H E A L T H I N N O V A T I O N , O V E R C O M I N G D I S E A S E S A N D P A N D E M I C S

S C I E N T I F I C H I G H L I G H T S

3 0 H I G H L I G H T S 2 0 2 3 I

X-ray crystallography determines the quaternary structure of the glucocorticoid receptor

Crystallographic data has been used to determine the quaternary structure of the glucocorticoid receptor (GR), a ligand-activated transcription factor regulating the transcription of a wide array of genes. The structure provides unique insights into the intramolecular allosteric communication pathways.

The glucocorticoid receptor (GR) is a ligand-activated transcription factor that binds DNA and assembles co- regulator complexes to regulate gene transcription. GR is activated through endocrine signalling by the glucocorticoid hormone cortisol, and is of fundamental importance for development, skeletal growth, behaviour, glucose homeostasis and inflammation. The first GR- targeting medicine was discovered more than 70 years ago. Since then, several GR ligands have been developed, and GR agonists are widely prescribed to treat people with inflammatory and autoimmune diseases. However, severe adverse effects, including diabetes and osteoporosis, have driven a continued interest to identify safer alternatives

The receptor consists of a ligand-binding domain (LBD), a DNA-binding domain (DBD) and a N-terminal domain (NTD). In this work, crystallographic data was collected at beamlines ID30A-1 and ID23-1 to determine the quaternary structure of a GR LBD-DBD protein construct in complex with DNA, ligand and a peptide from the coregulator PGC1α. This is the first high-resolution multidomain structure of a steroid receptor.

The structure reveals how the receptor forms an asymmetric dimer on the DNA and provides a detailed view of the domain interactions within and across the two

monomers (Figure 17). There are two potential positions for the LBD dimer partner (LBD2); in one, the dimer interactions are mediated though regions spanning the N-terminal end of H10/11 and H6-H7 (Figure 17a,b), and in the other, the LBD dimer interactions are mediated through helix 1 (H1) (Figure 17c). Based upon the size of the interfaces, it is proposed that the H10/11 and H6-H7 interaction is the relevant LBD dimer interface.

GR evolved from an estrogen receptor (ER)-like ancestor in a gene duplication event in vertebrates [1]. While isolated ER LBD forms a strong canonical dimer, the GR LBD is unable to form the same dimer due to a conserved C-terminal extension [2]. The GR LBD dimer interface in the quaternary structure is significantly smaller compared to the ER dimer interface. This agrees with biochemical data on isolated LBD constructs where the ER LBD dimer forms at sub nM concentrations [3], whereas the GR LBD dimer only forms at μM concentrations [4]. Unlike ER, GR has been reported to bind either as a monomer or as a dimer on the DNA. It is plausible that if GR had a strong LBD dimer interface, it would skew the oligomeric state towards the dimeric form, and that the evolution of the C-terminal extension therefore provided a path to diversify signalling.

To build the ligand-specific signalling knowledge, the quaternary structures of GR in complex with the non- steroidal ligand velsecorat was compared with the steroid fluticasone furoate (Figure 18). Looking at the overall domain arrangement of the structures, the different ligand pharmacophores appear to shift the position of LBD2 relative to LBD1, reducing the interaction surface with the DBD dimer. The ligand-specific structural rearrangements were further corroborated with hydrogen-deuterium exchange experiments and binding studies using labelled DNA.

Fig. 17: The GR quaternary structure of a GR dimer binding on DNA in complex with velsecorat (magenta) and a peptide from the coregulator PGC1α. In the crystal lattice, there are two potential interfaces for the LBD dimer: one mediated by

helices 6, 7 and 10/11 (a,b), and one mediated by helix 1 (c).