Contrasted evolution of the vomeronasal receptor repertoires in mammals and squamate reptiles

Abstract

The vomeronasal organ (VNO) is an olfactory structure that detects pheromones and environmental cues. It consists of sensory neurons that express evolutionary unrelated groups of transmembrane chemoreceptors. The predominant V1R and V2R receptor repertoires are believed to detect airborne and water-soluble molecules, respectively. It has been suggested that the shift in habitat of early tetrapods from water to land is reflected by an increase in the ratio of V1R/V2R genes. Snakes, which have a very large VNO associated with a sophisticated tongue delivery system, are missing from this analysis. Here, we use RNA-seq and RNA in situ hybridization to study the diversity, evolution, and expression pattern of the corn snake vomeronasal receptor repertoires. Our analyses indicate that snakes and lizards retain an extremely limited number of V1R genes but exhibit a large number of V2R genes, including multiple lineages of reptile-specific and snake-specific expansions. We finally show that the peculiar bigenic pattern of V2R vomeronasal receptor gene transcription observed in mammals is conserved in squamate reptiles, hinting at an important but unknown functional role played by this expression strategy. Our results do not support the hypothesis that the shift to a vomeronasal receptor repertoire dominated by V1Rs in mammals reflects the evolutionary transition of early tetrapods from water to land. This study sheds light on the evolutionary dynamics of the vomeronasal receptor families in vertebrates and reveals how mammals and squamates differentially adapted the same ancestral vomeronasal repertoire to succeed in a terrestrial environment.

Fig. 1.—

The VNO of the corn snake. (A) Corn snake. (B) Coronal section of a corn snake VNO stained with hematoxylin and eosin (HE). Full organ (left), scale bar 200 μm; close-ups of the columnar sensory epithelium (top right) and the zone in contact with the outside world (bottom right), scale bars 20 μm; SE, sensory epithelium; NE, nonsensory epithelium; L, lumen. (C) Schematic representation of a head hemi-section with a picture of the fluorescence detected in the VNO and the main olfactory epithelium (MOE) after application of a retrograde tracing dye onto the AOB and main olfactory bulb (MOB). (D,E) Coronal section of the VNO with fluorescence (red) detected after application of the tracing dye onto the AOB. Arrowheads and arrows in (D) indicate groups of potentially immature neurons at the edges and base, respectively, of the neuroepithelium; arrows and dotted frame in (E) indicate dendritic projections to the lumen. DNA is stained with Hoechst (blue). Scale bar in (D), 200 μm and in (E), 50 μm.

Fig. 2.—

The V2R repertoire of the corn snake. (A) Length distribution of 196 corn snake V2R contigs. (B) Distribution of the average read depth coverage in relationship to the length of 196 corn snake V2R contigs (those used for phylogenetic analyses are marked in red). (C) Distribution of BLASTX E values in relationship to the length of 196 snake V2R contigs. (D) Identity statistics (based on segments of a multiple alignment of 196 snake V2R contigs) for aa and nt sequence variability analyses and estimation of the number of distinct V2R transcripts; see text for details. (E) Absolute and relative sizes of VR gene repertoires, based on the present (red frame) and published studies (Grus et al. 2007; Shi and Zhang 2007; Young and Trask 2007; Date-Ito et al. 2008; Hashiguchi et al. 2008; Ji et al. 2009; Young et al. 2010; Alfoldi et al. 2011).

Fig. 3.—

Evolutionary history of snake and other vertebrate V2Rs. ML tree, based on a multiple alignment of 7TM-domain aa sequences from 66 corn snake V2Rs and 67 representative V2R sequences from five vertebrate species. Branch support values (under RaxML/MetaPIGA/MrBayes) are indicated for major clades. Branches with <50% support values are collapsed. Pink boxes highlight five of the snake-specific expansions. Arrows indicate sequences used for in situ hybridization (figs. 4 and 5). The tree is rooted with taste receptors (TAS1Rs) as an outgroup. Scale bar: Mean number of aa substitutions per site.

Fig. 4.—

Expression of V2R transcripts in the snake vomeronasal sensory epithelium. In situ hybridizations (green) of coronal sections of a female corn snake VNO, with antisense RNA probes for one pseudogene (V2R-EG204) and four likely functional V2R transcripts. DNA is stained with Hoechst (blue). Scale bar, 20 μm.

Fig. 5.—

Monogenic versus nonmonogenic expression of V2Rs. In situ hybridizations with antisense RNA probes for (A) two ABD-subfamily members V2R-EG154 (green) and V2R-EG107 (red), (B) one C (V2R-EG001; red), and one ABD (V2R-EG154; green) subfamily members; coexpression of V2Rs is only observed between family C and family A/B/D members. Left and middle panels show single channels in green and red, and right panel shows the merge plus Hoechst staining of DNA in blue. Scale bar, 10 μm.

Fig. 6.—

Evolutionary history of snake and other vertebrate V1Rs. ML tree, based on a 230-aa multiple alignment of reptilian, mouse, frog, and fish V1Rs. The topology of the RaxML analyses is shown, and branch support values (under RaxML/MetaPIGA/MrBayes) are indicated for major clades. Branches with <50% support values are collapsed. The tree is unrooted and has been arbitrarily oriented for display purposes (there is no outgroup). The dashed arrow indicates the Anolis single V1R. Scale bar: Mean number of aa substitutions per site.