Molecular and functional characterization of somatostatin-type signalling in a deuterostome invertebrate

Abstract

Somatostatin (SS) and allatostatin-C (ASTC) are structurally and evolutionarily related neuropeptides that act as inhibitory regulators of physiological processes in mammals and insects, respectively. Here, we report the first molecular and functional characterization of SS/ASTC-type signalling in a deuterostome invertebrate-the starfish Asterias rubens (phylum Echinodermata). Two SS/ASTC-type precursors were identified in A. rubens (ArSSP1 and ArSSP2) and the structures of neuropeptides derived from these proteins (ArSS1 and ArSS2) were analysed using mass spectrometry. Pharmacological characterization of three cloned A. rubens SS/ASTC-type receptors (ArSSR1-3) revealed that ArSS2, but not ArSS1, acts as a ligand for all three receptors. Analysis of ArSS2 expression in A. rubens using mRNA in situ hybridization and immunohistochemistry revealed stained cells/fibres in the central nervous system, the digestive system (e.g. cardiac stomach) and the body wall and its appendages (e.g. tube feet). Furthermore, in vitro pharmacological tests revealed that ArSS2 causes dose-dependent relaxation of tube foot and cardiac stomach preparations, while injection of ArSS2 in vivo causes partial eversion of the cardiac stomach. Our findings provide new insights into the molecular evolution of SS/ASTC-type signalling in the animal kingdom and reveal an ancient role of SS-type neuropeptides as inhibitory regulators of muscle contractility.

Keywords: evolution; muscle; neuropeptide; receptor; somatostatin; starfish.

Figure 1.

Sequences of the A. rubens neuropeptide precursors ArSSP1 and ArSSP2 and alignment of neuropeptides derived from these precursors (ArSS1 and ArSS2) with related peptides from other taxa. (a) Amino acid sequence of ArSSP1, with the predicted signal peptide shown in blue, a dibasic cleavage site shown in green and the neuropeptide ArSS1 shown in red. The underlined cysteine residues are predicted to form a disulfide bridge in the mature ArSS1 peptide. (b) Amino acid sequence of ArSSP2, with the predicted signal peptide shown in blue, a dibasic cleavage site shown in green and the neuropeptide ArSS2 shown in red. The underlined cysteine residues are predicted to form a disulfide bridge in the mature ArSS2 peptide. (c) Alignment of echinoderm SS1-type neuropeptides with protostome ASTC-type peptides. Note that in addition to conserved cysteine residues, there is a conserved FXP motif (where X is variable). A GR motif is, more specifically, a shared feature of the echinoderm peptides. (d) Alignment of echinoderm SS2-type neuropeptides with chordate SS-type peptides. Note that in addition to conserved cysteine residues, there is a conserved FWK/IWK motif. An aspartate (D) residue and two glycine (G) residues are, more specifically, shared features of the echinoderm peptides. In (c,d), the aligned amino acids are highlighted in black if the residue is present in at least 70% of the sequences or highlighted in grey if conservative amino acid substitutions are present in at least 70% of the sequences. The position of a disulfide bridge between the two conserved cysteine residues is shown above the alignments. Species and peptide names are highlighted in taxon-specific colours: yellow (Echinodermata), green (Protostomes), pink (Cephalochordata), blue (Vertebrata). Species name abbreviations are as follows: Ajap, Apostichopus japonicus; Apla, Acanthaster planci; Arub, Asterias rubens; Bflo, Branchiostoma floridae; Cele, Caenorhabditis elegans; Ctel, Capitella teleta; Dmel, Drosophila melanogaster, Hsap, Homo sapiens, Lgig, Lottia gigantea; Ovic, Ophionotus victoriae; Spur, Strongylocentrotus purpuratus.

Figure 2.

Comparison of the exon/intron structure of genes encoding somatostatin/allatostatin-C-related precursors in echinoderms and other taxa. The figure shows representations of the gene structures. The protein-coding exons are shown as rectangles and introns are shown as lines with intron length stated underneath. The N-terminal signal peptide, the neuropeptide and monobasic or dibasic cleavage sites are shown in blue, red and green, respectively, with amino acid residue positions shown underneath. Other regions of the precursor protein are shown in grey. The numbers above each precursor diagram show the phase of introns with respect to the reading frame, with 0 representing an intron located between two consecutive codons, 1 representing an intron between the first and second base of a codon and 2 representing an intron between the second and third base of a codon. Note that a shared characteristic of the echinoderm SSP1-type and SSP2-type genes and SS/CST/ASTC-type genes in other taxa is the presence of a phase 0 intron interrupting the coding region for the N-terminal part of the encoded proteins; this distinguishes these genes from UII/URP-type genes that have a phase 1 intron in an equivalent position. Species names are as follows: Apla, Acanthaster planci; Cele, Caenorhabditis elegans; Dmel, Drosophila melanogaster; Hsap, Homo sapiens; Myes, Mizuhopecten yessoensis; Spur, Strongylocentrotus purpuratus. The neuropeptide precursor names are as follows: SS, somatostatin; CST, cortistatin; UII, urotensin II; URP, urotensin II-related peptide; ASTC, allatostatin-C. Accession numbers for the sequences of the precursors shown in this figure are listed in electronic supplementary material, table S3.

Figure 3.

Phylogenetic analysis identifies three A. rubens receptor proteins as somatostatin/allatostatin-C-type receptors. The tree generated using the maximum-likelihood method (1000 bootstrap replicates, LG + F + I + G4 substitution model) comprises three distinct receptor clades—SS/ASTC/opioid-type receptors, MCH-type receptors and urotensin II-type receptors. Galanin-type receptors and kisspeptin-type receptors were included here as outgroups to root the tree. The three A. rubens receptor proteins characterized in this study (labelled with red arrows) are positioned in the clade containing SS/opioid/ASTC-type receptors, demonstrating that they can be classified as SS/ASTC-type receptors. The round dots represent bootstrap support and the different coloured backgrounds highlight different taxonomic groups (see legend). The scale bar represents the average residue substitution per site. Receptor names shown in red indicate that cognate ligands for these receptors have been identified experimentally. Species names are as follows: Acal, Aplysia californica; Ajap, Apostichopus japonicus; Amel, Apis mellifera; Apla, Acanthaster planci; Arub, Asterias rubens; Bflo, Branchiostoma floridae; Cele, Caenorhabditis elegans; Cgig, Crassostrea gigas; Ctel, Capitella teleta; Dmel, Drosophila melanogaster; Dpul, Daphnia pulex; Hrob, Helobdella robusta; Hsap, Homo sapiens; Lgig, Lottia gigantea; Ovic, Ophionotus victoriae; Pdum, Platynereis dumerilii; Pmar, Petromyzon marinus; Skow, Saccoglossus kowalevskii; Spur, Strongylocentrotus purpuratus; Tcas, Tribolium castaneum. Accession numbers of receptor sequences and associated references used to generate this figure are listed in electronic supplementary material, table S4.

Figure 4.

ArSS2, but not ArSS1, acts as a ligand for three A. rubens SS/ASTC-type receptors, ArSSR1, ArSSR2 and ArSSR3. The graphs show that ArSS2 causes dose-dependent activation of ArSSR1 (a), ArSSR2 (c) and ArSSR3 (e) expressed with Gα16 in CHO-K1 cells stably expressing a calcium-sensitive bioluminescent GFP-aequorin fusion protein G5A. Luminescence is expressed as a percentage of the maximal response observed in each experiment (a,c,e) and the ArSS2 mean (±s.e.m.) EC₅₀ values for each receptor are shown in red lettering. ArSS1 has no effect when tested over the same concentration range as ArSS2, demonstrating the specificity of the activation of ArSSR1–3 by ArSS2 (a,c,e). Each point represents mean values (±s.e.m.) from at least three independent experiments, with three repeats per experiment. (b,d,f) Comparison of the total bioluminescence responses of ArSSR1 (b), ArSSR2 (d) and ArSSR3 (f) expressing cells when exposed to ArSS2 and ArSS1 at a concentration of 10⁻⁵ M for 30 s. Cells exposed to BSA media were used as a control group. The mean luminescence responses to 10⁻⁵ M ArSS2 are 144 620 (b), 166 303 (d) and 46 575 (f), respectively, showing that ArSS2 has higher efficacy as ligand for ArSSR1 and ArSSR2 than for ArSSR3. Analysis of the data in (b,d,f) by one-way ANOVA with Bonferroni's multiple comparisons post hoc test revealed that luminescence in cells exposed to 10⁻⁵ M ArSS2 is significantly higher (****p < 0.0001, **p < 0.01) than in cells exposed to 10⁻⁵ M ArSS1 or BSA media (control). Luminescence is not significantly different between receptor-expressing cells exposed to 10⁻⁵ M ArSS1 and receptor-expressing cells exposed to BSA media.

Figure 5.

Localization of ArSSP2 mRNA in the nervous system of A. rubens using in situ hybridization. (a) Transverse section of a radial nerve cord incubated with antisense probes showing stained cells (ArSSP2-expressing cells) in both the hyponeural (arrow) and ectoneural (arrowheads) regions. The inset shows the absence of staining in a transverse section of radial nerve cord incubated with sense probes, demonstrating that staining observed with antisense probes can be attributed to the detection of ArSSP2 transcripts. (b) Higher magnification image of the boxed region in (a), showing stained cells in the hyponeural region (arrows) and in the subcuticular epithelial layer of the ectoneural region (arrowheads). (c) Transverse section of the central disc region showing stained cells in the hyponeural and ectoneural regions of the circumoral nerve ring; stained cells can also be seen in an adjacent peri-oral tube foot. (d) Higher magnification image of the boxed region in (c), showing stained cells in the hyponeural region (arrow) and the subcuticular epithelial layer of the ectoneural region (arrowheads). (e) Longitudinal parasagittal section of a radial nerve cord showing stained cells in both the hyponeural (arrow) and ectoneural (arrowhead) regions. Note that in the hyponeural region, the stained cells are clustered on either side of the transverse hemal strand. (f) Higher magnification image of the boxed region in (e), showing stained cells in the hyponeural region (arrows) and the subcuticular epithelial layer of the ectoneural region (arrowheads). (g) Stained cells in the body wall external epithelium proximal to the marginal nerve and in an adjacent tube foot (see also figure 7). CONR, circumoral nerve ring; Ec, ectoneural region; Hy, hyponeural region; MN, marginal nerve; RHS, radial hemal strand; RNC, radial nerve cord; TF, tube foot; THS, transverse hemal strand. Scale bars: 18 µm in (f); 30 µm in (b,d,g); 60 µm in (a), (a) inset, (c,e).

Figure 6.

Localization of ArSSP2 mRNA in the digestive system of A. rubens using in situ hybridization. (a) Transverse section through the central disc region showing stained cells in the peristomial membrane and in the circumoral nerve ring that surrounds the peristomial membrane. (b) Higher magnification image of the boxed region shown in (a), showing stained cells (arrowheads) in the external epithelium of the peristomial membrane. (c) Transverse section through the central disc region showing sparsely distributed stained cells in the cardiac stomach and pyloric stomach. (d) Higher magnification image of the boxed region of the pyloric stomach in (c), showing elongate shaped stained cells (arrowhead) in the mucosal layer and roundish shaped stained cells (arrow) close to the position of basi-epithelial nerve plexus. (e) Higher magnification image of the boxed region of the cardiac stomach in (c), showing elongate shaped stained cells (arrowhead) in the mucosal layer and roundish shaped stained cells (arrows) close to the position of the basi-epithelial nerve plexus. (f) Transverse section of a pyloric duct showing elongate shaped stained cells (arrowhead) and roundish shaped stained cells (arrow) located on both the aboral (upper) and oral (lower) sides. (g) Higher magnification image of the boxed region of (f) showing elongate shaped stained cells (arrowheads) in the mucosal layer. (h) Transverse section of an arm showing stained cells in a pyloric caecum. (i) Higher magnification image of the boxed region in (h), showing elongate shaped stained cells (arrowhead) in the mucosal layer and roundish shaped stained cells (arrow) close to the position of basi-epithelial nerve plexus. BNP, basi-epithelial nerve plexus; CONR, circumoral nerve ring; CS, cardiac stomach; Lu, lumen; Mu, mucosal layer; PC, pyloric caecum; PM, peristomial membrane; PS, pyloric stomach. Scale bars: 18 µm in (b); 30 µm in (a,e); 36 µm in (g); 60 µm in (d,f,i); 120 µm in (c,h).

Figure 7.

Localization of ArSSP2 mRNA in the tube feet and body wall of A. rubens using in situ hybridization. (a) Longitudinal section of a tube foot showing staining in both the stem and disc regions. (b) Higher magnification image of the stem region of (a), showing stained cells (arrowheads) in a subepithelial position. (c) Higher magnification image of the boxed region of (a) at the junction between the stem and the disc, showing a cluster of stained cells around the basal nerve ring. (d) Transverse section of the aboral body wall of an arm showing stained cells in the circular muscle layer proximal to the apical muscle. (e) Higher magnification image of the boxed region of (d), showing stained cells (arrowheads) in the circular muscle layer, which is located beneath the coelomic epithelium. (f) Transverse section of an arm showing staining in the external epithelium of the body wall. (g) Higher magnification image of the boxed region of (f), showing stained cells (arrowheads) in the external epithelium of the body wall. AM, apical muscle; BNR, basal nerve ring; CE, coelomic epithelium; CT, collagenous tissue; Di, disc; Ep, epithelium; ML, muscle layer; SNP, subepithelial nerve plexus; St, stem. Scale bars: 32 µm in (b); 60 µm in (c,e,f,g); 120 µm in (a,d).

Figure 8.

Localization of ArSS2 in the nervous system of A. rubens using immunohistochemistry. (a) Transverse section of the radial nerve cord showing ArSS2-immunoreactivity (immunostaining) in the hyponeural (arrows) and ectoneural (arrowheads) regions. In the ectoneural region, immunostained cells are located in the subcuticular epithelial layer, largely concentrated laterally, and immunostained fibres are present in the neuropile. The inset shows an absence of immunostaining in sections of radial nerve cord incubated with ArSS2 antiserum pre-absorbed with the ArSS2 peptide antigen, demonstrating the specificity of immunostaining observed in sections incubated with the ArSS2 antiserum. (b) Higher magnification image of the boxed region of (a), showing immunostained bipolar shaped cells in the subcuticular epithelium of the ectoneural region (arrowheads), immunostained roundish shaped monopolar cells in the hyponeural region (arrows) and immunostained fibres in the neuropile (asterisk). (c) Immunostaining in the marginal nerve (arrowhead) and immunostained fibres in the lateral motor nerve (arrows). (d) Longitudinal parasagittal section of a radial nerve cord showing immunostained cells in the hyponeural region (arrows) and ectoneural region (arrowheads). In the hyponeural region, immunostained cells are largely clustered near to the transverse hemal strand, whereas in the ectoneural region, immunostained cells are distributed evenly along the length of the radial nerve cord. Immunostained fibres are present throughout the ectoneural neuropile (asterisk). (e) Higher magnification image of the boxed ectoneural region of (d), showing immunostained bipolar shaped cells in the subcuticular epithelium (arrowheads) and immunostained fibres in the ectoneural neuropile (asterisk). (f) Higher magnification image of the boxed hyponeural region of panel (d), showing immunostained roundish monopolar cells. (g) Transverse section of the circumoral nerve ring showing immunostained cells in the hyponeural and ectoneural regions. Immunostained fibres are present in the ectoneural neuropile (asterisk) and are continuous with immunostained fibres in the basi-epithelial plexus of the adjacent peristomial membrane. (h) Higher magnification image of the boxed ectoneural region of (g), showing immunostained bipolar shaped cells (arrowheads) in the subcuticular epithelium. (i) Higher magnification image of the boxed hyponeural region of (g), showing immunostained roundish monopolar cells (arrows) and immunostained fibres in the adjacent ectoneural neuropile (asterisk). CONR, circumoral nerve ring; Ec, ectoneural region; Hy, hyponeural region; MN, marginal nerve; PM, peristomial membrane; THS, transverse hemal strand; RHS, radial hemal strand; RNC: radial nerve cord; TF, tube foot. Scale bars: 4.8 µm in (h); 8 µm in (b,e,f,i); 32 µm in (a), (a) inset, (d,g); 36 µm in (c).

Figure 9.

Localization of ArSS2 in the digestive system of A. rubens using immunohistochemistry. (a) Transverse section through the central disc showing ArSS2-immunoreactivity (immunostaining) in the peristomial membrane and the circumoral nerve ring. Immunostained cells are distributed sparsely in the external epithelial layer of the peristomial membrane and stained fibres are present in the underlying basi-epithelial nerve plexus and contiguous with the immunostained ectoneural region of the circumoral nerve ring. (b) Higher magnification image of the boxed region of peristomial membrane of (a), showing immunostained cells (arrowheads) in the external epithelial layer and immunostained fibres in the underlying basi-epithelial nerve plexus (arrow). (c) Transverse section through the central disc region showing immunostained fibres in the cardiac stomach and pyloric stomach regions. (d) Higher magnification image of the boxed region of the pyloric stomach in (c), showing immunostained fibres in the basi-epithelial nerve plexus. (e) Higher magnification image of the bottom boxed region of the cardiac stomach in (c), showing one bipolar shaped cell (arrowhead) in the mucosa layer and immunostained fibres in the basi-epithelial nerve plexus. (f) Transverse section of a pyloric duct showing immunostained fibres located on the aboral and oral sides, but with immunostaining more prominent on the oral side. (g) Higher magnification image of the boxed region of (f), showing immunostained fibres in the basi-epithelial nerve plexus on the oral side of the pyloric duct. (h) Transverse section of a starfish arm showing immunostaining present in a pyloric caecum, with region-specific variation in the intensity of immunostained fibres in the basi-epithelial nerve plexus. (i) Higher magnification image of the boxed region of the pyloric caecum in (h), showing immunostained cells (arrow) in the mucosal layer and immunostained fibres in the basi-epithelial nerve plexus. BNP, basi-epithelial nerve plexus; CONR, circumoral nerve ring; CS, cardiac stomach; CT, collagenous tissue; Lu, lumen; Mu, mucosa; PC, pyloric caecum; PD, pyloric duct; PM, peristomial membrane; PS, pyloric stomach. Scale bars: 12 µm in (b,d,eg); 24 µm in (i); 48 µm in (a), (h); 60 µm in (f); 120 µm in (c).

Figure 10.

Localization of ArSS2 in the tube feet, body wall and terminal tentacle of A. rubens using immunohistochemistry. (a) Longitudinal section of a tube foot showing ArSS2-immunoreactivity (immunostaining) in the subepithelial nerve plexus along the length of the stem and extending into the basal nerve ring of the disc region. (b) Higher magnification image of the lower boxed region of (a), showing immunostaining in the basal nerve ring of the disc region. (c) Higher magnification image of the upper boxed region of (a), showing immunostaining in the subepithelial nerve plexus along the length of the stem. (d) Transverse section of an arm tip, showing immunostaining in the terminal tentacle, lateral lappet, optic cushion and the surrounding body wall epithelium (arrow). (e) High magnification image of the lower boxed region of (d), showing immunostained bipolar shaped cells in the folded external epithelial layer of the terminal tentacle (arrowheads). (f) High magnification image of the upper boxed region of (d), showing immunostained fibres in the basi-epithelial nerve plexus (arrow) of the body wall epithelium adjacent to the terminal tentacle and immunostained cells in a lateral lappet (arrowheads). (g) Transverse section of an arm showing immunostaining in a pedicellaria and the nerve plexus beneath the external epithelium of the body wall. (h) High magnification image of the upper boxed region of (g), showing immunostained fibres between muscles of a pedicellaria. (i) High magnification image of the lower boxed region of (g), showing immunostained fibres in the nerve plexus beneath the external epithelium of the body wall. BNR, basal nerve ring; BW, body wall; CT, collagenous tissue; CuL, cuticular layer; Ep, epithelium; LL, lateral lappet; ML, muscle layer; OC, optic cushion; Pe, pedicellaria; SNP, subepithelial nerve plexus; TT, terminal tentacle. Scale bars: 12 µm in (b,c,e); 24 µm in (f,h,i); 60 µm in (d,g); 120 µm in (a).

Figure 11.

ArSS2 causes relaxation of in vitro preparations of tube feet and cardiac stomach from A. rubens. (a) Graphs showing the dose-dependent relaxing effect of ArSS2 on tube foot preparations at concentrations ranging from 10⁻¹⁰ to 10⁻⁶ M. The responses are expressed as the mean percentage reversal (±s.e.m.; n = 6) of the contraction induced by 10⁻⁶ M acetylcholine (ACh). (b) Upper: representative recording showing that ArSS2 causes dose-dependent relaxation of a cardiac stomach preparation that was pre-contracted with seawater containing 3 × 10⁻² M added KCl. Lower: graph showing the dose-dependent relaxing effect of ArSS2 on cardiac stomach preparations at concentrations ranging from 10⁻¹⁰ to 10⁻⁶ M. The responses are expressed as the mean percentage reversal (±s.e.m.; n = 10) of the contraction induced by 3 × 10⁻² M KCl. (c) Representative recording of a cardiac stomach preparation that compares the relaxing effects of ArSS2, the SALMFamide-type neuropeptide S2 and asterotocin (Ast) at a concentration of 10⁻⁷ M. As in (b), the cardiac stomach preparation was pre-contracted with KCl-supplemented seawater prior to application of the neuropeptides. The graph below compares the effects of S2, ArSS2 and Ast on cardiac stomach preparations at a concentration of 10⁻⁷ M, expressed as mean percentages (±s.e.m.; n = 6) with the relaxing effect of S2 defined as 100%. The relaxing effect of Ast is significantly larger than the effect of S2 (two-tailed Student's t-test; p = 0.0055; n = 6) and the effect of ArSS2 (two-tailed Student's t-test; p = 0.0250; n = 6). The relaxing effect of ArSS2 is significantly larger than the effect of S2 (two-tailed Student's t-test; p = 0.0303; n = 6). (d) Representative recording of a cardiac stomach preparation that compares the relaxing effects of S2, ArSS2 and Ast at a concentration of 10⁻⁶ M. As in (c), the cardiac stomach preparation was pre-contracted with KCl-supplemented seawater prior to application of the neuropeptides. The graph below compares the effects of S2, ArSS2 and Ast on cardiac stomach preparations at a concentration of 10⁻⁶ M, expressed as mean percentages (±s.e.m.; n = 6) with the relaxing effect of S2 defined as 100%. The relaxing effect of Ast is significantly larger than the effect S2 (two-tailed Student's t-test; p = 0.0255; n = 6) and the effect of ArSS2 (two-tailed Student's t-test; p = 0.0483; n = 6). There is not a significant difference in the magnitude of the relaxing effects of ArSS2 and S2.

Figure 12.

In vivo injection of ArSS2 triggers cardiac stomach eversion in A. rubens. (a) Temporal dynamics of ArSS2 (red; 10 µl of 10⁻³ M) and asterotocin (Ast) (black; 10 µl of 10⁻³ M) induced cardiac stomach eversion. The graph shows the mean area (±s.e.m.; n = 10) of the cardiac stomach everted expressed as a percentage of the area of the central disc region at 1 min intervals over a 10 min period following injection of ArSS2 or Ast. (b) Photographs from video recordings of the experiment in (a) showing representative control (water-injected) starfish (i–iii), ArSS2-injected starfish (iv–vi) and Ast-injected starfish (vii–ix) at 0, 5 and 10 min after injection. The cardiac stomach eversion area is labelled with a dashed line in (v,vi,viii,ix).