Persistence of duplicated PAC1 receptors in the teleost, Sparus auratus

Abstract

Background: : Duplicated genes are common in vertebrate genomes. Their persistence is assumed to be either a consequence of gain of novel function (neofunctionalisation) or partitioning of the function of the ancestral molecule (sub-functionalisation). Surprisingly few studies have evaluated the extent of such modifications despite the numerous duplicated receptor and ligand genes identified in vertebrate genomes to date. In order to study the importance of function in the maintenance of duplicated genes, sea bream (Sparus auratus) PAC1 receptors, sequence homologues of the mammalian receptor specific for PACAP (Pituitary Adenylate Cyclase-Activating Polypeptide), were studied. These receptors belong to family 2 GPCRs and most of their members are duplicated in teleosts although the reason why both persist in the genome is unknown.

Results: : Duplicate sea bream PACAP receptor genes (sbPAC1A and sbPAC1B), members of family 2 GPCRs, were isolated and share 77% amino acid sequence identity. RT-PCR with specific primers for each gene revealed that they have a differential tissue distribution which overlaps with the distribution of the single mammalian receptor. Furthermore, in common with mammals, the teleost genes undergo alternative splicing and a PAC1Ahop1 isoform has been characterised. Duplicated orthologous receptors have also been identified in other teleost genomes and their distribution profile suggests that function may be species specific. Functional analysis of the paralogue sbPAC1s in Cos7 cells revealed that they are strongly stimulated in the presence of mammalian PACAP27 and PACAP38 and far less with VIP (Vasoactive Intestinal Peptide). The sbPAC1 receptors are equally stimulated (LOGEC50 values for maximal cAMP production) in the presence of PACAP27 (-8.74 +/- 0.29 M and -9.15 +/- 0.21 M, respectively for sbPAC1A and sbPAC1B, P > 0.05) and PACAP38 (-8.54 +/- 0.18 M and -8.92 +/- 0.24 M, respectively for sbPAC1A and sbPAC1B, P > 0.05). Human VIP was found to stimulate sbPAC1A (-7.23 +/- 0.20 M) more strongly than sbPAC1B (-6.57 +/- 0.14 M, P < 0.05) and human secretin (SCT), which has not so far been identified in fish genomes, caused negligible stimulation of both receptors.

Conclusion: : The existence of functionally divergent duplicate sbPAC1 receptors is in line with previously proposed theories about the origin and maintenance of duplicated genes. Sea bream PAC1 duplicate receptors resemble the typical mammalian PAC1, and PACAP peptides were found to be more effective than VIP in stimulating cAMP production, although sbPAC1A was more responsive for VIP than sbPAC1B. These results together with the highly divergent pattern of tissue distribution suggest that a process involving neofunctionalisation occurred after receptor duplication within the fish lineage and probably accounts for their persistence in the genome. The characterisation of further duplicated receptors and their ligands should provide insights into the evolution and function of novel protein-protein interactions associated with the vertebrate radiation.

Figure 1

Nucleotide and predicted protein sequence of sbPAC₁A receptor. Seven TM domains are highlighted with black lines and the conserved N-terminal cysteine residues and putative N-glycosylation sites are annotated respectively by a circle and by "+". The microsatellite identified in the 5'UTR is underlined and the stop codon is indicated by "*".

Figure 2

Nucleotide and predicted protein sequence of sbPAC₁B receptor. The seven TM domains are highlighted with black lines and the conserved N-terminal cysteine residues and putative N-glycosylation sites are annotated respectively by a circle and by "+". The microsatellite identified in the 5'UTR is underlined and stop codon is indicated by "*".

Figure 3

Multiple sequence alignment of vertebrate PAC₁ receptors. The seven TM domains are boxed, the P-D-I/M motif indicated by "○"and the R-L-A-R motif by "*". The alternative receptor hop splice isoform is indicated by a dotted box and the putative signal peptide sequence with a double ended arrow. Conserved cysteines and putative N-glycosylation sites in the N-terminal domain are indicated by "●" and "+", respectively. Accession numbers of sequences used in the multiple amino acid sequence alignment were; human (Homo sapiens, P41586), rat (Rattus norvegicus, P32215), bovine (Bos taurus, Q29627), Rana (Rana ridibunda, Q90Y07), Xenopus (Xenopus laevis, Q9PTK1), Goldfish (Carassius auratus, O7376, Takifugu (Takifugu rubripes, AJ494861 for PAC₁A and AJ490804 for PAC₁B) and sea bream sbPAC₁A (AJ514930) and sbPAC₁B (AJ514931).

Figure 4

Phylogenetic analysis of the vertebrate PAC₁ members. This figure represents the upper quartile of the consensus tree, for full image and details please refer to Additional file 2. Sea bream PAC₁ receptors are in bold and the novel teleost PAC₁ receptors identified are refereed by their Ensembl nomenclature. PAC₁ accession numbers are: Human, P41586; Mouse, P70205; Rana, Q90Y07; Takifugu1A, AJ494861, Takifugu1B, AJ490804; zebrafish1A, NM_001013444 and zebrafish1B, XM_701685.

Figure 5

Short-range linkage analysis of the PAC₁ receptors. The gene environment of Takifugu, chicken and human PAC₁ regions is represented. Homologue genes present in all three study organisms are represented by open boxes and black boxes represent flanking genes that had no corresponding homologues. Genes are named according to the HUGO annotation and PAC₁ genes are in bold. Takifugu scaffolds (assembly 4) are represented using the NIX annotation [40]. The gene environment of PAC₁ in Xenopus is not represented since it was very incomplete. The Takifugu gene environment is represented and the scale corresponds approximately to 10 Kb and the relative position of linked genes in the chromosomes of the chicken and human is given.

Figure 6

Tissue expression of the duplicate sbPAC₁A and sbPAC₁B receptors. Expression was carried out by RT-PCR using specific primers to amplify potential receptor transcripts. PCR products of approximately 1200 bp, 1000 bp and 1280 bp were obtained and corresponded to sbPAC₁A, sbPAC₁B and sbPAC₁Ahop1 isoform, respectively.

Figure 7

Production of maximal cAMP by the recombinant Cos7 cell expressing sbPAC₁A (A) and sbPAC₁B (B). Transfected cells expressing the receptors were incubated with different concentrations of PACAP₂₇ (□) PACAP₃₈ (△) and VIP (○). Data was normalized and receptor potency profiles is given as percentage of intracellular cAMP produced per well and error bars indicate the ± SEM of a minimum of three independent experiments performed in duplicate. Only the lower error bars are represented.