The serendipitous origin of chordate secretin peptide family members

Abstract

Background: The secretin family is a pleotropic group of brain-gut peptides with affinity for class 2 G-protein coupled receptors (secretin family GPCRs) proposed to have emerged early in the metazoan radiation via gene or genome duplications. In human, 10 members exist and sequence and functional homologues and ligand-receptor pairs have been characterised in representatives of most vertebrate classes. Secretin-like family GPCR homologues have also been isolated in non-vertebrate genomes however their corresponding ligands have not been convincingly identified and their evolution remains enigmatic.

Results: In silico sequence comparisons failed to retrieve a non-vertebrate (porifera, cnidaria, protostome and early deuterostome) secretin family homologue. In contrast, secretin family members were identified in lamprey, several teleosts and tetrapods and comparative studies revealed that sequence and structure is in general maintained. Sequence comparisons and phylogenetic analysis revealed that PACAP, VIP and GCG are the most highly conserved members and two major peptide subfamilies exist; i) PACAP-like which includes PACAP, PRP, VIP, PH, GHRH, SCT and ii) GCG-like which includes GCG, GLP1, GLP2 and GIP. Conserved regions flanking secretin family members were established by comparative analysis of the Takifugu, Xenopus, chicken and human genomes and gene homologues were identified in nematode, Drosophila and Ciona genomes but no gene linkage occurred. However, in Drosophila and nematode genes which flank vertebrate secretin family members were identified in the same chromosome.

Conclusions: Receptors of the secretin-like family GPCRs are present in protostomes but no sequence homologues of the vertebrate cognate ligands have been identified. It has not been possible to determine when the ligands evolved but it seems likely that it was after the protostome-deuterostome divergence from an exon that was part of an existing gene or gene fragment by rounds of gene/genome duplication. The duplicate exon under different evolutionary pressures originated the chordate PACAP-like and GCG-like subfamily groups. This event occurred after the emergence of the metazoan secretin GPCRs and led to the establishment of novel peptide-receptor interactions that contributed to the generation of novel physiological functions in the chordate lineage.

Figure 1

Phylogenetic position of the non-vertebrate genomes analysed. Simplified phylogeny of the metazoan evolution indicating the relative position of the early metazoa (Porifera and Cnidaria), protostome (Nematoda, Arthropoda, Platyhelminthes, Mollusca, Annelida) and early deuterostome (Echinodermata, Cephalochordata and Urochordata) genomes analysed (adapted from [96-98]). The tunicate Chelyosoma productum is also represented (*) since it is the only invertebrate in which secretin family members have been isolated [7].

Figure 2

Amino acid sequence conservation of vertebrate secretin family mature peptides. The mature peptide sequences were extracted by comparison with the human homologues and only the amino acid (aa) residues 1 to 27 are represented with the exception of the first 5 residues of human GLP1 (P01275). Takifugu GHRH was obtained from [93] (N000079, Assembly_4) and the lamprey GLP2 sequence for proglucagon II was not used since it was found to share very little similarity with other vertebrate members suggesting it has undergone species-specific evolution. Vertebrate peptides are grouped according to their potential origin from a PACAP-like exon or GCG-like exon. Percentage of identity (%ID) for the human homologue is given and the consensus sequences for each peptide group were deduced using the GeneDoc programme [90] and used to generate a PACAP-like and GCG-like subfamily peptide. The most frequent residues within the different peptide groups are annotated in bold and a prototype model sequence for the chordate secretin family members was derived by fusing the conserved PACAP-like and GCG-like subfamily amino acid sequence (overlapping residues are annotated in bold and italics). Accession numbers of the teleost and non-mammalian sequences used are indicated in Table 2. The human precursors are PHM/VIP, P01282; PRP/PACAP, P18509; GHRH, P01286; GCG/GLPs, P01275; GIP, P09681; and SCT, P09683 and mouse (Mus musculus) accession numbers are PHM/VIP, P32648; PRP/PACAP, O70176; GHRH, P16043; GCG/GLPs, P55095; GIP, P48756; and SCT, Q08535.

Figure 3

Proposed evolutionary model of chordate PACAP-like (A) and GCG-like (B) members. Percentage of amino acid sequence identity/similarity of the different peptide groups is indicated and gene organisation of the coding region (excluding occasional species-specific gene organisation) is represented. Secretin family members are proposed to have evolved via exon and gene/chromosome duplication events from a common ancestor exon in the chordate radiation. Similarity between the deduced consensus sequences of the peptide groups in the same subfamily is higher than 62% within the vertebrate GCG-like members and 66% for the PACAP-like subfamily with the exception of SCT in which only the mammalian and chicken members have been identified. Boxes represent exons and lines introns and coding exons are indicated by the peptide abbreviation. Dashed lines indicate undefined evolutionary pathways. (A) Chordate PACAP and PRP and vertebrate VIP and PH share the same gene precursor and GHRH and SCT are encoded by a single exon. PACAP and VIP share the highest amino acid conservation and SCT is the most divergent and to date has only been identified in tetrapods. (B) Vertebrate GCG, GLP1 and GLP2 are encoded in the same gene precursor which arose by exon duplication events. GIP is encoded by a single exon in a different precursor which has a similar gene organisation with GCG/GLP precursor.

Figure 4

Evolutionary analysis of the chordate secretin family members. The maximum likelihood (ML) optimal tree topology is presented and was constructed with Phyml 3.0 [99]. ML bootstrap values higher than 50% are indicated at nodes and to facilitate interpretation a hypothetical root was added to the tree between the PACAP-like and GCG-like clades based upon gene structure evidence and proposed models for secretin family evolution. The different peptide groups are indicated and teleost duplicate genes are marked by a and b; Xenopus GLP1 exons by a, b and c. Accession numbers of the sequences used are described in Table 2 and for human and mouse members are: PHM/VIP (P01282 and P32648); PRP/PACAP (P18509 and O70176); GHRH (P01286 and P16043); GCG/GLPs (P01275 and P55095); GIP (P09681 and P48756); and SCT (P09683 and Q08535), respectively.

Figure 5

Gene environment comparisons of the GCG/GLPs, PRP/PACAP, GHRH and SCT genes in Takifugu, Xenopus, chicken and human. Homologue genes were identified using sequence similarity approaches with the Takifugu genes. Takifugu scaffolds are named according to the Assembly 4 available at [93] and have a direct correspondence with ENSEMBL (eg: N000046 corresponds to Takifugu Ensembl scaffolds_46). Genes were named based on HUGO annotation and the size of the genome regions analysed indicated within brackets. Genes are represented by boxes and genomic regions are indicated by lines. The figure is not drawn to scale and genes are positioned according to their relative distance in the genome assembly. For simplicity, only homologue genes are represented and GCG/GLP, PRP/PACAP, GHRH and SCT genes are edited in bold and underlined.

Figure 6

Comparisons of conserved flanking genes of human PRP/PACAP, GHRH and GCG/GLP with the putative homologue regions in Ciona, Drosophila and C. elegans. Non-vertebrate genomes were accessed using the ENSEMBL annotation. Accession numbers of the human neighbouring genes: KCNH7 (EAX11346); interferon induced with helicase C domain 1 (IFIH1, EAX11352); Solute carrier family 4, sodium bicarbonate transporter, member 10 (SLC4A10, AAI36270), viral oncogene yes-1 homolog 1 (YES1, NP_005424), Methyltransferase like 4 (METTL4, AAI36768), ribophorin II, (RPN2, NP_002942).