Analysis of the goldfish Carassius auratus olfactory epithelium transcriptome reveals the presence of numerous non-olfactory GPCR and putative receptors for progestin pheromones

Abstract

Background: The goldfish (Carassius auratus) uses steroids and prostaglandins as pheromone cues at different stages of the reproductive cycle to facilitate spawning synchronization. Steroid progestin pheromone binding has been detected in goldfish olfactory membranes but the receptors responsible for this specific binding remain unknown. In order to shed some light on the olfactory epithelium transcriptome and search for possible receptor candidates a large set of EST from this tissue were analysed and compared to and combined with a similar zebrafish (Danio rerio) resource.

Results: We generated 4,797 high quality sequences from a normalized cDNA library of the goldfish olfactory epithelium, which were clustered in 3,879 unique sequences, grouped in 668 contigs and 3,211 singletons. BLASTX searches produced 3,243 significant (E-value < e(-10)) hits and Gene Ontology (GO) analysis annotated a further 1,223 of these genes (37.7%). Comparative analysis with zebrafish olfactory epithelium ESTs revealed 1,088 identical unigenes. The transcriptome size of both species was estimated at about 16,400 unigenes, based on the proportion of genes identified involved in Glucose Metabolic Process. Of 124 G-protein coupled receptors identified in the olfactory epithelium of both species, 56 were olfactory receptors. Beta and gamma membrane progestin receptors were also isolated by subcloning of RT-PCR products from both species and an olfactory epithelium specific splice form identified.

Conclusion: The high similarity between the goldfish and zebrafish olfactory systems allowed the creation of a 'cyprinid' olfactory epithelium library estimated to represent circa 70% of the transcriptome. These results are an important resource for the identification of components of signalling pathways involved in olfaction as well as putative targets for pharmacological and histochemical studies. The possible function of the receptors identified in the olfactory system is described. Moreover, the role of olfactory epithelium specific isoforms of classical membrane progestin receptor genes as candidates for preovulatory pheromone sensing is discussed.

Figure 1

Frequency of common transcripts between goldfish and zebrafish olfactory EST libraries. Semi-logarithmic plot of counts of each transcript present in both the goldfish normalized and zebrafish non-normalized EST library. The almost uniform relationship (R = 0.009) is a strong indication of absence of amplification bias during normalization.

Figure 2

Classification of EST according to Gene Ontology. Distribution of goldfish olfactory epithelium unigenes classified using GeneOntology categories of (A) Molecular Function and (B) Biological Process.

Figure 3

Alignment of membrane progestin receptor splice variants from the fish olfactory epithelium. Exon composition of PAQR5b and mPRγ-1 (A, alternating colours) and alignment of two mPRγ-2 splice variants from goldfish olfactory epithelium (B, colours represent corresponding exons 5 and 6 in PAQR5b).

Figure 4

Predicted secondary structure of membrane progestin receptors from the fish olfactory epithelium. Snake-plot of complete PAQR5b and two mPRγ-2 splice variants from fish olfactory epithelia built using SOSUI prediction. Note that TM architecture remains despite significant rearrangement of transmembrane helices in case of mPRγ2b.