Structure-function relationships of retinoic acid receptors from genome-wide to atomic scales. – RAR-SCALES

We will pursue the use of ChIP-seq and RNA-seq to widen the repertoire of natural RAR-occupied sites and to examine the dynamics of paralogue-specific RAR promoter occupancy and gene activation during differentiation of ES and F9 cells. We will use EMSA to accurately identify RAR/RXR binding sequences in regions devoid of identifiable DR elements. The nucleotides required for binding will be precisely determined by EMSA assays and the elements further characterized by fluorescence spectroscopy and FRET to obtain quantitative binding affinity and information on the polarity of the heterodimer on the DNA. We will analyze the functional properties of the non-canonical sites both as isolated elements and in the context of their natural promoters by a series of mixing and matching exchanges of canonical and non-canonical elements. We will use genomic profiling and siRNA-mediated knockdown to determine their cofactor requirements. X-ray crystallography will be used to determine the atomic resolution structure of the RAR/RXR DBDs on natural non-consensus DR0-5 and IR0 binding sites to understand the effects of variations in the hexanucleotide repeats, flanking and spacer sequences on recognition. Small angle X-ray scattering (SAXS) combained with X-ray crystallography will be used to determine the topological organization of the full length RAR/RXR/cofactor heterodimers on this diverse series of binding elements. This proposal will allow the characterisation of RAR/RXR binding from the genome-wide to nucleotide and atomic scales. Through the combination of functional and structural analysis, we expect to identify and understand the regulation of RAR activity through allosteric control by the primary DNA sequence and topological organisation of the binding elements and by cofactor interactions.

See annex in report

We will pursue this work in order to determine the structures of the RAR-RXR bound to the DR0, IR0 and DR8 response elements. This will allow us to understand the structural basis of the observed allosteric control imparted by the half-site spacing in the response element.

Moutier et al., JBC in press , see annex in report

All-trans retinoic acid (RA), the naturally active vitamin A metabolite exerts a wide range of effects on vertebrate development and plays a critical role in the homeostasis and physiopathology of adult tissues. This simple molecule exerts most of its pleiotropic effects through interaction with three members of the nuclear receptor (NR) superfamily, the all-trans retinoic acid receptors (RARs) that heterodimerise with the retinoid X receptors (RXR) to act as RA-dependent transcriptional regulators activating or repressing target genes. The RAR/RXR heterodimers regulate gene expression by binding to response elements, the best characterised of which comprise direct repeats of the consensus 5’-RGKTCA-3’ separated by 1, 2 or 5 nucleotides. We have used chromatin immunoprecipitation coupled to array hybridisation (ChIP-chip) and sequencing (ChIP-seq) to identify RAR binding sites and target genes in mouse embryo fibroblasts, embryonic stem cells and F9 embryonal carcinoma cells. These results identify a large set of natural RAR binding sites and illustrate their diversity in primary sequence and half-site topology. Coupling this information to gene expression profiling during neuronal ES or F9 cell diferentiation identified target promoters that are either activated, repressed or non-responsive to RA, however there is no obvious relationship between the type of RAR binding element and the transcriptional response.
The objective of this proposal is to profile RAR-and RXR genomic occupancy during differentiation, identify and characterise non-canonical binding elements and determine if and how the primary sequence and topological organisation of the RAR binding sites may act as allosteric regulators of the RAR/RXR. To test this idea we propose a complementary series of functional and structural experiments. We will pursue the use of ChIP-seq and RNA-seq to widen the repertoire of natural RAR-occupied sites and to examine the dynamics of paralogue-specific RAR promoter occupancy and gene activation during differentiation of ES and F9 cells. We will use EMSA to accurately identify RAR/RXR binding sequences in regions devoid of identifiable DR elements. The nucleotides required for binding will be precisely determined by EMSA assays and the elements further characterized by fluorescence spectroscopy and FRET to obtain quantitative binding affinity and information on the polarity of the heterodimer on the DNA. We will analyze the functional properties of the non-canonical sites both as isolated elements and in the context of their natural promoters by a series of mixing and matching exchanges of canonical and non-canonical elements. We will use genomic profiling and siRNA-mediated knockdown to determine their cofactor requirements. X-ray crystallography will be used to determine the atomic resolution structure of the RAR/RXR DBDs on natural non-consensus DR0-5 and IR0 binding sites to understand the effects of variations in the hexanucleotide repeats, flanking and spacer sequences on recognition. Small angle X-ray scattering (SAXS) combained with X-ray crystallography will be used to determine the topological organization of the full length RAR/RXR/cofactor heterodimers on this diverse series of binding elements.
This proposal will allow the characterisation of RAR/RXR binding from the genome-wide to nucleotide and atomic scales. Through the combination of functional and structural analysis, we expect to identify and understand the regulation of RAR activity through allosteric control by the primary DNA sequence and topological organisation of the binding elements and by cofactor interactions.