ECB-ART-45894Nat Commun 2017 Nov 27;81:1799. doi: 10.1038/s41467-017-01961-y.
Show Gene links Show Anatomy links
Evolutionary recruitment of flexible Esrp-dependent splicing programs into diverse embryonic morphogenetic processes.
Epithelial-mesenchymal interactions are crucial for the development of numerous animal structures. Thus, unraveling how molecular tools are recruited in different lineages to control interplays between these tissues is key to understanding morphogenetic evolution. Here, we study Esrp genes, which regulate extensive splicing programs and are essential for mammalian organogenesis. We find that Esrp homologs have been independently recruited for the development of multiple structures across deuterostomes. Although Esrp is involved in a wide variety of ontogenetic processes, our results suggest ancient roles in non-neural ectoderm and regulating specific mesenchymal-to-epithelial transitions in deuterostome ancestors. However, consistent with the extensive rewiring of Esrp-dependent splicing programs between phyla, most developmental defects observed in vertebrate mutants are related to other types of morphogenetic processes. This is likely connected to the origin of an event in Fgfr, which was recruited as an Esrp target in stem chordates and subsequently co-opted into the development of many novel traits in vertebrates.
PubMed ID: 29180615
PMC ID: PMC5703972
Article link: Nat Commun
Species referenced: Echinodermata
Genes referenced: fgfr3 gcml hes hes1 hnf6 impact LOC100887844 LOC100893907 LOC105444881 LOC115919910 LOC115924199 LOC588026 LOC594566 LOC753325 mmut onecut2 pole ptch1 srpl
Morpholinos: LOC588026 MO1 LOC588026 MO2
Article Images: [+] show captions
|Fig. 1. Expression and developmental roles of esrp1 and esrp2 in zebrafish. a WMISH for esrp1 and esrp2 in Danio rerio WT embryos. At 14 h.p.f., esrp1 transcripts were observed in embryonic epidermis, while esrp2 expression was only detected in the polster (po). At 16 h.p.f., esrp1 was restricted to the posterior and tailbud (tb) epidermis, whereas esrp2 persisted in the hatching gland rudiment (hr) and mild expression started to be detected in the otic placode (ot). By 20 h.p.f., esrp1 was found in the tailbud epidermis and more subtly in the olfactory placode (ol), while esrp2 appeared in new territories, such as pronephros (pn) and ectodermal cells of tailbud fin fold (ff). At 24 h.p.f., expression of both paralogs presented a similar pattern including olfactory and otic placodes, cloaca (cl), and epidermis (ep), although esrp2 was also observed in the hatching gland (hg). By 36 h.p.f., both genes were detected in the inner ear epithelium (ie), notochord (nt), and phanynx (ph), and esrp2 was also observed in the heart (he). At 48 h.p.f., expression was found predominantly in inner ear and pharynx. b Schematic representation of the genomic and transcriptomic impact of the selected esrp1 and esrp2 mutations. Blue boxes/lines represent genomic deletions in the mutants, while the red line depicts an altered splice junction in the esrp1 mutant allele. TSS, transcription start site; PTC, premature termination codon; del, deletion. Standard and fluorescent (green dot) primers used during genotyping are represented by arrows. c Left: genotyping of embryos by fluorescent PCR readily distinguished between WT and MUT alleles. Right: Representative 5 d.p.f. larvae for wild type (WT), esrp1 mutant (esrp1 MUT/MUT), esrp2 mutant (esrp2 MUT/MUT), and double mutant (DMUT) genotypes. Deflated swim bladder in the DMUT embryo is indicated by a red asterisk. d–i Phenotypic differences between 6 d.p.f. WT (top) and DMUT (bottom) embryos in different embryonic structures. DMUT larvae showed impaired fin formation (arrows) and cleft palate (arrowhead) d, including malformation of the ethmoid bone, as shown by Alcian blue staining e. f–i Transversal histological sections stained with hematoxylin and eosin showing structural differences in pectoral fin f, esophagus g, inner ear h and olfactory epithelium i. Black arrowheads mark the dorso-lateral septum between semicircular canals in h. Proximal (pr) and distal (dl) parts of the fin are indicated in f. Scale bars: 1 mm a, 2 mm c–e, 100 µm f–h, 50 µm i|
|Fig. 2. Esrp is expressed in a subpopulation of TLCs in Ciona and is able to modulate cell motility in the mesenchymal lineage. a Esrp expression (green) is detected at 10 h.p.f. by fluorescent WMISH in the epidermis and in some cells within the mesenchymal lineage, as shown by co-staining with mChe driven by the Twist promoter, which labels the Twist-like1-derived cell lineage. b, c Esrp (green) is expressed in the TLC lineage as shown by co-staining with Hand-r > LHG/LexAOP > H2B::mCherry (red) in 13 h.p.f. and 15 h.p.f. embryos in lateral b and dorso-lateral c views, respectively. aTLC: Anterior TLCs, pTLC: posterior TLCs. d–l Double fluorescent WMISH for Esrp (red) and Twist-like2 (green), with the exception of j, which corresponds to colorimetric Esrp mRNA staining (purple). At middle tailbud stage, Esrp expression is detected in both epidermis and mesenchymal cell lineage (the latter is marked by arrows). m-m’ Mesenchymal lineage in WT larvae at 18 h.p.f. stained in red using the Twist > CD4::mCherry construct. Twist > GFP was used as co-electroporation control plasmid, showing full co-electroporation (green channel not shown for clarity). n–n’ Mesenchymal lineage stained in larvae co-electroporated with Twist > CD4::mCherry and Twist > Esrp constructs. The trunk region is shown in m–n, while tail segments are shown in m’–n’. o Quantification of the different phenotypes of mesenchymal cell lineage motility observed in Twist > Esrp and control larvae. These included individuals with abnormal migration in the trunk (‘Trunk’), with ectopic mChe-positive cells in the tail (‘Tail’) or both phenotypes (‘Trunk + Tail’). P-values correspond to a two-sided Fisher Exact tests (‘Normal’ vs rest). Scale bars correspond to 25 µm|
|Fig. 3. Esrp is expressed dynamically in the non-neural ectoderm during amphioxus embryo development. WMISH of Esrp in Branchiostoma lanceolatum embryos. Anterior is to the left in all cases. a 14 h.p.f. early neurula (dorsal view) showing expression in the ectodermal cells located in the border region (br) next to the neural plate (npl). b, c 16 h.p.f. and 18 h.p.f. mid-neurula embryos (dorsal view) stained most strongly in the ectodermal cells next to the neural plate border and that form the hinge points (hp) during neural tube closure. The location of the neuropore (npo) is indicated. d In 21 h.p.f. late neurula (dorsal view), Esrp expression is extended throughout the whole epidermis (ep). e, f Early (30 h.p.f.) and late (36 h.p.f.) pre-mouth stages (lateral views) showing Esrp-positive cells in anterior ectoderm, tailbud epithelia and in a few cells that likely corresponding to migrating sensory cells during epidermal incorporation or already integrated into the epithelium (black arrows). g Dorsal view of a 36 h.p.f. embryo shows the epidermal location those Esrp-positive cells (black arrows). Scale bars: 50 µm a–f, 25 µm g|
|Fig. 4. Esrp represses cilia formation in aboral ectoderm and is necessary for complete MET of pigment cells. a, b Colorimetric WMISH of Esrp in sea urchin embryos at 24 h.p.f., in lateral a and vegetal b views, showed expression in one side of the ectodermal territory (ec) and in some cells of the non-skeletogenic mesoderm (ns). vv, vegetal view. c Lateral view at 36 h.p.f. confirmed Esrp’s asymmetric expression in one side of the ns mesoderm. d, e Fluorescent WMISH at 30 h.p.f. (lateral view) confirmed ectodermal and mesodermal expression of Esrp. f Double fluorescent WMISH showed that Esrp expression (red) at 36 h.p.f. (animal pole view) did not overlap with the neurogenic ectoderm (ne) marker Hnf6 (green). nne, non-neural ectoderm. g Double fluorescent WMISH of Esrp (red) and Gcm (green) at 36 h.p.f. (animal pole view) revealed that Esrp was expressed in the aboral ectoderm (ae) and in the pigment cell precursors. Pigment cell precursors that are already in contact with the ectodermal epithelium are marked by arrowheads. oe, oral ectoderm. h, i Esrp (green) and Gcm (red) at 36 h.p.f. in two different stacks from same embryo (lateral view). The arrow indicates a representative migratory path of pigment cells from mesoderm to ectoderm. At the top right corner of each panel a–i, the orientation of the depicted embryo along the animal [a] - vegetal [v], oral [o] – aboral [ab], left [l] –right [r] axes are reported, when possible. j, k Early gastrula (42 h.p.f.) from uninjected control and Esrp trMO embryos showed differences in pigment cell location. Red/white arrowheads indicate pigment cells already integrated into the ectoderm or in the sub-ectodermal space, respectively. l, m Early pluteus larvae (68 h.p.f.) in abanal view show differences in pigment cell morphology upon trMO treatment (roundish instead of stellate/dendritic). n, o Ectopic embryonic cilia stained with acTubulin in the aboral ectoderm (ae) of 96 h.p.f. Esrp-trMO injected embryos. Apex is marked by a white arrowhead. cb, ciliary band. p RT-PCRs showing the levels of intron 2 retention in Esrp transcripts from control and Esrp splMO embryos; genomic DNA was used as a reference for intron inclusion. The asterisk marks the position of the first in-frame termination codon in intron-retained transcripts. q Quantification of pigment cell morphology in control, Esrp trMO and splMO knockdown 68 h.p.f. embryos (sum of three independent experiments). r Quantification of the ‘hairy’ phenotype in control, Esrp trMO and spMO knockdown 68 h.p.f. embryos (sum of two independent experiments). P-values correspond to 2-way q or 3-way r two-sided Fisher Exact tests. Hairy phenotype was considered “mild” when aboral ectoderm cells show long cilia (of similar or longer length than ciliary band cells) and “strong” when, in addition to the latter, larvae show particularly long cilia at the apex, as indicated by an arrowhead in panel o. Scale bars correspond to 20 µm|
|Fig. 5. Evolution of Esrp-dependent splicing programs. a Number Esrp-dependent AS events by type detected in human, mouse, zebrafish and sea urchin, RNA-seq samples. Red/blue bars correspond to Esrp-enhanced/silenced inclusion of the alternative sequence. IR, intron retention; Alt5, alternative 5′ splice site choice; Alt3, alternative 3′ splice site choice; AltEx, alternative cassette exons. b Venn diagram showing the overlap among homologous Esrp-dependent cassette exons in coding regions detected as regulated in the same direction in the studied species. Only Esrp-dependent exons with homologs in at least two species and sufficient read coverage in all the species in which they have homologs are included in the comparison. Esrp-dependent exons that lack homologous counterparts in all the other species are indicated for each species (NH). Similarly, the number of Esrp-dependent exons that do not have sufficient read coverage in at least one of the species with a homologous exon is displayed for each species (NC). c Summary of conservation at the level of genomic presence, alternative splicing and Esrp-dependency for all 321 clusters of human Esrp-dependent coding exons in other species. Shared Esrp-dependent exons between the previous phylogenetic group are classified in the test species into three categories: (i) the exon is not detected in the genome (top row, discontinuous line); (ii) the homologous exon is detected in the genome, but it is constitutively spliced (middle row, gray exon); and (iii) the homologous exon is alternatively spliced (bottom row, yellow exon, numbers in bold). Below this classification, the number of similarly regulated Esrp-dependent exons shared by the two species/lineages over the number of alternatively spliced exons with sufficient read coverage from (iii) (indicated by asterisks), is shown|
|Fig. 6. Fgfr AS is regulated by Esrp genes in vertebrates and amphioxus. a RT-PCR assays showing differential Fgfr exon IIIb and IIIc inclusion in WT versus DMUT 5 d.p.f. zebrafish embryos. b RT-PCR assays for Fgfr AS in different amphioxus adult tissues, depicted in a transversal section. nc, nerve cord, ms, muscle, gl, gills, hd, hepatic diverticulum, nt, notochord, sk, skin. Reverse primers were designed in both exons IIIb and IIIc (arrows) and used together in the same PCR reaction. c Top: schematic representation of pcDNA3.1-based minigene constructs containing the genomic region spanning the Fgfr AS event of Branchiostoma lanceolatum, with (pcDNA3.1-BlaFGFR) and without (pcDNA3.1-BlaFGFRΔIIIx) exon IIIx. Bottom: relative intensity of fluorescent RT-PCR bands supporting differential inclusion of exons IIIb and IIIc when transfecting the minigenes alone (Control) or together with a plasmid containing either amphioxus or zebrafish full-length Esrp transcripts (BlaEsrp and DreEsrp1, respectively). Despite significant mis-splicing of the minigene in all conditions, only the amphioxus construct was able to induce a dramatic switch toward exon IIIb inclusion. Primers were designed in the neighboring constitutive exons (arrows). d RT-PCR assays for endogenous human AS events in the same control, BlaEsrp or DreEsrp1 transfected 293T cells showing that the amphioxus and zebrafish Esrp constructs are able to modulate endogenous Esrp-dependent events in a similar manner. Error bars correspond to standard errors of three biological replicates. Esrp-enhanced isoforms are marked with an isoform cartoon|
|Fig. 7. Gene structure and AS at the IgIII domain of the Fgfr gene family in metazoans. Schematic representation of the AS diversity in the region encoding the homolog of the Fgfr IgIII domain for zebrafish (Danio rerio), vase tunicate (Ciona intestinalis), amphioxus (Branchiostoma lanceolatum), purple sea urchin (Strongylocentrotus purpuratus), acorn worm (Saccoglossus kowalevskii), honey bee (Apis mellifera), red flour beetle (Tribolium castaneum), fruit fly (Drosophila melanogaster), pacific oyster (Crassostrea gigas), and starlet sea anemone (Nematostella vectensis). Boxes represent exons, horizontal lines are introns, and diagonal lines connect splicing junctions. Homologous to vertebrate exons IIIa, IIIb, and IIIc are shown in purple, blue and red, respectively. The non-chordate pro-orthologous exon of exons IIIb and IIIc is colored half blue and half red. The amphioxus- and ambulacrarian-specific alternative exons are depicted in green and yellow, respectively. Light violet colors are used for the insect-specific mutually exclusive event involving exon IIIa, and brown and light blue for oyster-specific exons. Gray exons are constitutive in all species. Fruit fly orthologs (heartless and branchless) are intronless genes. Dupl: duplication; Amb: ambulacraria-specific exon; retrotransp.: retrotranscription|
References [+] :
Abitua, Identification of a rudimentary neural crest in a non-vertebrate chordate. 2013, Pubmed