Dylus DV et al. (2016), Large-scale gene expression study in the ophiur...

ECB-ART-44445

Evodevo 2016 Jan 01;7:2. doi: 10.1186/s13227-015-0039-x.

Show Gene links Show Anatomy links

Large-scale gene expression study in the ophiuroid Amphiura filiformis provides insights into evolution of gene regulatory networks.

Dylus DV , Czarkwiani A , Stångberg J , Ortega-Martinez O , Dupont S , Oliveri P .

???displayArticle.abstract???
BACKGROUND: The evolutionary mechanisms involved in shaping complex gene regulatory networks (GRN) that encode for morphologically similar structures in distantly related animals remain elusive. In this context, echinoderm larval skeletons found in brittle stars and sea urchins provide an ideal system. Here, we characterize for the first time the development of the larval skeleton in the ophiuroid Amphiura filiformis and compare it systematically with its counterpart in sea urchin. RESULTS: We show that ophiuroids and euechinoids, that split at least 480 Million years ago (Mya), have remarkable similarities in tempo and mode of skeletal development. Despite morphological and ontological similarities, our high-resolution study of the dynamics of genetic regulatory states in A. filiformis highlights numerous differences in the architecture of their underlying GRNs. Importantly, the A.filiformis pplx, the closest gene to the sea urchin double negative gate (DNG) repressor pmar1, fails to drive the skeletogenic program in sea urchin, showing important evolutionary differences in protein function. hesC, the second repressor of the DNG, is co-expressed with most of the genes that are repressed in sea urchin, indicating the absence of direct repression of tbr, ets1/2, and delta in A. filiformis. Furthermore, the absence of expression in later stages of brittle star skeleton development of key regulatory genes, such as foxb and dri, shows significantly different regulatory states. CONCLUSION: Our data fill up an important gap in the picture of larval mesoderm in echinoderms and allows us to explore the evolutionary implications relative to the recently established phylogeny of echinoderm classes. In light of recent studies on other echinoderms, our data highlight a high evolutionary plasticity of the same nodes throughout evolution of echinoderm skeletogenesis. Finally, gene duplication, protein function diversification, and cis-regulatory element evolution all contributed to shape the regulatory program for larval skeletogenesis in different branches of echinoderms.

???displayArticle.pubmedLink??? 26759711
???displayArticle.pmcLink??? PMC4709884
???displayArticle.link??? Evodevo
???displayArticle.grants??? [+]

Species referenced: Echinodermata
Genes referenced: ago1b alx1 arid3a erg ets1 foxa1 foxb1 gata3 gata6 gbx2l gcml irak1bp1 jun LOC100887844 LOC100893907 LOC115919910 LOC115921237 LOC115925415 LOC575170 LOC581907 LOC592057 pmar1 SpP19L tbr1 tgif2l

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Amphiura filiformis development and identification of skeletogenic cells. a Live imaging of embryos. Early cleavage stage shows tetrahedral arrangement of cells at 3 hpf, similar to Ophiopholis aculeata [28]. Mid-cleavage stage shows equally sized cells at 6 hpf. Early blastula stage shows spherical embryo with blastocoel at 10 hpf. Hatched blastula embryos have distinct blastocoel at 16 hpf and animal–vegetal orientation is visible through thickening at vegetal side of the embryo. At 23 hpf, mesenchyme blastula stage shows first ingressing cells from the vegetal side of the embryo that fill up the blastocoelar space by 27 hpf. At 30 hpf, calcein (green) stained gastrula embryo shows two newly formed spicules that extent to a tri-radiate structure as visible on a bright field by 36 hpf. b High-resolution time-courses for genes analyzed in this study shown as heatplot were obtained by QPCR. Expression values are relative to Afi-16S (see Additional file 1 for explanation of calculation and QPCR controls; exact numbers are shown in Table S1). c WMISH of skeletogenic marker genes identifies the vegetal plate and the primary ingressing cells as the SM cell lineage in A. filiformis. The two regulatory genes Afi-alx, Afi-jun and the skeleton matrix gene, Afi-p19, are expressed in the vegetal plate of the blastula embryos, then in the first ingressing mesenchymal cells and at later stage in a location congruent to where spicules are formed. Other two orthologs of sea urchin skeletogenic matrix genes, Afi-p58a and Afi-p58b, are also detected in the first ingressing cells of the mesenchyme blastula stage and at later stage in the same location where the spicules are formed. ECl early cleavage, Cl cleavage, EBl early blastula, Bl-VV blastula vegetal view, Bl blastula, MBI mesenchyme blastula, LMBl late mesenchyme blastula, G gastrula, LG late gastrula. Scale bars are 50 μm
	Fig. 2. Afi-pplx is expressed similar to sea urchin Spu-pmar1, but does not function as repressor. a Phylogeny of Pplx and Pmar1 proteins suggesting orthology of these genes. Other paired-like homeodomains are use as outgroup. First value on branch is bootstrap support and second value is posterior probability. Tree is the consensus of differently constructed trees, using different initial alignments as well as different methodologies (Additional file 1: Figure S3). b WMISH showing the expression of Afi-pplx during A. filiformis development. c Time-line comparison of Spu-pmar1 and Afi-pplx transcript abundance adjusted for the stages of development (see Additional file 1: Figure S10) and normalized to their individual maximum of expression shows high correlation of expression dynamic (cross-correlation: 0.801). For brittle star, error bars represent standard deviation of two biological replicas. Sea urchin data were obtained from [59]. d Schematic representation of the double negative gate in euechinoid, showing how the large micromeres are specified to be skeletogenic. e S. purpuratus embryos injected with synthetic mRNA for Spu-pmar1, Afi-pplx, and GFP control (see Additional file 1: Figure S4 for details). No phenotype is observable in Afi-pplx-mRNA embryos or GFP controls, while injection of Spu-pmar1-mRNA induces skeletogenic fate in all cells. WMISH of Spu-delta in embryos injected with Afi-pplx-mRNA or Spu-pmar1-mRNA, which show expansion of Spu-delta expression to the whole embryo in Spu-pmar1-mRNA-injected embryos. Bfl Branchiostoma floridae, Lva Lytechinus variegatus, Pli Paracentrotus lividus, Hpu Hemicentrotus pulcherrimus, Spu Strongyloncentrotus purpuratus, Afi Amphiura filiformis, Pmi Patiria miniata, Sko Saccoglossus kowalevskii, VV vegetal view, Cl cleavage, EBl early blastula, Bl blastula, MBl mesenchyme blastula
	Fig. 3. In A. filiformis hesC is co-expressed with its immediate downstream genes. a Single WMISH for blastula and mesenchyme blastula stage embryos. b Double fluorescent WMISH on blastula stage embryos. Afi-hesC expression is restricted to a ring of cells in the vegetal half and co-expressed with the endomesodermal marker Afi-foxA. Afi-ets1/2, Afi-tbr, and Afi-delta are co-expressed with Afi-hesC in one cell layer (yellow area) at the vegetal plate of the embryo at blastula stage and completely co-expressed at mesenchyme blastula stage. VV vegetal view, SVV semi vegetal view, Bl blastula, MBl mesenchyme blastula
	Fig. 4. Expression pattern of orthologs of late skeletogenic genes. a WMISH showing the expression pattern of of Afi-tgif, Afi-erg, Afi-hex, Afi-dri, Afi-foxB, and Afi-nk7 at different developmental stages. At blastula stage, orthologs of the three interlocking loop genes are expressed in the same domain; however, details of their vegetal plates show that Afi-tgif and Afi-hex are expressed in a wider domain than Afi-erg. At mesenchyme blastula, Afi-tgif, Afi-erg, and Afi-hex are co-expressed in NSM cells. From blastula to gastrula, Afi-dri and Afi-foxB shows no expression in SM lineage (skeletogenic mesodermal cells shown with black arrow). Afi-nk7 shows expression in SM lineage during blastula and mesenchyme blastula stage only. b Normalized timeseries comparison shows inverted sequence of onset of expression between brittle star (solid lines) and sea urchin (dashed lines). Sea urchin data were obtained from [59]. VV vegetal view, SVV semi vegetal view, Bl blastula, MBl mesenchyme blastula, G gastrula
	Fig. 5. Expression of A. filiformis non-skeletogenic mesodermal genes. a–i WMISH at different developmental stages as indicated in the bottom right corner of each image. WMISH probes used are indicated in the top right corner. a–c Afi-gcm is not detectable by WMISH at any of the stages analyzed, consistent with QPCR expression levels (compare Additional file 1: Table S1). d–f Afi-gataC expression becomes detectable at mesenchyme blastula stage in NSM cells in the vegetal plate and it stays active at the tip of the archenteron at gastrula stage. g–i Afi-gataE expression becomes active in a similar fashion to Afi-gataC at mesenchyme blastula stage. At the beginning of gastrulation, it marks the cells at the tip of the archenteron and in the blastopore region. Different to Afi-gataC, Afi-gataE is additionally expressed in cells at the base of the archenteron in the blastopore region. Both Afi-gataC, Afi-gataE are not expressed at blastula stage. Bl blastula, MBl mesenchyme blastula, G gastrula
	Fig. 6. Summary and comparison of regulatory states and minimal GRN rewiring between A. filiformis and sea urchin. a Regulatory states (left) and GRN model (right) for S. purpuratus. Regulatory states for the mesodermal territories are shown on cellular maps of cleavage (Cl), pre-hatching blastula (EBl), hatched blastula (Bl) and mesenchyme blastula (MBl). The dark green represents the skeletogenic mesoderm (SM), the light green represents the non-skeletogenic mesoderm (NSM), and the blue represents the endodermal territory. At mesenchyme blastula, the NSM is divided into precursors of pigment cells (orange) and other NSM precursors (light green). Genes expressed in each territory are listed. S. purpuratus SM-GRN architecture, modified from [2], of the genes analyzed in this study. Arrows indicate positive inputs (activation) and barred line negative inputs (repression). Dashed lines represent functional linkages inferred by perturbation data in [2] for 1 and [16] for 2. Ubq represents an inferred ubiquitous activator necessary for the expression of some of the genes downstream of the DNG. b Regulatory states (left) and potentially conserved GRN linkages (right) for A. filiformis. Regulatory states for the mesodermal territories are shown on cellular maps of pre-hatching blastula (EBl), hatched blastula (Bl), and mesenchyme blastula (MBl). The light green cells represent the mesoderm identified by the expression of pan-mesodermal genes. The dark green represents the here named SM cells expressing the skeletogenic marker genes alx1, jun, p58a, p58b, and p19; blue dots represent the expression of hesC and foxA within the mesoderm territory, while blue cells represent the endodermal territory that surrounds the mesoderm. At blastula stage, the SM is clearly separated from the rest of NSM, while the segregation of NSM and endoderm occurs at MBl stage. Genes expressed in each territory are listed. A hypothetical A. filiformis GRN including nodes and potentially conserved linkages represented as in (a) and based on the observations in this study. EM represents the minimal hypothetical positive input necessary for the hesC expression in the endomesoderm. M represents hypothetical pan-mesodermal positive input(s) necessary to drive the expression in all mesodermal cells. R represents hypothetical repressor(s) expressed in cells expressing only pan-mesodermal genes. SM represents the minimal hypothetical positive input necessary for erg to be expressed in the subset of mesodermal cells. This hypothetical model highlights the differences (nodes and linkages) based on the expression data. As discussed in the text, we parsimoniously assume that, in absence of functional data, linkages are conserved with sea urchin, but we cannot exclude alternative hypotheses

References [+] :

Adomako-Ankomah, P58-A and P58-B: novel proteins that mediate skeletogenesis in the sea urchin embryo. 2011, Pubmed, Echinobase