate the mediolateral patterning of the vertebrate neural plate and the nerve cords of insects and annelid worms [2,3]. These observations made in the mouse, frog, zebrafish, fly, and ragworm have contributed to the idea that the CNS emerged once, early during bilaterian evolution on the side of the body, where BMP signaling was low [4–8]. DV axis inversion at the base of chordates made this ancestrally ventral BMP minimum dorsal in chordates [9]. While the role of BMP signaling in regulating the DV patterning is clearly ancestral and, with few exceptions, present across all Bilateria [10], its strictly “anti-neural” function is contradicted by the variety in the degrees of centralization of bilaterian nervous systems—from diffuse, to being arranged into a varying number of nerve cords, to ganglionic, to fully centralized [11,12]. Similarly variable are the effects of BMP signaling perturbations on the nervous systems of different groups, from promoting to suppressing neurogenesis to having no effect at all [13–18]. Clearly, in order to understand the ancestral role of BMP signaling in neuron formation we need a wider phylogenetic sampling within and outside Bilateria, especially in animals with diffuse nervous systems and with centralized nervous systems forming in “nonstandard” locations. The evolutionary sister clade to Bilateria is the phylum Cnidaria, encompassing Anthozoa (sea anemones and corals) and four classes of Medusozoa: Hydrozoa (hydroids), Staurozoa (stalked jellyfish), Scyphozoa (true jellyfish), and Cubozoa (box jellyfish) (Fig 1A). Cnidaria possess a full repertoire of BMP pathway genes, including BMP ligands, BMP antagonists, BMP receptors and BMP effectors [19–25]. Cnidarian nervous systems are organized as diffuse nerve nets with local condensations, forming nerve rings and neurite tracts, but lacking a brain-like centralization [26–29]). Neurons are found both in the epidermis and in the gastrodermis of cnidarians, and, strikingly, work on the anthozoan sea anemone Nematostella vectensis showed that gastrodermal neurons are “born” in the gastrodermis rather than migrate there, as is the case in Bilateria [28]. Neuronal gene expression in Nematostella commences already in the blastula, which is before pSMAD1/5 becomes first detectable [30–32]. Neuronal markers were not affected by the disruption of BMP signaling during early phases of neurogenesis [32]. Only at planula larva stage, the upregulation and, surprisingly, the downregulation of BMP signaling both reduced neuronal markers and the number of RFamide- and GLWamide-positive neurons [32] suggesting that neural “induction” is independent of BMP signals, while later neurogenesis may involve a combination of BMP activation and repression. Taken together, it is largely unclear if and to what extent BMP signals are required during cnidarian neurogenesis. (A) Simplified phylogenetic tree of early branching metazoans indicating (i) the presence or absence of BMP pathway genes, (ii) the presence of a BMP-dependent secondary body axis, (iii) involvement of BMP signaling in neurogenesis. White circle – absent, blue circle – present; half blue/half white circle – present in some; blue and white circles with question marks – possibly present or absent, respectively. (B) Schematic of the BMP signaling pathway, BMP signaling is activated by the binding of BMP dimers to the BMP receptor complex, resulting in the nuclear translocation of phosphorylated SMAD1/5 (pSMAD1/5), acting as BMP effector and regulating BMP-responsive gene expression. (C) BMP signaling dynamics during early development of Nematostella: no activity in the blastula, pSMAD1/5 activity around the blastopore in the early gastrula, pSMAD1/5 gradient at 2-day planula stage and dispersed pSMAD1/5 activity in the 4-day planula. Blue dots indicate nuclear pSMAD1/5. (D) Overview of Nematostella polyp anatomy in lateral view and as a cross-section of the body column, the neuro-muscular domain of the mesentery gastrodermis is highlighted in blue. (E) Detailed schematic of an individual mesentery as a cross-section, highlighting the localization of pSMAD1/5-positive nuclei within the neuro-muscular domain. bwe-body wall epidermis, mg-mesoglea, bwg-body wall gastrodermis, mes-mesenteries, bc-body cavity, RMR-retractor muscle region, MS-mesentery stalk, PMR-parietal muscle region. Most of our knowledge about BMP signaling in Cnidaria comes from an anthozoan model—the sea anemone Nematostella vectensis. Unlike radially symmetric Medusozoa, Nematostella and other Anthozoa, display a bilaterally symmetric body plan with a secondary, “directive” body axis (Fig 1A). Similar to the bilaterian DV axis, the directive axis requires the formation of a BMP signaling gradient that is regulated by the interplay of multiple BMP pathway components, including bmp2/4, bmp5-8, gdf5-like (=gdf5-l), chordin, gremlin, rgm and zswim4-6 [30,33–35]. Active BMP signaling can be visualized by the spatial distribution of its transcriptional effector, phosphorylated SMAD1/5 (pSMAD1/5). In Nematostella, at late gastrula stage, BMP signaling forms a pSMAD1/5 gradient along the directive axis. This gradient persists while the directive axis is being patterned, but eventually disappears by 4 day planula larva stage, at which point BMP signaling activity becomes local and especially prominent in the mesenteries—the gastrodermal septa of the body wall [30,33,34]. Our recent study revealed pronounced BMP signaling activity in the mesenteries of the adult Nematostella polyp [36]. The mesenteries harbor a diversity of cell populations including epithelial, digestive, reproductive, muscular and neuronal cells. The two latter are concentrated in the “neuro-muscular” mesentery region, where the longitudinal muscles (the retractor and parietal muscle), the longitudinal neurite tracts, and sensory and ganglion mesentery neurons are located. While BMP signaling is strongest within the neuro-muscular domain, yet it was unclear if BMP signaling occurs in neuronal or muscular cell populations (Fig 1). Here, we analyze the potential involvement of BMP signaling in the diffuse nervous system of the anthozoan cnidarian Nematostella. Additionally, we use two other distantly related medusozoan cnidarians, Tripedalia and Stomolophus for comparison and show a pro-neural action of BMP signaling in these models. Finally, we also demonstrate distinct BMP activity in the central nervous system of the nonmodel spiralian Spadella. Together, our findings suggest that BMP signaling was ancestrally a pro-neural factor. Results BMP pathway genes are expressed across neuronal subtypes in the developmental single-cell atlas of Nematostella To address the expression of BMP components in the neuroglandular and muscular repertoire of Nematostella, we utilized the updated developmental single-cell atlas of Nematostella [37]. The analysis of the neuroglandular subset by Cole and colleagues yielded 47 developmentally closely related but distinct cell states, including insulinoma (insm)-positive (N1) and insm-negative (N2) neurons, secretory cells (S) and digestive gland cells (GD). Among the N1 and N2 neurons, several gastrodermis-derived types have been identified so-far, including six N1 (N1.g1-6) and a single N2 type (N2.g). The analysis of the retractor muscle subset by Cole and colleagues revealed four distinct cell states, including the early retractor muscle state (RM.1), the late tentacle retractor muscle state (TR.2), the early tentacle retractor muscle state (TR.1), and the late retractor muscle state (MR.1). We generated expression plots of BMP pathway genes, including BMP ligands (bmp2/4, bmp5-8, gdf5-l, admp), BMP effectors (smad1/5, smad4, smad4-like), BMP receptors (alk2, alk3/6, bmprII, actrII) and intracellular, membrane-bound, and secreted BMP antagonists (rgm, smad6, chordin, gremlinA, gremlinB, noggin1, noggin2, crossveinless2 and follistatin) [36]. The expression of BMP components was detectable in N1 and N2 neuron types, while it was reduced in putative sensory neurons (N1S) and secretory cells (S) and absent in digestive gland (GD) types (Fig 2A). Notably, the expression of BMP components in N1 and N2 neurons is not limited to the gastrodermal populations, despite the fact that our previous analysis revealed that BMP signaling activity is largely limited to the mesentery gastrodermis [36]. The expression of extracellular molecules such as BMP ligands and antagonists appeared to be sparse with only small fractions of cells ( 2dP), which overlaps with the time of the pSMAD1/5 gradient formation; ii) from 2d to 4d planula stage (2dP- > 4dP), when the pSMAD1/5 activity gradually switches from graded to dispersed and becomes located in the mesenteries, corresponding to the time of gastrodermal patterning and mesentery formation; and iii) from 4d to 5d (4dP- > 5dP), after the directive axis has been fully patterned and pSMAD1/5 is active in the mesenteries. For the different time windows of treatment, we again examined the expression of differentially expressed neural markers from the BMP2/4MO dataset, including soxC and soxB(2), as well as insulinoma (insm), which is expressed in neurosecretory and neuroglandular populations [50], prdm14d as marker for neural progenitors and ganglion neurons in the gastrodermis [40] and pou4 and elav1 as neuronal differentiation markers of gastrodermal and epidermal subpopulations [28,51]. K02288 treatments from blastula to 2d planula resulted in strong downregulation of all analyzed marker genes (Fig 8E), concurring with the results of the BMP2/4MO in situ and RNA-seq data. Similarly, BMP inhibition starting at 2d planula until 4d planula stage, strongly reduced neuronal marker genes (Fig 8E). In comparison, K02288 treatments from the 4d to 5d planula, which is after the directive axis has been patterned, and mesenteries have been formed, displayed only a mild reduction of neural marker expression (Fig 8E), suggesting that the neuronal regulation becomes less sensitive to BMP signaling input in the fully patterned late planula. While early studies indicated a combination of positive and negative effects of BMP signaling on neural patterning in the planula, our data support a predominantly pro-neural influence of BMP signaling in the nervous system of the Nematostella larva. (A) Schematic of the BMP signaling pathway and how it is affected by BMP2/4 morpholino oligo and pharmacological inhibition of the BMP receptor with K02288. (B) Timeline of Nematostella early development and BMP signaling dynamics from fertilization (F), Blastula (B) and 2-5 day Planula larvae (2dpP-5dP). The gray/cyan-colored bar indicates the absence/presence of pSMAD1/5 activity. Duration and fixation time points of morpholino knockdown experiments (purple) and K02288 treatments (lilac). (C) Bar plot of expression changes (log2 fold, padj. < 0.05;) of selected neuronal marker genes in 2 day planula upon bmp2/4MO knockdown by RNA-seq analysis [30], see Materials and methods). (D) Expression analysis of soxC and soxB(2) expression upon ctrlMO, bmp2/4MO, and gdf5-lMO knockdown at 2 day planula stage by in situ hybridization. (E) Selected neuronal marker expression upon K02288 treatments at 2 day planula, 4 day planula and 5 day planula stage visualized by in situ hybridization. Asterisks indicate blastopore, scale bars 100μm. Numerical data for the bar plot can be found in the sel.BMPMO_neuro sheet of the S2 Table. Putative neural lineage markers are downregulated by BMP signaling attenuation in the metaephyra of the scyphozoan Stomolophus Our immunohistochemical analyses suggested that BMP signaling is active in the neurons and cnidocytes of the cubozoan and scyphozoan jellyfish. Therefore, we asked what effect BMP signaling had on the expression of the putative neuronal and cnidocyte markers in the scyphozoan jellyfish Stomolophus. We performed RNA-Seq on 4 biological replicates per treatment of K02288- and DMSO-treated Stomolophus metaephyrae, assembled Stomolophus sp. transcriptome de novo using TRINITY, identified Stomolophus transcripts differentially expressed upon K02288 treatment with DeSeq2, and extracted their potential Nematostella orthologs with OrthoFinder. Using the published single-cell RNA-Seq dataset [37], we then identified which of the Nematostella orthologs of the differentially expressed Stomolophus genes were specifically expressed in neuronal or cnidocyte lineages. Due to limited material, we were unable to validate the expression of the Stomolophus orthologs of these genes by in situ hybridization and scRNA-Seq in metaephyrae, but we considered them to be putative neuronal or cnidocyte genes in Stomolophus and compared changes in their expression upon BMP signaling suppression to the changes in the expression of their Nematostella orthologs upon similar treatments (Fig 9A, 9B). (A) Schematic of BMP signaling inhibition in metaephyra using K02288 treatment and downstream processing and analysis. (B) Venn diagram illustrating the relationships between the combined set of DEGs in K02288 treated Stomolophus metaephyra, K02288 treated Nematostella polyps and previously published Nematostella datasets of DEG genes in the 2-day planula upon bmp2/4 morpholino knockdown and direct pSMAD1/5 targets at Late gastrula and 4d planula [30]. (C) Intersecting DEGs of K02288-treated Nematostella and Stomolophus, as well as of Nematostella BMP2/4 2 day morphants and K02288-treated Stomolophus. Nematostella DEGs and DE TFs were plotted using Nematostella developmental single-cell RNA-Seq data [37] to identify putative neuronal- and cnidocyte-specific gene orthologs in Stomolophus. DeSeq2 analysis revealed 7,661 Stomolophus transcripts differentially expressed in the K02288-treated and DMSO-treated metaephyra (padj <0.05). 5,803 of these 7,661 DEGs had a Nematostella ortholog according to OrthoFinder, corresponding to 4,862 Nematostella gene models (S3 Table). Multiple Stomolophus transcript hits to a single Nematostella gene model were likely due to truncated transcripts after de novo assembly matching different parts of the same Nematostella gene model, uncollapsed isoforms, or Stomolophus-specific gene duplications. We then asked whether any of the Nematostella orthologs of the differentially expressed Stomolophus transcripts were differentially expressed in the K02288 treated adult Nematostella polyp tissues and found 144 such DEGs (Fig 9C and S3 Table). We plotted their expression across Nematostella cell states using the published single-cell RNA-Seq dataset [37] and, along with many genes with broad expression, found 4 genes expressed specifically in neuronal cells and 11 specifically expressed in the cnidocytes (Fig 9C and S1 Plot). All 4 neuronal-specific genes were downregulated upon K02288 treatment in Nematostella and in Stomolophus, in line with the pro-neural function of BMP signaling in both species (Fig 9C and S3 Table). In contrast, 10/11 cnidocyte markers differentially expressed in both species were downregulated in Stomolophus metaephyrae but upregulated in the Nematostella polyp (Fig 9C and S3 Table), which is also in line with Stomolophus cnidocytes being pSMAD1/5-positive and Nematostella cnidocytes being pSMAD1/5-negative (Fig 6K’, [36]). Since metaephyra is a developing juvenile rather than a steady-state animal, we also intersected Nematostella orthologs of the Stomolophus differentially expressed transcripts with the set of DEGs from 2d old Nematostella larvae with BMP signaling abolished by BMP2/4 morpholino knockdown [30]. Here, 2421 genes were found to be differentially expressed in both species (Fig 9C and S3 Table). This very extensive DEG overlap is likely due the presence of not only direct but also indirect BMP target genes in the shared DEG list caused by the prolonged BMP signaling suppression in the Stomolophus metaephyra and morphant Nematostella. We analyzed the expression of these 2421 genes across the different Nematostella cell states and identified 86 neuronal-specific and 203 cnidocyte-specific genes (Fig 9C and S2 and S3 Plots). Upon BMP signaling inhibition, 63 out of 86 neuronal genes and 181 out of 203 cnidocyte genes were downregulated in both species (Fig 9C and S3 Table). This strongly suggested that during development, both the neural and the cnidocyte formation is under positive BMP control in both species. Next, we selected all 131 transcription factors (TFs) present among the 2,421 DEGs and identified 20 neuronal-specific TFs and 15 cnidocyte-specific TFs (Fig 9C and S4 Plot). 19/20 neuronal-specific and 14/15 cnidocyte-specific TFs were downregulated upon BMP signaling suppression in both species (Fig 9C and S3 Table). Thus, at the level of TFs, positive regulation of neuronal and cnidocyte differentiation by BMP signaling in Nematostella and in Stomolophus became even more pronounced. Taken together, we see a clear positive regulation of BMP signaling of neuronal genes in Nematostella and their putatively neuronal Stomolophus orthologs. In Nematostella, we also find a striking change in the modality of the response of the cnidocyte-specific genes to BMP signaling: from positive in the larva to negative in the adult, which suggests that BMP signaling—at least initially—positively regulates the development of the whole cnidocyte and neuronal lineage, which have been shown to share a common origin [43]. BMP signaling is active in the neurogenic domain of a spiralian Arrow worms (Chaetognatha) are planktonic predators named after their dart-shaped body and impressive jaw apparatus. Phylogenomic analyses place them within or as a sister group to Gnathifera, within the superphylum Spiralia [52,53]. Chaetognaths possess a brain composed of several ganglia and a centralized ventral nervous system (ventral nerve center). During embryogenesis, the ventral CNS develops from paired lateral neuroectodermal cell clusters, where cells proliferate to form “lateral somata clusters” and eventually fuse below the gut [54]. In line with observations in other protostomes, we find dorsal expression of bmp2/4 and ventral expression of chordin in the midgastrula of Spadella cephaloptera (S3 Fig). Antibody staining shows a dorsal-to-ventral gradient of nuclear pSMAD1/5 in the gastrula and early elongation stages (Fig 10A–10C’). Importantly, pSMAD1/5 signal is observed in the neuroectodermal regions that later contribute to lateral somata clusters, which will eventually fuse ventrally to form a CNS (Fig 10B–10C’). In the hatchling and juvenile, the pSMAD1/5 staining shows broader distribution, including in mesodermal and epidermal derivatives. Nonetheless, BMP signaling remains prominent in the lateral somata, as well as in a number of dorsal sensory organs such as the pax6-expressing corona ciliata in the head and ciliary sensory organs of the trunk (Fig 10D–10E’, [54]). These results show that parts of the central nervous system of Spadella, particularly the dorsal regions of the neuroectoderm, are located within the pSMAD1/5-positive domain, and that BMP signaling remains active in the ventral nerve center and in the forming sensory organs later in development. (A-A’) Transverse view of an early gastrula (A), and schematic representation of its anatomy (A’). Nuclear pSMAD1/5 forms a dorsal-to-ventral gradient. pSMAD1/5-positive nuclei are observed in the dorsal ectoderm and endoderm (ect, end on A’). (B-B’) Dorsal view of a late gastrula (B), and schematic representation of its anatomy (B’). Nuclear pSMAD1/5 is detected in the dorsal mesoderm (mes on B’) and in some neuroectodermal nuclei (arrowheads; nec on B’). (C-C’) Lateral view of the early elongation stage embryo (C) and schematic representation of its anatomy (C’). pSMAD1/5-positive nuclei are present in the dorsal epidermis (epi on C’), the developing digestive tract (in on C’), including the presumptive foregut (arrows), dorsal neuroblasts (white arrowheads) of the developing ventral nerve center (nbs on C’), and presumptive progenitor cells of the inner corona ciliata (red arrowheads). (D, D’) Dorsal view of the Spadella hatchling (D) and schematic representation of its anatomy (D’). (D) In the head of the hatchling, pSMAD1/5-positive nuclei are detected in the esophagus (es on D’), lateral plate (lp on D’). Additional signal is observed in sensory structures: corona ciliata (inner cells; cc on D’) and dorsal ciliary tuft/fence organs (co on D’). pSMAD1/5-positive nuclei are also present in the dorsal epidermal midline extending along the trunk and tail (dml on D’), the intestine (in on D’), and lateral somata clusters of the ventral nerve center (lsc on D’). Some pSMAD1/5-positive nuclei of unknown identity are also located lateral to the tail midline (arrows). (E, E’) Dorsal view of the juvenile (E) and schematic representation of its anatomy (E’). pSMAD1/5-positive nuclei remain present in the esophagus, lateral plate, and intestine (es, lp, in on E’), and are particularly prominent in the lateral somata clusters of the ventral nerve center (lsc on E’). Expression is also detected in mesoderm-derived lateral cells (lc on E’) and mesenterial cells (mm on E’). pSMAD1/5 signal is shown in yellow, nuclei are stained with DAPI (magenta). Orange outlines demarcate the neuroectoderm (B) and the developing ventral nerve center (C–E). The dashed outline in (B) marks the developing mesoderm and intestine. Dashed circles in (D) and (E) indicate the corona ciliata. Scale bars, 50 μm (A–C) and 100 μm (D, E). Abbreviations: cc, corona ciliata; cf, caudal fin; cg, cerebral ganglion; co, ciliary tuft/fence organs; dml, dorsal epidermal midline cells; ect, ectoderm; end, endoderm; epi, epidermis; es, esophagus; ey, eye; hd, head; in, intestine; lc, lateral cells; lmp, longitudinal muscle cell precursors; lp, lateral plate cells; lsc, lateral somata cluster; mes, mesoderm; mm, mesenterial cells; nbs, neuroblast of the developing VNC; nec, neuroectoderm; tl, tail bud. Discussion BMP signaling promotes neural gene expression in Nematostella Regulatory functions of BMP signaling are manifold in Bilateria, and this multifunctionality also extends to their closest relatives, the cnidarians [36]. In this study, we analyzed the activity of BMP signaling in a part of the gastrodermal nervous system of the sea anemone Nematostella. The previously described overall neural architecture of the polyp includes multiple types of sensory and ganglion neurons with stereotypical placement in the body column [26–28,31]. Different neuron types are present in the two body layers, the epidermis and the gastrodermis, and can be generated by the respective tissue layer [28]. In the gastrodermis of the polyp, subsets of neurons are regionalized and accumulate in proximity to the longitudinal musculature in the mesentery, which is especially pronounced in the longitudinal neurite tracts along the parietal muscle [26,28]. In this region of neural condensations, we observed pronounced BMP signaling activity, partially overlapping with specific subsets of neurons and neuronal progenitors (Fig 11A). We found elevated activity of BMP signaling in neurons of the gastrodermis, specifically within the retractor and the parietal muscle regions. BMP signaling was overlapping with neurons expressing elav1 and nanos1, but it was absent from other neurons located in the same regions, e.g., expressing prdm14d. BMP signaling was also present in two types of soxb(2)-positive neurons, firstly, in neurons with differentiated morphology in the epithelium, and secondly, in basiepithelial cells, likely representing neuronal progenitor cells, based on their localization in the mesoglea. In line with that, analysis of the single-cell RNA-Seq dataset shows the expression of BMP signaling components in the pSC.NPC.1 subset of putative stem cells/neuronal progenitor cells (S4 Fig). Notably, the occurrence of pSMAD1/5-positive basiepithelial cells, which were also soxb(2)-positive in most cases and rarely soxb(2)-negative, seemed to be restricted to the neuro-muscular mesentery region, while basiepithelial cells in other parts of the body were always pSMAD1/5-negative. These observations are especially interesting in light of the recent findings showing that basiepithelial vasa2/piwi1 double-positive cells represent a population of putative multipotent stem cells of the mesentery, which contributes to at least a portion of the neuroglandular/-secretory repertoire of the gastrodermis in the juvenile and adult polyp [41]. The progeny of these vasa2/piwi1-positive cells include soxB(2)-positive and prdm14d-positive neuronal progenitors, as well as elav1-positive neural and insm1-positive neuroglandular cells [41]. Previously, we could determine that basiepithelial vasa2-positive cells located in the medial mesentery and septal filament have no detectable BMP signaling activity, indicating BMP signaling is absent from putative multipotent stem cells [36]. In the context of the proposed model for cell lineage relationships in the gastrodermis [41], we can speculate that BMP signaling activity may correspond to specific branches of the neuroglandular lineage (Fig 11B). BMP signaling may be involved in the decision-making of neuronal progenitors and their derivatives, which is supported by the transcriptional response upon BMP signaling inhibition. (A) Localization of pSMAD1/5-positive and -negative neurons in the gastrodermal nervous system. (B) A model for BMP activity during neuronal differentiation in adult polyps. (C) BMP signaling-mediated regulation of neuronal genes in the planula larvae. (D) Comparative overview of phylogenetic relationships, regions of neural induction and nervous system anatomy in Bilateria and Nematostella. bwe-body wall epidermis, mg-mesoglea, bwg-body wall gastrodermis, mes-mesentery, DNC-dorsal nerve cord, VNC-ventral nerve cord, DNS-diffuse nervous system. Our bulk RNA-seq analysis of K02288 treated adult tissue revealed differential expression of multiple neuronal transcription factors including soxB(2), irx, ap2, ashB and paxB [55–58]. Several gene targets in the transcriptomic data from the adult also overlapped with target genes of our previously published RNA-seq data and anti-pSMAD1/5 ChIP data from early stages of development [30]. In the datasets from the early developmental stages, we found that many neuronal genes were differentially regulated in the 2d planula upon BMP signaling inhibition by BMP2/4 morpholino-mediated knockdown (BMP2/4MO) and/or direct gene targets of BMP signaling at Late gastrula and 4d planula stage. Based on the overlap of these datasets, BMP signaling appears to regulate neuronal genes both directly and indirectly across different stages of Nematostella development (Fig 11C). These findings are in agreement with previous analyses of the embryo and the planula larvae suggesting that BMP signaling is involved in patterning and maintenance of the planula nervous system [32]. Using BMP2/4 morpholino knockdown and reverse transcription (RT)-PCR, Watanabe et al.described the downregulation of genes encoding for a group of achaete-scute and atonal-related basic helix-loop-helix (bHLH) proteins and of the neuropeptides rf-amide and glw-amide [32]. Our RNA-seq analysis of BMP2/4MO planulae yielded similar results, expanding our list of neurogenic factors, the majority of which proved to be downregulated in the absence of BMP signaling. Downregulated genes include soxb(2), marking neuronal progenitors that give rise to nematocytes, sensory and ganglion neurons [42,57], as well as soxC, an upstream regulator of soxB(2) [43], which we have previously identified as a direct target gene of pSMAD1/5 [30]. Moreover, among the downregulated genes we find markers of neural progenitors and neural differentiation (pou4, insm, elav1, prdm14d, nanos1, otp; [28,50,51,59,40]), bHLH family members (ashA, ashB, ashC, ashD, arp6, ato7; [32,55,57]) and transcription factors associated with neurogenesis in bilaterians (gata, soxE1, ap2, irx, isl1, pou3, runx; [31,60]. In accordance, RNA detection by in situ hybridization in 2d planula upon BMP2/4 morpholino knockdown or K02288 BMP receptor inhibition showed significant downregulation of neurogenic factors. Notably, low levels of marker gene expression were still detectable, indicating that neurogenesis was not completely abolished, supporting the idea that BMP signaling is not required for the initiation of neurogenesis in the early embryo. Taken together, our data suggest a predominantly pro-neural role of BMP signaling in Nematostella larva and adult. BMP signaling activity overlaps with neuronal populations in Medusozoa To determine if BMP signaling also plays a role in the diffuse nervous system of medusozoans, we analyzed the activity of BMP in the box jellyfish Tripedalia and the cannonball jellyfish Stomolophus. Co-stainings of pSMAD1/5 and α-tubulin in Tripedalia revealed BMP signaling activity in different parts of the diffuse nerve net of the medusa, highlighting ganglion neurons in the umbrella and the ring nerve. Similar to the situation in Nematostella, where we observed that pSMAD1/5 activity differs between elav1-positive and prdm14d-positive neurons, we found both pSMAD1/5-positive and -negative neurons in the ring nerve of Tripedalia, indicating the presence of distinct neuronal subsets. In the ring nerve, the presence of specific neuronal populations labeled by different antibody combinations was demonstrated in the hydromedusa Cladonema [61] and in Tripedalia [45]. In Tripedalia and Stomolophus, we detected BMP signaling activity in distinct parts of the light- and gravity-sensing rhopalia. Similarly, BMP signaling was also active in the rhopalia of another scyphozoan, the moon jellyfish Aurelia (S5 Fig). The rhopalial nerve net features distinct subsets of regionalized sensory and ganglion cells [45,47,48,62,63]. In Stomolophus and Aurelia, BMP signaling activity was more pronounced around the rhopalial canal compared to the distal segment (S5 Fig). The regionalization of BMP signaling was even more pronounced in the rhopalia of Tripedalia, where areas of pSMAD1/5 activity were restricted to the region adjacent to the rhopalium stalk, pit eyes and the upper lens eye. Outside the eyes, pSMAD1/5 activity is found in the areas known to be populated by specific neurons, including PCNA-positive neurons innervating the pit eye and the upper lens eye, the giant neurons, FMRF-positive neurons around the stalk, but not in the flank neurons [48,47]. Transcriptomic analysis of the putative neuronal genes in Stomolophus showed that a vast majority of them, including soxC, anr7, ashA, ashC, ashD, and neurogenin are positively regulated by BMP signaling. Curiously, in all three medusozoan species, Tripedalia, Stomolophus and Aurelia, we identified pSMAD1/5 activity in the stinging cells (cnidocytes, S5 Fig), a highly specialized cell type of the neurosecretory lineage across Cnidaria [43,60]. In line with that, multiple genes responsible for cnidocyte differentiation in Nematostella (paxA, znf845, nr12, pou4, prdm6, prdm13, jun1, fos1 and minicollagens ncol1, ncol3, ncol5;) were suppressed upon BMP signaling inhibition in Stomolophus. In contrast, in the anthozoan Nematostella, where mature cnidocytes are always pSMAD1/5-negative [36], our transcriptome analyses show that the role of BMP signaling switches from pro-cnidogenic in the larvae to anti-cnidogenic in the adult. In summary, both in medusozoan and anthozoan cnidarians, we detect neurons receiving BMP signals, which suggests that pro-neural role of BMP signaling is an ancestral cnidarian trait predating the anthozoan-medusozoan divergence. BMP signaling promotes neurogenesis in parts of bilaterian nervous systems BMP signaling is associated with various stages of neurogenesis in Bilateria, displaying negative or positive regulatory effects depending on the specific location and timing within the developmental context. In vertebrates and arthropods, the specification of the centralized nervous system requires inhibition of BMP signaling, while differing levels of BMP activity regulate the mediolateral patterning of neurectoderm by homeobox factors nk2.1, nk2.2, nk6, pax6, pax3/7 and msx [64]. The question about neurogenic functions of BMP signaling becomes more complicated when we consider animal groups outside of vertebrates and arthropods. Spiralians, ambulacrarians and xenacoelomorphs display an assortment of centralized and decentralized neural architectures that, in many cases, differ in their developmental regulation from what is known from groups with centralized nervous systems. Little is known about neuronal roles of BMP signaling in xenacoelomorphs, whose phylogenetic position, either at the base of Bilateria or within the deuterostomes, is still unresolved [65–70]. The group displays divergent neural architectures, ranging from diffuse nets with simple nerve plexuses to organized nervous systems with multiple nerve cords [65,71]. BMP signaling inhibition in a xenacoelomorph nemertodermatid did not affect the establishment of the nerve cords but increased the number of serotonin/(5-HT)-positive commissures [16]. In nonchordate deuterostomes, different forms of decentralized nervous systems exist, including the pentaradial nerve cords and nerve ring in echinoderms and the epidermal nerve net in hemichordates, also featuring a ventral and the dorsal nerve cord [72]. Like in chordates, BMP signaling appears to initially repress neurogenesis in ambulacrarians [9], and later, may be required for local neural specification [73]. In the sea urchin, BMP signaling locally promotes the pro-neural markers soxC, soxB1 and sip1, as well as serotonergic neurogenesis only in the anterior neural ectoderm, but not in the ciliary band or in the region of the endomesoderm that later will give rise to the gut [17]. In contrast, BMP signaling disruption in the sea star had no effect on elav expression but affected the localization of elav-positive cells [74]. In the hemichordate Saccoglossus, exogenous BMP had no anti-neurogenic effects, based on the persisting expression of pan-neural marker elav [13]. Neural anatomies in spiralians are diverse, often displaying different numbers of nerve chords. In the context of neurogenesis, BMP signaling was found to have positive or negative effects on the formation or organization of nerve chords, the brain or eyes [18,75–80]. In the snail Ilyanassa, BMP signaling was shown to promote the neural ectoderm and the formation of eyes and the shell [75], while another gastropod, Crepidula, exhibits enlarged cerebral ganglia and ectopic ocelli when BMP signaling is blocked by the selective Activin receptor-like kinase-2 inhibitor Dorsomorphin Homolog 1 (DMH1) [15]. In the annelid worm Chaetopterus, BMP signaling inhibition had no effect on neural structures [76]. Similarly, a recent analysis in the early branching annelid Owenia, where a BMP signaling gradient is responsible for the DV patterning, showed that modulating BMP signaling also did not prevent the development of neural structures or differentiation of specific neuron types [81]. Until now, no information on the areas of BMP signaling activity was available for the nonlophotrochozoan spiralians. We analyzed BMP signaling activity in the embryos, hatchlings and juveniles of the chaetognath Spadella and showed that a dorsal-to-ventral BMP signaling gradient forms during its early development. Spadella has a highly centralized ventral nervous system, the ventral nerve center, which appears to be patterned mediolaterally by the same genes as the central nervous systems of vertebrates, insects and polychaetes [54]. Importantly, the chaetognath ventral nerve center forms from two bilateral patches of neurectoderm located in the lateral sides of the ectoderm of the embryo. These regions give rise to multiple neural components, including the lateral somata clusters, which contribute to the developing nervous system. Here we show that during Spadella development, dorsal parts of the neurogenic ectoderm are specified within the active BMP signaling domain in late gastrula. We detect BMP signaling in the dorsal parts of the lateral somata clusters (neuroectoderm) of the early elongation stage embryos and, subsequently, in parts of the ventral nerve center and transiently also in some of the sensory organs such as the corona ciliata in the hatchling and in the juvenile. Chaetognaths are not the only animals with highly centralized nervous systems, where BMP signaling is active in the neural structures—especially in the peripheral and enteric neurons. In the annelid worm Platynereis and the fly Drosophila, which both have a ventral nerve cord, the expression of pro-neural marker atonal, as well as the formation of peripheral sensory neurons were promoted by BMP signaling [2,82]. In the cephalochordate Amphioxus, BMP signaling inhibition is required for CNS formation [83], whereas high BMP signaling promotes the formation of elav-positive peripheral sensory neurons [14]. Similarly, BMP signaling in the tunicate Ciona is involved in the induction and differentiation of the peripheral nervous system [84,85]. In several vertebrate models (fish, chick, mouse and human), BMP signaling is associated with promoting the enteric nervous system [86,87]. In the zebrafish, pSMAD1/5 activity overlaps with phox2b-positive enteric progenitor cells and elav-positive differentiated enteric neurons, regulating either the generation of progenitor cells or their differentiation at different time points of embryonic and larval development [87]. But the role of BMP signaling in vertebrate nervous system development is not limited to the peripheral neurons. A ventral-to-dorsal Sonic hedgehog signaling gradient and a dorsal-to-ventral BMP signaling gradient in the vertebrate neural tube are responsible for its DV patterning [88]. However, the dorsalmost neural tube cells, in which BMP signaling is strongest, descend from the ventralmost neuroectodermal cells, which are closest to the BMP signaling maximum. Therefore, BMP-dependent patterning of the vertebrate neural tube appears to be not a separate, “late” role of BMP signaling kicking in once its initial anti-neural function has been fulfilled, but rather a continuous BMP-dependent DV patterning of the whole ectoderm, including the neuroectoderm (S6 Fig). Taken together, rather than being an unusual feature of cnidarians with their simple, diffuse nervous systems, pro-neural BMP signaling activity is also observed across Bilateria where it promotes neurogenesis of peripheral, sensory and enteric neurons and plays a part in the chordate CNS patterning. Conclusions and outlook Our analyses suggest that positive neuronal regulation by BMP signaling may represent an ancestral feature, predating the cnidarian-bilaterian divergence. We propose that the “anti-neural” role of BMP signaling documented in vertebrates and arthropods (Fig 11D) is a consequence of the global program of the BMP-dependent DV patterning of the ectoderm in the animals where neurectoderm is represented by a contiguous domain at the “low BMP signaling” side of the DV axis. In the future, it will be important to analyze the expression and the functions of the transcription factors responsible for the mediolateral patterning of the central nervous system in bilaterians with multiple nerve cords as well as in the ones with diffuse nervous systems. If we find evidence of involvement of the mediolateral patterning transcription factors in global patterning of the ectoderm, this will be a strong argument in favor of the “simple urbilaterian” hypothesis [11,12] and multiple independent centralizations of the nervous system across the bilaterian tree. In contrast, evidence of mediolateral TFs specifically controlling neuronal differentiation rather than general ectodermal patterning in the models with decentralized nervous systems would suggest that a high degree of nervous system centralization was already present in the “urbilaterian” [4,8] and was later independently lost in multiple bilaterian lineages. Materials and methods Animal culture Adult polyps of Nematostella vectensis were maintained in the dark at 18 °C in 16 ppt artificial seawater (Nematostella medium, NM) and spawned as described before [89,90]. Tripedalia medusae were collected at the aquarium of the Tiergarten Schönbrunn in Vienna and fixed and processed for antibody staining on the same day. Aurelia polyps were kept in Petri dishes in artificial sea water (ASW, 35‰) at 20 °C and fed with Artemia once a week. Strobilation was induced by keeping the polyps at 15 °C overnight or by incubation in 20 μM Indomethacin/ASW for 3 days. Ephyra were transferred to a small beaker with ASW and fed with Artemia once a week. Stomolophus sp. metaephyrae were provided by the “Haus des Meeres” aquarium (Vienna, Austria), and maintained as described for Aurelia. Live Spadella cephaloptera specimens were collected from the intertidal zone near Roscoff, France, and maintained under laboratory conditions until spawning, following established procedures [54,91]. Single-cell RNA-seq analysis Analysis of single-cell transcriptomic data was performed using previously published data [37]. Gene expression of genes of interest in the developmental dataset and the neuroglandular subset was examined using the Seurat::DotPlot function with maximum dot size fixed at 100%. R scripts used for the processing and analysis of the Nematostella single-cell data are available on GitHub at https://github.com/Genikhovich-Lab/Data-for-plots/. The R-object containing all the raw data used in this study is available at https://cells.ucsc.edu/sea-anemone-atlas/Nv2/all/AllData.Robj [37]. For each gene and cell group (ID.separate), all individual cell-level expression values can be obtained using the Seurat package by running an R-script available at https://github.com/Genikhovich-Lab/Data-for-plots/blob/main/scRNA-SeqDotPlots.R. Summary statistics used for visualization in dot plots were calculated as follows: mean expression represents the average normalized expression across all cells within a group; the percentage of expressing cells represents the fraction of cells with nonzero expression. Standard deviation (SD) and standard error of the mean (SEM) were calculated across cells within each group. No additional normalization or filtering beyond Seurat defaults was applied. Additional data processing steps required to analyze the expression of our genes of interest in stem cells are described in the R-script https://github.com/Genikhovich-Lab/Data-for-plots/blob/main/scRNA-seqDotplots_Fig2_FigS4.R. The content of the https://github.com/Genikhovich-Lab/Data-for-plots/ is additionally archived at 10.5281/zenodo.19686897. Immunostaining and vibratome sectioning Antibody staining in Nematostella polyps, Stomolophus metaephyra, Aurelia ephyra and Tripedalia medusa was performed as described previously [36]. For co-staining pSMAD1/5 and neuronal markers, Aurelia and Stomolophus were fixed for 2 min in cold 0.25% glutaraldehyde/3.7% formaldehyde in PTx (1x PBX, 0.2% Triton-X100), followed by 1 h fixation in 3.7% formaldehyde in PTx at 4 °C. After several washes with PTx, samples were incubated in Blocking solution (1% BSA, 5% sheep serum, 1 × PBS, 0.2% Triton-X100, 20% DMSO) at room temperature for at least 2 h. The primary antibodies (1:200 rabbit anti-pSMAD1/5/9 (Cell signaling, #13820), 1:500 rabbit anti-GLWamide 1675 III p [49], 1:1000 mouse anti-beta-Tubulin (Proteintech, 66240-1-Ig), 1:500 rabbit anti-Serotonin/5-HT (Immunostar, CAT20080), 1:500 rabbit anti-FMRF (Neuromics, RA20002)) were pre-absorbed in Blocking solution (1% BSA, 5% sheep serum, 1 × PBS, 0.2% Triton-X100, 0.1% DMSO) and incubated with the sample at 4 °C overnight. Samples were washed 10 times in 1 × PBS/ 0.2% Triton-X100, incubated in Blocking solution at room temperature for 1 h and stained with the secondary antibodies at room temperature for 2 h or at 4 °C overnight. Secondary antibodies were diluted 1:1000 in Blocking solution (goat α-rabbit IgG-Alexa633 (Invitrogen A21070), goat α-mouse IgG-Alexa488 (Invitrogen A11001), goat α-mouse IgG-Alexa568 (Invitrogen A11004) and 5 μg/ml DAPI) Samples were washed 10 times in 1 × PBS/ 0.2% Triton-X100, and either infiltrated with and mounted in Vectashield Antifade Mounting Medium (H-1000–10, VectorLabs) or processed for vibratome sectioning. For the sectioning, samples were embedded in 10% gelatin/PBS, fixed in 4% Formaldehyde in PBS at 4 °C overnight and sectioned at a Leica VT1200 vibratome as described before. All stainings were performed at least two times independently with three or more animals imaged. Images processing and figure preparation were performed using Fiji [92] and Affinity Designer (https://affinity.serif.com/en-us/designer/). Spadella embryos and post-hatching stages were fixed in 4% paraformaldehyde in MOPS buffer (0.1 M MOPS pH 7.4, 2 mM EGTA, 1 mM MgSO4, 2.5M NaCl). For pSMAD1/5 immunostaining, samples were stored in 70% ethanol at −20 °C. pSMAD1/5 immunostaining was performed on early and late gastrulae, early elongation stage individuals, hatchlings, and early juveniles (5 days post-hatching). Samples were rehydrated through an ethanol/PBS series (50%, 25%, PBS) and washed three times in PBS-TX (0.3% Triton X-100 in PBS). Blocking was carried out overnight at 4 °C in PBS-TX with 3% normal goat serum (NGS), followed by overnight incubation with anti-pSMAD1/5/9 antibody (1:100; Cell Signaling #13820) in PBS-TX + 3% NGS. After a series of washes in PBS-TX for 2 hours, specimens were incubated overnight at 4 °C with goat anti-rabbit Alexa Fluor 488 (1:1000; Invitrogen #A-11094) and DAPI (1 μg/ml) in PBS-TX + 1% NGS. After a final series of washes, samples were mounted in Vectashield Antifade Mounting Medium (Vector Laboratories). Morpholino knockdowns, inhibitor treatments and in situ hybridization in embryos For knockdowns of bmp2/4 and gdf5-like in Nematostella, previously published morpholino oligos were injected into fertilized eggs as described before [33,35]. For chemical BMP signaling inhibition, embryos were incubated in Nematostella Medium (NM) with 6 μM K02288 or, as control treatment, with equal volumes of DMSO. Treatments were carried out starting from blastula stage (18 h post fertilization, hpf). For the wash-out at 2-day (48 hpf), 4-day (96 hpf) or 5-day (120 hpf) planula stage, embryos were washed three times with NM. Morpholino and K02288 treated embryos were fixed for in situ hybridization as described before [30,93] with minor modifications. Nematostella planulae were fixed in 4% PFA in 1 × PBS and 0.1% Tween 20 (PTW) for 1 hr at room temperature. For permeabilization, samples were incubated in 10 μg/ml Proteinase K/PTW for 20 min at room temperature. Following the 2 × SSC wash, 4 d planulae were incubated in 1 unit/μl RNAseT1 in 2 × SSC at 37 °C for 40 min and then washed with 0.075 × SSC to reduce unspecific “collar” staining around the pharynx. Spadella embryos used for HCR-FISH with bmp2/4 and chordin probes were fixed at gastrula stage as described above and stored in 100% methanol at −20 °C (rather than in 70% ethanol). HCR probes were designed with the insitu_probe_generator.py script (https://github.com/rwnull/insitu_probe_generator; [94] and synthesized by Integrated DNA Technologies. Hairpins were purchased from Molecular Instruments (USA). HCR-FISH followed the protocol of Choi et al. [95], with additional modifications from Bruce et al.[96]. Cell nuclei were counterstained with DAPI during the final incubation step. Samples were then incubated in Vectashield for 20 min, cleared in a graded TDE (2,2′-thiodiethanol) series in PBS (30%, 60%, and 80%), and mounted in 80% TDE for imaging. Imaging was performed with a Leica TCS SP5 confocal microscope (Leica Microsystems, Germany). Image processing and figure preparation were carried out using Fiji [92] and Inkscape (https://inkscape.org). Inhibitor treatments and bulk RNA-seq of adult Nematostella polyps Adult Nematostella polyps were incubated in 5mM lidocaine in NM for 5 min or until fully relaxed and transferred to a 6-well plate containing 20 μM K02288 in NM or same volume of DMSO in NM as control. For a more even exposure, the solution was injected into the gastrovascular cavity through the mouth using a 1 ml syringe and a blunt Sterican needle with a bent tip. Animals were treated for 5 h, relaxed for 10 min using 5 mM lidocaine in treatment solution and then transferred to 100% methanol for dissection. Total RNA extraction was performed using the Zymo Research Quick-RNA Miniprep Kit Plus, following the manufacturer's instructions for sample preparations of tissues. Total RNA was extracted from the female mesenteries, male mesenteries, female body wall and female head of the K02288 and DMSO-treated polyps in biological triplicates and shipped to Novogene for polyA-enrichment, library preparation and paired-end 150 bp Illumina sequencing. Adapters were trimmed using Cutadapt 4.4 [97], and the reads were aligned to the Nematostella vectensis genome [98] using star/2.7.10b [99]. Expression values were obtained using featureCounts in Subread/2.1.1 [100] and differential expression analysis was performed with DeSeq2 [101]. For the analysis, expression changes with an adjusted p-value of <0.05 were used. Datasets generated in this study were compared with the previously published anti-pSMAD1/5 ChIP dataset (late gastrula and 4-day planula stage) and the BMP2/4 morpholino (BMP2/4MO) RNA-seq data (2-day planula) [30]. The formerly used NVE gene models were matched to the new NV2 gene annotations. NVE gene models with no clear NV2 ID were not considered for further analysis. NVE-to-NV2 matching reduced the number of gene models used for the comparison, e.g., of 254 NVE models identified as direct targets of BMP signaling by ChIP-Seq [30], 210 had a one-to-one NV2 counterpart. Inhibitor treatments, bulk RNA-seq, de novo transcriptome assembly and differential gene expression analysis in Stomolophus Stomolophus metaephyrae were treated for 16 h with DMSO or 6 μM K02288 and total RNA was extracted from 4 biological replicates for each treatment. The samples were shipped to Novogene, where nondirectional Illumina sequencing libraries were prepared from polyA-selected samples and paired-end 150 bp Illumina sequencing was performed. Trimmomatic v0.40 [102] was used to remove adapter leftovers and low quality stretches using a sliding window of 4 bp and a minimum average quality of 20 and a minimum read length after clipping of 75 bp. Trimmed reads were then assembled using Trinity v2.15.2 [103] using default parameters. Orthologous sequences between Nematostella, human and Stomolophus transcriptomes were identified with Orthofinder v3.1.1 using Diamond as the default search algorithm, famsa for the multiple sequence alignment and fasttree for fast phylogenetic tree construction. Pseudomapping of the reads was performed using Kallisto v0.52.0 [104] with default parameters. Differential gene expression analysis was performed in R using tximport v 1.34.0 [105] for importing and consolidating read counts and DESeq2 v1.46.0 [101] for differential gene expression. R scripts for reproduction of the analysis are available in GitHub https://github.com/Genikhovich-Lab/Stomolophus and additionally archived at 10.5281/zenodo.19686882. Supporting information S1 Fig. BMP signaling is detectable in nematocyst batteries of the Tripedalia exumbrella. ncb - nematocyst battery, Scale bars 50μm. https://doi.org/10.1371/journal.pbio.3003803.s001 (TIFF) S2 Fig. GO-enrichment analysis of the differentially expressed genes upon K02288 treatment of adult polyp tissues. Numerical data for the bar plot can be found in the K02288_adj.GOenrich sheet of the S1 Table. https://doi.org/10.1371/journal.pbio.3003803.s002 (TIFF) S3 Fig. bmp2/4 is expressed dorsally and chordin—ventrally in the midgastrula of the chaetognath Spadella. (A-A’) Lateral view and (B-B’) frontal view and the corresponding sketches of the anatomy of the embryo. Orange outlines demarcate the neuroectoderm (A, B) and dashed outlines demarcate the endoderm (B). Scale bars 50 μm. Abbreviations: ect, ectoderm; eme, endomesoderm; nec, neuroectoderm. https://doi.org/10.1371/journal.pbio.3003803.s003 (TIFF) S4 Fig. The expression of BMP signaling components in the developmental subset of primordial stem cells and neuronal progenitors is enriched in the soxB2-positive subcluster pSC.NPC.1. Average scaled expression of 0 or below is indicated in gray. Numerical data for the plot can be extracted by running the R scripts found here 10.5281/zenodo.19686897. https://doi.org/10.1371/journal.pbio.3003803.s004 (TIFF) S5 Fig. BMP signaling activity in the ephyra of the moon jellyfish Aurelia is present in the muscle field of the central disc, cnidocytes and parts of the rhopalia. (A) Life cycle schematic of Aurelia coerulea, details of the ephyra juvenile stage. (B) Rhopalia in the lappets exhibit pSMAD1/5 staining at the rhopalic canal (rc). (C) pSMAD1/5 staining in the central disc of the ephyra. Pha - phalloidin. (C’) Detail showing pSMAD1/5-positive crescent-shaped nuclei typical for cnidocytes. cd - central disc, rh - rhopalium, rc - rhopalial canal, bs-basal segment, in-intermediate segment, te-terminal segment; Scale bar (B-B’) 25 μm, (C) 50 μm, (C’) 5 μm. https://doi.org/10.1371/journal.pbio.3003803.s005 (TIFF) S6 Fig. Scheme of the BMP signaling-dependent patterning of the vertebrate ectoderm in case of primary neurulation. The dorsalmost part of the neural tube originates from the ventralmost neurectoderm. Blue circles indicate pSMAD1/5 positive nuclei. https://doi.org/10.1371/journal.pbio.3003803.s006 (TIFF) S1 Plot. Dot plot sorted by similar expression of Nematostella orthologs of differentially expressed Stomolophus transcripts with differential expression in K02288 treated adult Nematostella polyps. The R-object containing all the raw scRNA-Seq data used in this study is available at https://cells.ucsc.edu/sea-anemone-atlas/Nv2/all/AllData.Robj. The list of gene models and the R script “scRNA-SeqDotPlots_S1_Plot-S4_Plot.R” allowing to reproduce this plot and extract the underlying data can be found at 10.5281/zenodo.19686897. https://doi.org/10.1371/journal.pbio.3003803.s007 (PDF) S2 Plot. Dot plot of Nematostella orthologs of differentially expressed Stomolophus transcripts with differential expression in BMP2/4 2d Nematostella morphants, specifically expressed in neurons. The R-object containing all the raw scRNA-Seq data used in this study is available at https://cells.ucsc.edu/sea-anemone-atlas/Nv2/all/AllData.Robj. The list of gene models and the R script “scRNA-SeqDotPlots_S1_Plot-S4_Plot.R” allowing to reproduce this plot and extract the underlying data can be found at 10.5281/zenodo.19686897). https://doi.org/10.1371/journal.pbio.3003803.s008 (PDF) S3 Plot. Dot plot of Nematostella orthologs of differentially expressed Stomolophus transcripts with differential expression in BMP2/4 2d Nematostella morphants, specifically expressed in cnidocytes. The R-object containing all the raw scRNA-Seq data used in this study is available at https://cells.ucsc.edu/sea-anemone-atlas/Nv2/all/AllData.Robj. The list of gene models and the R script “scRNA-SeqDotPlots_S1_Plot-S4_Plot.R” allowing to reproduce this plot and extract the underlying data can be found at 10.5281/zenodo.19686897). https://doi.org/10.1371/journal.pbio.3003803.s009 (PDF) S4 Plot. Dot plot sorted by similar expression of Nematostella orthologs of differentially expressed Stomolophus transcripts encoding TF with differential expression in BMP2/4 2d Nematostella morphants. The R-object containing all the raw scRNA-Seq data used in this study is available at https://cells.ucsc.edu/sea-anemone-atlas/Nv2/all/AllData.Robj. The list of gene models and the R script “scRNA-SeqDotPlots_S1_Plot-S4_Plot.R” allowing to reproduce this plot and extract the underlying data can be found at 10.5281/zenodo.19686897). https://doi.org/10.1371/journal.pbio.3003803.s010 (PDF) S1 Table. List of differentially expressed genes upon K02288 treatments of adult polyps. https://doi.org/10.1371/journal.pbio.3003803.s011 (XLSX) S2 Table. Comparison of K02288 RNA-Seq, BMP2/4MObmp2/4MO RNA-Seq and LG4dP anti-pSMAD1/5 ChIP-Seq datasets. https://doi.org/10.1371/journal.pbio.3003803.s012 (XLSX) S3 Table. List of Stomolophus differentially expressed transcripts upon K02288 treatment (padj < 0.05), Nematostella gene models orthologous to the differentially expressed Stomolophus transcripts and comparison of the differential expression in Stomolophus with the DEG lists of K02288 treatments of adult Nematostella polyps and 2dpf BMP2/4MO-injected Nematostella embryos. https://doi.org/10.1371/journal.pbio.3003803.s013 (XLSX) Acknowledgments Confocal microscopy was performed at the Core Facility Cell Imaging and Ultrastructure Research, University of Vienna - member of the Vienna Life-Science Instruments (VLSI). Computational work was performed on the Life Science Compute Cluster (LiSC) of the University of Vienna. We thank Fabian Rentzsch (University of Bergen) for providing soxb(2)::mOrange and prdm14d::gfp transgenic lines of Nematostella, Anders Garm (University of Copenhagen) and Yulia Kraus (Lomonosov Moscow State University) for help with interpreting the morphology of the Tripedalia rhopalium, and Sanjay Narayanaswami for the discussions. References - 1. Mizutani CM, Bier E. EvoD/Vo: the origins of BMP signalling in the neuroectoderm. Nat Rev Genet. 2008;9(9):663–77. pmid:18679435 - 2. Denes AS, Jékely G, Steinmetz PRH, Raible F, Snyman H, Prud’homme B, et al. Molecular architecture of annelid nerve cord supports common origin of nervous system centralization in bilateria. Cell. 2007;129(2):277–88. pmid:17448990 - 3. Sprecher SG, Reichert H. The urbilaterian brain: developmental insights into the evolutionary origin of the brain in insects and vertebrates. Arthropod Struct Dev. 2003;32(1):141–56. pmid:18089000 - 4. Arendt D, Denes AS, Jékely G, Tessmar-Raible K. The evolution of nervous system centralization. Philos Trans R Soc Lond B Biol Sci. 2008;363(1496):1523–8. pmid:18192182 - 5. Arendt D, Tosches MA, Marlow H. From nerve net to nerve ring, nerve cord and brain—evolution of the nervous system. Nat Rev Neurosci. 2016;17(1):61–72. pmid:26675821 - 6. Arendt D, Nübler-Jung K. Common ground plans in early brain development in mice and flies. Bioessays. 1996;18(3):255–9. pmid:8867740 - 7. De Robertis EM. Evo-devo: variations on ancestral themes. Cell. 2008;132(2):185–95. pmid:18243095 - 8. Lichtneckert R, Reichert H. Insights into the urbilaterian brain: conserved genetic patterning mechanisms in insect and vertebrate brain development. Heredity (Edinb). 2005;94(5):465–77. pmid:15770230 - 9. Su Y-H, Chen Y-C, Ting H-C, Fan T-P, Lin C-Y, Wang K-T, et al. BMP controls dorsoventral and neural patterning in indirect-developing hemichordates providing insight into a possible origin of chordates. Proc Natl Acad Sci U S A. 2019;116(26):12925–32. pmid:31189599 - 10. Mörsdorf D, Knabl P, Genikhovich G. Highly conserved and extremely evolvable: BMP signalling in secondary axis patterning of Cnidaria and Bilateria. Dev Genes Evol. 2024;234(1):1–19. pmid:38472535 - 11. Martín-Durán JM, Hejnol A. A developmental perspective on the evolution of the nervous system. Dev Biol. 2021;475:181–92. pmid:31610146 - 12. Moroz LL. On the independent origins of complex brains and neurons. Brain Behav Evol. 2009;74(3):177–90. pmid:20029182 - 13. Lowe CJ, Terasaki M, Wu M, Freeman RM Jr, Runft L, Kwan K, et al. Dorsoventral patterning in hemichordates: insights into early chordate evolution. PLoS Biol. 2006;4(9):e291. pmid:16933975 - 14. Lu T-M, Luo Y-J, Yu J-K. BMP and Delta/Notch signaling control the development of amphioxus epidermal sensory neurons: insights into the evolution of the peripheral sensory system. Development. 2012;139(11):2020–30. pmid:22535413 - 15. Lyons DC, Perry KJ, Batzel G, Henry JQ. BMP signaling plays a role in anterior-neural/head development, but not organizer activity, in the gastropod Crepidula fornicata. Dev Biol. 2020;463(2):135–57. pmid:32389712 - 16. Martín-Durán JM, Pang K, Børve A, Lê HS, Furu A, Cannon JT, et al. Convergent evolution of bilaterian nerve cords. Nature. 2018;553(7686):45–50. pmid:29236686 - 17. McClay DR, Miranda E, Feinberg SL. Neurogenesis in the sea urchin embryo is initiated uniquely in three domains. Development. 2018;145(21):dev167742. pmid:30413529 - 18. Webster NB, Corbet M, Sur A, Meyer NP. Role of BMP signaling during early development of the annelid Capitella teleta. Dev Biol. 2021;478:183–204. pmid:34216573 - 19. Genikhovich G, Technau U. On the evolution of bilaterality. Development. 2017;144(19):3392–404. pmid:28974637 - 20. Gold DA, Katsuki T, Li Y, Yan X, Regulski M, Ibberson D, et al. The genome of the jellyfish Aurelia and the evolution of animal complexity. Nat Ecol Evol. 2019;3(1):96–104. pmid:30510179 - 21. Kayal E, Bentlage B, Sabrina Pankey M, Ohdera AH, Medina M, Plachetzki DC, et al. Phylogenomics provides a robust topology of the major cnidarian lineages and insights on the origins of key organismal traits. BMC Evol Biol. 2018;18(1). - 22. Khalturin K, Shinzato C, Khalturina M, Hamada M, Fujie M, Koyanagi R, et al. Medusozoan genomes inform the evolution of the jellyfish body plan. Nat Ecol Evol. 2019;3(5):811–22. pmid:30988488 - 23. Leclère L, Horin C, Chevalier S, Lapébie P, Dru P, Peron S, et al. The genome of the jellyfish Clytia hemisphaerica and the evolution of the cnidarian life-cycle. Nat Ecol Evol. 2019;3(5):801–10. pmid:30858591 - 24. Lewis Ames C, Ryan JF, Bely AE, Cartwright P, Collins AG. A new transcriptome and transcriptome profiling of adult and larval tissue in the box jellyfish Alatina alata: an emerging model for studying venom, vision and sex. BMC Genomics. 2016;17:650. pmid:27535656 - 25. Ohdera A, Ames CL, Dikow RB, Kayal E, Chiodin M, Busby B, et al. Box, stalked, and upside-down? Draft genomes from diverse jellyfish (Cnidaria, Acraspeda) lineages: Alatina alata (Cubozoa), Calvadosia cruxmelitensis (Staurozoa), and Cassiopea xamachana (Scyphozoa). Gigascience. 2019;8(7):giz069. pmid:31257419 - 26. Havrilak JA, Faltine-Gonzalez D, Wen Y, Fodera D, Simpson AC, Magie CR, et al. Characterization of NvLWamide-like neurons reveals stereotypy in Nematostella nerve net development. Dev Biol. 2017;431(2):336–46. pmid:28888696 - 27. Marlow HQ, Srivastava M, Matus DQ, Rokhsar D, Martindale MQ. Anatomy and development of the nervous system of Nematostella vectensis, an anthozoan cnidarian. Dev Neurobiol. 2009;69(4):235–54. pmid:19170043 - 28. Nakanishi N, Renfer E, Technau U, Rentzsch F. Nervous systems of the sea anemone Nematostella vectensis are generated by ectoderm and endoderm and shaped by distinct mechanisms. Development. 2012;139(2):347–57. pmid:22159579 - 29. Zhong R, Seidel CW, Klompen AML, Gibson MC. Morphological, molecular, and functional evidence for a CNS-like oral nerve ring in the sea anemone Nematostella vectensis. Evol Biol. 2026. - 30. Knabl P, Schauer A, Pomreinke AP, Zimmermann B, Rogers KW, Čapek D, et al. Analysis of SMAD1/5 target genes in a sea anemone reveals ZSWIM4-6 as a novel BMP signaling modulator. Elife. 2024;13:e80803. pmid:38323609 - 31. Rentzsch F, Layden M, Manuel M. The cellular and molecular basis of cnidarian neurogenesis. Wiley Interdiscip Rev Dev Biol. 2017;6(1):e257. pmid:27882698 - 32. Watanabe H, Kuhn A, Fushiki M, Agata K, Özbek S, Fujisawa T, et al. Sequential actions of β-catenin and Bmp pattern the oral nerve net in Nematostella vectensis. Nat Commun. 2014;5:5536. pmid:25534229 - 33. Genikhovich G, Fried P, Prünster MM, Schinko JB, Gilles AF, Fredman D, et al. Axis patterning by BMPs: Cnidarian network reveals evolutionary constraints. Cell Rep. 2015;10(10):1646–54. pmid:25772352 - 34. Leclère L, Rentzsch F. RGM regulates BMP-mediated secondary axis formation in the sea anemone Nematostella vectensis. Cell Rep. 2014;9(5):1921–30. pmid:25482565 - 35. Saina M, Genikhovich G, Renfer E, Technau U. BMPs and chordin regulate patterning of the directive axis in a sea anemone. Proc Natl Acad Sci U S A. 2009;106(44):18592–7. pmid:19833871 - 36. Knabl P, Mörsdorf D, Genikhovich G. A whole-body atlas of BMP signaling activity in an adult sea anemone. BMC Biol. 2025;23(1):49. pmid:39984987 - 37. Cole AG, Steger J, Hagauer J, Denner A, Ferrer Murguia P, Knabl P, et al. Updated single cell reference atlas for the starlet anemone Nematostella vectensis. Front Zool. 2024;21(1):8. pmid:38500146 - 38. Renfer E, Amon-Hassenzahl A, Steinmetz PRH, Technau U. A muscle-specific transgenic reporter line of the sea anemone, Nematostella vectensis. Proc Natl Acad Sci U S A. 2010;107(1):104–8. pmid:20018670 - 39. Denner A, Steger J, Ries A, Morozova-Link E, Ritter J, Haas F, et al. Nanos2 marks precursors of somatic lineages and is required for germline formation in the sea anemone Nematostella vectensis. Sci Adv. 2024;10(33):eado0424. pmid:39151009 - 40. Lemaître QIB, Bartsch N, Kouzel IU, Busengdal H, Richards GS, Steinmetz PRH, et al. NvPrdm14d-expressing neural progenitor cells contribute to non-ectodermal neurogenesis in Nematostella vectensis. Nat Commun. 2023;14(1):4854. pmid:37563174 - 41. Miramón-Puértolas P, Pascual-Carreras E, Steinmetz PRH. A population of Vasa2 and Piwi1 expressing cells generates germ cells and neurons in a sea anemone. Nat Commun. 2024;15(1):8765. pmid:39384751 - 42. Richards GS, Rentzsch F. Transgenic analysis of a SoxB gene reveals neural progenitor cells in the cnidarian Nematostella vectensis. Development. 2014;141(24):4681–9. pmid:25395455 - 43. Steger J, Cole AG, Denner A, Lebedeva T, Genikhovich G, Ries A, et al. Single-cell transcriptomics identifies conserved regulators of neuroglandular lineages. Cell Rep. 2022;40(12):111370. pmid:36130520 - 44. Layden MJ, Johnston H, Amiel AR, Havrilak J, Steinworth B, Chock T, et al. MAPK signaling is necessary for neurogenesis in Nematostella vectensis. BMC Biol. 2016;14:61. pmid:27480076 - 45. Garm A, Poussart Y, Parkefelt L, Ekström P, Nilsson D-E. The ring nerve of the box jellyfish Tripedalia cystophora. Cell Tissue Res. 2007;329(1):147–57. pmid:17340150 - 46. Kass-Simon G, Scappaticci, Jr. AA. The behavioral and developmental physiology of nematocysts. Canadian Journal of Zoology. 2002;80(10):1772–94. - 47. Nielsen SKD, Koch TL, Wiisbye SH, Grimmelikhuijzen CJP, Garm A. Neuropeptide expression in the box jellyfish Tripedalia cystophora-New insights into the complexity of a “simple” nervous system. J Comp Neurol. 2021;529(11):2865–82. pmid:33660861 - 48. Skogh C, Garm A, Nilsson D-E, Ekström P. Bilaterally symmetrical rhopalial nervous system of the box jellyfish Tripedalia cystophora. J Morphol. 2006;267(12):1391–405. pmid:16874799 - 49. Schmich J, Trepel S, Leitz T. The role of GLWamides in metamorphosis of Hydractinia echinata. Dev Genes Evol. 1998;208(5):267–73. pmid:9683742 - 50. Tournière O, Gahan JM, Busengdal H, Bartsch N, Rentzsch F. Insm1-expressing neurons and secretory cells develop from a common pool of progenitors in the sea anemone Nematostella vectensis. Sci Adv. 2022;8(16):eabi7109. pmid:35442742 - 51. Tournière O, Dolan D, Richards GS, Sunagar K, Columbus-Shenkar YY, Moran Y, et al. NvPOU4/Brain3 Functions as a Terminal Selector Gene in the Nervous System of the Cnidarian Nematostella vectensis. Cell Rep. 2020;30(13):4473-4489.e5. pmid:32234481 - 52. Laumer CE, Fernández R, Lemer S, Combosch D, Kocot KM, Riesgo A, et al. Revisiting metazoan phylogeny with genomic sampling of all phyla. Proc Biol Sci. 2019;286(1906):20190831. pmid:31288696 - 53. Marlétaz F, Peijnenburg KTCA, Goto T, Satoh N, Rokhsar DS. A new spiralian phylogeny places the enigmatic arrow worms among Gnathiferans. Curr Biol. 2019;29(2):312-318.e3. pmid:30639106 - 54. Ordoñez JF, Wollesen T. Unfolding the ventral nerve center of chaetognaths. Neural Dev. 2024;19(1):5. pmid:38720353 - 55. Layden MJ, Boekhout M, Martindale MQ. Nematostella vectensis achaete-scute homolog NvashA regulates embryonic ectodermal neurogenesis and represents an ancient component of the metazoan neural specification pathway. Development. 2012;139(5):1013–22. pmid:22318631 - 56. Matus DQ, Pang K, Daly M, Martindale MQ. Expression of Pax gene family members in the anthozoan cnidarian, Nematostella vectensis. Evol Dev. 2007;9(1):25–38. pmid:17227364 - 57. Richards GS, Rentzsch F. Regulation of Nematostella neural progenitors by SoxB, Notch and bHLH genes. Development. 2015;142(19):3332–42. pmid:26443634 - 58. Sinigaglia C, Busengdal H, Leclère L, Technau U, Rentzsch F. The bilaterian head patterning gene six3/6 controls aboral domain development in a Cnidarian. PLoS Biology. 2013;11(2):e1001488. - 59. Nakamura R, Nakamura R, Watanabe H. Cnidarian pharyngeal nervous system illustrates prebilaterian neurosecretory regulation of feeding. Evol Biol. 2023. - 60. Galliot B, Quiquand M, Ghila L, de Rosa R, Miljkovic-Licina M, Chera S. Origins of neurogenesis, a cnidarian view. Dev Biol. 2009;332(1):2–24. pmid:19465018 - 61. Koizumi O, Hamada S, Minobe S, Hamaguchi-Hamada K, Kurumata-Shigeto M, Nakamura M, et al. The nerve ring in cnidarians: its presence and structure in hydrozoan medusae. Zoology (Jena). 2015;118(2):79–88. pmid:25498132 - 62. Nakanishi N, Hartenstein V, Jacobs DK. Development of the rhopalial nervous system in Aurelia sp.1 (Cnidaria, Scyphozoa). Dev Genes Evol. 2009;219(6):301–17. pmid:19543911 - 63. Parkefelt L, Ekström P. Prominent system of RFamide immunoreactive neurons in the rhopalia of box jellyfish (Cnidaria: Cubozoa). J Comp Neurol. 2009;516(3):157–65. pmid:19598151 - 64. Holland LZ, Carvalho JE, Escriva H, Laudet V, Schubert M, Shimeld SM, et al. Evolution of bilaterian central nervous systems: a single origin? Evodevo. 2013;4(1):27. pmid:24098981 - 65. Álvarez-Presas M, Ruiz-Trillo I, Paps J. Novel genomic approaches support Xenacoelomorpha as sister to all Bilateria. In Review. 2024. - 66. Cannon JT, Vellutini BC, Smith J 3rd, Ronquist F, Jondelius U, Hejnol A. Xenacoelomorpha is the sister group to Nephrozoa. Nature. 2016;530(7588):89–93. pmid:26842059 - 67. Kapli P, Natsidis P, Leite DJ, Fursman M, Jeffrie N, Rahman IA, et al. Lack of support for Deuterostomia prompts reinterpretation of the first Bilateria. Sci Adv. 2021;7(12):eabe2741. pmid:33741592 - 68. Martinez P, Bailly X, Sprecher SG, Hartenstein V. The Acoel nervous system: morphology and development. Neural Dev. 2024;19(1):9. pmid:38907301 - 69. Philippe H, Brinkmann H, Copley RR, Moroz LL, Nakano H, Poustka AJ, et al. Acoelomorph flatworms are deuterostomes related to Xenoturbella. Nature. 2011;470(7333):255–8. pmid:21307940 - 70. Schiffer PH, Natsidis P, Leite DJ, Robertson HE, Lapraz F, Marlétaz F, et al. Insights into early animal evolution from the genome of the xenacoelomorph worm Xenoturbella bocki. eLife. 2024;13. - 71. Hartenstein V, Sprecher S, Martinez P. The ancestry of bilaterian neural architecture: cell types, patterning and developmental mechanisms in pre-bilaterian and early bilaterian groups. In: Reference module in neuroscience and biobehavioral psychology. Elsevier; 2025. p. B9780443273803000439. - 72. Adameyko I. Evolutionary origin of the neural tube in basal deuterostomes. Curr Biol. 2023;33(8):R319–31. pmid:37098338 - 73. Hinman VF, Burke RD. Embryonic neurogenesis in echinoderms. Wiley Interdiscip Rev Dev Biol. 2018;7(4):e316. pmid:29470839 - 74. Yankura KA, Koechlein CS, Cryan AF, Cheatle A, Hinman VF. Gene regulatory network for neurogenesis in a sea star embryo connects broad neural specification and localized patterning. Proc Natl Acad Sci U S A. 2013;110(21):8591–6. pmid:23650356 - 75. Lambert JD, Johnson AB, Hudson CN, Chan A. Dpp/BMP2-4 mediates signaling from the D-quadrant organizer in a spiralian embryo. Curr Biol. 2016;26(15):2003–10. pmid:27397892 - 76. Lanza AR, Seaver EC. Activin/Nodal signaling mediates dorsal-ventral axis formation before third quartet formation in embryos of the annelid Chaetopterus pergamentaceus. Evodevo. 2020;11:17. pmid:32788949 - 77. Lanza AR, Seaver EC. Functional evidence that Activin/Nodal signaling is required for establishing the dorsal-ventral axis in the annelid Capitella teleta. Development. 2020;147(18):dev189373. pmid:32967906 - 78. Lewin TD, Sakagami T, Shimizu K, Kao L-J, Chiu Y-L, Liao IJ-Y, et al. Brachiopod genome unveils the evolution of BMP signalling in bilaterian body patterning. Nat Commun. 2026;17(1):3856. pmid:41820373 - 79. Tan S, Huan P, Liu B. Molluskan dorsal-ventral patterning relying on BMP2/4 and chordin provides insights into spiralian development and evolution. Mol Biol Evol. 2022;39(1):msab322. pmid:34751376 - 80. Webster NB, Meyer NP. Capitella teleta gets left out: possible evolutionary shift causes loss of left tissues rather than increased neural tissue from dominant-negative BMPR1. Neural Dev. 2024;19(1):4. - 81. Carrillo-Baltodano AM, Haillot E, Meha SM, Luqman I, Pashaj A, Montagne J, et al. Developmental system drift in dorsoventral patterning is linked to transitions to autonomous development in Annelida. Nat Commun. 2026. - 82. Rusten TE, Cantera R, Kafatos FC, Barrio R. The role of TGF beta signaling in the formation of the dorsal nervous system is conserved between Drosophila and chordates. Development. 2002;129(15):3575–84. pmid:12117808 - 83. Onai T, Yu J-K, Blitz IL, Cho KWY, Holland LZ. Opposing Nodal/Vg1 and BMP signals mediate axial patterning in embryos of the basal chordate amphioxus. Dev Biol. 2010;344(1):377–89. pmid:20488174 - 84. Pasini A, Amiel A, Rothbächer U, Roure A, Lemaire P, Darras S. Formation of the ascidian epidermal sensory neurons: insights into the origin of the chordate peripheral nervous system. PLoS Biol. 2006;4(7):e225. pmid:16787106 - 85. Waki K, Imai KS, Satou Y. Genetic pathways for differentiation of the peripheral nervous system in ascidians. Nat Commun. 2015;6:8719. pmid:26515371 - 86. Chalazonitis A, Pham TD, Li Z, Roman D, Guha U, Gomes W, et al. Bone morphogenetic protein regulation of enteric neuronal phenotypic diversity: relationship to timing of cell cycle exit. J Comp Neurol. 2008;509(5):474–92. pmid:18537141 - 87. Moore JA, Moreno-Campos R, Noah AS, Singleton EW, Uribe RA. BMP signaling pathway member expression is enriched in enteric neural progenitors and required for zebrafish enteric nervous system development. Dev Dyn. 2025;254(3):272–87. pmid:39297504 - 88. Zagorski M, Tabata Y, Brandenberg N, Lutolf MP, Tkačik G, Bollenbach T, et al. Decoding of position in the developing neural tube from antiparallel morphogen gradients. Science. 2017;356(6345):1379–83. pmid:28663499 - 89. Fritzenwanker JH, Technau U. Induction of gametogenesis in the basal cnidarian Nematostella vectensis (Anthozoa). Development Genes and Evolution. 2003;212(2):99–103. - 90. Genikhovich G, Technau U. Induction of Spawning in the Starlet Sea Anemone Nematostella vectensis, In Vitro Fertilization of Gametes, and Dejellying of Zygotes. Cold Spring Harbor Protocols. 2009;2009(9):pdb.prot5281. - 91. Wollesen T, Rodriguez Monje SV, Oel AP, Arendt D. Characterization of eyes, photoreceptors, and opsins in developmental stages of the arrow wormSpadella cephaloptera(Chaetognatha). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution. 2023;340(5):342–53. - 92. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676–82. pmid:22743772 - 93. Lebedeva T, Aman AJ, Graf T, Niedermoser I, Zimmermann B, Kraus Y, et al. Cnidarian-bilaterian comparison reveals the ancestral regulatory logic of the β-catenin dependent axial patterning. Nat Commun. 2021;12(1):4032. pmid:34188050 - 94. Kuehn E, Clausen DS, Null RW, Metzger BM, Willis AD, Özpolat BD. Segment number threshold determines juvenile onset of germline cluster expansion in Platynereis dumerilii. J Exp Zool B Mol Dev Evol. 2022;338(4):225–40. pmid:34793615 - 95. Choi HMT, Calvert CR, Husain N, Huss D, Barsi JC, Deverman BE, et al. Mapping a multiplexed zoo of mRNA expression. Development. 2016;143(19):3632–7. pmid:27702788 - 96. Bruce SH, Jerz G, Kelly SR, McCarthy J, Pomerantz A, Senevirathne G, et al. Hybridization Chain Reaction (HCR) In Situ Protocol v1. 2021. 10.17504/protocols.io.bunznvf6 - 97. Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnetJournal. 2011;17(1):10. - 98. Zimmermann B, Montenegro JD, Robb SMC, Fropf WJ, Weilguny L, He S, et al. Topological structures and syntenic conservation in sea anemone genomes. Nature Communications. 2023;14(1):null. - 99. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21. pmid:23104886 - 100. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30(7):923–30. pmid:24227677 - 101. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281 - 102. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. - 103. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nature Biotechnology. 2011;29(7):644–52. - 104. Bray NL, Pimentel H, Melsted P, Pachter L. Near-optimal probabilistic RNA-seq quantification. Nature Biotechnology. 2016;34(5):525–7. - 105. Soneson C, Love MI, Robinson MD. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research. 2015;4:1521.