The repertoire of olfactory C family G protein-coupled receptors in zebrafish: candidate chemosensory receptors for amino acids

Abstract

Background: Vertebrate odorant receptors comprise at least three types of G protein-coupled receptors (GPCRs): the OR, V1R, and V2R/V2R-like receptors, the latter group belonging to the C family of GPCRs. These receptor families are thought to receive chemosensory information from a wide spectrum of odorant and pheromonal cues that influence critical animal behaviors such as feeding, reproduction and other social interactions.

Results: Using genome database mining and other informatics approaches, we identified and characterized the repertoire of 54 intact "V2R-like" olfactory C family GPCRs in the zebrafish. Phylogenetic analysis - which also included a set of 34 C family GPCRs from fugu - places the fish olfactory receptors in three major groups, which are related to but clearly distinct from other C family GPCRs, including the calcium sensing receptor, metabotropic glutamate receptors, GABA-B receptor, T1R taste receptors, and the major group of V2R vomeronasal receptor families. Interestingly, an analysis of sequence conservation and selective pressure in the zebrafish receptors revealed the retention of a conserved sequence motif previously shown to be required for ligand binding in other amino acid receptors.

Conclusion: Based on our findings, we propose that the repertoire of zebrafish olfactory C family GPCRs has evolved to allow the detection and discrimination of a spectrum of amino acid and/or amino acid-based compounds, which are potent olfactory cues in fish. Furthermore, as the major groups of fish receptors and mammalian V2R receptors appear to have diverged significantly from a common ancestral gene(s), these receptors likely mediate chemosensation of different classes of chemical structures by their respective organisms.

Figure 1

OlfC genes are clustered in the zebrafish genome. Zebrafish OlfC genes were mapped to the Zv6 draft genome assembly. Shown are the positions of predicted OlfC genes in three clusters on chromosomes 11, 17 and 18 as well as the two singlets on chromosomes 16 and 20. Subfamily members are in most cases found in contiguous clusters and often share the same transcriptional orientation. The interrupted coding sequence of OlfCg7 (in brackets) is likely due to the unfinished state of scaffold 2506 in the Zv6 genome assembly. Also shown are non-OlfC RefSeq sequences and zebrafish mRNAs from the UCSC genome browser [78].

Figure 2

Phylogeny of the zebrafish OlfC family. A phylogenetic tree of 57 full-length OlfC sequences (54 genes and 3 pseudogenes) plus other C family GPCRs was constructed using a maximum likelihood (ML) algorithm (see Methods). Bootstrap support is indicated at each node. The full-length OlfC genes were placed into 17 subfamilies, forming clades with 100% bootstrap support. Average percent identity between subfamilies is 46%. CaSR: putative calcium sensing receptor; mGluR: metabotropic glutamate receptors; T1R: T1R-like putative taste receptors.

Figure 3

Phylogeny of C family GPCRs from zebrafish, fugu, goldfish and mouse. A phylogenetic tree of selected zebrafish, fugu, goldfish and mouse C family receptors was constructed using a maximum likelihood algorithm and displayed in radial format. Groups I – V all form clades distinct from the CaSR, T1R, mGluR and GABA-B families. Most of the fish OlfC receptors form a clade (Group I) distinct from the mouse V2R receptors (mmV2R) found in Groups II, IV and V. Group II contains both fish OlfC receptors (zebrafish: drOlfCc1; goldfish: ca5.3; fugu: 735220) and mouse V2R2-like (mmV2R2) sequences. Group III contains receptors from zebrafish (drOlfCa1), goldfish (ca5.24), fugu (179742) and mouse (mmGprc6a). Scale bar indicates the number of amino acid substitution per site.

Figure 4

Sequence logo of the zebrafish OlfC family. Conservation of predicted amino acid sequence for the zebrafish OlfC repertoire is displayed as a sequence logo. In this representation, the relative frequency with which an amino acid appears at a given position is reflected by the height of its one-letter amino acid code in the logo, with the total height at a given position proportional to the level of sequence conservation. For this analysis, the subset of 54 full-length intact zebrafish OlfC amino acid sequences was extracted from the alignment of all 62 OlfCs and gaps were removed with respect to OlfCq6. The regions corresponding to the signal peptide, lobe 1, lobe 2, the cysteine-rich domain and the transmembrane (TM) domains are indicated. Y axis, information content. X axis, residue position with respect to OlfCq6.

Figure 5

Conservation of amino acid binding signature in the OlfC family. Sequence logos were constructed for predicted binding pocket residues in the zebrafish OlfC (drOlfC) and mouse V2R (mmV2R) receptor families, based on an alignment of full-length, intact zebrafish OlfC and mouse V2R amino acid sequences. The over-representation of amino acids at positions equivalent to those predicted to interact with amino acid ligands in zebrafish OlfCa1 [28] is shown graphically. Y axis, information content; X axis, residue position with respect to OlfCa1 (note that positions are non-contiguous). Conserved proximal pocket signature motif sites that are thought to make contact with the ligand are highlighted in bold black, whereas predicted distal binding pocket sites are shown in grey. Additional signature motif residues thought to play a structural role but not contact ligand directly are shown in bold blue (bottom of figure; see the text for details).

Figure 6

Amino acid conservation and selective pressure displayed on a structural model of the OlfC N-terminal ligand-binding domain. Percent identity (a, c, e) and inferred selective pressure (b, d, f) for the zebrafish OlfC family is displayed on a structural model of OlfCa1 [28]. The extracellular N-terminal domain (a, b) is represented by a ribbon. The inner, ligand-binding faces of lobe 1 (c, d) and lobe 2 (e, f) are shown as surface representations. Bound glutamate – the cognate ligand for OlfCa1 [28] – is shown docked in the binding pocket with the alpha carbon and proximal pocket oriented to the left and distal pocket to the right. Residues are colored by percent identity, from 2% (red) to 100% (blue) (a, c, e) or by selective pressure (positive selection shown in red (p < 0.1) or orange (p < 0.2), neutral selection in white, and negative selection in blue-green (p < 0.2) or blue (p < 0.1) (b, d, f).

Figure 7

Sites under positive and negative selection in OlfC coding sequences. A schematic representation of site-by-site selective pressure is shown on the OlfCa1 receptor sequence. Nucleotide alignments were generated from the corresponding amino acid alignment [see Additional file 4]. SLAC analysis shows the probability of sites being under selective pressure (positive selection shown in red (p < 0.1) or orange (p < 0.2), neutral selection in yellow and negative selection in blue-green (p < 0.2) or blue (p < 0.1). The null hypothesis is that a site is neutrally evolving with dN/dS = 1.