Courtesy of the Kobal Collection.

Signalling through the retinoid receptor pathway is important for a wide variety of developmental processes in vertebrates, including the patterning of the nervous system. There are two families of receptor, the retinoic-acid receptors (RARs) and the retinoid-X receptors (RXRs), which combine to form heterodimers that function as ligand-activated transcription factors. There is strong evidence to suggest that retinoids are the endogenous ligands for at least some of these receptors but, in some cases, the ligands that activate them in vitro have proved difficult to find in the living organism. For example, the RXRs are activated by 9-cis retinoic acid in an in vitro reporter assay, but this compound is barely detectable in vivo, leading to doubts about its eligibility for the role of endogenous RXR activator. Because of this uncertainty, the RXRs have until recently been classified as 'orphan' receptors. Now, in a new study reported in Science , Mata de Urquiza et al. present compelling evidence to show that the polyunsaturated fatty acid docosohexaenoic acid (DHA) is an endogenous RXR ligand.

The researchers began by using a cell-based reporter assay to identify brain extracts that were able to activate the mouse RXR. They found that medium conditioned with adult brain could do this, and that the strongest activity resided in the hippocampus, striatum, motor cortex and cerebellum. Using a series of extraction techniques, they concluded that the ligand responsible was a negatively-charged lipophilic molecule. They purified the activating fraction by HPLC, then analysed the peak fractions using mass spectrometry. Using this technique, they identifed the active component as DHA, and this was further confirmed by testing the purified compound in the cell-based reporter assay.

There are several lines of evidence that make DHA a strong candidate to be an endogenous RXR ligand. First, unlike other ligands that have been previously implicated, it is highly abundant in the brain, making up 30–50% of the total fatty acid content. Second, the effects of DHA deficiency, which include defects in brain maturation and spatial learning, are remarkably similar to those of RXR knockouts. Last, it was found that the activating ability of DHA is affected by mutations that disrupt the ligand binding site of RXR. Therefore, in addition to identifying an endogenous RXR ligand, these observations provide important new information regarding the mode of action of DHA in brain development and function.