Decision letter: Single-cell RNA sequencing of the Strongylocentrotus purpuratus larva reveals the blueprint of major cell types and nervous system of a non-chordate deuterostome

Abstract

Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Identifying the molecular fingerprint of organismal cell types is key for understanding their function and evolution. Here, we use single-cell RNA sequencing (scRNA-seq) to survey the cell types of the sea urchin early pluteus larva, representing an important developmental transition from non-feeding to feeding larva. We identify 21 distinct cell clusters, representing cells of the digestive, skeletal, immune, and nervous systems. Further subclustering of these reveal a highly detailed portrait of cell diversity across the larva, including the identification of neuronal cell types. We then validate important gene regulatory networks driving sea urchin development and reveal new domains of activity within the larval body. Focusing on neurons that co-express Pdx-1 and Brn1/2/4, we identify an unprecedented number of genes shared by this population of neurons in sea urchin and vertebrate endocrine pancreatic cells. Using differential expression results from Pdx-1 knockdown experiments, we show that Pdx1 is necessary for the acquisition of the neuronal identity of these cells. We hypothesize that a network similar to the one orchestrated by Pdx1 in the sea urchin neurons was active in an ancestral cell type and then inherited by neuronal and pancreatic developmental lineages in sea urchins and vertebrates. Editor's evaluation This work provides a comprehensive analysis of cell state specification of a whole deuterostome organism, the sea urchin Strongylocentrotus purpuratus. It is also vigorous example for the use of single-cell sequencing to identify cell type homologies across evolution. The paper is thus of significant interest to scientists within the broad fields of developmental biology and evolution, as well as to the more specific communities of researchers that use the sea urchin as a model system or those interested in employing the single-cell mRNA-sequencing technology for "non-conventional" (and marine) molecular model systems. https://doi.org/10.7554/eLife.70416.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Multicellular organisms consist of numerous cell types, specialized in performing different tasks that guide growth and survival. During embryonic development, cells go through rounds of proliferation, specification and differentiation into cell types with distinct functions. The information for this developmental diversification lies in the genome and the spatio-temporal expression of regulatory genes that specifies the molecular fingerprint of a given cell type (Fu et al., 2017). The identity of each cell type is established, controlled, and maintained by distinct Gene Regulatory Networks (GRNs). GRNs are logical maps of the regulatory inputs and outputs active in a cell at a given place and time, and are enacted by transcription factors, signaling molecules and terminal differentiation genes (Davidson et al., 2003; Davidson and Erwin, 2006). GRNs have been studied in a variety of organisms ranging from plants to animals in order to analyze the gene interactions at a specific time and place during the life of an organism (Krouk et al., 2013), and have been used for understanding the relationship between genome, development and evolution (Davidson and Erwin, 2006). Therefore, understanding the genetic mechanisms that provide cell types with a specific identity, and the conservation of this identity across animal taxa, is essential for understanding cell type function and evolutionary history (Arendt, 2008; Arnone et al., 2016). Until recently, most approaches for comparing cell types relied on the description of distinct morphological features, that is linked to the functionality of a given cell type, the identification of molecular markers, perturbation of gene expression and fate mapping. However, technological advances in microfluidics and nucleic acid barcoding now allow the high-throughput recognition of an organism’s cell types at a single-cell level. In particular, single-cell RNA sequencing (scRNA-seq) technology, developed during the last decade, is a powerful method used to unravel the transcriptional content of individual cells, resulting in the identification of distinct cell types in an unbiased manner (Tang et al., 2009; ; Klein et al., 2015). ScRNA-seq involves dissociation of an organism, organ, or tissue into single cells, isolation and capture of the single cells into droplets, specific barcoding of individual mRNAs, and sequencing of transcriptomic content of each cell. Computational analysis can then identify putative cell type families by clustering cells with similar transcriptional profiles. Further analysis of such cell clusters can lead to the identification of distinct cell types. Within deuterostomes, echinoderms are a member of the phylogenetic sister group to chordates, making them an ideal model for understanding the origin and diversification of deuterostome and chordate cell types (Arnone et al., 2015). Sea urchin embryos and larvae have also been extensively used to unravel the general mechanisms of cell type specification and differentiation during development (Cameron and Davidson, 1991; Davidson et al., 1998; McClay, 2011; Lyons et al., 2012; McClay et al., 2020). The main reason for this lies in the ease with which different cell types and biological processes can be observed in the optically transparent embryos and larvae. Among the most well-studied sea urchin cell types are those comprising the nervous (Bisgrove and Burke, 1987Burke et al., 2006a, McClay et al., 2018), immune (Rast et al., 2006; Ho et al., 2017) and digestive systems (Annunziata et al., 2014; Annunziata and Arnone, 2014; Perillo and Arnone, 2014; Perillo et al., 2016), and of both musculature (Andrikou et al., 2013; Andrikou et al., 2015) and skeleton (Okazaki, 1965; Duloquin et al., 2007; Rafiq et al., 2012; Sun and Ettensohn, 2017). For these, the developmental origins and gene regulatory wiring has been described in great detail, making the sea urchin an ideal model for GRN comparative analyses in development and evolution (Cary et al., 2020). Here, we take advantage of the detailed characterization of the sea urchin cell types performed over the years, the available cell-type-specific molecular markers, and the ease with which the sea urchin larvae are dissociated into single cells, to perform scRNA-seq and generate a comprehensive atlas of sea urchin larval cell type families. Our findings suggest that the larva consists of 21 genetically distinct cell clusters, representing distinct cell type families (Shekhar and Menon, 2019), which we validate using fluorescent in situ hybridization (FISH) and immunohistochemistry (IHC). Based on previous studies tracing developmental lineage, we assign cell type families to the specific embryonic germ layers they derived from. Comparing the transcription factor content of the grouped cell clusters reveals that most transcription factors are expressed pleiotropically independently of their germ layer origins, yet tend to be cell type family-specific within a germ layer. In addition, we illustrate how single-cell data complement and validate previously studied GRNs, and also reveal novel cellular domains where these GRNs are likely also activated. Lastly, we investigate neuronal diversity in the sea urchin larva, identifying 12 distinct neuronal cell types. Among these, we recover a unique neurosecretory type controlled by Sp-Pdx1 and Sp-Brn1/2/4 exhibiting a pancreatic-like gene expression signature (Perillo et al., 2018). Our results confirm and extend this pancreatic-like signature, suggesting that an ancestral neuron in early deuterostomes may have given rise to the endocrine cells in the vertebrate pancreas. Supporting this, knockdown of Sp-Pdx1 shows it is necessary for differentiation of this pancreatic-like neuronal endocrine population, indicating it has an evolutionary conserved role as a mediator of endocrine fate. Results Building a cell type atlas of the sea urchin larva with single-cell transcriptomics Sea urchin early pluteus larvae were cultured and collected at three dpf. We performed single-cell RNA sequencing on six samples from four independent biological replicates. Individual samples were dissociated into single cells using a gentle enzyme-free dissociation protocol and using the 10 x Chromium scRNA-seq system (Figure 1A). In total, transcriptomes from 19,699 cells were included in the final analysis. To identify sea urchin larval cell types, we used Louvain graph clustering as implemented in the Seurat pipeline (see Materials and methods). This revealed 21 genetically distinct cell clusters (Figure 1B, Figure 1—figure supplement 1A,B), each representing an individual cell type or a set of closely related cell types in the early pluteus larva. Figure 1 with 1 supplement see all Download asset Open asset Cell type family atlas of the three dpf S. purpuratus larva. (A) Single-cell RNA sequencing pipeline from gamete fertilization to computational analysis. (B) UMAP showing three dpf larval cells colored by their assignment to the initial set of 21 distinct cell clusters. (C) UMAP with cells colored by germ layer of origin: endoderm (yellow), mesoderm (red), and ectoderm (blue). (D) Dotplot of gene markers specific to cell clusters. (E) Illustration depicting location of cell type families on different larval domains. Color-code is the same as in (B). Next, we set out to explore the identity of our initial 21 cell clusters. We first assigned preliminary identities to each cluster based on the expression of previously described cell type markers, benefiting from the unique and rich knowledge on sea urchin developmental lineages: ciliary band (Btub2) (Harlow and Nemer, 1987), apical plate (Hbn) (Burke et al., 2006a), aboral ectoderm (Spec2a) (Yuh et al., 2001), lower oral ectoderm (Bra) (Wei et al., 2012), upper oral ectoderm (Gsc) (Wei et al., 2012), neurons (SynB) (Burke et al., 2006a), esophageal muscles (Mhc) (Andrikou et al., 2013), coelomic pouches (Nan2) (Juliano et al., 2010), blastocoelar cells (185/333) (Ho et al., 2017), immune cells (Gcm) (Materna et al., 2013), skeleton (Msp130) (Harkey et al., 1992), anus (Hox11/13b), intestine (Cdx), pyloric sphincter (Pdx-1), different stomach domains (Chp, ManrC1a, Endo16) (Annunziata and Arnone, 2014; Annunziata et al., 2019), exocrine pancreas-like domain (Ptf1a) (Perillo et al., 2016), cardiac sphincter (Trop1) (Yaguchi et al., 2017), and esophagus (Brn1/2/4) (Cole and Arnone, 2009). Further, we grouped the cell clusters according to their embryonic germ layer origin (Figure 1C) using knowledge from previous lineage tracing experiments (Angerer and Davidson, 1984; Cameron et al., 1987). To validate these identities, we identified all genes expressed in each cell type, totaling in 15,578 WHL genes (transcriptome models) and 12,924 genes (SPU gene models) (Supplementary file 1), and performed in situ hybridization on a selected set of these together in combination with previously described markers. Based on this, we mapped 20 of the 21 clusters to distinct larval domains and confirmed their identity (Figure 1E). Notably, the resulting expression patterns validated the initial predictions (Figure 2—figure supplement 2), verifying the high quality of the single-cell dataset. Importantly, this approach identified various new markers for each cell type family, including Sp-FbsL_2 (ciliary band; Figure 2A1), Sp-hypp_2386 (skeletal cells; Figure 2A9), and Sp-Serp2/3 (exocrine pancreas-like cells; Figure 2A14). The 21st cluster, which had a poorly-defined molecular signature as judged by the low number and expression level of total and marker genes combined with the lack of specific localization, likely represents not fully differentiated cells (Figure 1—figure supplement 1D,F) and we refer to this cluster as undefined. Figure 2 with 4 supplements see all Download asset Open asset Validation of scRNA-seq predictions and novel expression domains. (A) FISH of S. purpuratus three dpf larvae with antisense probes for Sp-FbsL_2 (A1), Sp-Chrna9_4 (A2), Sp-Frizz5/8 (A3), Sp-FoxABL (A4), Sp-Bra (A5), SPU_006199 (A6), Sp-FcolI/II/III (A7), Sp-Hypp_1249 (A8), Sp-Hypp_2386 (A9), Sp-Mlckb (A10), Sp-MsxI (A11), SPU_008104 (A12), Sp-Ahrl (A13), Sp-Serp2; Sp-Serp3 (A14), Sp-Hnf4 (A15), Sp-Cyp2L42 (A16), Sp-Rfxc1l (A17), Sp-Pdx1 (A17) and Sp-Cdx (A18). Color-code indicates germ layer embryonic origin: endoderm (yellow), mesoderm (red), ectoderm (blue). Immunofluorescent detection of acetylated tubulin in ciliary band (green). (B) Dotplot of Sp-Sip1, Sp-SoxC, Sp-Hbn, and Sp-Fgf9/16/20 expression showing previously described and novel expression domains. (C) FISH of S. purpuratus three dpf larvae with antisense probes for Sp-SoxC (C1–C2), Sp-Hbn (C4–C5), and Sp-Fgf9/16/20 (C7–C8). Illustrations depicting all the expression domains of Sp-SoxC (C3), Sp-Hbn (C6), and Sp-Fgf9/16/20 (C9). Illustrations depicting all the expression domains of Sp-SoxC (C3), Sp-Hbn (C6), and Sp-Fgf9/16/20 (C9). Nuclei are labeled with DAPI (in blue). All images are stacks of merged confocal Z sections. (D) FISH of S. purpuratus three dpf larvae with antisense probes for Sp-SoxC and Sp-Fgf9/16/20 (D1–D4), for Sp-Hbn combined with immunohistochemical detection for the skeletal cells marker Msp130 (D5–D9) and Sp-Fgf9/16/20 with immunohistochemical detection for the midgut and hindgut protein Endo1 (D10–D13). A, Anus; Cs, Cardiac sphincter; I, Intestine; M, Mouth; PMCs, Primary mesenchyme cells; St, Stomach. Our scRNA-seq analysis and in situ hybridization protocol unraveled novel expression domains for several previously described cell type markers. For instance, the transcription factors Sp-SoxC and Sp-Hbn, previously described in early neuronal specification (Garner et al., 2016; Wei et al., 2016; Yaguchi et al., 2016), were predicted by our scRNA-seq analysis to also be expressed in skeletal cells (Figure 2B). Likewise, the FGF signaling ligand, Sp-Fgf9/16/20, is known to be involved in skeletal formation and is expressed in specific populations of PMCs (Adomako-Ankomah and Ettensohn, 2014). ScRNA-seq indicates it is also expressed in oral ectoderm, cardiac sphincter, intestine, and anus (Figure 2B). Using in situ hybridization and immunohistochemical detection methods we confirmed the predictions and we found Sp-SoxC expressed in Sp-Fgf9/16/20 positive PMCs and Sp-Hbn expressed in skeletal cells located in the vertex (Figure 2 2D1-4 & 2D5-9). Moreover, we found Sp-Fgf9/16/20 transcripts localized in distinct gut domains corresponding to the cardiac sphincter, intestinal, and anal regions (Figure 2). We compared the limits of detection by in situ hybridization versus single cell RNA sequencing, using the coelomic pouch cell cluster as a case study. The coelomic pouches are derived from the mesoderm and the left coelomic pouch contributes to the formation of the rudiment and juvenile sea urchin after metamorphosis (Strathmann, 1987; Smith et al., 2008). The formation of the coelomic pouches is complex, and includes contributions from the small micromeres, a mesodermal cell population that is set aside during early development (Pehrson and Cohen, 1986; Strathmann, 1987). Previous attempts to characterize this population had involved screening of genes active in germline determination and maintenance in other species and revealed that, while some germ-line-specific transcripts and proteins were found exclusively expressed in the small micromeres and the coelomic pouch of the sea urchin embryo (Juliano et al., 2006), the majority of the genes tested by in situ hybridization were not enriched in this cell type. Interestingly, plotting the Juliano and co-authors’ gene list, alongside previously described coelomic pouch specific genes (Luo and Su, 2012; Martik and McClay, 2015), they were all found in our analysis to be expressed in the same cell cluster (Figure 2—figure supplement 3). This suggests a higher sensitivity of single-cell RNA sequencing compared to the in situ hybridization, adding crucial missing information on the molecular fingerprint of a complex cell type. Lastly, to determine which cells in the larva were undergoing active proliferation, we plotted expression of cell division markers in sea urchin, including pcna, DNA polymerases, DNA ligases, condensins, and centromere proteins (Perillo et al., 2020). The majority of cell proliferation genes were found to be enriched in the ciliary band, apical plate, coelomic pouch, immune, and skeletal cell clusters (Figure 2—figure supplement 4A). We also observed Cdk genes enriched in several endodermally derived cell populations (Figure 2—figure supplement 4A). Validating this, we observed S-phase cells in ciliary band, apical plate, oral ectoderm, endodermal and skeletal cells using Edu pulse labeling (Figure 2—figure supplement 4B). In contrast, we did not observe Edu fluorescence in cell populations that lacked expression of proliferation markers, such as the aboral ectoderm, suggesting that indeed they may not be proliferating. Shared lineage information of the larval cell type families We next compared transcription factor profiles of early larval cell type families. In general, cell populations derived from the same germ layer shared more factors with each other than with cell populations from other germ layers (Figure 3, Figure 3—figure supplement 1), consistent with the finding that cell type expression programs often retain information about their developmental lineage (Sladitschek et al., 2020). Further analysis, however, revealed very few transcription factors specific to cells from a single germ layer, contrasting with many regulators expressed in derivatives of more than one germ layer (Figure 3A), and nearly one third (n = 187) shared by cell type families from all three layers. Unexpectedly, mesodermal cell populations share expression of more transcription factors with ectodermal than with endodermal cell populations, even though they are more closely linked to endodermal lineage in development. We also noted that cell type families derived from the same germ layer share up to one third of transcription factors, while the majority are cell type-specific (Figure 3B, D and E). In general, neighboring cell populations and those with common developmental origins share a larger number of TFs (Figure 3D), compared to cell populations with different developmental histories (Figure 3C). Figure 3 with 1 supplement see all Download asset Open asset Regulatory states of the three dpf S. purpuratus larva. (A) Comparison of the transcription factor content per germ layer. Venn diagram showing the shared and unique transcription factors per germ layer. Ectodermally derived cell type families are shown in blue, mesodermally derived in red, and endodermally derived in yellow. (B) Comparison of the transcription factor content across mesodermal lineages and cell type families. Venn diagram showing the shared and unique transcription factors per comparison. (C) Transcription factor content comparison of pyloric sphincter (endodermally derived) and skeletal cells (mesodermally derived), used as a negative control of our comparison. (D) Comparison of the transcription factor content per endodermal lineage and endodermally derived cell type families. Venn diagram showing the shared and unique transcription factors per comparison. (E) TF signature comparison of ectodermally derived cell type families. Venn diagram showing the shared and unique transcription factors per comparison. Cartoons indicated the relative position of each cell type family/lineage. Mesodermal cell type families/lineages are shown in shades of red, endodermal ones in shades of yellow and endodermal ones in shades of blue. To further characterize the regulatory profile of larval cell type families we set out to identify the expression profiles of members of major transcription factor families (Figure 4, Figure 4—figure supplement 1). In S. purpuratus, most homeobox transcription factors were previously found expressed at the gastrula stage (two dpf), with several members expressed in domains derived from all three germ layers (Howard-Ashby et al., 2006). Our single-cell analysis, although at a later developmental time point, supports these findings, and further refines our understanding of their expression to specific cell type families. In the early pluteus larva, most homeobox class transcription factors are enriched in ectodermally derived cell populations, such as the apical plate and neurons. In contrast, ANTP Class and HNF class transcription factors are enriched in endodermal derivatives (Figure 4). Other major transcription factor families, such as the Forkhead, Ets, and Zinc-finger families, members of which are expressed throughout sea urchin embryogenesis (Tu et al., 2006; Rizzo et al., 2006; Materna et al., 2006), are also expressed across a spectrum of cell populations. Forkhead and zinc-finger transcription factors are highly expressed in specific cell type families of all three germ layer derivatives, whereas Ets family TFs are enriched in ectodermal and mesodermal derivatives (Figure 4). Figure 4 with 2 supplements see all Download asset Open asset Localization of major transcription factor family members. Dotplot showing the average scaled expression of members of the Homeobox, Forkhead and Ets transcription factor families. The developmental origins of each cell type family are shown in blue for ectodermally derived, red for mesodermally derived and yellow for endodermally derived ones. The active regulatory state of a given cell type family is an immediate consequence of the gene regulatory network active at this time point. Previous research in sea urchin has described in detail many regulatory networks active during embryonic and larval development. Our scRNA-seq data is broadly consistent with previous studies, yet also identifies new domains and cell populations in which these regulatory networks may be active. For instance, we plotted all transcription factors known to be active in specifying coelomic pouch cells (Martik and McClay, 2015). Our data confirmed their co-expression in coelomic pouch, but also revealed their co-expression in the apical plate (Figure 5A). Similarly, plotting genes involved in the aboral ectoderm gene regulatory network (Ben-Tabou de-Leon et al., 2013), we found all genes in both the aboral ectoderm cluster as well as in the apical plate cells (Figure 5B). On the other hand, plotting members of the pre-gastrula skeletogenic mesoderm regulatory network (Figure 5C) revealed most were still active in the pluteus larva and specific to skeletal cells (Oliveri et al., 2008; Shashikant and Ettensohn, 2019; Khor et al., 2019; Ettensohn, 2020). These findings illustrate the immediate benefit of our dataset to drastically expand our knowledge of larval regulatory networks. Finally, our scRNA-seq recreates a nearly identical three dpf endoderm expression pattern atlas as that published previously by our group using more traditional methods (Annunziata et al., 2014), providing additional information on each gene’s average expression level and the percentage of cells expressing each marker (Figure 5—figure supplement 1). Figure 5 with 1 supplement see all Download asset Open asset Validation of preexisting GRNs and putative novel function of specific gene regulatory modules. (A) Dotplot showing the mRNA localization of genes involved in the homing of small micromeres to the coelomic pouch and novel apical plate domain. (B) Dotplot of aboral ectoderm regulatory module genes showing novel apical plate expression. (C) Pre-gastrula gene regulatory network enriched in skeletal cells of the sea urchin pluteus larva. Asterisks indicate larval genes involved in biomineralization, putative members of this GRN. Unravelling the neuronal diversity and molecular signature of the nervous system The assessment of neuronal cell type diversity is an important step for unravelling the evolution and function of the nervous system. The sea urchin free swimming larva is equipped with a nervous system consisting of interconnected ganglia (Burke et al., 2006a) that allows the animal to respond to environmental stimuli and coordinate its swimming behavior (Soliman, 1983; Katow et al., 2010). Several neuronal types, including apical and ciliary band neurons, as well as neurons along the digestive tube, have been previously identified and their specification described in detail (Burke et al., 2006a, Burke et al., 2006b, Wei et al., 2009; Wei et al., 2011; Burke et al., 2014; Garner et al., 2016; Wei et al., 2016; McClay et al., 2018; Perillo et al., 2018; Wood et al., 2018). Our initial clustering analysis resolved single clusters for neuronal cells, as well as for PMCs and immune cells. However, expression of known markers suggested the presence of distinct subclusters in each of these cell type family groups. In order to investigate this, we independently performed subclustering and re-analysis of the neuronal, immune, and PMC cells. Subclustering of each of these initial major clusters revealed 12 neuronal, 8 immune, and 5 PMC subclusters, each likely representing distinct cell types (Figure 6, Figure 6—figure supplements 1 and 2). Two of the immune subclusters expressed polyketide synthase 1 (Sp-Pks1), suggesting these represent sea urchin pigment cell populations (Calestani and Rogers, 2010). We also found a subcluster of immune cells that expresses the membrane attack complex/perforin family gene (Sp-MacpfA2), suggesting this corresponds to immune system globular cells (Figure 6—figure supplement 2). Notably, our finding of 5 PMC subclusters corroborates previous reports showing five distinct groups of PMC cells along the syncytium (Sun and Ettensohn, 2014; Figure 6—figure supplement 1). Figure 6 with 3 supplements see all Download asset Open asset Neuronal complexity of the three dpf S. purpuratus larva. (A) (From left to right and top to bottom) UMAP highlighting the neurons cluster (green), immunohistochemical detection for the paneuronal sea urchin marker 1E11 (green), UMAP showing the 12 distinct neuronal subclusters. (B) Schematic representation of the three dpf pluteus larva showing the localization of neuronal subclusters (colors as in A). (C) Dotplot of signaling molecules, transcription factors, and neurotransmitters involved in sea urchin neuronal function and neurogenesis (colors as in A). (D) FISH of S. purpuratus three dpf larvae with antisense probes for the neuronal genes Sp-Delta (D1), Sp-SoxC (D2), Sp-Brn1/2/4 (D3), Sp-Ddc (D4), Sp-Ngn (D5), Sp-Prox1 (D6), Sp-Nacha6 (D7), Sp-Isl (D8), Sp-An (D8 and D14), Sp-Chrna9_4 (D9), Sp-Hbn (D10), Sp-SoxB2 (D11), Sp-Otx (D12), Sp-Tph (A13), Sp-Salmfap (D14), Sp-NeuroD1(D15), Sp-Six3 (D16), Sp-Trh (D17), Sp-Kp (D17), and Sp-Rhox3 (D18). FISH shown in figures D1-3 are paired with immunohistochemical detection of the neuropeptide Sp-An. Nuclei are labeled with DAPI (in blue). All images are stacks of merged confocal Z sections. A, anus; M, mouth. To annotate the 12 neuronal cell types revealed via subclustering, we took advantage of the extensive previous work investigating neurogenesis and neuronal differentiation in sea urchin. Plotting neuronal markers, we resolved unique molecular signatures for each subcluster and assigned each a putative identity and location in the larva (Figure 6B–D). To validate this, we then conducted in situ hybridization experiments for gene markers labeling these specific neuronal populations (Figure 6D), including genes encoding transcription factors (SoxC, Delta, Ngn, Prox1, Isl, Hbn, SoxB2, Otx, NeuroD1, Six3), as well as members of neurotransmitter (Ddc, Nacha6, Chrna9_4, Tph), and neuropeptidergic signaling pathways (An, Salmfap, Trh). The sea urchin larva neuronal differentiation proceeds stepwise, including transient expression of the Notch ligand Delta, followed by expression of the transcription factors SoxC and Brn1/2/4 (Garner et al., 2016). During the final stages of neurogenesis, the transcription factors Sip1, Z167, Ngn, and Otp regulate differentiation of diverse neuronal populations, including apical and ciliary band neurons (Wei et al., 2016; McClay et al., 2018). In our data, we observed Sp-Delta, Sp-SoxC, and Sp-Brn1/2/4, as well different combinations of the transcription factors mentioned above, co-localize in three neuronal populations (subclusters 1, 2, and 4), indicating neuronal differentiation is taking place in those three subclusters (Figure 6C). Interestingly, in one of these populations (subcluster 2) we found expression of the transcription factors Sp-Rx, Sp-Hbn (Figure 6D10) and Sp-Six3 (Figure 6D16), which are known to be expressed in the periphery of the larva’s apical domain (Burke et al., 2006a, Wei et al., 2009). This suggests that this population is located in the periphery of the apical plate and not within the apical organ. In the apical domain, we also detected a subcluster (number 6), which coexpress Sp-Trh and Sp-Salmfap neuropeptides (Wood et al., 2018), as well as Sp-Kp (Kissepeptin) (Figure 6D17). In total, we identified three neuronal subclusters located in the apical domain (subclusters 2, 3 and 4) of the larva that express Tryptophan hydroxylase (Tph), which encodes a key enzyme in the serotonin biosynthesis pathway, suggesting these represent serotonergic neurons in the larva. Within the ciliary band, which comprises t