**3.1 Sterically impede the agonistic orientation of helix-12 3.1.1 Case study: the progesterone receptor**

Steroidal anti-progestins are typically differentiated structurally from progestins by the presence of a bulky attachment at their position 11 (Madauss, Stewart, and Williams, 2007). Recent publications of the anti-progestin Mifepristone (Raaijmakers, Versteegh, and Uitdehaag, 2009) and the SPRM Asoprisnil (Madauss et al, 2007) clearly demonstrate that the role of this bulky attachment is to clash with helix-12 and preclude it from adopting its required agonistic position. Both studies also demonstrate an important role specifically for Met909 in the agonism/antagonism balance. Met909 sits within helix-12 at the Cterminal end of the ligand binding domain (LBD), and in the classic agonist conformation of the receptor, is oriented toward the ligand binding pocket. Met909 is typically the only helix-12 residue directly in contact with ligands. The nature of the ligand-Met909 interactions appears to be a key determinant of the receptors function (Petit-Topin et al, 2009). Clashes between Met909 and ligands are likely to destabilize helix-12 (Raaijmakers, Versteegh, and Uitdehaag, 2009), which results in a reduced agonistic response. It has even been suggested that the degree of clash with Met909 might correspond directly to the reduction in agonism (Madauss, Stewart, and Williams, 2007), but this has yet to be shown categorically.

Drug Design Approaches to Manipulate the Agonist-Antagonist Equilibrium in Steroid Receptors 225

The binding of testosterone and dihydrotestosterone to AR demonstrate the existence of crucial receptor stabilizing interactions mediated by agonistic ligands. As we will discuss later, the loop between helix-11 and helix-12 is a key region for mediating partial agonism. As shown in figure 2, AR is stabilized by a ligand mediated hydrogen-bond network from Thr877 in helix-11 to the 17β-OH group in the endogenous steroidal agonists to Asn705 in

Hydroxyflutamide is the active metabolite of the androgen receptor antagonist flutamide. Its antagonism appears to be a result of its inability to complete the entire network of stabilizing hydrogen-bonds (Bohl et al, 2005) also shown in figure 2. The result is that Thr877 is left buried in a predominately hydrophobic pocket, destabilizing the receptor and

Fig. 2. Left shows the X-ray structure of DHT bound to AR including full hydrogen-bond network. Right shows a model of hydroxyflutamide bound to AR based on the X-ray

The residue equivalent to AR residue Asn705 in MR is Asn770. Extensive X-ray, SAR and mutation studies have been conducted on Asn770 which demonstrate clearly the existence of a ligand-mediated hydrogen bonding network which is critical for the activation of MR in a similar fashion to the one described for AR (Bledsoe et al, 2005;Hellal-Levy et al, 2000). Agonistic steroidal ligands for GR and MR are typified by 11β-hydroxyl groups which hydrogen bond to Asn564 in GR and Asn770 in MR respectively. Despite the similarity between MR and PR, the endogenous PR agonist progesterone behaves as an antagonist of PR. This appears to at least in part be due to a lack of an 11β-hydroxyl group on progesterone. It is interesting how the lack of the hydroxyl group doesn't disturb the

Another important example of disrupting the function of stabilizing interactions can be seen in the estrogen receptors. In addition to their role in sterically precluding helix-12, SERM side-chains also contain an important basic amine function which is almost ubiquitous

structure of hydroxyflutamide bound to an AR-T877A mutant.

**3.2.2 Additional examples** 

agonistic activity of PR but does MR.

helix-3 and finally to the backbone of Asp890 in the loop itself (Matias et al, 2000).

**3.2 Disrupt the function of direct stabilizing interactions** 

**3.2.1 Case study: the androgen receptor** 

shifting the agonist-antagonist equilibrium.

Introducing bulky groups onto PR modulating non-steroidal scaffolds has also been demonstrated to result in partial agonists on a number of occasions (Jones et al, 2005;Kallander et al, 2010;Thompson et al, 2009;Washburn et al, 2009).
