Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 35 (12), 2940-2956

Integrating Embryonic Development and Evolutionary History to Characterize Tentacle-Specific Cell Types in a Ctenophore

Affiliations

Integrating Embryonic Development and Evolutionary History to Characterize Tentacle-Specific Cell Types in a Ctenophore

Leslie S Babonis et al. Mol Biol Evol.

Abstract

The origin of novel traits can promote expansion into new niches and drive speciation. Ctenophores (comb jellies) are unified by their possession of a novel cell type: the colloblast, an adhesive cell found only in the tentacles. Although colloblast-laden tentacles are fundamental for prey capture among ctenophores, some species have tentacles lacking colloblasts and others have lost their tentacles completely. We used transcriptomes from 36 ctenophore species to identify gene losses that occurred specifically in lineages lacking colloblasts and tentacles. We cross-referenced these colloblast- and tentacle-specific candidate genes with temporal RNA-Seq during embryogenesis in Mnemiopsis leidyi and found that both sets of candidates are preferentially expressed during tentacle morphogenesis. We also demonstrate significant upregulation of candidates from both data sets in the tentacle bulb of adults. Both sets of candidates were enriched for an N-terminal signal peptide and protein domains associated with secretion; among tentacle candidates we also identified orthologs of cnidarian toxin proteins, presenting tantalizing evidence that ctenophore tentacles may secrete toxins along with their adhesive. Finally, using cell lineage tracing, we demonstrate that colloblasts and neurons share a common progenitor, suggesting the evolution of colloblasts involved co-option of a neurosecretory gene regulatory network. Together these data offer an initial glimpse into the genetic architecture underlying ctenophore cell-type diversity.

Figures

Fig. 1.
Fig. 1.
Tentacle morphology varies across ctenophores. (A) Mnemiopsis leidyi has reduced tentacles as an adult, Beroe forskalii lacks tentacles completely, and Haeckelia rubra has long unbranched tentacles (white arrow). (B) The cydippid (larval) stage of M. leidyi has long branched tentacles (T). (C) DIC micrograph of a cydippid with tentacles retracted into the tentacle sheath (TS). (D) Higher magnification of the boxed area in C showing a branched tentacle (T) emerging from the tentacle sheath. (E) Emerging tentacle showing partially contracted side branches (tentilla, Tt). (F) Fully extended tentacle showing colloblast islets (C) along the tentacle. (G) Tentacle cell types—neurons (N) and smooth muscle cells (SM) are in the mesogleal core (M); colloblasts (C; yellow), covering cells (CC), ciliated sensory cells (CS), granular gland cells (GG), hoplocytes (HC), and mucous gland cells (MG) are in the epidermis (Ep). Images in panel A courtesy of Bruno Vellutini (M. leidyi) and Steve Haddock (B. forskalii, and H. rubra).
Fig. 2.
Fig. 2.
Beroe and Haeckelia are nested within Ctenophora. An 18S phylogeny of 36 species of ctenophore. Bootstrap support values are indicated at the nodes and branch length indicates rate of substitution. Clades with <50% bootstrap support have been collapsed. Shaded boxes indicate the presence of the indicated trait. *Indicates the presence of tentacles only in the larval stage.
Fig. 3.
Fig. 3.
Identifying colloblast and tentacle candidate genes. (A) Summarized 18S tree. Colloblast candidates (N = 189) were present in ≥70% of the sampled taxa but absent from Beroe and Haeckelia (magenta boxes). Tentacle candidates (N = 165) were present ≥70% of the sampled taxa, including at least one species of Haeckelia (dagger), but were absent from Beroe (teal boxes). (B) Developmental transcriptome sampling for Mnemiopsis leidyi. Early cleavage occurs during the first 8 h post fertilization (hpf). A thickened tentacle epithelium (TE) is visible at 9 hpf and invagination of the epithelium to form the tentacle sheath (TS) is complete by 12 hpf. Larval tentacles (T) grow continuously from the tentacle bulb (TB). (C) Candidate data sets are enriched for genes expressed during tentacle morphogenesis, relative to gene models from M. leidyi (ML2.2) (P < 0.0001 for both). (D, E) The two largest clusters of gene expression profiles identified from candidate genes; colored lines represent different genes. Grey bars denote the 9–12 hpf window highlighted in panel B. White arrows indicate peaks in expression during tentacle bulb invagination, magenta arrows indicate peaks during tentacle morphogenesis. TPM—transcripts per million mapped reads.
Fig. 4.
Fig. 4.
Colloblast and tentacle candidates are expressed together in adult Mnemiopsis leidyi. (A) Tissue sampling protocol for tentacle bulb (TB) and comb row (CR) transcriptome sequencing. (B) Over 70% of the colloblast and tentacle candidate genes were expressed in adult tissues (N = 138/189 colloblast candidates, N = 130/165 tentacle candidates). In both data sets, a significant proportion of the expressed genes (P < 0.0001 for both colloblast and tentacle candidates) were upregulated in the tentacle bulb, relative to the comb row (≥2 log2-fold change, padj < 0.05). (C) There was a significant cluster of colloblast candidates in cell C52 (P < 0.0001), with a second cluster in cell C53. Tentacle candidates clustered significantly in cell C54 (P = 0.0015). Cell IDs refer to single-cell sequencing results reported in Sebe-Pedros, Chomsky, et al. (2018). (D) Seventeen transcription factors were identified from colloblast candidates (C), tentacle candidates (T), or cells C52, C53, or C54 using GO terms GO: 0003677, GO: 0003700, GO: 0006351, and GO: 0006355. E-values represent the reciprocal best BLAST hit in the human genome, except for genes already identified from M. leidyi (Eval = 0.0). Only significant hits (E ≤ 1e–03) are shown. Genes previously characterized by in situ hybridization are indicated in bold. (E) Summarized expression of MlNR1, MlNR2, MlBsh, and MlIslet in the presumptive tentacle epithelium (TE), the invaginated tentacle bulb (TB), and near the apical organ (AO) during embryonic development. (F) Read counts for previously characterized transcription factors in tissues from adult M. leidyi. MlNR1, MlBsh, MlIslet, and MlPRD10A are significantly upregulated (≥2 log2-fold change) in the tentacle bulb (MlNR1 P = 0.004, MlBsh P = 0.010, MlIslet P < 0.001, MlPRD10A P = 0.026). MlNR2 is not differentially expressed (P = 0.893).
Fig. 5.
Fig. 5.
Ontology of candidate genes. (A) Cladogram of the taxa used to identify ctenophore-specific genes. Genes with significant BLAST hits (E ≤ 1e–02) to taxa other than Ctenophora are assumed to have been present in the common ancestor of all animals (grey arrow). Genes lacking significant hits to other metazoan taxa are assumed to have originated after ctenophores diverged from the rest of animals or to have been lost in the stem lineage giving rise to the rest of animals (orange arrow). (B) Ctenophore-specific genes (orange) are significantly enriched in the set of late-expressed colloblast candidates (P = 0.0007), but comprise similar proportions of the tentacle candidate genes and complete set of protein models from Mnemiopsis leidyi (ML 2.2). (C, D) Overrepresented annotations from colloblast and tentacle candidates, relative to ML2.2 protein model annotations.
Fig. 6.
Fig. 6.
Identification of features associated with subcellular localization. (A) Diagram of an immature colloblast after Storch and Lehnert-Moritz (1974) showing developing secretory vesicles. (B) Both colloblast and tentacle candidates were significantly enriched for genes encoding signal peptides (blue) and transmembrane domains (yellow) (P < 0.0001 for all). (C) Number of genes encoding single- and multi-pass transmembrane domains from colloblast and tentacle candidates.
<sc>Fig</sc>. 7.
Fig. 7.
Neurons and colloblasts share a common progenitor. (A) Embryonic cleavage stages in Mnemiopsis leidyi. Random micromeres (red) were injected at the late gastrula stage. (B) Optical section of a live gastrula immediately following injection of DiI (red); nuclei are labeled with Hoechst. (C) High magnification image of the boxed area in B showing a single injected micromere. (D) High magnification image of a clone of labeled cells 30 min after DiI injection. (EG) A live cydippid larva with two labeled populations of cells: in the floor plate of the apical organ (AO) and the colloblasts of the tentacle (T). (E) Both cell populations are depicted in a single focal plane. (F) A different focal plane showing labeling of cells and dome cilia (arrow) on one side of the apical organ. (G) High magnification image of the extended tentacle; colloblasts (arrows) are the only labeled cells in the tentacle. (H) Cydippid larva with DiI-labeled neurons of the subepidermal nerve net in the body wall (white arrows) and colloblasts (red arrows); the animal is viewed from the tentacular plane (black arrow) as summarized in the inset. TB—tentacle bulb, T—tentacle, Tt—tentilla.
Fig. 8.
Fig. 8.
A possible scenario for the origin of secretory cell diversity in animals. (A) The common ancestor to ctenophores, cnidarians, and bilaterians may have already had a secretory cell progenitor (white) giving rise to neurons and some other unspecialized secretory cell. This pathway may have been independently co-opted (colored lines) to give rise to novel secretory cell types in ctenophores, cnidarians, and bilaterians. (B) The common origin of the neural and secretory cells responsible for making the sensory apparatus in flies supports this hypothesis in bilaterians. Sponges and placozoans have been excluded for simplicity.

Similar articles

See all similar articles

References

    1. Alie A, Leclere L, Jager M, Dayraud C, Chang PR, Le Guyader H, Queinnec E, Manuel M. 2011. Somatic stem cells express Piwi and Vasa genes in an adult ctenophore: ancient association of “germline genes” with stemness. Dev Biol. 350(1):183–197. - PubMed
    1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol. 215(3):403–410. - PubMed
    1. Anderluh G, Podlesek Z, Macek P. 2000. A common motif in proparts of cnidarian toxins and nematocyst collagens and its putative role. Biochim Biophys Acta-Protein Struct Molec Enzymol. 1476(2):372–376. - PubMed
    1. Babonis LS, Martindale MQ, Ryan JF. 2016. Do novel genes drive morphological novelty? An investigation of the nematosomes in the sea anemone Nematostella vectensis. BMC Evol Biol. 16(1):1–22. - PMC - PubMed
    1. Borisenko I, Ereskovsky AV. 2013. Tentacular apparatus ultrastructure in the larva of Bolinopsis infundibulum (Lobata: ctenophora). Acta Zool. 94(2):193–202.

Publication types

Feedback