A whole-body atlas of BMP signaling activity in an adult sea anemone

Knabl, Paul; Mörsdorf, David; Genikhovich, Grigory

doi:10.1186/s12915-025-02150-w

Research
Open access
Published: 21 February 2025

A whole-body atlas of BMP signaling activity in an adult sea anemone

Paul Knabl^1,2,
David Mörsdorf¹ &
Grigory Genikhovich¹

BMC Biology volume 23, Article number: 49 (2025) Cite this article

892 Accesses
19 Altmetric
Metrics details

Abstract

Background

BMP signaling is responsible for the second body axis patterning in Bilateria and in the bilaterally symmetric members of the bilaterian sister clade Cnidaria—corals and sea anemones. However, medusozoan cnidarians (jellyfish, hydroids) are radially symmetric, and yet their genomes contain BMP signaling components. This evolutionary conservation suggests that BMP signaling must have other functions not related to axial patterning, which keeps BMP signaling components under selective pressure.

Results

To find out what these functions might be, we generated a detailed whole-body atlas of BMP activity in the sea anemone Nematostella. In the adult polyp, we discover an unexpected diversity of domains with BMP signaling activity, which is especially prominent in the head, as well as across the neuro-muscular and reproductive parts of the gastrodermis. In accordance, analysis of two medusozoan species, the true jellyfish Aurelia and the box jellyfish Tripedalia, revealed similarly broad and diverse BMP activity.

Conclusions

Our study reveals multiple, distinct domains of BMP signaling in Anthozoa and Medusozoa, supporting the versatile nature of the BMP pathway across Cnidaria. Most prominently, BMP signaling appears to be involved in tentacle formation, neuronal development, and gameto- or gonadogenesis.

Background

BMP signaling is initiated by the binding of dimeric BMP ligands to a tetrameric BMP receptor complex of two type II and two type I receptors. The receptor complex recruits and phosphorylates the transcriptional effector SMAD1/5 that binds the common-mediator SMAD4 to form a trimeric complex of two SMAD1/5 molecules and one SMAD4. The SMAD complex translocates to the nucleus, where it modulates BMP-responsive gene expression (Fig. 1A). Since the initial discovery of BMPs in the context of bone and cartilage morphogenesis [1, 2], BMP signaling has been studied in various vertebrate and invertebrate species, revealing far more versatile roles of the pathway. BMP signaling has been demonstrated to be involved in processes related to tissue homeostasis [3], apoptosis [4,5,6], cell fate commitment [7, 8], as well as gametogenesis, gonadogenesis, and reproduction [9,10,11]. Moreover, BMP signaling is vital for instructing pattern formation of various tissues and organs, such as the limb bud [12], the insect imaginal discs [13], the gastrointestinal tract [14, 15], the nervous system [8], and arguably most famously, the dorsoventral body axis across Bilateria [16, 17]. While the roles of BMP signaling in Bilateria are multifaceted, its functions in non-bilaterian Metazoa are much less explored.

BMP pathway members are not exclusive to Bilateria but highly conserved in the bilaterian sister clade Cnidaria, a phylum comprising Anthozoa (sea anemones and corals) and Medusozoa (jellyfish and hydroids) (Fig. 1B), the latter encompassing Hydrozoa (hydroids), Staurozoa (stalked jellyfish), Scyphozoa (true jellyfish), and Cubozoa (box jellyfish). All cnidarian classes possess the full complement of the intracellular BMP machinery, consisting of bona fide BMP ligands, receptors, and effectors [19,20,21,22,23,24,25]. Most of our knowledge about cnidarian BMP signaling to-date stems from Anthozoa, where it has been mainly studied in the sea anemone Nematostella vectensis and only in the context of axial patterning. Similar to Bilateria, Nematostella, and other Anthozoa form a bilaterally symmetric body plan with a secondary, “directive” body axis regulated by a BMP signaling gradient [26,27,28]. In the embryo, graded BMP signaling was shown to be responsible for the formation and patterning of the directive axis, requiring the interaction of multiple BMP signaling components. Knockdown of BMP signaling pathway members including bmp2/4, bmp5-8, gdf5-like (= gdf5-l), chordin, gremlin, and rgm resulted in defects or complete loss of the directive axis and, consequently, impaired patterning of the gastrodermal body layer^{Footnote 1} [26,27,28,29]. The dynamics of BMP signaling have been studied during early Nematostella development, using an antibody against the phosphorylated form of the transcriptional effector of BMP signaling, pSMAD1/5 [26,27,28]. In the Nematostella late gastrula, pSMAD1/5 forms a bilaterally symmetric activity gradient, which disappears in the late planula larva (4 dpf = days post-fertilization) once the directive axis is established and patterned [27] (Fig. 1C). In the 4 dpf Nematostella planula, BMP signaling activity becomes radially symmetric and localizes to the oral epidermis, as well as to the newly formed gastrodermal folds termed the mesenteries [27] (Fig. 1C). To this point, it remains undetermined if and to which extent BMP signaling is required in other developmental processes beyond larval development of Nematostella.

Unlike bilaterally symmetric Anthozoa, Medusozoa lack the directive body axis and feature a radially symmetric body plan.^{Footnote 2} Expression and transcriptomic data suggest that BMP signaling may be active in different medusozoan life stages, yet it is unknown which processes, if not the formation of a secondary axis, are regulated by BMP signaling in these animals. Previous analyses in several medusozoan species revealed generally broad expression profiles of BMP pathway genes across different body regions: The mRNA encoding the BMP ligand BMP5-8 has been found in the tentacle zone of the polyp of the scyphozoan Aurelia and the hydrozoans Hydra, Podocoryne, and Clytia, as well as the radial canals and sensory structures (rhopalia) in the Aurelia ephyra and in the medusae of Clytia and Podocoryne [23, 30,31,32]. In the hydroid Hydractinia, the expression of a putative bmp receptor was found in developing oocytes and in the male gonophore gastrodermis [33]. To our knowledge, in Medusozoa, the activity of BMP signaling has not yet been directly addressed by visualizing phosphorylated SMAD1/5 in any published work.

Given that BMP signaling remains active in the sea anemone after directive axis formation and that BMP components are present in medusozoan species lacking the secondary axis, we aimed to determine what roles other than directive axis patterning may be regulated by BMP signaling in Cnidaria. In this work, we analyzed the activity of BMP signaling under homeostatic conditions in several cnidarian species, with a special focus on the adult polyp of Nematostella. We analyzed the publicly available scRNA-seq data [18, 34, 35] and found a broad expression of BMP signaling components across many cell types. Then, using an antibody against the activated form of the BMP effector SMAD1/5, we generated a morphological atlas of pSMAD1/5 activity in the whole adult polyp of Nematostella, highlighting the surprisingly diverse activity domains across different body regions, especially in the inter-tentacle epidermis and specific portions of the mesentery gastrodermis (Fig. 1D). In the latter, we detected a population of pSMAD1/5-positive basiepithelial cells that, however, was distinct from previously described vasa2-positive basiepithelial stem-like cells. Together with a first characterization of BMP signaling in the jellyfish Aurelia and Tripedalia, we find evidence for the activity of BMP signaling in multiple cell types across different body regions, suggesting versatile roles of BMP signaling in Cnidaria.

Results

Single-cell RNA-seq data analysis suggests broad expression profiles of BMP pathway components across multiple cell populations in the adult Nematostella polyp

Thus far, expression studies of BMP pathway components and BMP signaling activity in Nematostella have been largely limited to early stages of development and the context of directive axis patterning [26,27,28,29, 36,37,38,39]. To get some information about where we can expect BMP signaling in the adult polyp, we interrogated the publicly available single-cell transcriptomic datasets for the expression of the BMP pathway genes [18, 34, 35]. Here, we focused on datasets of female and male polyps (“adult tissues”) with a coarse clustering approach recovering 13 distinct clusters based on the study by Cole et al. [18] (Fig. 1E). The Nematostella genome contains gene orthologues for four BMP ligands (bmp2/4, bmp5-8, gdf5-l and admp [19, 36, 38]), four BMP receptors (two type I receptors, alk2 and alk3/6 and two type II receptors, bmprII and actRII; Additional file 1:Fig. S1), and two BMP effectors (smad1/5 and smad4, as well as a “truncated” smad4-like [38]). In addition, several orthologues of potential BMP signaling antagonists are present, including transmembrane (rgm [28]), intracellular (smad6 [38]), and extracellular or secreted molecules (chordin, gremlinA, gremlinB, noggin1, noggin2, crossveinless-2, and follistatin [19, 37,38,39]; Additional file 1:Fig. S1).

BMP receptors and BMP effectors displayed low expression levels but with surprisingly broad expression profiles and transcripts detectable in multiple cell clusters. Intracellular components were present in mature cnidocytes (“mat.cnido”), “retractor muscle,” “gastrodermis,” epithelial ectoderm (“epithelia.ect”) and embryonic ectoderm (“ectoderm.embryonic”), and upregulated in putative stem cell (“pSC”), and neurosensory and neurosecretory (“neuronal”) cell types. In contrast, the expression of BMP ligands was spatially segregated, with bmp2/4, bmp5-8, and gdf5-l exhibiting similar expression profiles in the “gastrodermis,” “retractor muscle,” and “ectoderm.embryonic,” whereas admp expression was restricted to the “epidermis.” For the expression of BMP receptor types, no obvious cell type-specific segregation was apparent. Together, the expression of BMP pathway genes in the single-cell transcriptomic data suggests that BMP signaling is potentially active in various cell populations, involving few “BMP secreting” cell types that can signal to a wider range of “BMP receiving” cell types. BMP pathway genes display low expression levels and are limited to a low number of cells per cluster (transcripts expressed in < 16% of cells of each cluster), which might be due to diverging expression profiles within the different clusters. Indeed, expression across the 91 adult-restricted transcriptomic states shows that the expression of BMP pathway genes can vary between different sub-clusters in each coarse cluster (Additional file 1:Fig. S2).

BMP signaling forms distinct activity domains in the inner and outer epithelial layers of the adult polyp

To explore BMP signaling unrelated to axial patterning, we analyzed the activity of BMP signaling in the sexually mature polyp (Fig. 2A), utilizing the cross-reactive commercial antibody against the phosphorylated form of the BMP signaling effector SMAD1/5, which we have successfully used in Nematostella embryos previously [27]. In the adult, we detected distinct domains of pSMAD1/5 activity in both epithelial body layers (Fig. 1D), comprising the “outer” epidermis and the “inner” gastrodermis, which are separated by an extracellular matrix layer (mesoglea). Most of the epidermal epithelium was free of pSMAD1/5 activity, except for the oral domain (Fig. 2B, Additional file 1:Fig. S3). Here, BMP signaling was detectable in the inter-tentacular spaces (Fig. 2C–E’), as well as in longitudinal, epidermal stripes (Fig. 2I–L’). BMP signaling between the tentacles was present at each tentacle base with pSMAD1/5 activity mostly evident in the epidermis (Fig. 2C–E’), slightly reduced in the gastrodermis (Fig. 2D), and completely absent from other parts of the tentacles, except for the unspecific, non-nuclear staining of the capsules and threads of the tentacle-specific cnidocyte type—the spirocytes (Additional file 1: Fig. S4). In the heterozygous bmp2/4::mCherry transgenic animals, in which mCherry reporter is expressed under control of the 4.5-kb regulatory region upstream of the bmp2/4 start codon [40], strong mCherry expression marks one side of the pharynx forming the ciliated groove, the siphonoglyph (Fig. 2F–H,Additional file 1:Fig. S5). Previous work also showed that bmp2/4 mRNA is present in the siphonoglyph cells of the planula larva [36]; however, we did not detect above-background pSMAD1/5 staining in this area in the adult.

In the body wall epidermis, at the level of the pharynx, epidermal clusters of pSMAD1/5-positive cells assemble into eight longitudinal stripes (Fig. 2I–L’), correlating with the positions of each mesentery attachment site in the gastrodermis (Fig. 2J). Notably, discernible BMP signaling activity in the stripes was restricted to the pharyngeal and subpharyngeal region (Fig. 2K–K’), while absent from more aboral positions (Fig. 2L–L’). There, at the levels of the gonadal, trophic, and physa region, BMP antagonist gremlin was expressed in the epidermis vis-à-vis the mesentery attachment sites, and this area was free of pSMAD1/5 activity (Fig. 2M).

In contrast to the orally restricted BMP signaling in the epidermis, BMP signaling in the gastrodermis was evident along the entire primary, oral-aboral (OA) body axis (Fig. 3), and especially prominent in the mesenteries—the eight folds of the gastrodermal epithelium partitioning the gastric cavity (Figs. 1D, 3A), where pSMAD1/5 activity was highly pronounced. We did not observe any differences in pSMAD1/5 staining between different mesenteries, however, when comparing BMP signaling in the basal, medial, and apical portion of the mesentery, we observed differences in activity depending on the position along the OA body axis (Fig. 3A–D’). BMP signaling in the basal part of the mesentery (neuro-muscular mesentery, closest to the body wall) was the most consistent along the OA axis, with continuous and pronounced signaling activity from pharynx to physa (Fig. 3A–C’, blue solid arrowheads). The apical domain of the mesentery (septal filament, cnidoglandular tract) displayed little nuclear pSMAD1/5, but often showed unspecific cytoplasmic or vesicular signal. Unspecific signal was repeatedly observed in putative muco-glandular cells that are most abundant in the trophic region (Fig. 3C–C’, white arrowheads in the sf region; Additional file 1:Fig. S4). Unlike in the basal and apical domain, there were distinct differences in pSMAD1/5 staining in the medial mesentery portion, changing according to the body region along the primary body axis (Fig. 3A–D’, mm region). While BMP signaling was broadly active in the medial mesentery of the pharynx and upper subpharynx region (Fig. 3B–B’, solid blue arrowheads in the mm region), pSMAD1/5 was reduced or absent from the medial gastrodermis at lower positions (lower subpharynx region, gonad region, and trophic region) (Fig. 3C–D’, hollow white arrowheads in the mm region).

Gastrodermal BMP signaling occurs in distinct domains along the basal–apical axis of the mesentery

To get a more detailed picture of BMP signaling in different parts of the mesentery, we analyzed the pSMAD1/5 activity in the individual mesentery domains, starting at the body wall and the basal part of the mesentery (neuro-muscular domain, Fig. 4), moving along the medial mesentery domain and into the septal filament at the mesentery tip pointing towards the center of the gastric cavity (Fig. 5). In the mesenteries, we could differentiate between (1) pSMAD1/5-positive epithelial cell populations in different domains of the mesentery and (2) pSMAD1/5-positive interstitial or basiepithelial cells, located in the extracellular matrix (mesoglea).

The body wall gastrodermis between the mesenteries was generally free of BMP signaling (Fig. 4B–B’), although individual pSMAD1/5-positive cells were occasionally observed. At the base of each mesentery, BMP signaling was active in two epithelial cell clusters located on both sides of the parietal muscle (Fig. 4C–C’, blue arrowheads), while other cells of the parietal muscle region were pSMAD1/5-negative. The adjoining mesentery stalk contained pSMAD1/5-positive and -negative epithelial cells, distributed throughout the stalk (Fig. 4D–D’). The gastrodermal epithelium of the retractor muscle region displayed pronounced pSMAD1/5 activity, also with intermixed communities of pSMAD1/5-positive and pSMAD1/5-negative cells (Fig. 4E–G’).

Aside from epithelial cell populations, we also observed interstitial, basiepithelial cells that were present in the mesoglea of the mesentery and the body wall, across different locations of the body column, as previously described by others [41, 42]. Notably, many basiepithelial cells located in the parietal muscle region, along the mesentery stalk and in the retractor muscle region were pSMAD1/5-positive (Fig. 4C–D’, G–G’, yellow arrowheads), although we observed basiepithelial cells without BMP signaling activity in the same region as well. Strikingly, BMP signaling activity in these cells appeared to be restricted to the neuromuscular region of the mesentery but was absent from other portions of the mesentery or other locations of the body column. This indicates that there are different subpopulations of basiepithelial cells present in the mesoglea that can be either pSMAD1/5-positive or -negative, and/or that BMP signaling in these cells is only active in a specific context, which may possibly be linked to their differentiation choices.

In the medial and apical mesentery (Figs. 5A,Additional file 1:Fig. S6), activity domains of BMP signaling were less pronounced compared to the basal mesentery (Fig. 5B–F’). In the medial domain, BMP signaling was mostly absent (Fig. 5B, D’–E’), except at the levels of the pharynx and subpharynx (Fig. 3A–A’), as described above, and in a narrow stripe of cells along the periphery of the medial domain at its border to the reticular tract (rt) in the gonadal region or to the cnidoglandular tract (cgt) in the trophic region (Fig. 5E’). Other exceptions, observed in the somatic gonad, are outlined in a separate section below. The cnidoglandular tract was mostly devoid of BMP signaling; however, sporadically, pSMAD1/5-positive cells were also observed in different locations of the septal filament (Fig. 5F’, Additional file 1:Fig. S6).

Overall, we observed intermixed communities of pSMAD1/5-positive and -negative cells, with epithelial or basiepithelial localization, that form multiple BMP signaling domains in the mesentery. pSMAD1/5 activity domains of the basal, neuro-muscular mesentery are highly pronounced, compared to the more reduced or completely absent activity in the medial and apical mesentery. This heterogeneity in composition and location of pSMAD1/5 domains suggest that BMP signaling activity is not limited to a single BMP-responsive cell population but rather can be active in multiple cell types or cell states, which is supported by the broad expression of intracellular BMP components in the single-cell transcriptomic data.

Medial and apical mesentery domains display reduced pSMAD1/5 activity but increased BMP expression

Our characterization of BMP signaling domains in the mesentery revealed the intriguing reduction of signaling activity in the medial and apical mesentery including the gonad region (Fig. 6A,B) that we aimed to address in more detail (Fig. 6,Additional file 1:Fig. S7). First, we analyzed the expression of BMP receptors using colorimetric in situ hybridization, and detected alk2, alk3/6 and actrII but no bmprII mRNA in parts of the septal filament, with an upregulated expression in the reticular and ciliated tracts in the gonad region of the polyp (Fig. 6C–F). We also detected BMP signaling activity there, although pSMAD1/5 staining appeared weaker than in the areas described in the previous chapters. In the septal filament, BMP signaling was active in the epithelial cells of the reticular tract (Fig. 6G–G’, blue solid arrowheads), while it was absent from basiepithelial cells. Similarly, low levels of BMP signaling were sometimes detectable in the ciliated tract of the gonad region (Fig. 6H–I’), but often completely absent (Fig. 6G–G’, white hollow arrowheads), and BMP signaling activity was never observed in the ciliated tract of the subpharynx region. Active BMP signaling in the ciliated tract seemed to be localized to the cells at the top rather than at the bottom of the folds (Fig. 6I–I’). It remains unclear if these differences in the detection and localization of pSMAD1/5 signals reflect dynamic changes of BMP signaling in the ciliated tract or they are due to technical limitations of the antibody staining.

In line with the expression of the BMP receptors and pSMAD1/5 immunoreactivity (Fig. 6C-F; H-I’), in situ hybridization analysis also showed the expression of BMP ligand genes in the septal filament. We found bmp2/4 (Fig. 6J–P) and bmp5-8 (Fig. 6Q–T) expression in the ciliated tract, curiously, at the bottom of the folds, i.e., in a domain complementary to the areas of pSMAD1/5 activity at the top of the ciliated track folds. Unlike bmp2/4, in addition to the ciliated tract expression, bmp5-8 is also strongly expressed in the reticular tract of the septal filament (Fig. 6Q-S). In agreement with in situ hybridization data, transgenic adult polyps of the bmp2/4::mCh reporter line displayed mCherry expression in the mesenterial gastrodermis and at the basis of the ciliated tracts, both in the subpharynx and the gonadal region (Fig. 6M–P,Additional file 1:Fig. S5).

In the gonads (Fig. 7A,B), transcripts of BMP ligands bmp2/4, bmp5-8, and gdf5-like, and BMP receptors alk2, alk3/6, actrII, and bmprII were detectable in the gastrodermis of the somatic gonad (Fig. 7C–I’), and several of the BMP signaling components were deposited in maturing gametes. In female polyps, in situ hybridization suggested maternal deposition of mRNAs of bmp5-8, alk2, alk3/6, actrII, and bmprII but not of bmp2/4 or gdf5-like (Fig. 7C–I). This was further supported by the RNA-Seq data from the NvERTx database [43,44,45], detecting maternal transcripts of bmp5-8, alk2, alk3/6, actrII, and bmprII in unfertilized Nematostella eggs. According to the NvERTx, low bmp2/4 and gdf5-like appear to be also provided maternally, however at much lower levels than bmp5-8 and undetectable by in situ hybridization (Additional file 1:Fig. S8). In situ hybridization analysis of the male polyps showed expression of bmp2/4, bmp5-8, gdf5-l, alk2, alk3/6, and actrII in the somatic gonad, but no detectable staining of BMP RNAs in the germ cells. In contrast, mRNA of the BMP receptor alk2 appears to accumulate in the periphery of the spermaries (i.e., in spermatogonia) (Fig. 7C’–I’).

BMP signaling is absent from vasa-positive germ cells and stem cells

Single-cell transcriptomic analyses suggested the upregulation of BMP pathway genes in the previously identified stem cell cluster [18]. While the Nematostella stem cell system is still poorly characterized, vasa2 protein was reported to not only mark primordial germ cells, maturing oocytes, and the spermatogonia [46,47,48] (see Fig. 8A,B) but also a recently described population of putative stem cells in the mesentery contributing to germline and somatic lineages [41]. To determine if BMP signaling is active in any of these populations in the mesentery, we performed immunostaining against pSMAD1/5 in combination with the antibody staining against Nematostella vasa2 [48]. In agreement with previous studies, we detected vasa2-positive basiepithelial cells in the reticular and gonad tracts in the subpharyngeal (Fig. 8C–C’) and gonadal region (Fig. 8D–D’), yet there appears to be no overlap with pSMAD1/5 activity (Fig. 8C”, 8D”). In rare instances, we detected pSMAD1/5-positive basiepithelial cells located in the gonad region (Fig. 8D”, blue arrowhead) that were, however, distinct from vasa-positive cells (Fig. 8D”, white hollow arrowhead). In line with what has previously been described by us and others, vasa2 protein was present in maturing oocytes and basiepithelial germ cells in the female gonad, and in the outermost region of the spermaries populated by the spermatogonia [41, 46, 48, 49], yet gametes and putative germ cells displayed no pSMAD1/5 immunoreactivity (Fig. 8E–F”). Occasionally, we observed pSMAD1/5-positive cells that were negative for vasa2 protein in close proximity to the vasa-positive spermatogonia (Fig. 8F”-F”’).

In summary, we did not observe an overlap between vasa2-positive basiepithelial cells and pSMAD1/5-positive basiepithelial cells. Strikingly, BMP signaling is active in basiepithelial cells located in the basal mesentery region. These cells are vasa2-negative; however, it is conceivable that they may represent a progeny of vasa2-positive stem cells and cluster together with the stem cells in the single-cell transcriptomic data.

BMP signaling in the gonad region is active in female trophonemata and male ciliated plugs

Another exception to the low BMP signaling activity of the medial mesentery we discovered in the accessory cells of the somatic gonad. These accessory cells are the trophonema cells specific for the female, and ciliated plug cells specific for the male [49]. They form small islands in the somatic gonad gastrodermis in close proximity to maturing gametes (Fig. 9A,B). In the accessory cells of both, male and female polyps, we detected low levels of pSMAD1/5 activity (Fig. 9C–F). In our in situ hybridization analysis of the somatic gonad, we could not observe detectable levels of BMP receptor expression there; however, trophonema and ciliated plug cells were positive for the BMP ligand gdf5-l and the BMP antagonist chordin RNA (Fig. 9G-L).

While accessory cells have been addressed in histological and ultrastructure studies [51, 52], they could not be clearly assigned to any cell cluster by single-cell transcriptomic analyses. Since no distinct molecular signatures are known for the trophonemata and ciliated plug cells, we aimed to further analyze these cell populations. Trophonemata and ciliated plugs are apparent due to their tight aggregation and apical constriction of the cells in them compared to the circumjacent gastrodermal epithelium easily discernible in classical cadherin and fibrillar actin stainings (Fig. 9M–O’). Nematostella has two classical cadherins: cadherin1 (cdh1) and cadherin3 (cdh3). cdh1 is expressed in the gastrodermis of the mesenteries and in both layers of the body wall, but it is depleted from the epidermis of the tentacles; cdh3 is expressed in the epidermis of the body wall and, especially strongly, in the epidermis of the tentacles, as well as in the pharynx-derived tissue including the septal filament [53] (Additional file 1:Fig. S9). Surprisingly, in the otherwise cdh1-positive/cdh3-negative epithelium of the gonad, cdh3 was clearly detectable in the contacts between the apically constricted cells of the trophonemata and ciliated plugs (Fig. 9M–N”). Side views of the trophonema (Fig. 9N–N”) showed colocalization of cdh3 and cdh1 at the apical side of the cells facing the gastric cavity, while only cdh1 was present at the basal trophonema side facing the animal pole of the oocyte, where the germinal vesicle is located (Fig. 9N–N”). In line with the differential expression of cadherins, we observed the downregulation of mCherry expression in the trophonemata in the snailA::mCherry reporter line compared to the rest of the gonad gastrodermis (Fig. 9O–O’), suggesting the repression of cdh3 in the mesentery gastrodermis by SnailA, similar to the situation reported in the embryo [53]. Moreover, recently, trophocytes have been shown to express the RNA-binding protein Nanos2 [49], which we have recently identified as a direct target of pSMAD1/5 during early development [27]. In line with that, we found pSMAD1/5 activity overlapping with the mCherry expression in the nanos2::mCherry transgenic reporter line (Fig. 9P–P”).

The medusozoan species Aurelia coerulea and Tripedalia cystophora display broad BMP signaling activity across different body regions

The expression of several BMP components, but mostly of BMP ligands, has been previously analyzed in several members of Medusozoa, including Hydra, Clytia, Podocoryne, and Aurelia [30,31,32, 54], yet nothing is known about the activity and localization of BMP signaling in jellyfish and hydroids. Therefore, we aimed to test the cross-reactivity of the pSMAD1/5 antibody in multiple medusozoan representatives. We failed to detect specific anti-pSMAD1/5 staining in any of the assayed hydrozoan species including Clytia hemisphaerica, Hydra vulgaris, Hydractinia sp., and Ectopleura sp. and a scyphozoan species, Sanderia malayensis (Additional file 1:Fig. S10). In contrast, we successfully detected BMP signaling by pSMAD1/5 staining in the scyphozoan jellyfish Aurelia coerulea and the cubozoan jellyfish Tripedalia cystophora. Both species feature a triphasic life cycle common for most Medusozoa, with a free-swimming larva, a sessile polyp, and a free-swimming medusa stage (Fig. 10A, E).

In Aurelia, we focused on the developmental stages that were most accessible for whole-mount immunofluorescence staining: the polyp stage and early juvenile medusa stages (ephyra and metaephyra). Aurelia polyp (scyphistoma) undergoes serial transverse fission and releases multiple genetically identical juvenile jellyfish (ephyrae), which develop into the metaephyra and, subsequently, into adult medusae (Fig. 10A). No active BMP signaling was observed in the unsegmented Aurelia polyp (Additional file 1:Fig. S10), while at the ephyra stage, BMP signaling activity was present across different portions of the body. Here, mixed populations of pSMAD1/5-positive and -negative cells were found in the epithelium of the central disc, the lappets, and the sensory organs called rhopalia (Fig. 10B,C). In contrast, the manubrium appeared to be free of BMP signaling, although we observed a strong background signal and unspecific (non-nuclear) staining of putative muco-glandular cells (Fig. 10B), similar to the unspecific staining observed in the Nematostella pharynx and septal filament. At the metaephyra stage, which is characterized by the stepwise fusion of the individual lappets into a bell-shaped umbrella, BMP signaling was observed in similar areas as in the ephyra. In addition, pSMAD1/5-positive cells were found at the base of the gastric filaments (Fig. 10D), which are located in four distinct gastric pouches around the manubrium. In contrast, gastric filaments, which are very rich in gland cells, showed only unspecific (non-nuclear) staining. Later during development, the gastric pouches also contain four horseshoe-shaped gonads sitting underneath the gastric filaments. Co-immunostaining with anti-pSMAD1/5 and anti-Nematostella vasa2 antibody revealed vasa-positive cells with a typical perinuclear vasa signal at the base of the gastric filaments that were distinct from the pSMAD1/5-positive cells located immediately next to the vasa expressing cells—similar to the situation in Nematostella (Fig. 10D’).

In Tripedalia, antibody staining revealed a broad pSMAD1/5 activity in juvenile and adult jellyfish. Mixed populations of pSMAD1/5-positive and -negative cells were detectable in the umbrella, velarium, ring nerve, parts of the nematocyst warts, and rhopalia of the jellyfish, while the activity was absent from the manubrium, pedalia, and tentacles (Fig. 10F). In sexually mature females, four pairs of wing-shaped gonads are located on the inside of the medusa bell next to the radial canals of the gastrovascular system. Strikingly, germ cells in the female gonad of Tripedalia were pSMAD1/5-positive (Fig. 10G–H’), and co-staining with anti-Nematostella vasa2 antibody showed an overlap with the accumulation of vasa protein (Fig. 10H–H”), which was never observed for germ cells in Nematostella.

Discussion

Diversity of BMP signaling domains in Cnidaria indicates a variety of functions

To date, most of our knowledge on BMP signaling in Cnidaria comes from expression-based studies in Medusozoa, as well as expression-based and functional studies in anthozoan embryos [26,27,28,29,30,31,32, 36,37,38,39, 55]. Research in anthozoans, however, only addressed the role of BMP signaling in directive axis establishment and patterning. In this work, we aimed to expand this understanding by investigating the activity of BMP signaling unrelated to axial patterning in the adult polyp of the anthozoan Nematostella, and two radially symmetric medusozoans, the juvenile ephyra of the scyphozoan Aurelia and the cubozoan jellyfish Tripedalia, utilizing an antibody against the BMP effector pSMAD1/5 [26,27,28]. In all three species, BMP signaling was active in distinct cell populations in different locations, suggesting that BMP signaling must be involved in the regulation of multiple processes.

We centered our analysis on Nematostella and systematically dissected BMP activity domains in this bilaterally symmetric model sea anemone. We showed that BMP signaling was strongest in distinct areas of the epidermis of the oral region, as well as in specific cell populations in the gastrodermis of the mesenteric folds (Fig. 11A–D). In the mesentery, pronounced BMP activity was detected in the basally located neuro-muscular region, while it appeared reduced, although not absent, in the more apical parts including the gonad region. In line with that, our analysis of the published single-cell transcriptomic data indicates broad expression of the BMP pathway genes [18]. The diversification of BMP activity in multiple specialized cell types may already start during larval development, after patterning of the directive axis is completed, as indicated by our previous anti-pSMAD1/5 ChIP analysis. We identified direct BMP signaling target genes at late gastrula, when the axis patterning gradient of BMP signaling activity has just formed, and in 4d planula, when the axis is already patterned, and the pSMAD1/5 gradient is no longer present. While at both stages, many BMP targets are developmental regulators involved in the patterning of the directive axis, in 4d planula we observed an increasing number of targets with other putative functions such as metabolism, extracellular matrix dynamics, or neurogenesis [27]. In the future, it will be important to follow the dynamics of BMP signaling between 4d planula and adult and figure out how adult BMP signaling domains appear.

In medusozoans, BMP activity and the expression of BMP signaling components appear to be similarly versatile. Our study revealed pSMAD1/5 being broadly active in the epithelium of the central disc, radial canal and lappets, as well as in the rhopalia of Aurelia and Tripedalia. In line with that, earlier studies in the ephyra of Aurelia and medusa of Clytia and Podocoryne detected bmp5/8 expression in the radial canal and sensory structures (rhopalia) [30]. In Hydra, HySmad1 mRNA is absent from the tentacles and the peduncle but broadly detectable along the entire body column, where it is expressed in a variety of cell types including epithelial cells, neurons, gland cells, interstitial stem cells, and endodermal epithelium, as determined by northern blot on mRNA from fractionated cells [54]. In contrast, a bmp2/4-like gene and three copies of bmp5-8 are expressed at the tentacle base, as well as in the body column [32, 55,56,57,58,59].

Together, our data on BMP activity domains in three cnidarian representatives, combined with existing cnidarian expression data, indicate that BMP signaling may be relevant in at least three different contexts, besides axial patterning: (1) in tentacle formation, (2) in the neuro-muscular region, and (3) in the reproductive region.

Potential role of BMP signaling in tentacle formation

Among the most pronounced domains of BMP signaling in Nematostella polyp was the pSMAD1/5 activity between the tentacles (Fig. 11B, D). A similar pattern of BMP activity can be observed during postembryonic development, when BMP signaling forms a cross-like activity domain in the oral epidermis [27] (Additional file 1:Fig. S3), with four pSMAD1/5-free territories at the localization of the four primary tentacle buds. Starting from the 4d planula, pSMAD1/5 activity is also accompanied by the expression of several BMP components and BMP target genes in the tentacle area. The pSMAD1/5-negative tentacle buds express genes encoding BMP2/4, BMP5-8 (Additional file 1:Fig. S3B-C), BMP antagonists Noggin and Gremlin [37], a metalloprotease Tld [38] and a BMP and BMP receptor binding molecule RGM [28] (Additional file 1:Fig. S3D-D’). In contrast, the pSMAD1/5-positive intertentacular area expresses genes encoding a BMP-binding protein CV2 [60] (Additional file 1:Fig. S3E-E’) and another BMP ligand, ADMP (Additional file 1:Fig. S3F-F’). Single-cell transcriptomic data indicates that admp is the only BMP gene, whose expression persists in the epidermis of the adult polyp making it the prime candidate for signaling in the intertentacular domain at this stage.

Intertentacular pSMAD1/5 suggested potential role of BMP signaling in preventing tentacle formation, and this notion is supported by the phenotype observed in bmp2/4 mutants we generated using CRISPR/Cas9-mediated gene editing. While most injected F0 embryos died after developing a “worm phenotype” described previously for the bmp2/4 morphants [28], some embryos survived until adulthood. Many of these adults formed large bubble-like structures on the sides of their bodies. Genotyping showed that in the F0 animals, the bubbles contained bmp2/4 mutant cells, and once we obtained heterozygous bmp2/4^± F1, these animals also started sporadically developing bubbles (Additional file 1:Fig. S11B-C). Morphological analysis confirmed the presence of epidermal muscles and spirocytes in the bubble tissue (Additional file 1:Fig. S11D-E’), both of which are unique features of a normal Nematostella tentacle. Moreover, qPCR analyses showed that cdh3 (a cadherin specifically upregulated in the tentacles [53]) and nem64 (a tentacle muscle-specific bHLH transcription factor [34]) are upregulated in the bubble of the F0 mutants 7.7-fold and 7.6-fold, respectively, while cdh1 (a cadherin specifically upregulated in the body wall and downregulated in the tentacle [53]) is 2.6-fold downregulated (Additional file 1:Fig. S11F-G). Thus, the bubbles developing in the mosaic and heterozygous bmp2/4 mutants appear to be products of an ectopically and improperly activated tentacle formation program.

Expression data suggest the association of BMPs with the tentacles in several Medusozoa. In the polyps of the hydrozoans Hydra, Podocoryne, and Clytia, as well as in the polyp of the scyphozoan Aurelia, BMP ligand bmp5-8 is expressed in developing tentacle buds [30,31,32]. In the case of Hydra, several other BMPs, including bmp2/4-like and bmp5-8c, also show increased expression at the tentacle base in a homeostatic polyp, and all bmp genes react dynamically to budding and/or head regeneration [32, 58].

Taken together, we see evidence for the involvement of BMP signaling in regulating patterning and morphogenesis of the tentacle domain in Nematostella as well as in several medusozoans. However, the exact role of BMP signaling in this and the way BMP signaling acts in concert with other pathways, e.g., with FGF and Hedgehog signaling in Nematostella [46, 61], remain to be discovered.

Potential role of BMP signaling in neurons, muscles, and cnidocytes

Our analysis of the single-cell transcriptomic data shows upregulation of BMP pathway genes in Nematostella neurons, muscles, and cnidocytes. Neurons and cnidocytes are generated by a common pool of SoxB2-positive multipotent progenitors during embryonic, larval, and post-larval development [35, 62], while muscles originate from other cells. Our data in the adult polyp demonstrate that domains of pronounced BMP signaling activity coincide with the basal part of the mesentery (Fig. 11C,D) and include two of the three neuron-rich areas of the gastrodermis: the longitudinal nerve tracts at base of the mesentery located next to the parietal muscle, and the intramesenterial neurons, some of which are embedded into the retractor muscle, but not the neurons of the body wall and the circular muscles.

ChIP-seq with an anti-pSMAD1/5 antibody on late gastrula and 4d planula of Nematostella revealed multiple genes as BMP targets including arp6, ashB, atoh7, hmx3, isl, irx, lmx, soxC, rgm, ephrinB, and netrin, which are associated with neural function in vertebrates and invertebrates [27]. In contrast, we do not find many muscle-related genes among the direct BMP signaling targets, which suggests mesenterial neurons rather than parietal or retractor muscle cells as more likely candidates for being pSMAD1/5-positive. Despite being vital during formation and patterning of the centralized nervous system in many Bilateria, the roles of BMP signaling in the formation of the diffuse nerve net common in Cnidaria are largely unknown. In Nematostella, early neurogenesis is considered to be independent of BMP signaling [63, 64], with pSMAD1/5 becoming detectable in the oral domain during gastrulation, which is after the onset of the neural gene expression at the blastula stage and in a different location in comparison to where the first RF-amide-positive and elav-positive neurons are found in the gastrula [27, 62, 65]. In line with that, knockdown and upregulation of BMP signaling showed no effects on early neurogenesis; however, an anti- as well as a pro-neurogenic potency of BMP signaling has been observed during oral nerve net formation in the planula larvae [64].

While some neurons appear to be pSMAD1/5-positive, the situation with cnidocytes is different. Transcripts of the BMP pathway genes are expressed in mature cnidocytes in the single-cell data (Fig. 1E); however, we did not observe nuclear pSMAD1/5 in any stinging cell subtype, besides the unspecific, non-nuclear signal in spirocytes (Additional file 1:Fig. S4A–A’). Notably, the cnidocyte-specific expression of some of the BMP components may be shared with other cnidarians, given the fact that HySmad1 could be detected in mature nematocytes of the gastric region in Hydra [54]. Co-staining of pSMAD1/5 in different existing transgenic reporter lines [35, 62, 66] or together with antibody staining for neuronal, muscle, and nematocyte markers will help elucidate the role of BMP signals in the development, regional patterning, and maintenance of neurons, cnidocytes, and muscles in Nematostella and, eventually, other cnidarians.

Potential role of BMP signaling in the reproductive region

While we have a comprehensive understanding of the stem cell system in hydrozoan species [57, 67,68,69], most efforts to characterize stem cells in non-hydrozoan cnidarians remained unsuccessful. Ultrastructural studies in Anthozoa and Scyphozoa revealed extraepithelial, amoeboid cells, which were proposed to be stem cells [42, 70] but lack further characterization. A recent study in Nematostella provided evidence for a vasa2/piwi1 double-positive basiepithelial putative stem cell population in the mesentery in the basal portion of the septal filament [41]. As these adult stem cells contribute to vasa2-positive primordial germ cells (PGCs) as well as to the soxB2-positive neuronal progenitors, they were proposed to be homologous to the interstitial stem cells (i-cells) of Hydrozoa [41].

Our analysis of Nematostella scRNA-seq data indicated the conspicuous upregulation of BMP pathway members in the previously identified stem cell cluster [18]. In accordance with the observations by Miramon et al. [41], we detected vasa2-positive basiepithelial cells in the septal filament (Fig. 8C,D); however, these cells were pSMAD1/5-negative. Instead, we found an upregulated expression of BMP ligands bmp2/4 and bmp5-8 as well as pSMAD1/5-positive cells in the epithelial parts of the septal filament, where vasa-positive basiepithelial cells concentrate. Curiously, pSMAD1/5 is detectable in vasa2-negative basiepithelial cells located in the neuromuscular domain of the mesentery. Considering that single-cell transcriptomic data indicated the upregulation of BMP pathway genes in putative stem cells, these findings suggest that either vasa2/piwi1-positive adult stem cells express BMP components but do not receive BMP signals or that there are other, vasa-negative stem cell populations that may be pSMAD1/5-positive.

Curiously, expression data of BMP components in Hydrozoa also implies possible involvement of BMP signaling in interstitial stem cell and germ cell development. Hydrozoan i-cells constitute a pool of pluri- or multipotent stem cells that contributes to somatic cells and the germline [67, 69, 71,72,73]. In Hydra, Northern hybridization of smad1/5 indicated that it is upregulated in i-cells [54], while transcriptomic analysis showed the upregulation of the BMP receptor actr1 in multipotent interstitial cells and smad1/5 and BMP antagonist dand in germline stem cells [74]. In Hydractinia, the expression of a putative bmp receptor has been detected in developing oocytes and the male gonophore gastrodermis [33]. bmp5-8 has been also found to be differentially regulated in female and male tfap2-positive germ cells in Hydractinia [75]. Another TGFβ molecule, gonadless (Gls), which may or may not be a BMP, is expressed in tfap2-positive germ cells, and gls knockout does not affect the germ cells but results in the absence of the sexual zooids carrying gonads in Hydractinia [76].

All these observations in hydrozoans suggest a role of BMP signaling in gonado- or gametogenesis. This role appears to be highly conserved across Bilateria [10], and our analysis suggests that BMP signaling related to reproductive processes might be conserved in Nematostella, Tripedalia, and Aurelia. Single-cell transcriptomic data from the adult Nematostella polyp suggest that BMP genes are not expressed in primordial germ cell (PGC) clusters, and we never observed pSMAD1/5 activity in the vasa2-positive stem and germ cell lineage, including maturing gametes (Fig. 8C,D); however, RNA of several BMP signaling components was detected in the developing oocytes (bmp5-8 and alk6) and spermatogonia (alk2). Similar to Nematostella, the juvenile Aurelia metaephyra displayed pSMAD1/5-positive cells in proximity to the developing reproductive region in the gastric pouches, but without showing an overlap with vasa-positive cells. Notably different is the situation in Tripedalia, where BMP signaling was specifically active in vasa-positive germline cells in the gonad.

One interesting cell population in this context is the cnidarian-specific accessory cells described in Anthozoa and Scyphozoa [49, 51, 52, 77,78,79,80,81,82]. Nematostella accessory cells, termed trophocytes in the female and ciliated plug cells in the male polyp, are located in the somatic gonad in a close contact with maturing oocytes and the spermatogonia (Fig. 9C–F). Little is known about their function; however, unlike previously suggested, trophonemata appear not to be involved in nutritional transport [79] but may play a role during germ cell development [49] or, given the differential cadherin expression we found, serve as release points for the gametes. While the somatic gonad gastrodermis is generally pSMAD1/5-negative, we detected active BMP signaling in accessory cells of the female and male polyp, also expressing the BMP ligand gdf5-l and the BMP antagonist chordin. Accessory cells display an upregulation of RNA-binding zinc finger molecule nanos2, a gene expressed in the piwi1/vasa2/nanos2-positive multipotent stem cell population and in a broad progeny of the neuroglandular lineage [49]. Furthermore, nanos2^−/− knockouts not only lack accessory cells but also fail to establish the germline [49]. While we cannot comment on the regulatory relationship of BMP signaling and nanos2 expression in the adult polyp besides them being active together in accessory cells, nanos2 is a direct target gene of pSMAD1/5 during larval development [27].

Conclusions

In this study, we provide a whole-body atlas of BMP signaling activity in the adult bilaterally symmetric cnidarian model Nematostella and a brief overview of BMP signaling activity in two radially symmetric cnidarian models, Aurelia and Tripedalia. Currently, we are unable to state with certainty which cell types/states from single-cell transcriptomic data are the ones receiving BMP signals in the adult Nematostella polyp; however, in the future, as more and more adult cell states become “de-orphanized,” this question will no doubt be resolved. The same approach will then have to be expanded to our two medusozoan models, Aurelia and Tripedalia. Our analyses show that BMP signaling is involved in a variety of processes beyond axial patterning, which explains the presence of the BMP signaling components in non-bilateral Cnidaria. Among these processes, the role of BMP signaling in regulating tentacle formation, neuronal differentiation, and gameto- or gonadogenesis clearly stand out and will form focal points of the future studies.

Methods

Animal culture

Nematostella vectensis adult polyps, separated by sexes, were maintained in the dark at 18 °C in 16 ppm artificial seawater (Nematostella medium, NM) and spawned as described before [83, 84]. Aurelia polyps (scyphistomae) were cultured in 35‰ ASW (artificial sea water) in small Petri dishes at 20 °C. Strobilation was induced by temperature change (polyps were transferred to ~ 15 °C for several hours and then again kept at 20°). Clytia polyp colonies and jellyfish were maintained in 35‰ ASW in kreisel tanks. Tripedalia jellyfish and Sanderia ephyra were obtained from the Tiergarten Schönbrunn, Vienna, Hydractinia from the Haus des Meeres, Vienna, and directly processed for immunostaining. Ectopleura polyps were collected in Grado, Italy, and processed for immunostaining in Vienna.

Nematostella bmp2/4 mutagenesis

To generate bmp2/4 mutants, the genomic sequence encoding for the N-terminus of the mature BMP2/4 ligand was targeted using the ALT-R CRISPR-Cas9 system (IDT). crRNAs targeting two (overlapping) sequences were ordered: AAACGGAGCCTGCGGTCCGGCGG and AGAAAACGGAGCCTGCGGTCCGG, and both proved to work efficiently. Genotyping was performed using the primers GCAGACAACGATGGGATTGACGCTAGTG and GTATTGCCCGTTCTAATCATCCTTGAAGGC.

Quantitative PCR on wild type and bubble tissue cDNA was performed with the following primers: for cadherin3 (NV2.4928 or NVE8896), cdh3_qF GCTACATCTTCCGTCTCGGG and cdh3_qR TAGTTGCCCTCACGGTTCAC; for cadherin1 (NV2.4715 or NVE818), cdh1_qF AGTCGTGAATGCCCCAGAAG and cdh1_qR ACACGCAAATGCTGTCCCTA; for nem64 (NV2.10867 or NVE19721), nem64_qF TGACGCCTTCTCCTGAACCA, and nem64_qR CGTTCTCGTGCGTTCGCTAA. Primers against GAPDH were used as normalization control. Boxplots were created using the boxplot web tool (http://shiny.chemgrid.org/boxplotr/).

Fixation for immunohistochemistry

Adult Nematostella were starved for up to 3 days and left in 0.1 M MgCl₂ in NM until fully relaxed (5–15 min). For fixation, polyps were transferred to a new dish containing ice-cold fixation solution 1 (3.7% Formaldehyde in PBS). For antibody staining of pSMAD1/5, fixation solution 1 (3.7% Formaldehyde, 1 × PBS, 0.2% Triton-X100) was used. For antibody staining of pSMAD1/5, vasa2, cdh1, and cdh3, fixation solution 2 (3.7% Formaldehyde, 0.4% glyoxal, 0.1% MeOH, 1 × PBS, 0.2% Triton-X100) was used. To ensure consistent penetration, the fixation solution was injected slowly through the mouth opening, using a 1-ml syringe and a Sterican blunt needle with a bent tip. Polyps were transferred to a new tube with fresh, ice-cold fixation solution and fixed for 60 min under rotation at 4 °C. After 20–30 min fixation time, samples were transferred to a dish with fresh fixation solution for further dissection. For vibratome sectioning, samples were cut into ~ 5-mm pieces transversely through the body column. For whole mount imaging, the body column was opened longitudinally, and mesenteries were extracted from the body wall. After dissection, the fixation was continued under rotation at 4 °C. The fixative was removed in 3 washes with 1 × PBS / 0.2% Triton-X100. To remove pigments, samples were repeatedly washed with 100% methanol until the liquid remained clear. Samples were washed 3 × with 1xPBS, 0.2% Triton-X100, and directly further processed. For anti-pSMAD1/5 staining, samples were incubated in Bloxall™Blocking Solution (Vector Laboratories) before the antibody blocking step (see “Immunohistochemistry”) to reduce unspecific signal. Samples were incubated with Bloxall for 10–30 min at room temperature, followed by 3 × washes with 1 × PBS.

Ephyra of Aurelia and Sanderia, polyps of Aurelia, Hydractinia, Ectopleura, and jellyfish of Clytia and Tripedalia were relaxed in 0.1 M MgCl₂. Hydra polyps were relaxed in 2% urethane for 2 min. Samples were fixed using 3.7% Formaldehyde in 1 × PBS and 0.2% Triton-X100 for 90 min (except Tripedalia with 60 min of fixation) at 4 °C under rotation. After fixation, samples were washed 3 times with 1 × PBS / 0.2% Triton-X100.

Fixation for in situ hybridization

Sample preparation for in situ hybridization of mesentery pieces or pieces of the body column was carried out as described before (see “Fixation for immunohistochemistry”). Sample preparation and fixation was performed at room temperature using fixation solution 1 (3.7% FA, 1 × PBS, 0.2% Triton-X100), fixation was carried out on a rotator for 1 h at room temperature. ISH samples were stored in 100% MeOH at − 20 °C.

Immunohistochemistry

The same antibody staining protocol was used for all cnidarian species. Samples were blocked in Blocking solution I (1% BSA, 5% sheep serum, 1 × PBS, 0.2% Triton-X100, 20% DMSO) for at least 2 h at room temperature. In parallel, the primary antibodies were pre-absorbed in Blocking solution II (1% BSA, 5% sheep serum, 1 × PBS, 0.2% Triton-X100, 0.1% DMSO). Primary antibody incubation of the sample was carried out overnight at 4 °C with the following primary antibody dilutions: rabbit anti-pSMAD1/5/9 1:200 (mAb #13,820, Cell signaling, [26,27,28], mouse anti-vasa2 1:1000 [48], rabbit anti-cadherin1 (1:500) and mouse anti-cadherin3 (1:1000) [53], mouse anti-Tyrosinated Tubulin (TUB-1A2, T9028 Sigma, 1:1000). Samples were washed 10 times with 1 × PBS / 0.2% Triton-X100, blocked in Blocking solution II for 1 h at room temperature and incubated with the secondary antibodies 1:1000 (goat α-rabbit IgG-Alexa568 (Invitrogen A11008), goat α-rabbit IgG-Alexa633 (Invitrogen A21070), goat α-mouse IgG-Alexa488 (Invitrogen A11001), goat α-mouse IgG-Alexa568 (Invitrogen A11004) and DAPI, 1:1000) in Blocking solution II at room temperature for 2 h or at 4 °C overnight. After 10 washes in 1 × PBS / 0.2% Triton-X100, they were either infiltrated with and mounted in VECTASHIELD® Antifade Mounting Medium (H-1000–10, VectorLabs) or further processed for vibratome sectioning.

Vibratome sectioning

Vibratome sectioning was performed after antibody staining or in situ hybridization. Embedding medium (10% gelatin in PBS) was dissolved at 50 °C and stored at − 20 °C in 2 ml tubes. Before use, the embedding medium was warmed at 37 °C until liquified and pieces of the body column were left for infiltration until settling down to the bottom of the tube. Samples in the embedding medium were poured into mounting molds (cut 1-ml pipette tip), pieces were orientated, and molds were transferred to the fridge for hardening for 30 min. Hardened gelatine blocks were fixed in 4% Formaldehyde in PBS at 4 °C overnight. Samples were washed 3 × with 1 × PBS and sectioned at a Leica VT1200 vibratome (speed = 0.5, amplitude = 0.7 and auto feed thickness = 50–100 μm). Sectioned in situ hybridization samples were infiltrated with and mounted in glycerol, immunohystochemistry samples in VECTASHIELD® Antifade Mounting Medium (H-1000–10, VectorLabs).

In situ hybridization

Urea-based in situ hybridization on adult tissues was carried out as described previously [79], with minor changes. Pieces of Nematostella body column or of individual mesenteries were dissected and fixed for 1 h at room temperature (as described in Fixation for immunohistochemistry and Fixation for in situ hybridization). Anti-Digoxigenin-AP (anti-DIG/AP) Fab fragments (Roche) were diluted 1:4000 in 0.5% blocking reagent (Roche) in 1 × Blocking buffer solution (Roche).

Single-cell transcriptomic analysis

Nematostella gene IDs of BMP signaling pathway molecules were obtained from the Nv2 genome browser (https://simrbase.stowers.org/starletseaanemone): NV2.19847 (bmp2/4), NV2.6269 (bmp5-8), NV2.6489 (gdf5-l), NV2.2978 (admp), NV2.2977 (smad1/5), NV2.10628 (smad4), NV2.15308 (smad4-l), NV2.13963 (smad6), NV2.11643 (alk2), NV2.13851 (alk3/6), NV2.13010 (bmprII), NV2.12106 (actrII), NV2.24312 (rgm), NV2.14185 (chordin), NV2.122 (gremlinA), NV2.12016 (gremlinB), NV2.23832 (follistatin), NV2.12730 (cv2), NV2.7393 (noggin2), NV2.7394 (noggin1). Single-cell transcriptomic analysis of BMP pathway genes was performed using previously published data by Cole et al. [18] (https://github.com/technau). The R-package Seurat vs4 [85] was used to process the count matrices to include adult wild type libraries. The full dataset from Cole et al. [18] was partitioned into samples derived only from adult tissues, and all clusters with less than 10 remaining cells were dropped. Expression of genes of interest across the available clustering was examined using the Seurat::DotPlot function, holding maximum dot size fixed at 100%. R scripts used for the analysis are available on GitHub (https://github.com/technau/Nv2_Atlas, https://github.com/technau/BMPatlas/releases/tag/Vs.BioRxiv) and Zenodo (https://doiorg.publicaciones.saludcastillayleon.es/10.5281/zenodo.14765821).

Protein IDs and phylogenetic analyses

For the phylogenetic analyses, Nematostella sequences (see “Single-cell transcriptomic analysis”) were translated using the expasy webtool (https://web.expasy.org/translate/). Sequences of human BMP receptors and cv2 sequences of different species were obtained from Uniprot: P37023 (ACVRL1), Q04771 (ACVR1), P36894 (BMPR1A), P36896 (ACVR1B), P36897 (TGFR1), Q8NER5 (ACVR1C), O00238 (BMPR1B), P37173 (TGFR2), Q13873 (BMPR2), P27037 (ACVR2A), Q13705 (ACVR2B); A0A8J0VHU6 (BMPER_XENLA), Q8CJ69 (BMPER_MUSMU), Q8N8U9 (BMPER_HOMSA), Q9W2H2 (BMPER_DROME), Q9W494 (BMPER_ DROME). Oscarella carmela Chordin model (m.95558) was obtained by BLAST at https://compagen.unit.oist.jp/. Trichoplax adherens Chordin-like EST (TriadT52199) was retrieved by BLAST from https://metazoa.ensembl.org/Trichoplax_adhaerens/. Oyster Chordin protein sequence CGI_10022848 was taken from [86], and Aurelia Chordin sequence was obtained by BLAST from the published transcriptome [87]. Other Chordin and Chordin-like sequences used in the tree were obtained by NCBI BLAST and are as follows: XP_001625161.2, NP_001081778.1, NP_001259576.1, NP_034023.1, XP_009058760.1, AMY99568.1, AMY99573.1, NP_001036036.1, AAC41250.1, AAI62594.1, XP_019630752.1, NP_001079355.1.

Data availability

All data generated or analysed during this study are included in this published article, its supplementary information files and publicly available repositories. R scripts used for the analysis are available on GitHub (https://github.com/technau/Nv2_Atlas, https://github.com/technau/BMPatlas/releases/tag/Vs.BioRxiv) and Zenodo (https://doiorg.publicaciones.saludcastillayleon.es/10.5281/zenodo.14765821).

Notes

In cnidarian developmental biology literature, the terms „ectoderm “, „(mes)endoderm “, and “pharyngeal ectoderm” are used to describe embryonic tissues. Since this paper primarily deals with adult tissues, we will be using the morphological terms „epidermis “ for the outer cell layer and „gastrodermis “ for the lining of the gut. We will also extrapolate the use of these terms to embryonic and larval tissues.
Scyphozoa, Cubozoa and Staurozoa all have radially symmetric body plans, however, in Hydrozoa, there are known exceptions. Few hydrozoan species display a clear bilaterality, with an additional plane of symmetry created by the tentacle arrangement in the polyp or the medusa [90], and in the case of the colonial siphonophores, by the distinct functional specialization of the units of the colony (zooids), but also in the embryo [91]. To this point, it is not clear if the secondary body axes of Hydrozoa and Anthozoa are homologous or evolved independently.

Abbreviations

ac:: Accessory cells
bwe:: Body wall epidermis
bwg:: Body wall gastrodermis
cd:: Central disc
ct:: Ciliated tract
cgt:: Cnidoglandular tract
epi:: Epidermis
em:: Epidermal muscle
gd:: Gastrodermis
gf:: Gastric filament
gfb:: Gastric filament base
gon:: Gonad
gv:: Germinal vesicle
itr:: Inter-tentacular region
mla:: Marginal lappet
ma:: Manubrium
mes:: Mesentery
mg:: Mesoglea
mo:: Mouth
mm:: Medial mesentery
ms:: Mesentery stalk
nc:: Nematocyst wart
oo:: Oocyte
obw:: Outer body wall
pe:: Pedalium
pha:: Pharynx
pm:: Parietal muscle
pmr:: Parietal muscle region
rm:: Retractor muscle
rmr:: Retractor muscle region
rt:: Reticular tract
sgg:: Somatic gonad gastrodermis
si:: Siphonoglyph
sf:: Septal filament
sg:: Somatic gonad
si:: Siphonoglyph
sp:: Spermary
spc:: Spermatocytes
spg:: Spermatogonia
spt:: Spermatids and sperm
tb:: Tentacle base
tb-ep:: Tentacle base epidermis
tb-gd:: Tentacle base gastrodermis
te:: Tentacle
trg:: Trophic gastrodermis
um:: Umbrella
vel:: Velarium
vla:: Velar lappet

References

Urist MR. Bone: formation by autoinduction. Science. 1965;150(3698):893–9.
Article CAS PubMed Google Scholar
Wu M, Wu S, Chen W, Li YP. The roles and regulatory mechanisms of TGF-beta and BMP signaling in bone and cartilage development, homeostasis and disease. Cell Res. 2024;34(2):101–23.
Article CAS PubMed PubMed Central Google Scholar
Wang S, Chen YG. BMP signaling in homeostasis, transformation and inflammatory response of intestinal epithelium. Sci China Life Sci. 2018;61(7):800–7.
Article PubMed Google Scholar
Herrera AM, Shuster SG, Perriton CL, Cohn MJ. Developmental basis of phallus reduction during bird evolution. Curr Biol. 2013;23(12):1065–74.
Article CAS PubMed Google Scholar
Kaltcheva MM, Anderson MJ, Harfe BD, Lewandoski M. BMPs are direct triggers of interdigital programmed cell death. Dev Biol. 2016;411(2):266–76.
Article CAS PubMed PubMed Central Google Scholar
Zou H, Niswander L. Requirement for BMP signaling in interdigital apoptosis and scale formation. Science. 1996;272(5262):738–41.
Article CAS PubMed Google Scholar
Friedrichs M, Wirsdoerfer F, Flohe SB, Schneider S, Wuelling M, Vortkamp A. BMP signaling balances proliferation and differentiation of muscle satellite cell descendants. BMC Cell Biol. 2011;12:26.
Article CAS PubMed PubMed Central Google Scholar
Hegarty SV, O’Keeffe GW, Sullivan AM. BMP-Smad 1/5/8 signalling in the development of the nervous system. Prog Neurobiol. 2013;109:28–41.
Article CAS PubMed Google Scholar
Itman C, Loveland KL. SMAD expression in the testis: an insight into BMP regulation of spermatogenesis. Dev Dyn. 2008;237(1):97–111.
Article CAS PubMed Google Scholar
Lochab AK, Extavour CG. Bone Morphogenetic Protein (BMP) signaling in animal reproductive system development and function. Dev Biol. 2017;427(2):258–69.
Article CAS PubMed Google Scholar
Regan SLP, Knight PG, Yovich JL, Leung Y, Arfuso F, Dharmarajan A. Involvement of Bone Morphogenetic Proteins (BMP) in the Regulation of Ovarian Function. Vitam Horm. 2018;107:227–61.
Article CAS PubMed Google Scholar
Vieira WA, Wells KM, Raymond MJ, De Souza L, Garcia E, McCusker CD. FGF, BMP, and RA signaling are sufficient for the induction of complete limb regeneration from non-regenerating wounds on Ambystoma mexicanum limbs. Dev Biol. 2019;451(2):146–57.
Article CAS PubMed PubMed Central Google Scholar
Tripathi BK, Irvine KD. The wing imaginal disc. Genetics. 2022;220(4):iyac020.
Driver I, Ohlstein B. Specification of regional intestinal stem cell identity during Drosophila metamorphosis. Development. 2014;141(9):1848–56.
Article CAS PubMed PubMed Central Google Scholar
Zhang Y, Que J. BMP Signaling in Development, Stem Cells, and Diseases of the Gastrointestinal Tract. Annu Rev Physiol. 2020;82:251–73.
Article CAS PubMed Google Scholar
Bier E, De Robertis EM. EMBRYO DEVELOPMENT. BMP gradients: A paradigm for morphogen-mediated developmental patterning. Science. 2015;348(6242):aaa5838.
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;234:1-19.
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.
Article CAS PubMed PubMed Central Google Scholar
Genikhovich G, Technau U. On the evolution of bilaterality. Development. 2017;144(19):3392–404.
Article CAS PubMed Google Scholar
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.
Article PubMed Google Scholar
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):68.
Article PubMed Central Google Scholar
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.
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.
Article PubMed Google Scholar
Leclère L, Horin C, Chevalier S, Lapebie 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.
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.
Article PubMed PubMed Central Google Scholar
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.
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.
Leclère L, Rentzsch F. RGM regulates BMP-mediated secondary axis formation in the sea anemone Nematostella vectensis. Cell Rep. 2014;9(5):1921–30.
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.
Article CAS PubMed PubMed Central Google Scholar
Kraus JE, Fredman D, Wang W, Khalturin K, Technau U. Adoption of conserved developmental genes in development and origin of the medusa body plan. EvoDevo. 2015;6:23.
Article PubMed PubMed Central Google Scholar
Reber-Muller S, Streitwolf-Engel R, Yanze N, Schmid V, Stierwald M, Erb M, et al. BMP2/4 and BMP5-8 in jellyfish development and transdifferentiation. Int J Dev Biol. 2006;50(4):377–84.
Article CAS PubMed Google Scholar
Reinhardt B, Broun M, Blitz IL, Bode HR. HyBMP5-8b, a BMP5-8 orthologue, acts during axial patterning and tentacle formation in hydra. Dev Biol. 2004;267(1):43–59.
Article CAS PubMed Google Scholar
Sanders SM, Shcheglovitova M, Cartwright P. Differential gene expression between functionally specialized polyps of the colonial hydrozoan Hydractinia symbiolongicarpus (Phylum Cnidaria). BMC Genomics. 2014;15(1):406.
Article PubMed PubMed Central Google Scholar
Cole AG, Jahnel SM, Kaul S, Steger J, Hagauer J, Denner A, et al. Muscle cell-type diversification is driven by bHLH transcription factor expansion and extensive effector gene duplications. Nat Commun. 2023;14(1):1747.
Article CAS PubMed PubMed Central Google Scholar
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.
Article CAS PubMed Google Scholar
Finnerty JR, Pang K, Burton P, Paulson D, Martindale MQ. Origins of bilateral symmetry: Hox and dpp expression in a sea anemone. Science. 2004;304(5675):1335–7.
Article CAS PubMed Google Scholar
Matus DQ, Pang K, Marlow H, Dunn CW, Thomsen GH, Martindale MQ. Molecular evidence for deep evolutionary roots of bilaterality in animal development. Proc Natl Acad Sci U S A. 2006;103(30):11195–200.
Article CAS PubMed PubMed Central Google Scholar
Matus DQ, Thomsen GH, Martindale MQ. Dorso/ventral genes are asymmetrically expressed and involved in germ-layer demarcation during cnidarian gastrulation. Curr Biol. 2006;16(5):499–505.
Article CAS PubMed Google Scholar
Rentzsch F, Anton R, Saina M, Hammerschmidt M, Holstein TW, Technau U. Asymmetric expression of the BMP antagonists chordin and gremlin in the sea anemone Nematostella vectensis: implications for the evolution of axial patterning. Dev Biol. 2006;296(2):375–87.
Article CAS PubMed Google Scholar
Schwaiger M, Schönauer A, Rendeiro AF, Pribitzer C, Schauer A, Gilles AF, et al. Evolutionary conservation of the eumetazoan gene regulatory landscape. Genome Res. 2014;24(4):639–50.
Miramon-Puertolas 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.
Article CAS PubMed PubMed Central Google Scholar
Tucker RP, Shibata B, Blankenship TN. Ultrastructure of the mesoglea of the sea anemone Nematostella vectensis (Edwardsiidae). Invertebr Biol. 2011;130(1):11–24.
Article Google Scholar
Fischer AHL, Smith J. Nematostella High-density RNAseq time-course. https://hdl.handle.net/1912/59812012.
Helm RR, Siebert S, Tulin S, Smith J, Dunn CW. Characterization of differential transcript abundance through time during Nematostella vectensis development. BMC Genomics. 2013;14:266.
Article CAS PubMed PubMed Central Google Scholar
Warner JF, Guerlais V, Amiel AR, Johnston H, Nedoncelle K, Röttinger E. NvERTx: a gene expression database to compare embryogenesis and regeneration in the sea anemone Nematostella vectensis. Development. 2018;145(10):dev162867.
Chen CY, McKinney SA, Ellington LR, Gibson MC. Hedgehog signaling is required for endomesodermal patterning and germ cell development in the sea anemone Nematostella vectensis. Elife. 2020;9:e54573.
Extavour CG, Pang K, Matus DQ, Martindale MQ. vasa and nanos expression patterns in a sea anemone and the evolution of bilaterian germ cell specification mechanisms. Evol Dev. 2005;7(3):201–15.
Article CAS PubMed Google Scholar
Praher D, Zimmermann B, Genikhovich G, Columbus-Shenkar Y, Modepalli V, Aharoni R, et al. Characterization of the piRNA pathway during development of the sea anemone Nematostella vectensis. RNA Biol. 2017;14(12):1727–41.
Article PubMed PubMed Central Google Scholar
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.
Kirillova A, Genikhovich G, Pukhlyakova E, Demilly A, Kraus Y, Technau U. Germ-layer commitment and axis formation in sea anemone embryonic cell aggregates. Proc Natl Acad Sci U S A. 2018;115(8):1813–8.
Article CAS PubMed PubMed Central Google Scholar
Eckelbarger KJ, Hand C, Uhlinger KR. Ultrastructural features of the trophonema and oogenesis in the starlet sea anemone, Nematostella vectensis (Edwardsiidae). Invertebr Biol. 2008;127(4):381–95.
Article Google Scholar
Moiseeva E, Rabinowitz C, Paz G, Rinkevich B. Histological study on maturation, fertilization and the state of gonadal region following spawning in the model sea anemone, Nematostella vectensis. PLoS ONE. 2017;12(8): e0182677.
Article PubMed PubMed Central Google Scholar
Pukhlyakova EA, Kirillova AO, Kraus YA, Zimmermann B, Technau U. A cadherin switch marks germ layer formation in the diploblastic sea anemone Nematostella vectensis. Development. 2019;146(20):dev174623.
Hobmayer B, Rentzsch F, Holstein TW. Identification and expression of HySmad1, a member of the R-Smad family of TGFbeta signal transducers, in the diploblastic metazoan Hydra. Dev Genes Evol. 2001;211(12):597–602.
Article CAS PubMed Google Scholar
Rentzsch F, Guder C, Vocke D, Hobmayer B, Holstein TW. An ancient chordin-like gene in organizer formation of Hydra. Proc Natl Acad Sci U S A. 2007;104(9):3249–54.
Article CAS PubMed PubMed Central Google Scholar
Krishnapati LS, Khade S, Trimbake D, Patwardhan R, Nadimpalli SK, Ghaskadbi S. Differential expression of BMP inhibitors gremlin and noggin in Hydra suggests distinct roles during budding and patterning of tentacles. Dev Dyn. 2020;249(12):1470–85.
Article CAS PubMed Google Scholar
Siebert S, Farrell JA, Cazet JF, Abeykoon Y, Primack AS, Schnitzler CE, et al. Stem cell differentiation trajectories in <i>Hydra</i> resolved at single-cell resolution. Science. 2019;365(6451):eaav9314.
Watanabe H, Schmidt HA, Kuhn A, Hoger SK, Kocagoz Y, Laumann-Lipp N, et al. Nodal signalling determines biradial asymmetry in Hydra. Nature. 2014;515(7525):112–5.
Article CAS PubMed Google Scholar
Wenger Y, Buzgariu W, Perruchoud C, Loichot G, Galliot B. Generic and context-dependent gene modulations during Hydra whole body regeneration. bioRxiv. 2019:587147. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/587147.
Mörsdorf D, Prünster MM, Genikhovich G. Chordin-mediated BMP shuttling patterns the secondary body axis in a cnidarian. bioRxiv. 2024:2024.05.27.596067.
Ikmi A, Steenbergen PJ, Anzo M, McMullen MR, Stokkermans A, Ellington LR, et al. Feeding-dependent tentacle development in the sea anemone Nematostella vectensis. Nat Commun. 2020;11(1):4399.
Article CAS PubMed PubMed Central Google Scholar
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.
Article CAS PubMed Google Scholar
Rentzsch F, Layden M, Manuel M. The cellular and molecular basis of cnidarian neurogenesis. Wiley Interdiscip Rev Dev Biol. 2017;6(1):e257.
Watanabe H, Kuhn A, Fushiki M, Agata K, Özbek S, Fujisawa T, et al. Sequential actions of beta-catenin and Bmp pattern the oral nerve net in Nematostella vectensis. Nat Commun. 2014;5:5536.
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.
Article CAS PubMed Google Scholar
Renfer E, Amon-Hassenzahl A, Steinmetz PR, 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.
Article CAS PubMed Google Scholar
Houliston E, Momose T, Manuel M. Clytia hemisphaerica: a jellyfish cousin joins the laboratory. Trends Genet. 2010;26(4):159–67.
Article CAS PubMed Google Scholar
Seipel K, Yanze N, Schmid V. The germ line and somatic stem cell gene Cniwi in the jellyfish Podocoryne carnea. Int J Dev Biol. 2004;48(1):1–7.
Article CAS PubMed Google Scholar
Varley Á, Horkan HR, McMahon ET, Krasovec G, Frank U. Pluripotent, germ cell competent adult stem cells underlie cnidarian regenerative ability and clonal growth. Curr Biol. 2023;33(10):1883-92.e3.
Article CAS PubMed PubMed Central Google Scholar
Gold DA, Jacobs DK. Stem cell dynamics in Cnidaria: are there unifying principles? Dev Genes Evol. 2013;223(1–2):53–66.
Article PubMed Google Scholar
Bosch TC, David CN. Stem cells of Hydra magnipapillata can differentiate into somatic cells and germ line cells. Dev Biol. 1987;121(1):182–91.
Article Google Scholar
Holstein TW. The Hydra stem cell system - Revisited. Cells Dev. 2023;174: 203846.
Article CAS PubMed Google Scholar
Muller WA, Teo R, Frank U. Totipotent migratory stem cells in a hydroid. Dev Biol. 2004;275(1):215–24.
Article PubMed Google Scholar
Nishimiya-Fujisawa C, Petersen H, Koubková-Yu TC-T, Noda C, Shigenobu S, Bageritz J, et al. An ancient split of germline and somatic stem cell lineages in Hydra. bioRxiv. 2023:2023.07.04.546637.
DuBuc TQ, Schnitzler CE, Chrysostomou E, McMahon ET, Febrimarsa, Gahan JM, et al. Transcription factor AP2 controls cnidarian germ cell induction. Science. 2020;367(6479):757–62.
Curantz C, Doody C, Horkan HR, Krasovec G, Weavers PK, DuBuc TQ, et al. A positive feedback loop between germ cells and gonads induces and maintains sexual reproduction in a cnidarian. Sci Adv. 2025;11(2):eadq8220.
Eckelbarger KJ, Larson R. Ultrastructure of the ovary and oogenesis in the jellyfish Linuche unguiculata and Stomolophus meleagris, with a review of ovarian structure in the Scyphozoa. Mar Biol. 1992;114(4):633–43.
Article Google Scholar
Lauretta D, Wagner D, Penchaszadeh PE. First record of a trophonema in black corals (Cnidaria: Antipatharia). Coral Reefs. 2018;37(2):581–4.
Article Google Scholar
Lebouvier M, Miramon-Puertolas P, Steinmetz PRH. Evolutionarily conserved aspects of animal nutrient uptake and transport in sea anemone vitellogenesis. Curr Biol. 2022;32(21):4620–30 e5.
Scott A, Harrison PL. Gametogenic and reproductive cycles of the sea anemone. Entacmaea quadricolor Marine Biology. 2009;156(8):1659–71.
Article Google Scholar
Tan ES, Izumi R, Takeuchi Y, Isomura N, Takemura A. Molecular approaches underlying the oogenic cycle of the scleractinian coral, Acropora tenuis. Sci Rep. 2020;10(1):9914.
Article CAS PubMed PubMed Central Google Scholar
Wedi SE, Dunn DF. Gametogenesis and Reproductive Periodicity of the Subtidal Sea Anemone Urticina Lofotensis (Coelenterata: Actiniaria) in California. Biol Bull. 1983;165(2):458–72.
Article PubMed Google Scholar
Fritzenwanker JH, Technau U. Induction of gametogenesis in the basal cnidarian Nematostella vectensis(Anthozoa). Dev Genes Evol. 2002;212(2):99–103.
Article PubMed Google Scholar
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 Harb Protoc. 2009;2009(9):pdb prot5281.
Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM, 3rd, et al. Comprehensive Integration of Single-Cell Data. Cell. 2019;177(7):1888–902 e21.
Tan S, Huan P, Liu B. Expression patterns indicate that BMP2/4 and Chordin, not BMP5-8 and Gremlin, mediate dorsal-ventral patterning in the mollusk Crassostrea gigas. Dev Genes Evol. 2017;227(2):75–84.
Article CAS PubMed Google Scholar
Brekhman V, Malik A, Haas B, Sher N, Lotan T. Transcriptome profiling of the dynamic life cycle of the scypohozoan jellyfish Aurelia aurita. BMC Genomics. 2015;16(1):74.
Article PubMed PubMed Central Google Scholar
Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, et al. IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era. Mol Biol Evol. 2020;37(5):1530–4.
Article CAS PubMed PubMed Central Google Scholar
Steinmetz PRH, Aman A, Kraus JEM, Technau U. Gut-like ectodermal tissue in a sea anemone challenges germ layer homology. Nat Ecol Evol. 2017;1(10):1535–42.
Article PubMed PubMed Central Google Scholar
Malakhov VV. Symmetry and the tentacular apparatus in Cnidaria. Russ J Mar Biol. 2016;42(4):287–98.
Article Google Scholar
Mańko MK, Munro C, Leclère L. Establishing Bilateral Symmetry in Hydrozoan Planula Larvae, a Review of Siphonophore Early Development. Integr Comp Biol. 2023;63(5):975–89.
Article PubMed Google Scholar

Download references

Acknowledgements

We thank Alison Cole for advice and feedback on the single-cell transcriptomic data analysis. We are grateful to Oliver Link for proving Aurelia and Sanderia ephyra and Clytia jellyfish, Roland Halbauer (Tiergarten Schönbrunn, Vienna) for providing Tripedalia jellyfish and Daniel Abed-Navandi (Haus des Meeres, Vienna) for providing Hydractinia polyps. For the purpose of Open Access, the author has applied a CC BY public copyright license to any Author Accepted Manuscript (AAM) version arising from this submission. Confocal microscopy was performed at the Core Facility Cell Imaging and Ultrastructure Research, University of Vienna—member of the Vienna Life-Science Instruments (VLSI).

Funding

Open access funding provided by University of Vienna. This research was funded in whole or in part by the Austrian Science Fund (FWF) grant (DOI https://doiorg.publicaciones.saludcastillayleon.es/10.55776/P32705) to G.G.. P.K. is a recipient of the Dimitrov Fellowship of the Austrian Academy of Sciences (OeAW). D.M. is a recipient of the Austrian Science Fund (FWF) Lise-Meitner Fellowship (DOI https://doiorg.publicaciones.saludcastillayleon.es/10.55776/M3291).

Author information

Authors and Affiliations

Department of Neurosciences and Developmental Biology, University of Vienna, Vienna, Austria
Paul Knabl, David Mörsdorf & Grigory Genikhovich
Vienna Doctoral School of Ecology and Evolution (VDSEE), University of Vienna, Vienna, Austria
Paul Knabl

Authors

Paul Knabl
View author publications
You can also search for this author inPubMed Google Scholar
David Mörsdorf
View author publications
You can also search for this author inPubMed Google Scholar
Grigory Genikhovich
View author publications
You can also search for this author inPubMed Google Scholar

Contributions

P.K. designed and performed most of the experiments. D.M. generated and analyzed the bmp2/4 knockout. G.G. conceived the project, acquired funding, designed and performed experiments. P.K. and G.G. wrote the paper. All authors edited the paper. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Grigory Genikhovich.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12915_2025_2150_MOESM1_ESM.pdf

Additional file 1. Fig. S1 – Phylogenetic analyses of Nematostella TGFβ receptors and CV2. Maximum likelihood phylogenies were constructed using IQ-tree [88], the bootstrap values are indicated at the nodes. Protein IDs used in the tree are listed in the Methods. Additional File 1: Fig. S2 – Dot plot showing the expression of BMP pathway genes in single-cell transcriptomic data of adult Nematostella polyp tissues with fine clustering presenting 91 transcriptomic cell states (36). Percent Expressed – dot size indicates the percentage of cells in a cluster expressing the gene of interest, Average Expression – color indicates averaged expression value. Additional File 1: Fig. S3 – BMP signaling activity and expression of BMP signaling components in the head region. (A-A’) pSMAD1/5 activity between the forming tentacles and around the mouth of the 4d planula. (B) bmp2/4 expression in the oral gastrodermis and in the aboral epidermis of the planula; faint expression is discernable in the tentacle buds, (C) bmp5-8 is also expressed in the gastrodermis, tentacle buds and aboral epidermis, but appears stronger than bmp2/4. (D-D’) rgm is strongly expressed in the tentacle buds, siphonoglyph and aboral epidermis. (E-E’) cv2 is strongly and (F-F’) admp is weakly expressed in a cross-like pattern between the tentacles and then in eight epidermal stripes – a pattern identical to the epidermal pSMAD1/5 activity at this stage (see A-A’ and (26)). (G-H’) pSMAD1/5 activity in juvenile polyps (~ 6 weeks old), (G-G’) lateral view of the head region, intertentacular and epidermal pSMAD1/5 activity indicated by white arrowheads, (H–H’) oral view of the head region, intertentacular pSMAD1/5 activity is visible between individual tentacles (white arrowheads), unspecific, non-nuclear signal is visible in the cavity of individual tentacles (hollow white arrowheads); tb – tentacle bud, si – siphonoglyph, te – tentacle, ph – pharynx. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bars 50 µm (white) and 100 µm (black). Additional File 1: Fig. S4 – Unspecific staining of spirocytes and putative muco-glandular cells by pSMAD1/5 antibody. (A-A’) cross-section of a tentacle (te) shows unspecific staining of the coiled thread and capsule wall in the spirocytes (B-B’) close-up of the cnidoglandular tract (cgt) shows unspecific, cytoplasmic signal in putative muco-glandular cell types. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bar 25 µm. Additional File 1: Fig. S5 – Dynamics of mCherry expression in the bmp2/4::mCh transgenic reporter line. bmp2/4::mCherry expression (yellow) is bilaterally symmetric in the 2d planula larva (A). At primary polyp stage, it is confined to the mesentery gastrodermis (B). Additional aboral expression domain forms in late planula (C, also detectable on B; See also Figure S3 for in situ hybridization). In the adult, expression is observed in the mesentery gastrodermis. epi – epidermis, mes – mesentery, sf—septal filament. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bars 50 µm. Additional File 1: Fig. S6 – Overview of BMP signaling domains in the mesentery across optical sections. (A-C’’) show different optical sections of the same mesentery region in a longitudinal side view. A’’-C’’ shows the same images as A’-C’ but using a color gradient LUT making the differences in the pSMAD1/5 staining intensity more obvious. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bars 50 µm. rmr—retractor muscle region, mm—medial mesentery, cgt—cnidoglandular tract, ct – ciliated tract. Additional File 1: Fig. S7 – Expression of BMP ligands and receptors in the mesentery. (A) Schematic of a longitudinal section of an adult Nematostella polyp and (A’) female and (A’’) male mesentery whole-mounts after dissection, dashed line (purple) indicates the dissection site located between the mesentery stalk (not visible) and the retractor muscle region (rmr). (B-G’) Whole mount tissue pieces of individual female and male mesenteries stained by in situ hybridization for the expression of (B-B’) bmp2/4, (C–C’) bmp5-8, (D-D’) alk2, (E-E’) alk6, (F-F’) acrRII and (G-G’) bmpRII. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bar 100 µm. phr – pharynx region, rmr – retractor muscle region, cgt – cnidoglandular tract, ct—ciliated tract, oo—oocytes, sp – spermaries. Additional File 1: Fig. S8 – Expression dynamics of BMP pathway genes during embryonic development of Nematostella. Expression of (A) bmp2/4, bmp5-8, gdf5-l, admp, (B) alk2, alk6, actRII and bmpRII, (C) smad1/5 and smad4 in the egg and during early stages of Nematostella development according to the NvERTx database [43,44,45]. Additional File 1: Fig. S9 – Differential cadherin localization in the mesentery. (A-A’) Broad localization of cdh3 throughout the gastrodermis and distinct localization of cdh1 localization in the septal filament (ingrowth of the pharyngeal tissue [89]; and (B) in the accessory cells. All stainings were performed at least two times independently with 3 or more animals imaged. Scale bars 50 µm. Additional File 1: Fig. S10 – No detection of pSMAD1/5 in several medusozoan cnidarian species. (A’-F’) show no or non-nuclear signal. All stainings were performed two times independently with 5 or more animals imaged. Scale bar 50 µm. Additional File 1: Fig. S11 – bmp2/4 knockout animals sporadically develop tentacle-like bubbles. (A) Results of the genotyping of a mosaic F0 mutant and of a F1 heterozygous mutant animal with a single G deletion generating a frame shift. (B-C) Heterozygous bmp2/4 mutants with a normal morphology (B) and with multiple bubbles (C). (D-E’) Morphological features of the bubble. (D) Phalloidin staining of the bubble tissue shows epidermal muscle fibers exclusive for the normal tentacle tissue. (E-E’) Differential interference contrast image of the same area as on (D) shows multiple spirocytes (arowheads) – a cnidocyte type unique for the tentacles. (F) qPCR comparison of the bodywall marker cdh1 and tentacle markers cdh3 and nem64 in wild-type bodywall (wt_bw) tissue and bubble tissue of the F0 mosaic bmp2/4 mutants. (G) Overview of morphological and molecular differences between the adult wt polyp and the F0 mosaic bmp2/4 mutant. Scale bars 5 mm (B-C) and 50 µm (E). epi – epidermis, gastro – gastrodermis, mg – mesoglea (yellow shading), em – epidermal muscle (blue shading).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Knabl, P., Mörsdorf, D. & Genikhovich, G. A whole-body atlas of BMP signaling activity in an adult sea anemone. BMC Biol 23, 49 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12915-025-02150-w

Download citation

Received: 05 August 2024
Accepted: 05 February 2025
Published: 21 February 2025
DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12915-025-02150-w

A whole-body atlas of BMP signaling activity in an adult sea anemone

Abstract

Background

Results

Conclusions

Background

Results

Single-cell RNA-seq data analysis suggests broad expression profiles of BMP pathway components across multiple cell populations in the adult Nematostella polyp

BMP signaling forms distinct activity domains in the inner and outer epithelial layers of the adult polyp

Gastrodermal BMP signaling occurs in distinct domains along the basal–apical axis of the mesentery

Medial and apical mesentery domains display reduced pSMAD1/5 activity but increased BMP expression

BMP signaling is absent from vasa-positive germ cells and stem cells

BMP signaling in the gonad region is active in female trophonemata and male ciliated plugs

The medusozoan species Aurelia coerulea and Tripedalia cystophora display broad BMP signaling activity across different body regions

Discussion

Diversity of BMP signaling domains in Cnidaria indicates a variety of functions

Potential role of BMP signaling in tentacle formation

Potential role of BMP signaling in neurons, muscles, and cnidocytes

Potential role of BMP signaling in the reproductive region

Conclusions

Methods

Animal culture

Nematostella bmp2/4 mutagenesis

Fixation for immunohistochemistry

Fixation for in situ hybridization

Immunohistochemistry

Vibratome sectioning

In situ hybridization

Single-cell transcriptomic analysis

Protein IDs and phylogenetic analyses

Data availability

Notes

Abbreviations

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher’s Note

Supplementary Information

12915_2025_2150_MOESM1_ESM.pdf

Rights and permissions

About this article

Cite this article

Share this article

Keywords

BMC Biology

Contact us