Molecular evolution and functional characterisation of tunicate xenobiotic receptors
<p>Marine microorganisms generate a wide range of ’bioactive’ compounds that can have far-reaching effects on biological and ecological processes. Metazoans have developed specialised biochemical pathways that metabolise and eliminate potentially toxic chemicals (xenobiotics) from their bodies. The vertebrate xenobiotic receptor, pregnane X receptor (PXR), is a ligand-activated nuclear receptor transcription factor regulating expression of multiple detoxification genes. Ligand-binding domains (LBDs) of vertebrate PXR orthologues may have adaptively evolved to bind toxins typically encountered by these organisms. Marine invertebrate filter-feeders are exposed to relatively high concentrations of xenobiotics associated with their diet. Tunicates (phylum: Chordata) are of particular interest as they form the sister clade to the Vertebrata. Genomes of the solitary tunicate Ciona intestinalis and the colonial tunicate Botryllus schlosseri both encode at least two xenobiotic receptors that are orthologues to both the vertebrate vitamin D receptor (VDR) and PXR. Pursuing the idea that tunicate xenobiotic receptors (VDR/PXR) may adaptively evolve to bind toxic chemicals commonly present in an organism’s environment, this thesis aims to identify if: (i) adaptive evolution is acting on putative tunicate VDR/PXR orthologues to enhance binding of dietary xenobiotics; (ii) these receptors are activated by dietary xenobiotics (e.g. microalgal biotoxins) and; (iii) tunicate VDR/PXR LBDs can be used as sensor elements in yeast bioassays for the detection of both natural and synthetic bioactive compounds. To identify genetic variation and to search for evidence of positive selection, next-generation sequencing was performed on three tunicate VDR/PXR orthologues genes. Recombinant yeast (Saccharomyces cerevisiae) cell lines were developed for the functional characterisation of tunicate VDR/PXR LBDs. These tunicate VDR/PXR LBD-based yeast bioassays were utilised to detect known microalgal biotoxins, natural bioactive compounds, and environmental contaminants. Next-generation sequencing revealed both an unusually high genetic diversity and strong purifying selection in VDR/PXR orthologues from C. intestinalis and B. schlosseri. Single-base-deletion allelic variants were found in C. intestinalis VDR/PXR orthologues resulting in predicted proteins having a DNA-binding domain but lacking a LBD. The persistence of these variants may reflect constitutive expression of detoxification genes as a selective advantage in the marine environment. To assess the functional characteristics of tunicate VDR/PXR orthologues, recombinant yeast cell lines were developed that express VDR/PXRα LBDs from C. intestinalis and B. schlosseri. These chimeric proteins mediate liganddependent expression of a lacZ reporter gene which encodes an easily assayed enzyme (β-galactosidase). These yeast bioassays were highly sensitive towards both synthetic and natural toxins (coefficients of variance, CV <25%). Microalgal biotoxins (okadaic acid and portimine) were two orders of magnitude more potent than synthetic chemicals, which was consistent with the hypothesis that tunicate xenobiotic receptors can bind marine bioactive compounds frequently present in a filter-feeder’s diet. Following these functional studies, the yeast bioassays were tested in a more applied context by screening the following compounds: (i) natural bioactive compounds that represent promising compounds for drug development and; (ii) synthetic chemicals that are common environmental pollutants. Of the 34 compounds tested, 30 were active in the tunicate yeast bioassays. The yeast bioassays were particularly sensitive towards a small number (n = 11) of marine and terrestrial bioactive compounds (CJ-13-014, CJ-13-104, thysanone and naringin) and emerging contaminants such as pharmaceuticals (ketoconazole), antifungals (radicicol), preservatives (butylated hydroxtoluene) and surfactants (oil dispersants), generating CV values <25%. Activities of the remaining 19 compounds were highly variable and appeared to depend on several factors, such as solvent used, duration of exposure and type of recombinant protein expressed (e.g. C. intestinalis versus B. schlosseri VDR/PXRα). In conclusion, the yeast bioassay developed in this thesis, with further development, may provide a template for novel bioassays that may find application in routine microalgal biotoxin testing and environmental monitoring. These bioassays may also assist in the identification of marine bioactive compounds as drug lead compounds.</p>