Med 21275

Beyond the Ligand-Binding Pocket: Targeting Alternate Sites in Nuclear Receptors
Laura Caboni and David G. Lloyd

School of Biochemistry and Immunology, Trinity Biomedical Sciences Institute, Trinity College Dublin, Dublin 2, Ireland Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/med.21275
Abstract: Nuclear receptors (NRs) are a family of ligand-modulated transcription factors with signicant therapeutic relevance from metabolic disorders and inammation to cancer, neurodegenerative, and psychiatric disorders. Drug discovery efforts are typically concentrated on modulating the natural ligand action within the ligand-binding pocket (LBP) in the C-terminal ligand-binding domain (LBD). Drawbacks of LBP-based strategies include physiological alterations due to disruption of ligand binding and difculties in achieving tissue specicity. Furthermore, the lack of a pure and predictable mechanism of action predisposes such intervention toward drug resistance. Recent outstanding progress in our understanding of NR biology has shifted the focus of drug discovery efforts from inside to outside the LBP, affording consideration to the interaction between NRs and coactivator proteins, the interaction between NRs and DNA and the NRs ligand-independent functions. This review encompasses such currently available NR non-LBP-based interventions and their potential application in therapy or as specic tools C 2012 Wiley Periodicals, Inc. Med. Res. Rev., 00, No. 0, 138, 2012 to probe NR biology. Key words: non-LBP; nuclear receptor; alternative site
1. NUCLEAR RECEPTORS: AN IMPORTANT PHARMACEUTICAL TARGET Nuclear receptors (NRs) are a family of ligand-regulated transcription factors that directly bind to DNA and modulate a wide variety of physiological functions. The essential role covered by NRs makes them ideal drug targets, mostly for cancer and metabolic disease.1 Since research on NRs began in the early 1960s, 48 human NRs have been characterized and divided in seven different subfamilies according to their amino acidic sequence homology. For approximately
Contract grant sponsor: Irish Health Research Board; Contract grant number: HRB/2007/2; Contract grant sponsor: Enterprise Ireland; Contract grant number: CFTD/06/110; Contract grant sponsor: Irish Higher Education Authoritys Programme for Research in Third Level Institutions (PRTLI). Correspondence to: David G. Lloyd, School of Biochemistry and Immunology, Trinity Biomedical Sciences Institute, Trinity College Dublin, Dublin 2, Ireland, E-mail: [email protected] Medicinal Research Reviews, 00, No. 0, 138, 2012 C 2012 Wiley Periodicals, Inc.
r CABONI AND LLOYD
Figure 1. Modular structure of nuclear receptors (NRs). NRs share the same structure, starting from a lessconserved N-terminal domain (NTD), also called A/B domain and containing the transcriptional activation function 1 (AF-1); a DNA-binding domain (DBD), also called the C domain. The close-up picture represents a glucocorticoid (GR) dimer bound to the DNA (pdb id: 1GLU6 ; a hinge region (H) and a C-terminal ligand-binding domain (LBD), containing the transcriptional activation function 2 (AF-2). The close-up picture represents the androgen receptor (AR) LBD 12 helical structure (pdb id: 1T7R7 with bound ligand. Three-dimensional images were generated with MOE 2011.108 .
half of them, their endogenous ligands have been well characterized, and they mostly belong to the steroidal NRs subfamily (NR3)2 ; however, for some NRs the ligand is unknown, and these receptors are dened as orphan NRs.3 Even though they are implicated in different physiological processes, all NRs share the same modular structure, comprising an N-terminal domain (NTD), a central DNA-binding domain (DBD), a connecting hinge region and a C-terminal ligand-binding domain (LBD; Fig. 1). The NTD is the least conserved domain in NRs and it is an intrinsically disordered region. This region contains an activation function (AF-1), which per se regulates gene transcription in a ligand-independent fashion. The DBD is the most conserved domain, mediating NR binding to recognition elements of specic genes in DNA. Some NRs bind the DNA as monomers, others as homodimers (e.g., steroid receptors) or as heterodimers, often with the retinoid X receptor (RXR). The DBD is composed of two cysteine-rich zinc-nger motifs; the rst one contains the highly conserved P box, which determines the sequence specicity for binding of response elements. The response elements are short sequences composed of two hexameric core half sites motifs that position each receptor and transcriptional equipment by them recruited close to the target gene. The second zinc nger contains the D box, which is involved in dimerization. The hinge region bridges between the DBD and the LBD and it also harbors a nuclear localization signal (NLS). Another important dimerization interface is present in the C-terminal LBD, which is composed of a 12-alpha-helical structure enclosing a central portion, called the ligand-binding pocket (LBP) wherein the natural ligands bind. The LBP greatly varies in shape and volume in different NRs, but nonetheless it is highly conserved in both superstructure and amino acidic sequence. LBP helix 12 (H12) is the most exible part of the molecule, as it is not engaged in any salt bridges. Conformational changes of this specic helix are unambiguously associated with the molecular mechanism of action of ligands bound
Medicinal Research Reviews DOI 10.1002/med
BEYOND THE LIGAND-BINDING POCKET
r 3
to the LBP. When an agonist is bound, following receptor conformational change, a second AF (AF-2) is generated on the surface of the NR LBD, involving interplay of helices 3, 4, 5, and 12. AF-2 regulates NR ligand dependent transcriptional activity by recruiting auxiliary protein components such as coregulatorswhich activate (coactivators) or repress (corepressors) NR transcriptional activity once the NR has associated with the DNA. Some NR LBDs are bound to a set of transcriptional corepressors in the absence of ligand.1 The aforementioned liganddriven conformational changes at the LBD also inuence the interdomain interaction between the NTD AF-1 and the LBD AF-2, as described in the androgen receptor (AR).4 Some NRs, notably within the steroid receptors, have a direct tethering to other DNA-bound transcription factors such as nuclear factor B (NF B) and activator protein-1 (AP-1), which, for example, are believed to mediate anti-inammatory actions for the glucocorticoid receptor (GR) and the estrogen receptor (ER) without a requirement for the NR binding to the DNA.5
2. THINKING OUTSIDE THE BOX: PRINCIPLES OF ALTERNATIVE INHIBITION OF NUCLEAR RECEPTORS For several decades, the focus of drug discovery in NRs has been the LBPrationally designing agonists, antagonists, partial agonists, or inverse agonists that would compete with natural ligands for binding and induce conformational changes of the H12. These changes ultimately allosterically disrupt the NRs interaction with coactivator proteins. This strategy has brought forward some of the most important drugs currently in the clinic for cancer and metabolic related diseases.1 However, there are many limitations of LBP directed strategies. First of all, displacing the natural ligand from the binding pocket may alter physiological homeostasis eliciting unwanted side effects. The physical nature of LBP itself has consequences in limiting the nature of molecular diversity (in terms of size and shape of binding ligands), which can result in subsequent lack of tissue specicity. Furthermore, drug resistance is a common outcome of LBP-based regimens, partly due to mutations leading to functional changes and in many instances antagonist conversion to agonism in a physiological context. Examples include tamoxifen-based therapy for breast cancer or antiandrogen-based therapy for prostate cancer, where drug resistance is inevitable and its onset is associated with poor prognosis and high mortality.9 In addition, LBP-based strategies may not be applied in those cases where the endogenous ligand might not be accommodated within the LBP (true orphans). Current LBP drug discovery strategy is based on the design of super antagonists10 and modulators with superior potency than the natural ligands or on the design of pure antagonists that lack intrinsic agonist activity and bear tissue selective properties. Examples include enzalutamide (MDV3100),11, 12 which demonstrates effectiveness in castration-resistant prostate cancer (CRPC) in phase III clinical trials,13 and the pure antiestrogen fulvestrant (ICI 182,780),14 an FDA-approved drug for metastatic breast cancer.15 Non-LBP approaches are accordingly focused on circumventing drug resistance associated with classic LBP-based regimens. This could be achieved by directly blocking the crucial steps in NR transcriptional activity, such as the interaction between NR and cofactors, ligandindependent interactions or the NR interaction with the responsive elements of target genes in the DNA. The NR-cofactor interaction has been successfully validated as a bona de pharmaceutical target in the last decade,1622 inspired by successes in the design of inhibitors of protein protein interactions.23 This approach is indeed very challenging from the point of view of specicity, potency, and exibility of the protein surface, and the different conformations (and the possibility of different sites generation)24 elicited by different ligands in different in vivo
r CABONI AND LLOYD
Figure 2. AF-2 amino acid analyses. Amino acids of the AF-2 site are shown for estrogen alpha receptor (above left), estrogen beta receptor (above, right), androgen receptor (below, left), progesterone receptor (below, middle) and glucocorticoid receptor (below, right) receptors (pdb ids: 3ERD26 , 3OLL,27 1T63,28 1E3K,29 and 1M2Z30 respectively). Green represents lipophilic residues, blue represents basic residues, red represents acid residues, and white represents neutral residues.
conditions. The specicity challenge arises from the overlapping of coregulators and NRs. To exert their physiological role as transcription factors, the 48 human NRs bind to more than 300 coregulators,25 with complex combinations of overlapping specicities and afnities. Moreover, coregulators interact within the AF-2 surface with conserved or common short alpha-helical motifs. The so-called NR box is composed of ve amino acids that fold into amphipathic helices mostly of the type LXXLL (where L is a leucine and X, any amino acid). This interaction is based both on t (depending on the shape of the AF-2) and also on electrostatic interactions with the NR charge clamp composed of highly conserved, oppositely charged amino acids anking both ends of AF-2. Shape and electrostatic complementarity suggest that the AF-2 site could be a druggable target. X-ray crystallography studies at this interface reveal that despite the high sequence homology between NRs AF-2, NRs could indeed present different surface shapes and different electrostatic characteristics, which may be exploited to achieve selectivity (Figs. 2 and 3). Phage display technology combined with X-ray crystallography has also advanced the coregulator preference theory. Specic sequences interacting at the coregulator interface were identied, through variation of residues anking the common LXXLL NR box. Interestingly, in the case of AR, there is a clear preference toward bulkier motifs, of the type
r 5
Figure 3. AR versus ER AF-2 shape and electrostatic comparison. AR (A and A ) pdb id: 1T7R7 cocrystallized with a FXXLF coactivator motif; ER (B and B ) pdb id: 2B1Z31 cocrystallized with a LXXLL coactivator motif. Subtle differences in electrostatic and shape may be exploited to achieve selectivity. Surfaces generated with MOE2011.108 ; (A and B) electrostatic surface, red: negative; blue: positive; white: neutral; (A and B ) molecular surface, color set to constant.
FXXLF.7, 32, 33 These motifs occur in natural coactivators such as ARA70 or ARA55. The intramolecular cross-talk between the N-terminal and C-terminal domains (N/C) interaction is based on NTD motifs of the type FQNLF (residues 2327) and WHTLF (residues 433437) that bind to the AR AF-2 surface.4, 34 To accommodate aromatic rich motifs at positions +1 and +5, the AR AF-2 cleft rearranges through an induced t mechanism to form a long-, deep-, and narrow-binding groove7, 35 when compared to other NRs such as ER, which are shallower.36 The signicance of AR coregulator inhibition is mostly based on the inhibition of the N/C interaction as it bridges between the AF-1 and AF-2 transactivation domains and may be important for stabilization of the LBD-ligand complex in an active state.35 Given these premises, the NR-cofactor interaction is a challenging but attractive pharmaceutical target. This approach is believed to overcome traditional LBP drug resistance issues by direct blockade of NR transcriptional activation. It is predicted that this interface would have lower incidence of adaptive mutations compared to the LBP, as any mutations at the NR-cofactor interface that renders the drug ineffective would also block the cofactor recruitment.10 Interestingly, instead of looking at the NR-protein interface to block coregulator recruitment, direct coactivator inhibition with small molecules was also recently explored.37 As coactivator overexpression is believed to be one of the hallmarks of drug resistance, this strategy would offer several advantages, such as the possibility of tailoring each inhibitor for a specic coactivator.
r CABONI AND LLOYD
Moving beyond the LBP (inside and out), the interaction between NRs and DNA is also an essential step for transcriptional activation. Inhibitors at this interface have also been developed as a possibility to overcome drug resistance. As DBD is the most conserved region in NRs, it is very challenging indeed to achieve specicity with such strategies. Since ligand independency is one of the hallmarks of drug resistance, one additional alternative approach has focused on the identication of NTD inhibitors. Drug discovery efforts at the NTD interface to date are hampered by the lack of structural information; therefore, high-throughput screening (HTS) has been the only approach employed to discover new inhibitors. The relevance of inhibition of the AF-1 has been only demonstrated for AR,38 as AR AF-1 has an essential role in transactivation, unlike in other steroid receptors.3941 In the body of this work, we intend to outline the alternative (non-LBP) drug discovery efforts progress in terms of the evolution from peptidic to small organic molecule modulators, and the challenges for the future. We will group strategies by site, starting from the LBD, DBD, and nally NTD.
3. NUCLEAR RECEPTORS ALTERNATIVE SITE MODULATORS A. Peptidic Inhibitors The validation of the NR-cofactor interaction as a pharmaceutical target was provided by the study and identication of several peptide sequences that interact on the NR AF-2 surface. McDonnell and colleagues42, 43 rst applied phage display methodology to identify protein sequences interacting with NRs (ER and ER ), in an effort to develop a novel class of ER antagonists. Phage display is a method to study proteinprotein interactions, where highly diverse peptide libraries are generated and displayed individually on the surface of phages. The peptide libraries thus generated are then exposed to a protein such as the full-length NR or the NR LBD in order to identify potential binding sequences, and obtain information about their specicity. Utilizing this approach, peptide libraries were screened for their ability to interact selectively with estradiol- or tamoxifen-bound ER and ER . The rationale behind this project was in rst instance to explore the validity of NR-cofactor interaction as a pharmaceutical target, and in second instance, given the promiscuity of coactivators and the similar structural requirements for their interaction within the AF-2, to assess the possibilities pertaining to specicity. Their work demonstrated that different ER conformations are elicited according to the nature of the ligand bound in the LBP (in this case, estradiol, 4-hydroxy tamoxifen or apo-ER),43 and a different set of high-afnity peptide probes are recruited as a result. An interesting implication is that the biological activity of 4-hydroxy tamoxifen and other ER modulators could be accounted for and differentiated based on the presentation of different sites outside the LBP that recruit a particular set of peptide sequences. The functional signicance of disrupting novel non-LBP binding sites was conrmed at a cellular level using a mammalian two-hybrid assay. The study has interesting implications regarding the NR-cofactor interface being an accessible target, but introduced the concept that NR signaling was strongly dependent on environmental factors, such as the cellular environment and the nature of ligand bound. It was known that the NR-cofactor interaction is largely conserved around common motifs such as LXXLL but it was not clear how specicity in such intervention might be achieved across NRs. Additional phage display studies suggested that selectivity of peptide sequences between different NRs is dependent on the residues anking the core LXXLL motif, so that even within this specic and common core, functional variability can be achieved. This was demonstrated for ER and ER , by designing a large (more than 108 members) focused combinatorial phage library. The library consisted of a constant LXXLL core anked on each
r 7
side by a 7-mer of random amino acids (X7 -LXXLL-X7 ).44 Screening identied three distinct classes of LXXLL interacting motifs that functionally differed in their ability to interact within the AF-2 of ER LBD mutants in a mammalian two-hybrid assay. Not all the phages display identied peptides containing the LXXLL motif shared the same interaction pattern within ER AF-2. When the peptide sequences were coexpressed in cells, they were able to mimic the interaction between ER and endogenous cofactors and block ER transcriptional activity. Residues anking the LXXLL core at 2 or 1 position were found to play a critical role in specicity and in determining the motif orientation on the protein surface. Remarkably, peptide #293 showed ER selectivity compared to ER , indicating that residues in addition to the 1 or 2 positions surrounding the core may also inuence or determine receptor selectivity. A focused library was then advanced to further investigate ER subtype specicity and develop different classes of potent ER peptide antagonists, effective in presence or even in absence of the hormone,45 with interesting implications for chemotherapy-resistant breast cancer. In addition, such combinatorial phage display approaches were also applied to understand which specic sequences anking the core motif confer peroxisome proliferator-activated receptor (PPAR) specicity.46 These studies identied an extended PPAR consensus motif (HPLLXXLL), and selective peptides such as NBM131, which binds (KD 80 nM for PPAR and PPAR ) to PPAR LBD with higher afnity than the endogenous coactivators PGC-1 or TRAP220. Peptides identied were able to compete with coactivators and to repress PPAR-mediated transcriptional activity. Recently, using the same approach, a 108 set of different 19-mer peptides was screened in the mineralocorticoid receptor (MR) in presence of different ligands.47 The report identied potent and selective MR antagonists of MR mediated transactivation. An MR unique interacting motif, MPXLXXLL, was identied in around 50% of the peptide sequences that interacted with MR in a ligand-dependent fashion. It is generally accepted that coregulator and NRs function overlap, therefore different NRs may bind the same set coactivators, but they utilize different LXXLL NR boxes within the same coactivator for interaction, thus achieving specicity. For example, although thyroid receptor (TR) and ER bind to the same steroid receptor coactivator (SRC)-1, they have different NR box preferences within SRC-1. TR or ER specicity is achieved by directed mutations on the residues around and within (at positions +2 and +3) of the LXXLL motif of the NR box.48 These studies contributed in elucidating the specicity of interaction of the common LXXLL NR box with different NRs and the potential required features in the design of specic peptidomimetics. A great advance in our understanding NR selectivity was provided by the identication of NR box motifs other than LXXLL that specically bind AR AF-2 with higher afnity. AR has been shown to preferentially bind (F/W)XXL(F/W)-, FXXF(F/Y)-, and FXXLY-type motifs, in addition to the classic LXXLL NR motif.7, 33, 36, 49 Chang et al. identied high-afnity peptide motifs mostly containing aromatic side chains in the NR box and have designed selective AR peptidic inhibitors directed at the coactivator interface.50 The aim of their work was also to nd other possible coactivator interacting surfaces on the full-length AR, but despite the expectations of the authors, only sequences binding at the LBD were identied. Interaction of the peptides with full-length AR was evaluated in a mammalian two-hybrid assay. Peptides were found to be AR-specic over GR and ER, but had moderate afnity for PR, consistent with our understanding of the higher similarity between the AF-2 regions of AR and PR. This was the rst report validating the AR-cofactor interaction as a drug target and afforded a good starting point for the design of novel inhibitors. It is known that aromatic rich motifs are found in naturally occurring coactivators such as ARA55 and ARA70 and in the AR NTD (FQNLF and WXXLF), where they mediate the N/C interaction.4, 34 Recently, additional specic interaction motifs in AR cofactors such as FXXFF
r CABONI AND LLOYD
(gelsolin) and FXXMF (PAK6) were also identied.51 Gelsolin promotes AR transactivation both in presence of androgens and antiandrogens such as hydroxyutamide.52 Increased expression of gelsolin was found after androgen ablation and is thought to be associated with prostate cancer progression. In a recent report, the FXXFF motif from gelsolin was found to inhibit AR function53 and to also affect PR function. A gelsolin-based screen was employed by Joseph et al.54 to identify novel AR small molecule non-LBP inhibitors and will be discussed in Section 3.B.5. The selectivity and afnity of AR for bulky hydrophobic residues at positions i + 1 and i + 5 in the NR box was conrmed by mutational studies, in silico models36, 55 and by X-ray crystallography combined with phage display technology.7 A clear role in the determination of AR selectivity was established for residues anking the FXXLF or LXXLL core.33, 56 Recently, another approach57 studied the ideal alpha-helical conformation and length required for AR-coactivator interaction, through the computational design of a series of miniproteins optimizing the amino acids around the central biomotif and identifying, which were the critical features for AR binding. Other miniproteins have been successfully designed for ER.58 Expanding this approach to other NRs could underpin the design of selective inhibitors based on the distinct molecular recognition signatures recognized between AR, ER, and other NRs. Phage display screening can be also informative about the NRs mechanism of transactivation. It was applied in probing the orphan receptor ROR interaction with LXXLLtype coactivators59 and in understanding VDR mechanism of transactivation linked to RXR heterodimerization.60, 61 In the latter study, subsets of previously identied small LXXLL peptides44 were selected for their ability to interact with VDR or RXR and showed that the inhibition of RXR transcriptional activity also indirectly affects VDR transactivation, reinforcing the idea that RXR is a necessary heterodimeric partner for ligand mediated VDR activity. Collectively, such phage display studies contribute to validate the NR-cofactor interaction as a drug target and to our understanding of the principles of selectivity in cofactor recognition, describing the rst potent and selective inhibitors of NR-cofactor interactions functionally disrupting NR transcriptional activity. These studies have important implications in assay development, in determining appropriate peptide sequences or combination of sequences to be used in screening campaigns. It has been demonstrated that the use of a full-length NR protein as opposed to NR LBD can lead to different phage display screening outcomes,49, 50 hence the need to combine this technique with X-ray crystallography, mutational analysis, and in silico approaches for more powerful results. Many issues associated with phage display peptidic inhibitors limit their application in clinical settings. In particular, these include the challenge to achieve efcient intracellular delivery in vivo, due to poor stability and poor permeability, while short plasma half-life reduces their efcacy in vivo. Formulation advances may ameliorate such adverse pharmacokinetic characteristics, with extensive investigations ongoing in the choice of optimal potential carriers, conjugates or tags to improve adsorption, solubility and tissue-specic targeting.21, 53, 62 1. Improving the Drug-Like Properties of Peptide Inhibitors Phage display studies provided proof of concept for the possibility of selectively targeting NRcoactivator interactions through functionally blocking NRs transcriptional activity. The design of nonnatural cyclic and stapled peptides incrementally provided a valid step toward a more drug-like intervention, with the ultimate goal to design small molecule peptidomimetic antagonists. Cyclization is frequently used to stabilize peptides in an alpha-helical conformation. Macrolactam constrained peptides were rst described as inhibitors of the interaction between
r 9
TR and the coactivator glucocorticoid interacting protein 1 (GRIP-1).63 This approach was expanded in the computational design of a library of peptidomimetics bearing different substitutions on the common L1 XXL2 L3 NR box motif with nonnatural amino acids to understand the basis of selectivity in the inhibition of TR and ER /ER interaction with SRC2-2.64 Cyclic peptides were studied in presence of different hormones, as they are able to affect the conformation of the AF-2 surface. Leduc et al.65 designed disulde bridged cyclic peptides based on the cocrystal structure of ER with a coactivator,26 yielding a novel class of peptidomimetic estrogen modulators (PERMs). This approach combined the formation of a stabilized cyclic alpha-helix by i and i + 4 amide formation (macrolactam cyclization) and the combination of D,L-dicysteine to form an i and i + 3 disulde bridge. The latter was found to give maximal helix stabilization, demonstrated by high potency and conrmed by X-ray structural studies (pdb id: 1PCG). The most potent peptide identied (Ki = 25 nM), a disulde-linked nonapeptide containing the LXXLL motif (PERM-1), was isoform selective (around 15 times more potent for ER than ER ). Synthetic analogs of PERM-1 were investigated by evaluating nonnatural analogs of cysteine at i, i + 3 positions, such as lanthionine and cystathionine (redox stable compared to its counterpart cysteine, therefore more applicable for evaluation at an intracellular level) that form a thioether cyclization,66 and furthermore penicillamine and homocysteine substituents that give variable conformational exibility to the disulde bridge.67 The introduction of nonnatural amino acids such as tertiary leucine or neopentylglycine within the NR-recognition motif LXXLL improved afnity dramatically.67 The introduction of neopentylglycine in the NR box at the i + 4 (L2 ) position, replacing leucine in i, i + 3 disulde constrained cyclic peptides, yielded a potent and selective ER inhibitor (potency 70 pM as determined by time-resolved uorescence resonance energy transfer [TR-FRET]). This is of particular interest because the leucine at i + 4 is oriented outside the coregulator groove. The side chain of the neopentylglycine, being the same length as leucine side chain, would not be expected to make extra van der Waals hydrophobic contacts that could account for this dramatic potency improvement. However, structural studies demonstrated that intramolecular van der Waals contacts are formed with the side chain of isoleucine at the i position, and are responsible for the stability of the peptide. The activity of these nonnatural peptides was correlated with their conformational features, suggesting that other factors in addition to the peptide alpha-helical character are important in determining the activity and selectivity of the peptides. The functional inhibition of NR-mediated transcription mediated by these peptides has not yet been described, suggesting that the poor permeability could be an issue for effective delivery at a cellular level. Thus, these studies encouraged efforts in improving peptide permeability and selectivity. In a recent report, a nona-arginine tag (R9 ) was found to facilitate cellular permeability when attached to natural high-afnity LXXLL motif sequences that were previously found to inhibit ER transcriptional activity when overexpressed in cells.68 The nona-arginine LXXLL peptide coactivator inhibitors disrupt ER-mediated transcriptional activity at a cellular level and have been used as molecular probes to inhibit and evaluate ER -coactivator interaction in living cells and to monitor ER cellular localization (in fact, ER is displaced from the nucleus to the nucleoli upon binding of peptides). In order to improve peptide permeability, another approach, the design of stapled peptides, has been applied for ER and ER .69 Stapled peptides are thought as helix stabilizers, and afford increased permeability and resistance to proteases.70 An all hydrocarbon link functions as a staple between alpha-helix successive turns through the addition of olen bearing tethers that link the i position with the i + 3, i + 4, or i + 7 positions of the peptide. X-ray crystal structures (pdb id: 2YJA or 2YJD) demonstrate that the hydrocarbon linker modies the binding properties of the NR box LXXLL to the ER or ER so that the staple tether drives the interaction on the coactivator groove through van der Waals hydrophobic
10
r CABONI AND LLOYD
contacts. The consequences of this modication, in terms of increased hydrophobicity, on the NR-cofactor interaction are not easy to predict and further studies are required to verify the cellular permeability of the stapled peptides, their functional inhibition of ER transcriptional activation and their NR-subtype selectivity. 2. Peptides Directed at the NR LBD Dimerization Interface In parallel to the NR-cofactor approach, other peptides have been designed to block NR function by targeting the dimerization interface in ER LBD. A 12 amino acid phosphotyrosyl peptide (Yp537) from the H12 region of ER (amino acids 532543) inhibited ER binding to estrogen-responsive elements (ERE). While further studies are needed to validate the peptide mechanism of action, this is possibly due to disruption of ER dimerization.71 Yudt and Koide72 rationally designed a 16 amino acid oligopeptide (also called the I-box) derived from the alpha-helical region of ER involved in homodimerization, encompassing the majority of the helix 10/11 region (amino acids 503518). This peptide specically precipitates ER from cell extracts depending on its alpha-helical nature, but with an unclear mechanism, possibly due to misfolding. Protein misfolding and aggregation are not likely to be the best options to inhibit NR activity in a therapeutic context, and the functional in vivo consequences are not straightforward, as NRs are associated with molecular chaperones in their inactive state, intent on preventing misfolding. Recent in silico studies73 have elucidated the molecular recognition principles at the ER dimerization interface in different conditions (unliganded, agonist-, or antagonist-bound ER) and designed a peptide that interacts with ER independently of the liganded state. This particular approach could be useful in the rational design of peptide or small molecule inhibitors for the treatment of selective ERs modulators (SERM) resistant breast cancers. B. From Peptides to Irreversible and/or Selective Small Molecules 1. Thyroid Receptor (TR) The thyroid receptor is involved in regulation of metabolism, growth, development and cardiac function74 and therefore constitutes an important pharmaceutical target. Like all other NRs, strategies to modulate TR function have largely been focused on the LBD. Inhibitors of the natural thyroid hormone (T3 ) and their clinical utility in hyperthyroidism conditions have previously been described.75, 76 Quantitative HTS campaigns coupled by uorescence polarization (FP) assays utilizing SRC-2 as a uorescent probe led to the discovery by Arnold et al. of the rst series of irreversible small molecule coactivator binding inhibitors (CBIs) of the TR interaction with SRC-277, 78 (Fig. 4). The most potent molecules identied, exhibiting IC50 values of 2 M and showing full inhibition of TR transcriptional activity, belong to the class of aromatic -aminoketones.
Figure 4. Small molecules inhibitors of TR /SRC2-2 interaction. From left to right, a -aminoketone chemotype: 3-(dibutylamino)-1-(4-hexylphenyl)-propan-1-one (DHPPA), the ML151 probe from the MSNB chemotype and one of the most potent compounds of the SNPT chemotype 2.
r 11
These compounds are Mannich bases, and function as pro-drugs that covalently interact with nucleophilic cysteine residues at the AF-2 surface after in situ deamination and liberation of , -unsaturated ketones, the active principle responsible for TR -SRC-2 inhibition. The presence of cysteine residues in TR AF-2 surface is unique and affords selectivity over other classes of NR. Compounds evaluated were still active in presence of the natural ligand T3 and strongly suggest a differential mechanism of interaction. The covalent interaction between the molecules and TR was conrmed with mass spectrometry and mutagenesis studies to identify which cysteine was involved in binding, leading to the hypothesis that the compounds interact with Cys298 on the AF-2 surface. This hypothesis was validated by the X-ray crystal structure of a -aromatic aminoketone 3-(dibutylamino)1-(4-hexylphenyl)-propan-1-one (DHPPA), where the deaminated active principle was shown in the AF-2 region, supporting the specic mechanism of alkylation and a deamination in situ catalyzed by Cys298.79 This report represents a very signicant advance in the study and design of inhibitors at the NR-coactivator interface, providing structural evidence of the mechanism of action of a small molecule at a NR AF-2 site. These compounds present distinctive characteristics of an irreversible mechanism of inhibition, such as time dependency and binding to TR with 1:1 stoichiometry. Furthermore, to demonstrate functional TR inhibition, the authors used a combination of full-length SRC2 (containing three NR boxes) and full-length TR, which did not affect the compounds inhibition. Extensive structureactivity relationship (SAR) and optimization studies for the aminoketone class were subsequently reported.80, 81 Compounds were evaluated for their potency using FP techniques in TR and TR isoforms, and also for their cytotoxicity, solubility, and permeability with a view to in vivo studies. Selectivity was conrmed by lack of inhibition in AR, ER , and PPAR . The in vitro potency did not change compared to that reported for the initial hits, but the pharmacological properties in terms of cytotoxicity and ion-channel off-target effects were improved. The above-mentioned adverse characteristics of -aminoketones in vivo led to a second quantitative high-throughput screen (qHTS) in the search of a novel-improved chemotype. This screen identied a novel class of irreversible binders, methylsulfonylnitrobenzoates (MSNB)82, 83 (Fig. 4), which also target Cys298 in the AF-2. The irreversible mechanism was again conrmed by mass spectrometry analysis and by the time dependency of the interaction. MSNB were selective for TR when compared to AR, VDR, or PPAR . This class of compounds constitutes a signicant improvement regarding off-target effects (the lack of the basic tertiary amino group eliminates the off-target activity toward ion channels) but present similar solubility and cytotoxicity proles to the -aminoketones. The most promising molecule from this screen, ML151 was reported as a selective probe to block TR -SRC-2 interaction, with a potency of 1.8 M.84 A third class of irreversible inhibitors was recently advanced to improve the in vivo properties of the MSNB probe, belonging to the class of sulfonylnitrophenylthiazoles (SNPTs), designed by substituting the ester linker with a cyclic bioisostere such as thiazole85 (Fig. 4). SNPTs demonstrated selectivity in inhibiting the TR -SRC-2 interaction in FP and functional cellular assays, leading to an improved promising class of inhibitors for future in vivo studies. The most potent inhibitor of the series, 2{4,1,5}, exhibited an IC50 of 0.3 M, but had very weak transcriptional inhibitory activity. In contrast, a less-potent analog, 2{3,1,2} (IC50 = 2.4 M; Fig. 4) signicantly inhibited T3 -mediated luciferase activity at 5 M. The results demonstrated for the different classes of inhibitors discussed suggest specic interaction within AF-2 site and specic alkylation. Alternative targeting of TR is useful to obtain probes capable to modulate TR activity in vitro and in vivo. The curtailment of activity in presence of natural ligand such as T3 is benecial as a potential treatment for hyperthyroidism
12
r CABONI AND LLOYD
without negatively affecting the endogenous hormones binding, but specic toxicity concerns arising from irreversible covalent inhibition remain to be addressed. 2. Vitamin D Receptor (VDR) Vitamin D Receptor (VDR) is involved in calcium homeostasis regulation, bone mineralization, and immune system modulation. VDR heterodimerizes with RXR upon binding of 1,25(OH)2 vitamin D3 to the LBD, which promotes the recruitment of coactivators necessary for VDRmediated transcriptional activity. Inhibition of VDR transcriptional activity has been applied in Pagets disease of the bone by using VDR LBP secosteroidal inhibitors. A de novo strategy to design small molecule VDR peptidomimetic inhibitors was based on the structural requirements of the alpha-helical interaction of LXXLL-type motifs with the receptor. The rst nonsecosteroidal small molecule VDR inhibitors, and also the rst inhibitors of the interaction of VDR with coactivators, were benzodiazepines, bearing three alkyl side chains designed to mimic the three leucines of the LXXLL motifs86 (Fig. 5). The most potent substituted benzodiazepines reported exhibit an IC50 of 20 M on VDR in a TR-FRET assay and selectivity compared to ER in inhibiting VDR transcriptional activity. The non-LBP nature of such antagonists was conrmed in a cellular reporter assay by monitoring the effects of the benzodiazepines in VDR-mediated transcription in the presence of increasing concentration of the natural ligand 1,25(OH)2 D3 . No rightward shift on the doseresponse curve could be detected, excluding a competitive LBP mechanism of action. The benzodiazepine scaffold has already been validated as an alpha-helix peptidomimetic inhibitor87 and can be thought as a rigid substructure that mimics the alpha-helical arrangement of the side chains in the NR box of coactivators. Further studies are in progress to elucidate the SAR of the benzodiazepine scaffold and investigate selectivity. Recently, 3-indolyl-methanamines were advanced as the rst irreversible small molecule inhibitors of the interaction of VDR with the third nuclear interaction domain of SRC-2 (SRC2-3)88 (Fig. 5), with selectivity over other classes of NRs, such as AR, ER , TR , TR , and PPAR . This class of molecules was identied through an FP-based high-throughput screen, and characterized for their solubility, toxicity, and permeability, and their ability to inhibit VDR-mediated transcription at a cellular level. The most potent compound reported, compound 1, presented an on target IC50 of 3.3 M but only a very weak inhibition of transcriptional activity at 62.5 M. A good selection of secondary conrmatory assays coupled by an exhaustive SAR analysis proved the usefulness of the 3-indolyl-methanamine scaffold
Figure 5. Benzodiazepine and 3-Indolyl-methanamine inhibitors of VDR-coactivator interaction. Substituted benzodiazepine scaffold (2) mimic of LXXLL-mediated DRIP205 interaction with VDR and 3-indolyl-methanamine (31b) as selective and irreversible inhibitor of VDR interaction with SRC2-3.
r 13
to selectively probe the VDR-coactivator interaction. In particular, the synthetic compound 31b (IC50 = 36.7 M for VDR-SRC2-3 inhibition, Fig. 5) was extensively characterized for its mechanism of action. The authors propose an irreversible mechanism involving reaction of an intermediate azafulvenium salt (more usually formed in acidic or high-temperature conditions by elimination reactions), which then alkylates VDR. Remarkably, the high specicity of VDR alkylation was conrmed by no reactivity on other classes of NRs, where other electrophilic inhibitors have been described. However, cell toxicity for compound 31b at a similar concentration required for its transcriptional inhibitory activity was reported, which could be a side effect of its irreversible mechanism of action. The potential for clinical utility and cross reactivity of these VDR inhibitors has yet to be conrmed in vivo. 3. Pregnane X Receptor (PXR) Pregnane X receptor (PXR) and constitutive androstane receptor (CAR) are orphan NRs, which do not conform to the classical denition of NRs as ligand-dependent transcription factors but instead function as xenobiotic sensors, and have an important role in drug metabolism.89 Antagonists of xenobiotic receptors such as PXR would be useful not only to probe the biological functions of orphan NRs, but also to improve therapeutic efciency of drugs, in cancer treatment and metabolic diseases.90 Computational design of PXR LBP antagonists has been challenged by the size and exibility of PXR LBP, which is activated by a wide range of ligands.91 In 2002 Takeshita et al.92 repurposed ketoconazole (Fig. 6), a known azole antifungal drug, as a novel PXR antagonist that was able to antagonize rifampicin, one of the known agonists of PXR. Ketoconazoles mechanism of PXR inhibition was conrmed by later reports, where it disrupts the interaction between activated PXR and activated CAR with SRC-1 in a noncompetitive fashion. The non-LBP nature of SRC-1 inhibition was demonstrated by scintillation proximity assays by comparing the IC50 of ketoconazole with that of rifampicin in
Figure 6. Non-LBP modulators of PXR/SRC-1 interaction. Ketoconazole was the rst PXR cofactor/inhibitor identied, SPB03255 and leunomide where derived from a ligand-based rational approach. Coumestrol was also found to antagonize coregulator recruitment, but lacks absolute selectivity and its mechanism of action remains to be claried.
14
r CABONI AND LLOYD
competing with the PXR LBP ligand [3 H] SR12813.93 The estimated KB for ketoconazole was 55.3 M, much higher than its known biological effective concentrations (ranging from 6 to 25 M), excluding a potential LBP competitive mechanism. Furthermore, a specic non-LBP inhibitory action on the AF-2 surface was suggested by site-directed mutagenesis, where some mutants preserved ligand binding and activation but were not inhibited by ketoconazole.94 Ketoconazole and other azoles were found to be pan-antagonists for PXR, CAR, liver X receptor (LXR), and farnesoid X receptor (FXR) in disrupting SRC-1 binding, but did not affect SRC-1 binding to ER or PPAR . Interestingly, ketoconazole presents eclectic roles across NR biology, by indirect or direct actions. Ketoconazole inhibits 1,25(OH)2 vitamin D3 metabolizing enzymes, enhancing VDR signaling,95 and inhibits cytochrome P450-dependent enzymes suppressing adrenal steroidogenesis.96 Furthermore, it has been reported as a GR-competitive antagonist97, 98 (where it inhibits [3 H]dexamethasone binding with an IC50 of 100 M) and as a direct AR binder, with weak antiandrogen activity (IC50 64 M in competing with [3 H]R1881).99 Additionally, analogs of ketoconazole, such as sertaconazole and oxiconazole have been shown to competitively antagonize DHT binding to the AR LBP with IC50 of 1 M.100 This raises the possibility that ketoconazole may also directly affect SRC-1 interaction with AR and other NRs. Phytochemicals were also identied as PXR antagonists, such as coumestrol (Fig. 6), a phytoestrogen that was found to bind on the surface of PXR and antagonize coregulator recruitment in functional gene reporter assay.101 Although competition-binding experiments suggest that coumestrol competes for the binding of the ligand in PXR and CAR LBP, coumestrol also antagonizes PXR ligand dependent SRC-1 recruitment to the same extent regardless of the concentration of rifampicin. This result supports the hypothesis of a second binding site outside the LBP, putatively the AF-2 region, as suggested by FP competition experiment with uorescently labeled SRC-1 LXXLL in ligand-bound PXR-LBD. Coumestrol was tested across a panel of NRs and it was found to be PXR selective, except for cross-reactivity with ER and ER . The existence of a second binding site for coumestrol can only be ultimately conrmed by X-ray crystal structure elucidation. Using the azole data set, a PXR antagonist pharmacophore was developed from ligand alignment.102 Docking studies on AF-2 suggested that ketoconazole only occupies two of the three subsites of the LXXLL motif (corresponding to where the three leucines sit), leaving open the possibility of additional rational design approaches to further improve inhibition. The azole pharmacophore was validated for its ability to retrieve other known PXR antagonists (although not necessarily binding at the same site) and it was also applied in the identication of novel antagonists that could also target the AF-2 surface103 (Fig. 6). Interestingly, one of the candidates identied was leunomide, a known antirheumatic drug, which was repurposed as a PXR/SRC-1 inhibitor. Leunomide disrupts PXR/SRC-1 interaction with an IC50 of 6.8 M, possibly by direct binding within PXR AF-2, as suggested by site-directed mutagenesis and molecular docking studies. This report constitutes a signicant improvement toward the design of smaller and more efcient PXR non-LBP antagonists. However, selectivity issues and mechanism of action of some of the PXR antagonists (some of which may bind at multiple sites or multiple orientations within and outside the LBP) remain to be claried, together with their in vivo utility, which is still a matter of debate. 4. Estrogen Receptors (ERs) De novo structure-based drug design successfully promoted the transition from peptide inhibitors to the rst small molecule inhibitors for NR-coactivator interaction by considering the structural arrangement of the LXXLL motif binding in ER AF-2 surface.104 Generally, this approach entailed the design of a central rigid central core corresponding to the peptide
r 15
Figure 7. Estrogen receptor coactivator binding inhibitors. Series of small molecules CBI identied for the ER through de novo and structure-based design such as pyrimidines, amphipathic benzenes, bicyclo octanes, pyridylpyridones, and biphenyl proteomimetics, virtual (ERI-5 and ERI-7) and high-throughput screening (1 and 4).
backbone (such as a pyrimidine ring), and three substituents that mimic the leucines spatial disposition of the NR LXXLL box (Fig. 7). Small molecules bearing different cores were tested for their ability to disrupt ER -coactivator interaction using an FP assay. The non-LBP nature of the interaction was conrmed by a competitive radiometric binding assay. The best binders were found to be pyrimidine molecules. A exible docking approach was applied to
16
r CABONI AND LLOYD
predict the binding mode of the best pyrimidine binders into the AF-2 site, which is seemingly based on hydrophobic interactions. Given this premise, optimization of the molecule to maximize the contacts with the charge clamp could afford more potent inhibitors. Although the molecules identied were not very potent (the best pyrimidine had a Ki of around 30 M), this report identied the rst small molecule NR CBI, and conrmed that the direct mechanism of coactivator inhibition is an alternative feasible approach. Selectivity studies and functional cellular inhibition of ER transcriptional activity needs to be investigated further. A SAR for the pyrimidine core CBI was successively reported105 in an effort to improve potency. The effect of different substitutions on the pyrimidine core yielded a rened pharmacophore that summarized the required characteristics for binding to AF-2 (with the aid of molecular docking), improved their potency (best pyrimidine Ki was around 2 M as determined by a TR-FRET assay) and conrmed their activity in a functional gene reporter assay. The inhibition of ER transcriptional activity was not affected by increasing concentrations of estradiol, conrming these compounds as non-LBP antagonists. Nearly all CBIs evaluated were selective for ER over ER . Based on the pyrimidine de novo design, another class of molecules, amphipathic benzenes, was designed to inhibit the ER -coactivator interaction106 (Fig. 7). Amphipathic benzenes were classied as non-LBP antagonists based on TR-FRET assays and cellular reporter gene assays. These molecules have advantages over the pyrimidines in terms of solubility due to their amphipathic nature, and therefore could be better suited to interact in the shallow AF-2 surface, as the coactivator alpha-helix is also amphipathic. Another report combined high-throughput mammalian two hybrid assays with virtual screening in order to identify novel alternative ER antagonists that mimic the structural characteristics of the LXXLL motif cocrystallized with ER .107 ERI-5 (Fig. 7), derived from HTS, was the most potent molecule (IC50 5.5 M) that binds to ER and ER but not to PR. ERI-7 (Fig. 7) was entirely derived from virtual screening and it is ER selective over ER and PR. However, ERI-7 could not be evaluated for functional ER inhibition in cellular assays due to poor permeability. The alternative mechanism of inhibition of both ERI-5 and ERI-7 was conrmed by their inability to displace LBP-bound estradiol and their direct binding to ER was demonstrated by NMR experiments. It was then suggested that these compounds could directly or allosterically disrupt the ER interaction with coactivators. Based on this report, in an effort to improve the characteristics of ERI-5, a series of 16 guanylhydrazone analogs were evaluated in functional gene reporter assays and the non-LBP nature was conrmed by unsurmountable binding with increasing concentrations of estradiol. -Tetralone guanylhydrazone-based compounds were found to have improved potency (IC50 0.9 M for the best compound).108 Interestingly, while it has been excluded that these compounds react covalently with nucleophilic residues in ER , it may be possible that the guanidine group, which is protonated under physiological conditions, could interact with the negative patch of the AF-2 charge clamp. NR selectivity for this class remains to be explored. De novo structure-based design to mimic the structural disposition of the LXXLL motif generated two more classes of antagonists, the bicyclooctanes109 and the pyridylpyridones110 (Fig. 7). The rst class was rationally designed by an inside out approach by concentrating on mimicking the deeply buried leucines and then building the core. While this report affords potential structural mimics of the coactivator helix, the molecules exhibit lower potency compared to their pyrimidine counterpart (Ki between 7 and 40 M) and do not exhibit a clear SAR. Furthermore, the result from the TR-FRET assay indicates only partial inhibition of the coactivator recruitment to ER . The pyridylpyridones were designed based on the alpha-helix mimetic strategy, previously applied for other inhibitors of proteinprotein interactions.111 The pyridylpyridone scaffold inspired a new class of biphenyl proteomimetics,112 by adding functional groups that are able to mimic the intrinsic dipole of the helix and that are complementary to the charge clamp of the AF-2 site (Fig. 7). However, this approach still requires improvement in potency (Ki of the best compound was 30 M) and
r 17
selectivity. The biphenyl molecules presented in the report only partially inhibit coactivator recruitment to ER . To date, small molecule CBIs identied have only moderate potencies, in the micromolar range, which could limit their potential clinical application. A recent report113 identied and explored SAR around two new scaffolds (1 and 4, Fig. 7) as ER CBIs. Instead of de novo design, thus far the method of choice to design potent ER CBIs, the authors used HTS combined with TR-FRET to advance compounds in the 530 M potency range, that were functionally conrmed by a luciferase reporter assay. The study also underlines the importance of compound purity to exclude false positives (two of the four scaffolds identied were inactive after resynthesis). A new paradigm to improve potency is proposed, based on water displacement contributions to enhance binding free energy, using a combination of induced t docking and MM-GBSA approaches. Such small molecule CBIs conforming to the Lipinski rule of ve (MW < 500 Da) would be insufcient to effectively displace water from the binding site in a manner comparable to that of the peptide, and also only partially occupy the cleft. The authors suggest that we should move from a groove-based approach to cover other regions of the peptide. Further studies, based on X-ray crystallography and NRs selectivity, would validate the CBI-enhanced potency hypothesis and design-targeted libraries, which penalize the Lipinski rule but improve potency and NR selectivity. An interesting hypothesis emerged when the X-ray crystal structure of 4-hydroxy tamoxifen was solved for ER and showed its binding at two different sites, the LBP (high afnity) and a second site outside the LBP, the AF-2 region (low afnity).114 The implication of this second site further conrms the importance of the AF-2 in modulating NR transcriptional activity and support the design of small molecules directed at this interface.115 Furthermore, alternative conformations of the H12 in response to the LBP-bound ligand suggest the possibility of additional modulatory protein surfaces, whose biological signicance is unclear and yet to be claried.24, 116 5. Androgen Receptor (AR) Based on advances in the design of ER CBIs104, 105 and on the AR preference for bulkier side chains at the i + 1 and i + 5 positions in the NR box of the type FXXLF7, 28, 50 Gunther et al. designed a series of trisubstituted peptidomimetic pyrimidines117 (Fig. 8), which are characterized by a rigid pyrimidine central core and three substituents, again following the classical peptidomimetic approach. From SAR analysis, it is clear that through systematically increasing the bulk and lipophilicity of the pyrimidines substituents, it is possible to achieve AR selectivity over ER. In fact, the presence of more than two aromatic rings seems to confer AR specicity, as determined by an AR transcriptional activation in a cellular assay. However, increasing lipophilicity has adverse consequences in solubility; hence the pyrimidines designed could not be evaluated in a
Figure 8. Peptidomimetics pyrimidines as selective inhibitors of the AR-coactivator interaction. Structure-based design of bulky side-chain pyrimidines affords selectivity toward AR.
18
r CABONI AND LLOYD
Figure 9. Mixed AR AF2/BF3 inhibitors. Molecules identied by FP and X-ray screen including NSAIDs (ufenamic, tolfenamic, and meclofenamic acid) and thyroid hormones (T3 and TRIAC) both in AF-2 and in a novel allosteric site on the AR surface, named BF-3.
TR-FRET assay for direct AR LBD binding. Optimization is therefore required for these series of molecules to improve solubility and to generate an accurate SAR analysis for the optimal study of the required characteristic for AR coactivator binding inhibition. Nonetheless, this report found potent inhibitors of AR transcriptional activity with a demonstrated non-LBP mechanism and AR selectivity with IC50 values ranging from 6.6 to 1.5 M. Functional FP screens combined with X-ray screening provided an interesting series of small molecule inhibitors for the AR-coactivator interaction118 (Fig. 9). The molecules identied included off-patent nonsteroidal antiinammatory drugs (NSAIDs) and thyroid hormones such as TRIAC and T3 , which were repurposed in this work as potential candidates for the treatment of advanced prostate cancer by binding to alternative non-LBP sites. The most potent compound, tolfenamic acid, had IC50 of 47 M and all of them demonstrated inhibition of AR activity in cellular models. Additionally, two small molecules and two chemical fragments were uniquely identied through the X-ray screen. This study is particularly important for the identication of a novel surface, termed by the authors binding function 3 (BF-3). In fact, in contrast to what was initially expected, X-ray screens revealed ufenamic acid, Triac, T3 , and the two chemical fragments identied, 2-methylindole and indole-3-carboxylic acid reside in the BF-3 binding site instead of the AF-2 site. Moreover, some residues of AF-2 were conformationally affected by binding of the compounds at the BF-3 surface, inuencing coactivator recruitment as a consequence. From this structural evidence, it was possible to hypothesize that BF-3 is a novel allosteric surface and a potential non-LBP pharmaceutical target that could be exploited to nd alternative therapies for prostate cancer. In addition, BF-3 is a conserved surface amongst different NRs,119 and this offers an opportunity to understand the molecular mechanism of coactivator recruitment, and subsequently its inhibition. Combining routine in vitro assays such as FP with X-ray crystallography screening is a useful approach that allows the identication of small binding fragments that could be used as building blocks to build tight binding inhibitors. As the small molecules identied in this report were crystallized both at AF-2 and BF-3 sites, we can refer to them as mixed AF-2/BF-3 inhibitors (Fig. 9). Selective AR BF-3 inhibitors were recently identied in a study by Lack et al.120 (Fig. 10), where a virtual screening campaign was conducted on the Zinc database and led to the identication of several different scaffolds with potencies ranging from 0.5 to 50 M in inhibiting AR transcriptional activity in cellular models. AR binding was conrmed by surface plasmon resonance (SPR) and ultimately the small molecules binding mode at the BF-3
r 19
Figure 10. AR BF-3 selective inhibitors. Most potent molecules identied from virtual screening campaign on the Zinc database. All these molecules were cocrystallized in the AR BF-3 site.
Figure 11. Inhibitors of the AR FKBP52 association. MJC013 was identied in a recent screen as a possible binder at the BF-3 surface.
site was elucidated by X-ray crystal structures (pdb id: 3ZQT, 2YLO, 2YLP, and 2YLQ). Interestingly, it was observed that the binding at the BF-3 site does not always induce allosteric coactivator displacement from the AF-2 site. Another recent report developed a series of small molecules from a yeast screen which structurally resemble previously identied ufenamic acid as candidates to inhibit the 52kDa FK506 binding protein (FKBP52) association with the AR-Hsp90 complex.121 FKBP association with Hsp90-receptor complex is mediated by the FK1 domain that interacts with the AR LBD and enhances hormone binding to the LBP. Interestingly, mutational analysis suggests that amino acids in the AR LBD that enhance FKBP binding overlap with the recently identied surface BF-3. The compounds identied differ in their activity between wt and mutant receptors in yeast reporter assays, suggesting BF-3 as the potential target site. Moreover, the compounds tested were unable to compete with a uorescently labeled coactivator for the AF-2 site and they do not displace the dihydrotestosterone (DHT) from the LBP. In absence of direct evidence of interaction within BF-3 surface, the authors performed in silico docking simulation to predict the compounds binding mode. The most promising compound, MJC013 (Fig. 11), was found to specically inhibit FKBP52 mediated enhancement of AR-mediated expression of a luciferase reporter gene (with IC50 of 0.45 M) and to block AR-dependent gene expression and hormone-dependent cell proliferation in several prostate cancer cell lines with IC50 consistent with those exhibited in inhibiting luciferase reporter activity. Although direct binding evidence has not yet been provided, this report suggests an additional strategy of non-LBP intervention on the FKBP52/AR interaction, which could be specically modulated with small molecules, and it proposes a new interesting insight on the role of BF-3 in regulating AR function. In addition, the multifactorial role of BF-3 in regulating the AR function as an allosteric site or as a potential proteinprotein interaction site has recently been conrmed by mutational studies and molecular dynamics simulations,122 wherein different BF-3 mutants where predicted to inuence AF-2 subpockets conformational reorganization when compared to AR wt. Such effect varied accordingly to the mutants capacity to activate or inhibit AR transcriptional activity. This has signicant implications for future design strategies for alternative AR surface inhibitors.
20
r CABONI AND LLOYD
Figure 12. Diarylhydrazides as selective inhibitors of AR-coactivator interaction. Diarylhydrazides were designed using a 3D pharmacophore model based on the structural requirements of AR interaction with coactivators.
A structure-based drug-design approach was recently carried out by our own group,123 wherein the structural information available for AR interaction with different coactivator motifs was employed to build a 3D pharmacophore. The resultant model was used to screen commercially available small molecule databases. This approach yielded a novel class of diarylhydrazides (Fig. 12), presenting a rigid linker and two aromatic features that were designed to mimic the F (i) and F (i + 5) of the NR box. Although nonselective for PR, these small molecules are selective for AR over GR, ER , and ER , and act as true AR antagonists, as demonstrated by the lack of intrinsic agonistic activity in inducing prostate-specic antigen (PSA) expression in absence of androgens, and inhibiting both DHT or cyproterone acetate (CPA) stimulated PSA expression at 10 M. The diarylhydrazides identied had potencies between 13 and 26 M in a TR-FRET assay. This novel class of antagonists could potentially be useful in the design and optimization of a specic AR AF-2 pharmacophore model to identify selective inhibitors at the coactivator interface with potential application in CRPC treatment. In parallel to the research conducted on AR BF-3, a virtual screening campaign on the Zinc database identied selective AR AF-2 specic inhibitors124 (Fig. 13). Two classes of small molecules were identied, a class of 3-dihydro-1H-perimidine and a substituted 1H-pyrazol5-(4H)-one. Crystallographic information were obtained for compound 5 (pdb id: 2YHD), showing a potential induced t interaction with Lys720 in the charge clamp. Hits identied displace a uorescently labeled coactivator peptide with potency ranging from 8 to 32 M in an FP assay. Due to the high conservation between NRs within the charge clamp, it would be
Figure 13. Specic AR AF-2 inhibitors. Class of substituted 3-dihydro-1H-perimidine and a substituted 1Hpyrazol-5-(4H)-one identied through a virtual screening campaign on the Zinc database.
r 21
necessary to further explore selectivity. SAR studies would also be advantageous to understand the mechanism of action of the compounds and their in vivo application in CRPC. Nonetheless, this was the rst report where virtual screening was applied in the discovery of selective AR AF-2 binders.
AR non-LBP general approaches. In 2009 two different reports, one from Jones et al.125 and the other from Joseph et al.54 identied second site noncompetitive antiandrogens whose exact mechanism of action and binding site remains to be claried (Fig. 14). Jones et al.125 identied pyrviunium pamoate (PP), an FDA-approved drug as an anthelmintic and harmol hydrochloride (HH), a natural product, as noncompetitive antiandrogens using a conformation-based cellular screen previously reported.126 Differential mechanism of action of the reported compounds was conrmed by their synergistic action in prostate cancer cell lines with a known LBP antiandrogen, bicalutamide, and by in vivo models. PP and HH do not displace radiolabeled DHT from the LBP, and the inhibition of AR maximal activity is not affected in the presence of saturating concentrations of DHT. The compounds demonstrated higher potency than bicalutamide in inhibiting AR transcriptional activity in LNCaP cells, with a reported IC50 of 24 nM for PP and 128 nM for HH. Interestingly, PP and HH act synergistically; therefore, they each adopt a different mechanism of inhibition of AR transcriptional activity. To gain further insight to the mechanism of action, chromatin immunoprecipitation (ChIP) was employed to identify potential interaction with AR-binding sequences in the DNA near androgen responsive genes. In the presence of DHT, HH in synergy with bicalutamide reduces
Figure 14. AR non-LBP general approaches. Small molecules that inhibit AR though non-LBP mechanism of action with unidentied binding site. D36 and D80 were selected by a gelsolin displacement assay, while pyrviunium (PP) and harmol hydrochloride (HH) were identied by a conformational screen.
22
r CABONI AND LLOYD
AR occupancy at AR-binding sites located near androgen responsive genes. The combination of PP, HH, and bicalutamide also reduces RNA pol II occupancy at the PSA promoter (start site), but while HH blocks AR promoter occupancy, PP acts at a different level, by permitting promoter occupancy but blocking subsequent RNA pol II occupancy. Preliminary in vivo studies for PP (HH could not be evaluated due to rapid metabolism) indicate that it could be further explored in clinical studies as an antiandrogen, perhaps in combination with bicalutamide.125 These ndings demonstrate the possibility of modulating AR function without affecting ligand binding, conrmed by a synergistic mechanism of action with classical antagonist both in vitro and in vivo. However, the target-binding site of these compounds remains to be identied and the selectivity across NRs requires additional investigation. D36 and D80 were selected from a conformational screen of a 10,000 member library of small molecules for their ability to disrupt gelsolin, a known AR-interacting protein containing the FXXFF motif, coupled by a functional MMTV luciferase assay.54 Based on this study, D36 and D80 were clustered by inducing a conformational change which closely resembles the apo/unliganded form of the receptor and it is clearly distinguishable from the classical antagonist induced conformational change. Direct interaction with AR was conrmed by SPR and functional cellular assays reveal an inverse agonist mechanism of inhibition as the compounds inhibited basal levels of AR-dependent gene expression, repressed basal pol II, and cofactor recruitment at PSA enhancer below vehicle control. Compounds were found to compete with radiolabeled R1881 (a potent AR agonist) but to induce a unique AR conformation, suggesting the potential binding at a second site. Although the precise mechanism of inhibition remains unclear, the use of conformational screen on a full-length AR offers the advantage to identify unique inhibitors and it is a promising tool to probe ligand induced AR conformational ensembles. In conclusion, until the resolution of an X-ray crystal structure, the binding site for such molecules is only speculative. From the results obtained to date, it is possible to conrm that compounds action is mechanistically different than that of classical LBP-based antagonists and thus they present an additional opportunity to potentially overcome classical antiandrogen resistance. C. Direct Targeting of the Coactivator One of the molecular mechanisms of drug resistance associated with cancer is overexpression of coactivators such as SRC-1 and SRC-3. Therefore, an interesting recent approach proposes it is possible to target the coactivator itself37 instead of the interacting NR surface. Notably, the most commonly used screening assays such as TR-FRET or competition FP experiments are currently unable to distinguish whether a particular molecule targets the NR surface of interaction with coactivators or the coactivator itself. Hence, the authors reevaluated publicly available small molecule hits that disrupt coactivator interaction with NRs for their ability to directly degrade SRC-1 or SRC-3. Amongst these small molecules, gossypol (Fig. 15), was found to directly bind SRC-3 receptor interacting domain (RID) and to reduce SRC-1 and SRC-3 protein levelsindependently of ER expression levels and the ER -bound ligands in MCF-7 breast cancer cells. The downregulation of SRC-1 and SRC-3 induced by gossypol is proteasome independent and occurs at a posttranslational level. This investigation was extended to several cancer cell lines, wherein gossypol reduced SRC-3 protein levels and inhibited cell viability with IC50 values ranging from 2 to 4 M while demonstrating no measureable effect on normal cell lines at the same concentrations. Interestingly, this promiscuous natural polyphenol was originally studied for its effects on fertility and later developed as a male contraceptive.127 It is presently in phase II clinical trial for prostate cancer treatment in combination with androgen ablation therapy and
r 23
Figure 15. SRC direct inhibition. Gossypol was identied from reevaluation of available NR-coactivator inhibitors as a candidate molecule that degrades SRC-1 and SRC-3 coactivators.
it has been evaluated in metastatic breast cancer in phase I/II clinical trials.128 Among the targets by which gossypol exerts its biological effects are 5 reductase,129 11 -hydroxysteroid dehydrogenase,130 and the Bcl-2 family, due to its BH3 mimetic properties.131 Compounds that directly target coactivators again represent a novel strategy to overcome drug resistance. The strength of this approach is given by the essential role of SRC family of coactivators in NRs transcriptional activity, by their overexpression in many cancers and by their crosstalk with other regulatory pathways, such as growth factor signaling cascades, which contributes to acquired chemotherapy resistance development.37, 132 The design of small molecules that directly target the SRC coactivator proteins themselves is challenging due to both the absence of well-dened high-afnity ligand-binding sites and their large size.37, 133 Nevertheless, this was the rst report to demonstrate feasibility of direct SRC inhibition with smallx molecules. D. The DNA-Binding Domain as a Drug Target The DBD represents another alternative promising target for non-LBP-based NR drug discovery. This approach is thought to possibly overcome drug resistance, as all mechanisms of resistance studied to date still involve the NR binding to DNA. This is supported by the fact that NR-DBD interaction with NR-response elements (NRE) specic sequences in DNA is a key step in NR transcriptional activation. Drug discovery at the DBD is challenging, specically in regard to achieving selectivity between the NR family membersthe DBD is the most highly conserved domain in this protein family. Additionally, regulatory elements in the genes may be promiscuous in vitro (with some exceptions). Essentially, there are two main drug-design strategies for inhibition of NR binding to DNA. The rst is aimed at the development of drugs that bind specic DNA sequences and thus disrupt the NR binding. Drugs that are able to bind to a broad spectrum of DNA sequences with high afnity include pyrrole-imidazole (where pyrrole is Py and imidazole is Im) polyamides, which contain Py and Im units usually connected by a amino acid turn to generate a hairpin. A series of hairpin polyamides were designed to inhibit the NR-DBD interaction, in particular the interaction between hERR2 and ER with the ERE134 minor groove. Polyamides inhibit NR binding to DNA with two different mechanisms, one that targets the minor groove directly, and one indirect, where polyamides binding at the DNA minor groove allosterically disrupt the conformation of the opposing major groove within the ERE. This allosteric mechanism of action has been conrmed by X-ray crystallography.135 Hairpin polyamides were also designed to inhibit AR DNA binding interaction.136 Although NR selectivity was not raised before with polyamides, Nickols and Dervan136
24
r CABONI AND LLOYD
designed androgen-responsive element (ARE) specic hairpin polyamides. The designed hairpin polyamides were cell permeable and demonstrated binding to AR-specic ARE at the PSA promoter and at other AR regulated genes, such as transmembrane protease serine 2 (TMPRSS2) and FKBP5, downregulating their expression in a similar fashion to that achieved by bicalutamide when tested at the same concentration. Chenoweth et al.137 improved the DNA-binding afnity of the hairpin polyamides by designing and synthesizing a series of cyclic polyamides. In a cyclic polyamide, the classic hairpin previously described is closed with a second amino acid turn. Cyclization had no bearing on cell permeability and reduced mRNA levels of AR-dependent genes such as PSA in a comparable way to the hairpin polyamides. In vitro ADMET studies also showed favorable pharmacokinetic proles, characterized by good solubility, high liver stability, and low toxicity. An AR DNA binding inhibition approach could potentially be applied in CRPC, and also to study AR-dependent gene expression. However, some questions remain. A single polyamide would unlikely be able to affect all AR-dependent genes simultaneously, as the ARE is degenerate, but it is indeed possible that one, or a small mixture of, tailored polyamide(s) could selectively target groups of AR-dependent genes on the basis of their similarities in ARE.136 A similar approach was applied in the design of specic polyamides directed to GR glucocorticoid response elements (GRE), in an effort to uncouple transrepression (DNA-binding independent) and transactivation (DNA-binding dependent) mechanisms of GR transcriptional activity. Due the promiscuity of hormone response elements (HRE), the selectivity of polyamides has not been investigated fully and in vivo validation is outstanding. Alternatively to polyamides, the inhibition of ER binding to DNA was also achieved by the design of a novel inhibitor, XR5944, originally rationally designed as a topoisomerase inhibitor.138 This study conrmed how XR5944s mechanism of action is independent of topoisomerase binding but rather is related to inhibition of transcription through specic binding to the DNA major groove (the site of transcription factors binding). In this report, XR5944 specically inhibited ER transcriptional activity in MCF-7 cellular models cells by disrupting ER-ERE interaction directly. The second NR DNA binding inhibition strategy is to target the DBD itself, focusing on the zinc ngers. Zinc ngers have a crucial role in NR-DBD stability and NR dimerization. As demonstrated by Whittal et al.,139 disruption of zinc ngers has a critical impact in NR function. ER DNA-binding is reversibly inhibited by a range of oxidizing agents that disrupt the ZnS interaction by eliminating the zinc ion and forming a disulde bond. A structure-based drug-design strategy was encouraged by the disclosure of the rst X-ray crystal structures for GR6 and ER.140 The structural arrangement of the DBD, consisting of two zing ngers coordinating four cysteines inspired the design of (weak) electrophilic compounds which displace Zn2+ from the DBD and react with the sulfhydryl of the cysteines,141 or alternatively, metalchelating compounds. Through such strategies, it has been possible to design small molecule inhibitors, which have permeability advantages over polyamides. However, it remains challenging to achieve specicity for a particular NR or specicity across diverse range of cellular zinc ngers. Electrophilic compounds such as disulde benzamide (DIBA; Fig. 16), and benzisothiazolone derivatives (BITA) were evaluated for their ability to inhibit estrogen stimulated proliferation in MCF-7 ER positive cells in vitro and in vivo.142 DIBA and BITA selectively react within ER DBD zinc ngers, as monitored by a zinc-ejection assay, and block ER binding to its responsive elements and subsequent ER-mediated transactivation gene transcription.142, 143 The selectivity preference toward ER zinc ngers was evaluated in comparison to other NRs like PR and to other zinc-dependent and nonzinc-dependent transcription factors. By investigating the mechanism of action of DIBA and BITA in depth, it was hypothesized that DBD targeting can circumvent tamoxifen resistance and restore tamoxifen sensitivity in breast cancer cells143
r 25
Figure 16. Small molecules as NR DNA-binding domain inhibitors. Small molecules inhibitors of DBD identied through HTS. Electrophilic molecules such as TPBM and its analogs TPSF, DIBA, and most recently identied anacardic acid.
via modulation of interdomain communications with the NTD and coregulator recruitment as demonstrated by in vivo experiments. These reports provide proof of concept for the possibility of selectively disrupting NR-DNA interaction as an alternative mechanism for the treatment of advanced breast cancer to overcome drug resistance. Another class of DBD inhibitors was retrieved from an HTS screen using uorescence anisotropy microplates, evaluating the binding capability of small molecules for a uorescently labeled consensus ERE (cERE). A small molecule, theophylline, 8-[(benzylthio)methyl](7Cl,8Cl) (TPBM)144 (Fig. 16), was found to disrupt estradiol-ER complex binding to ERE with an IC50 of around 3 M. Unlike the majority of the small molecules identied by this HTS screen, TPBM is selective toward for ER over other members of the NR family (AR, GR, and PR). It exhibits low toxicity in ER-negative cell lines. The compound inhibits estradiol-induced transactivation in both estrogen-dependent cells and in tamoxifen-resistant cells. TPBM does not displace estradiol from the LBP and therefore inhibits estradiol-ER complex association to cERE with an alternative mechanism of action, which is not due to zinc chelation in the DBD, in contrast to the described mechanism of inhibition for electrophiles such as DIBA. Although it can be classied as an alternative non-LBP inhibitor, the actual binding site of TPBM remains to be claried. A cellular-based screen on a set of structural analogs of TPBM, identied p-uoro-4(1,2,3,6,-tetrahydro-1,3-dimethyl-2-oxo-6-thionpurin-8-ylthio) (TPSF)145 as a noncompetitive inhibitor of ER (Fig. 16), which was 15-fold more potent than TPBM in inhibiting ER mediated gene expression in cells. Intriguingly, although structurally similar to TPBM, TPSF has a different and unique mechanism of action in that it does not disrupt ER binding to cERE in vitro. TPSF affects ER protein levels in breast cancer cells, while TPBM does not. This effect was blocked by the proteasome inhibitor MG132, suggesting that the promotion of TPSF ER degradation is proteasome dependent. Just like TBPM, its mechanism of action is noncompetitive. The inhibitory doseresponse curve does not present a rightward shift in presence of increasing concentrations of estradiol. TPSF tethers estradiol-ER complex
26
r CABONI AND LLOYD
through DNA-bound transcriptional regulators, demonstrated by the inhibition of induction of cyclin D1 mRNA, a contributing factor to MCF-7 cell growth in vitro. TPSF demonstrated ER selectivity and only weakly inhibits AR and GR. Since the compound showed inhibition of ER transcriptional activity in estradiol-dependent and tamoxifen-resistant cell lines, and exhibited low toxicity in ER-negative cell lines, it is positioned as a promising agent for the treatment of advanced breast cancer. More recently, anacardic acid (C24:1), already known as a histone acetyltransferase inhibitor146 (Fig. 16), was shown to inhibit breast cancer cell proliferation regardless of their endocrine/tamoxifen sensitivity.147 Interaction with the ER DBD and in vitro ERE inhibition was supported by electrophoretic mobility shift assay (EMSA) and by molecular docking studies, where anacardic acid was predicted to bind between the zinc ngers of the DBD with much higher afnity when compared to the LBP. In support of an alternative non-LBP mechanism of action, anacardic acid does not displace radiolabeled estradiol from the LBP of ER or ER and ChIP data show that treatment of ER positive MCF-7 cells with anacardic acid blocks the estradiol-induced ER occupancy at the endogenous pS2 gene promoter. Anacardic acids predicted binding site in the DBD zinc ngers could be reasonably justied by its known antioxidant properties, due to its action as a metal-chelating agent.148 However, this hypothesis remains to be conrmed by a zinc displacement assay. Although anacardic acid was found to be approximately twofold selective toward ER in reducing estradiol-induced ERE reporter activity compared to ER , the selectivity against other NRs remains to be assessed and quantied. E. The N-Terminal Domain as a Drug Target Sadar and colleagues discovered the rst NR-NTD small molecule inhibitor, termed EPI001149 (Fig. 17). The AF-1 region in the NTD is deemed essential for ligand-dependent AR
Figure 17. AR NTD directed inhibitors. EPI-001 is the rst selective small molecule inhibitor. Sintokamides and niphatenones result from a screen of marine products extracts.
r 27
transcriptional activity,3941 but it also regulates ligand-independent transactivation.150152 Targeting this interface with small molecule inhibitors could be valuable for the treatment of CRPC. For such reasons, the AR-NTD has been dened as the Achilles heel of AR activity.38 Another putative advantage of targeting the NTD, is a greater potential to achieve selectivity over other NRs, due to low degree of sequence similarity (<15% homology across the NR family) in this site. Current drug discovery efforts are hindered by the lack of an X-ray crystal structure for the NTD, which is an intrinsically disordered region of high exibility. Nonetheless, in the case of AR, HTS approaches have been so far successful in discovering different classes of peptide and small molecule inhibitors.38 In a rst rational approach, a fragment corresponding to the NTD peptide itself was overexpressed to create decoy molecules (AR1_558 ) that competitively bind the interacting proteins required for activation of the endogenous full-length AR and inhibit specically AR transcriptional activity in androgen-dependent and androgen-independent conditions (Fig. 17). Decoy peptides were found to be AR selective, except for cross reactivity with PR.153 This provided the rst evidence for NR-NTD targeting as a novel therapeutic approach, whose functional implications were validated by in vivo studies. In an effort to probe the AR-NTD function with selective small molecule inhibitors, Sadar et al.154 screened a library of marine sponge extracts, and found small chlorinated peptides from the sponge Dysidea sp. as AR-NTD transactivation inhibitors. Sintokamides (Fig. 17) present a common peptidic rigid scaffold, which varies according to the number of chlorinated substituents. Sintokamides inhibit proliferation in androgen-sensitive prostate cancer cells and also to inhibit androgen-independent signaling, represented by forskolin-induced AR-NTD transcriptional activity. A similar screen was more recently conducted, leading to the discovery of another class of natural products, niphatenones, extracts of the marine sponge Niphates digitalis as NTD transactivation inhibitors155 (Fig. 17). Niphatenone B alkynyl ether analog was found to covalently interact within the AR-NTD AF-1. Niphatenones do not bear a rigid peptidic scaffold, rather they present more drug-like properties, and could inspire a new class of small molecule irreversible inhibitors relevant to the treatment of CRPC. EPI-001, a bisphenol A diglycidic ether analog (BADGE), was the rst small molecule identied to selectively block proteinprotein interactions at the AR-NTD by interacting with the AF-1 in vitro and in vivo,149 without interfering with the ligand binding in the LBP. EPI-001 was found to block both androgen-dependent and androgen-independent (forskolin and IL-6 induced) activation of AR-dependent genes, such as PSA and TMPRSS2. EPI-001 inhibited NTD transactivation in the 22Rv1 cellular model, which expresses both full-length AR and constitutively active splice variants that lack the LBD. Truncated splice variants that constitutively activate AR transcriptional activity are one of the feature characteristics of CRPC.156 In vivo studies demonstrate selective inhibition of androgen-dependent tumor growth and low toxicity. According to these reports, EPI-001 could potentially overcome the limitations of LBP antiandrogens and contribute toward the understanding of the molecular mechanism of CRPC and by selectively silencing the ligand-independent AR-transactivation pathway through the NTD. The strong correlation between CRPC and the AR AF-1 is supported by several lines of evidence, such as the presence of constitutively active splice variants lacking the LBD as mentioned above, and by evidence that the ligand-independent pathway does not rely on the AR AF-2.157 Thus, a successful candidate NTD inhibitor would block AR transcriptional activity regardless of the progression status of prostate cancer, both in ligand-dependent and ligand-independent conditions. To date, compounds identied have shown direct interaction within the AR-NTD and inhibition of NTD-mediated transcriptional activity, suggesting the feasibility of this alternative strategy. Further studies need to be carried out to validate the
28
r CABONI AND LLOYD
NTD inhibition and so exploit this target for other NR-mediated diseases where LBP-based therapy fails.
4. CONCLUSIONS During the past decade, NR drug discovery research has made considerable progress in identifying and validating alternative non-LBP sites as drug targets to overcome the limitations of classic LBP-based therapeutic regimens, which have been the mainstay for treatment for over 40 years.158 Alternative non-LBP drug discovery is driven by increasing structural studies efforts which further our understanding of how NRs interact with their coregulator protein partners and DNA to initiate transcriptional activity. While traditional approaches have identied molecules that indirectly inhibit NR activity through LBP ligand displacement, and elucidated the molecular pharmacology of agonism/antagonism activity and coregulator recruitment, non-LBP-based approaches aim at the direct inhibition of NR transcriptional interfaces. Thinking outside the LBP box could overcome drug resistance mechanisms, which, due to mutations and functional changes from antagonists to agonists, are far more prevalent when targeting the LBP. Several non-LBP strategies have been explored to date, across domains beyond the NR LBD, such as the NTD and the DBD. Very little is still known about the NTD ligandindependent transactivation mechanism and no X-ray crystal structures are available, and as a result, research in this eld has been promising but limited to only AR. The DBD approach has been applied only to AR, ER, and GR, but in vivo and functional studies to validate the NR-DBD as a possible candidate site to overcome drug resistance are still lacking. In contrast, NR-cofactor interaction has received the lions share of non-LBP attention as a druggable and promising target for the development of small molecule inhibitors. Studies to date have identied specic peptidic sequences required for coregulator interaction with different NRs and related potent peptide inhibitors, and have developed several small organic molecules as peptidomimetics, mostly using HTS or a de novo drug-design approach. Our understanding of the structural requirements for NR-cofactor interaction has also encouraged structure-based and ligand-based virtual screening approaches. X-ray crystallography has been successfully used as a complementary tool in HTS campaigns and in the identication of novel regulatory sites on the NR surface.118 NR selectivity at the coregulator interface is by no means fully explored, due to the interplay of numerous factors, including the nature of the ligand bound in the LBP, which generate different conformations, and perhaps different regulatory sites on the NR surface.24, 42, 159 Recently, Sun et al113 proposed the design of CBI targeted libraries with increased molecular weight to improve the (small) molecule coverage of the coactivator binding site. These parameters are often overlooked in routine screening campaigns that consider drug-like libraries with small molecules, striving to satisfy Lipinskis rule of ve for orally bioavailable drugs. CBI potency may also be improved by exploiting hydrogen bond networks on the AF-2 surface7 and the small molecules charge complementarity with the charge clamp. Beyond this, irreversible inhibitors have been identied for several targets, including TR, VDR, and AR AF-1. Irreversible inhibition may afford improvement of potency, but such an approach is not divorced from potential clinical drawbacks in vivo such as a toxicity and a potential lack of specicity.160 More comprehensive SAR studies are needed for compounds targeting other than the LBP in NRs, with a view to progressing from hit to lead and further into eventual clinical evaluation. Nonetheless, the future of non-LBP NRs drug discovery remains promising and is gaining pace, building on the rapid progress made in just the past 10 years.
BEYOND THE LIGAND-BINDING POCKET ACKNOWLEDGMENTS
r 29
The authors thank the Irish Health Research Board (HRB,; HRB/2007/2), Enterprise Ireland (EI, CFTD/06/110) for research funding. The Trinity Biomedical Sciences Institute is supported a capital infrastructure investment from Cycle 5 of the Irish Higher Education Authoritys Programme for Research in Third Level Institutions (PRTLI).
REFERENCES
1. Gronemeyer H, Gustafsson JA, Laudet V. Principles for modulation of the nuclear receptor superfamily. Nat Rev Drug Discov 2004;3(11):950964. 2. Imming P, Sinning C, Meyer A. Drugs, their targets and the nature and number of drug targets. Nat Rev Drug Discov 2006;5(10):821834. 3. Germain P, Staels B, Dacquet C, Spedding M, Laudet V. Overview of nomenclature of nuclear receptors. Pharmacol Rev 2006;58(4):685704. 4. He B, Kemppainen JA, Voegel JJ, Gronemeyer H, Wilson EM. Activation function 2 in the human androgen receptor ligand binding domain mediates interdomain communication with the NH(2)terminal domain. J Biol Chem 1999;274(52):3721937225. 5. De Bosscher K, Vanden Berghe W, Haegeman G. Cross-talk between nuclear receptors and nuclear factor kappaB. Oncogene 2006;25(51):68686886. 6. Luisi BF, Xu WX, Otwinowski Z, Freedman LP, Yamamoto KR, Sigler PB. Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 1991;352(6335):497 505. 7. Hur E, Pfaff SJ, Payne ES, Gron H, Buehrer BM, Fletterick RJ. Recognition and accommodation at the androgen receptor coactivator binding interface. PLoS Biol 2004;2(9):E274. 8. MOE 2011.10. Montreal: Chemical Computing Group. www.chemcomp.com; 2011. 9. Mohla S, Stearns V, Sathyamoorthy N, Rosenfeld MG, Nelson P. The biology of hormone refractory breast and prostate cancer: An NCI workshop report. Cancer Biol Ther 2009;8(21):1975 1985. 10. Estebanez-Perpina E, Jouravel N, Fletterick RJ. Perspectives on designs of antiandrogens for prostate cancer. Expert Opin Drug Discov 2007;2(10):13411355. 11. Tran C, Ouk S, Clegg NJ, Chen Y, Watson PA, Arora V, Wongvipat J, Smith-Jones PM, Yoo D, Kwon A, Wasielewska T, Welsbie D, Chen CD, Higano CS, Beer TM, Hung DT, Scher HI, Jung ME, Sawyers CL. Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science 2009;324(5928):787790. 12. Scher HI, Beer TM, Higano CS, Anand A, Taplin ME, Efstathiou E, Rathkopf D, Shelkey J, Yu EY, Alumkal J, Hung D, Hirmand M, Seely L, Morris MJ, Danila DC, Humm J, Larson S, Fleisher M, Sawyers CL. Antitumour activity of MDV3100 in castration-resistant prostate cancer: A phase 1-2 study. Lancet 2010;375(9724):14371446. 13. Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller MD, de Wit R, Mulders P, Chi KN, Shore ND, Armstrong AJ, Flaig TW, Flechon A, Mainwaring P, Fleming M, Hainsworth JD, Hirmand M, Selby B, Seely L, de Bono JS. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med 2012; in press. 14. Wakeling AE, Dukes M, Bowler J. A potent specic pure antiestrogen with clinical potential. Cancer Res 1991;51(15):38673873. 15. Morris C, Wakeling A. Fulvestrant (Faslodex)A new treatment option for patients progressing on prior endocrine therapy. Endocr Relat Cancer 2002;9(4):267276. 16. Chang CY, McDonnell DP. Androgen receptor-cofactor interactions as targets for new drug discovery. Trends Pharmacol Sci 2005;26(5):225228. Medicinal Research Reviews DOI 10.1002/med
30
r CABONI AND LLOYD
17. Moore TW, Mayne CG, Katzenellenbogen JA. Minireview: Not picking pockets: Nuclear receptor alternate-site modulators (NRAMs). Mol Endocrinol 2009;24(4):683695. 18. Shapiro DJ, Mao C, Cherian MT. Small molecule inhibitors as probes for estrogen and androgen receptor action. J Biol Chem 2011;286(6):40434048. 19. Vaz B, Mocklinghoff S, Brunsveld L. Targeting the nuclear receptor-cofactor interaction. In: Ottow E, Weinmann H, Eds. Nuclear Receptors as Drug Targets, Methods and Principles in Medicinal Chemistry. Wiley-VCH; 2008. p 2545. 20. van de Wijngaart DJ, Dubbink HJ, van Royen ME, Trapman J, Jenster G. Androgen receptor coregulators: Recruitment via the coactivator binding groove. Mol Cell Endocrinol 2012;352(1 2):5769. 21. Gao W. Peptide antagonist of the androgen receptor. Curr Pharm Des 2010;16(9):11061113. 22. Estebanez-Perpina E, Fletterick, RJ. The androgen receptor coactivator-binding interface. In: Tindall D, Mohler J, Eds. Androgen Action in Prostate Cancer. Springer; 2009. p 297311. 23. Arkin MR, Wells JA. Small-molecule inhibitors of protein-protein interactions: Progressing towards the dream. Nat Rev Drug Discov 2004;3(4):301317. 24. Kong EH, Heldring N, Gustafsson JA, Treuter E, Hubbard RE, Pike AC. Delineation of a unique protein-protein interaction site on the surface of the estrogen receptor. Proc Natl Acad Sci USA 2005;102(10):35933598. 25. Lonard DM, Lanz RB, OMalley BW. Nuclear receptor coregulators and human disease. Endocr Rev 2007;28(5):575587. 26. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL. The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 1998;95(7):927937. 27. Mocklinghoff S, Rose R, Carraz M, Visser A, Ottmann C, Brunsveld L. Synthesis and crystal structure of a phosphorylated estrogen receptor ligand binding domain. Chembiochem 2010;11(16):2251 2254. 28. Estebanez-Perpina E, Moore JM, Mar E, Delgado-Rodrigues E, Nguyen P, Baxter JD, Buehrer BM, Webb P, Fletterick RJ, Guy RK. The molecular mechanisms of coactivator utilization in ligand-dependent transactivation by the androgen receptor. J Biol Chem 2005;280(9):8060 8068. 29. Matias PM, Donner P, Coelho R, Thomaz M, Peixoto C, Macedo S, Otto N, Joschko S, Scholz P, Wegg A, Basler S, Schafer M, Egner U, Carrondo MA. Structural evidence for ligand specicity in the binding domain of the human androgen receptor. Implications for pathogenic gene mutations. J Biol Chem 2000;275(34):2616426171. 30. Bledsoe RK, Montana VG, Stanley TB, Delves CJ, Apolito CJ, McKee DD, Consler TG, Parks DJ, Stewart EL, Willson TM, Lambert MH, Moore JT, Pearce KH, Xu HE. Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition. Cell 2002;110(1):93105. 31. Hsieh RW, Rajan SS, Sharma SK, Greene GL. Molecular characterization of a B-ring unsaturated estrogen: Implications for conjugated equine estrogen components of premarin. Steroids 2008;73(1):5968. 32. He B, Minges JT, Lee LW, Wilson EM. The FXXLF motif mediates androgen receptor-specic interactions with coregulators. J Biol Chem 2002;277(12):1022610235. 33. Hsu CL, Chen YL, Yeh S, Ting HJ, Hu YC, Lin H, Wang X, Chang C. The use of phage display technique for the isolation of androgen receptor interacting peptides with (F/W)XXL(F/W) and FXXLY new signature motifs. J Biol Chem 2003;278(26):2369123698. 34. He B, Kemppainen JA, Wilson EM. FXXLF and WXXLF sequences mediate the NH2-terminal interaction with the ligand binding domain of the androgen receptor. J Biol Chem 2000;275(30):22986 22994. 35. He B, Gampe RT, Jr, Kole AJ, Hnat AT, Stanley TB, An G, Stewart EL, Kalman RI, Minges JT, Wilson EM. Structural basis for androgen receptor interdomain and coactivator interactions Medicinal Research Reviews DOI 10.1002/med
r 31
36.
37.
38. 39.
40. 41.
42.
43.
44.
45. 46.
47.
48.
49. 50.
51.
52.
suggests a transition in nuclear receptor activation function dominance. Mol Cell 2004;16(3):425 438. Dubbink HJ, Hersmus R, Verma CS, van der Korput HA, Berrevoets CA, van Tol J, Ziel-van der Made AC, Brinkmann AO, Pike AC, Trapman J. Distinct recognition modes of FXXLF and LXXLL motifs by the androgen receptor. Mol Endocrinol 2004;18(9):21322150. Wang Y, Lonard DM, Yu Y, Chow DC, Palzkill TG, OMalley BW. Small molecule inhibition of the steroid receptor coactivators, SRC-3 and SRC-1. Mol Endocrinol 2011;25(12):2041 2053. Sadar MD. Small molecule inhibitors targeting the achilles heel of androgen receptor activity. Cancer Res 2011;71(4):12081213. Jenster G, van der Korput HA, van Vroonhoven C, van der Kwast TH, Trapman J, Brinkmann AO. Domains of the human androgen receptor involved in steroid binding, transcriptional activation, and subcellular localization. Mol Endocrinol 1991;5(10):13961404. Simental JA, Sar M, Lane MV, French FS, Wilson EM. Transcriptional activation and nuclear targeting signals of the human androgen receptor. J Biol Chem 1991;266(1):510518. Rundlett SE, Wu XP, Miesfeld RL. Functional characterizations of the androgen receptor conrm that the molecular basis of androgen action is transcriptional regulation. Mol Endocrinol 1990;4(5):708714. Norris JD, Paige LA, Christensen DJ, Chang CY, Huacani MR, Fan D, Hamilton PT, Fowlkes DM, McDonnell DP. Peptide antagonists of the human estrogen receptor. Science 1999;285(5428):744 746. Paige LA, Christensen DJ, Gron H, Norris JD, Gottlin EB, Padilla KM, Chang CY, Ballas LM, Hamilton PT, McDonnell DP, Fowlkes DM. Estrogen receptor (ER) modulators each induce distinct conformational changes in ER alpha and ER beta. Proc Natl Acad Sci USA 1999;96(7):3999 4004. Chang C, Norris JD, Gron H, Paige LA, Hamilton PT, Kenan DJ, Fowlkes D, McDonnell DP. Dissection of the LXXLL nuclear receptor-coactivator interaction motif using combinatorial peptide libraries: Discovery of peptide antagonists of estrogen receptors alpha and beta. Mol Cell Biol 1999;19(12):82268239. Hall JM, Chang CY, McDonnell DP. Development of peptide antagonists that target estrogen receptor beta-coactivator interactions. Mol Endocrinol 2000;14(12):20102023. Mettu NB, Stanley TB, Dwyer MA, Jansen MS, Allen JE, Hall JM, McDonnell DP. The nuclear receptor-coactivator interaction surface as a target for peptide antagonists of the peroxisome proliferator-activated receptors. Mol Endocrinol 2007;21(10):23612377. Yang J, Chang CY, Sa R, Morgan J, McDonnell DP, Fuller PJ, Clyne CD, Young MJ. Identication of ligand-selective peptide antagonists of the mineralocorticoid receptor using phage display. Mol Endocrinol 2011;25(1):3243. Northrop JP, Nguyen D, Piplani S, Olivan SE, Kwan ST, Go NF, Hart CP, Schatz PJ. Selection of estrogen receptor beta- and thyroid hormone receptor beta-specic coactivator-mimetic peptides using recombinant peptide libraries. Mol Endocrinol 2000;14(5):605622. Chang CY, McDonnell DP. Evaluation of ligand-dependent changes in AR structure using peptide probes. Mol Endocrinol 2002;16(4):647660. Chang CY, Abdo J, Hartney T, McDonnell DP. Development of peptide antagonists for the androgen receptor using combinatorial peptide phage display. Mol Endocrinol 2005;19(10):2478 2490. van de Wijngaart DJ, van Royen ME, Hersmus R, Pike AC, Houtsmuller AB, Jenster G, Trapman J, Dubbink HJ. Novel FXXFF and FXXMF motifs in androgen receptor cofactors mediate high afnity and specic interactions with the ligand-binding domain. J Biol Chem 2006;281(28):19407 19416. Nishimura K, Ting HJ, Harada Y, Tokizane T, Nonomura N, Kang HY, Chang HC, Yeh S, Miyamoto H, Shin M, Aozasa K, Okuyama A, Chang C. Modulation of androgen receptor Medicinal Research Reviews DOI 10.1002/med
32
r CABONI AND LLOYD

transactivation by gelsolin: A newly identied androgen receptor coregulator. Cancer Res 2003;63(16):48884894. van de Wijngaart DJ, Dubbink HJ, Molier M, de Vos C, Jenster G, Trapman J. Inhibition of androgen receptor functions by gelsolin FxxFF peptide delivered by transfection, cell-penetrating peptides, and lentiviral infection. Prostate 2011;71(3):241253. Joseph JD, Wittmann BM, Dwyer MA, Cui H, Dye DA, McDonnell DP, Norris JD. Inhibition of prostate cancer cell growth by second-site androgen receptor antagonists. Proc Natl Acad Sci USA 2009;106(29):1217812183. Steketee K, Berrevoets CA, Dubbink HJ, Doesburg P, Hersmus R, Brinkmann AO, Trapman J. Amino acids 3-13 and amino acids in and anking the 23FxxLF27 motif modulate the interaction between the N-terminal and ligand-binding domain of the androgen receptor. Eur J Biochem 2002;269(23):57805791. Dubbink HJ, Hersmus R, Pike AC, Molier M, Brinkmann AO, Jenster G, Trapman J. Androgen receptor ligand-binding domain interaction and nuclear receptor specicity of FXXLF and LXXLL motifs as determined by L/F swapping. Mol Endocrinol 2006;20(8):17421755. Vaz B, Mocklinghoff S, Folkertsma S, Lusher S, de Vlieg J, Brunsveld L. Computational design, synthesis, and evaluation of miniproteins as androgen receptor coactivator mimics. Chem Commun (Camb) 2009; (36):53775379. Phan T, Nguyen HD, Goksel H, Mocklinghoff S, Brunsveld L. Phage display selection of miniprotein binders of the estrogen receptor. Chem Commun (Camb) 2010;46(43):82078209. Kurebayashi S, Nakajima T, Kim SC, Chang CY, McDonnell DP, Renaud JP, Jetten AM. Selective LXXLL peptides antagonize transcriptional activation by the retinoid-related orphan receptor RORgamma. Biochem Biophys Res Commun 2004;315(4):919927. Pathrose P, Barmina O, Chang CY, McDonnell DP, Shevde NK, Pike JW. Inhibition of 1,25dihydroxyvitamin D3-dependent transcription by synthetic LXXLL peptide antagonists that target the activation domains of the vitamin D and retinoid X receptors. J Bone Miner Res 2002;17(12):21962205. Pike JW, Pathrose P, Barmina O, Chang CY, McDonnell DP, Yamamoto H, Shevde NK. Synthetic LXXLL peptide antagonize 1,25-dihydroxyvitamin D3-dependent transcription. J Cell Biochem 2003;88(2):252258. Lu Y, Yang J, Sega E. Issues related to targeted delivery of proteins and peptides. AAPS J 2006;8(3):E466478. Geistlinger TR, Guy RK. An inhibitor of the interaction of thyroid hormone receptor beta and glucocorticoid interacting protein 1. J Am Chem Soc 2001;123(7):15251526. Geistlinger TR, Guy RK. Novel selective inhibitors of the interaction of individual nuclear hormone receptors with a mutually shared steroid receptor coactivator 2. J Am Chem Soc 2003;125(23):6852 6853. Leduc AM, Trent JO, Wittliff JL, Bramlett KS, Briggs SL, Chirgadze NY, Wang Y, Burris TP, Spatola AF. Helix-stabilized cyclic peptides as selective inhibitors of steroid receptor-coactivator interactions. Proc Natl Acad Sci USA 2003;100(20):1127311278. Galande AK, Bramlett KS, Burris TP, Wittliff JL, Spatola AF. Thioether side chain cyclization for helical peptide formation: Inhibitors of estrogen receptor-coactivator interactions. J Pept Res 2004;63(3):297302. Galande AK, Bramlett KS, Trent JO, Burris TP, Wittliff JL, Spatola AF. Potent inhibitors of LXXLL-based protein-protein interactions. Chembiochem 2005;6(11):19911998. Carraz M, Zwart W, Phan T, Michalides R, Brunsveld L. Perturbation of estrogen receptor alpha localization with synthetic nona-arginine LXXLL-peptide coactivator binding inhibitors. Chem Biol 2009;16(7):702711. Phillips C, Roberts LR, Schade M, Bazin R, Bent A, Davies NL, Moore R, Pannifer AD, Pickford AR, Prior SH, Read CM, Scott A, Brown DG, Xu B, Irving SL. Design and structure of stapled peptides binding to estrogen receptors. J Am Chem Soc 2011;133(25):96969699.
53.
54.
55.
56.
57.
58. 59.
60.
61.
62. 63. 64.
65.
66.
67. 68.
69.
r 33
70. Bird GH, Madani N, Perry AF, Princiotto AM, Supko JG, He X, Gavathiotis E, Sodroski JG, Walensky LD. Hydrocarbon double-stapling remedies the proteolytic instability of a lengthy peptide therapeutic. Proc Natl Acad Sci USA 2010;107(32):1409314098. 71. Arnold SF, Notides AC. An antiestrogen: A phosphotyrosyl peptide that blocks dimerization of the human estrogen receptor. Proc Natl Acad Sci USA 1995;92(16):74757479. 72. Yudt MR, Koide S. Preventing estrogen receptor action with dimer-interface peptides. Steroids 2001;66(7):549558. 73. Chakraborty S, Cole S, Rader N, King C, Rajnarayanan R, Biswas PK. In silico design of peptidic inhibitors targeting estrogen receptor alpha dimer interface. Mol Divers 2012; in press. 74. Yen PM. Physiological and molecular basis of thyroid hormone action. Physiol Rev 2001;81(3):1097 1142. 75. Baxter JD, Goede P, Apriletti JW, West BL, Feng W, Mellstrom K, Fletterick RJ, Wagner RL, Kushner PJ, Ribeiro RC, Webb P, Scanlan TS, Nilsson S. Structure-based design and synthesis of a thyroid hormone receptor (TR) antagonist. Endocrinology 2002;143(2):517524. 76. Webb P, Nguyen NH, Chiellini G, Yoshihara HA, Cunha Lima ST, Apriletti JW, Ribeiro RC, Marimuthu A, West BL, Goede P, Mellstrom K, Nilsson S, Kushner PJ, Fletterick RJ, Scanlan TS, Baxter JD. Design of thyroid hormone receptor antagonists from rst principles. J Steroid Biochem Mol Biol 2002;83(15):5973. 77. Arnold LA, Estebanez-Perpina E, Togashi M, Jouravel N, Shelat A, McReynolds AC, Mar E, Nguyen P, Baxter JD, Fletterick RJ, Webb P, Guy RK. Discovery of small molecule inhibitors of the interaction of the thyroid hormone receptor with transcriptional coregulators. J Biol Chem 2005;280(52):4304843055. 78. Arnold LA, Estebanez-Perpina E, Togashi M, Shelat A, Ocasio CA, McReynolds AC, Nguyen P, Baxter JD, Fletterick RJ, Webb P, Guy RK. A high-throughput screening method to identify small molecule inhibitors of thyroid hormone receptor coactivator binding. Sci STKE 2006;2006(341):pl3. 79. Estebanez-Perpina E, Arnold LA, Jouravel N, Togashi M, Blethrow J, Mar E, Nguyen P, Phillips KJ, Baxter JD, Webb P, Guy RK, Fletterick RJ. Structural insight into the mode of action of a direct inhibitor of coregulator binding to the thyroid hormone receptor. Mol Endocrinol 2007;21(12):2919 2928. 80. Arnold LA, Kosinski A, Estebanez-Perpina E, Fletterick RJ, Guy RK. Inhibitors of the interaction of a thyroid hormone receptor and coactivators: Preliminary structure-activity relationships. J Med Chem 2007;50(22):52695280. 81. Hwang JY, Arnold LA, Zhu F, Kosinski A, Mangano TJ, Setola V, Roth BL, Guy RK. Improvement of pharmacological properties of irreversible thyroid receptor coactivator binding inhibitors. J Med Chem 2009;52(13):38923901. 82. Hwang JY, Huang W, Arnold LA, Huang R, Attia RR, Connelly M, Wichterman J, Zhu F, Augustinaite I, Austin CP, Inglese J, Johnson RL, Guy RK. Methylsulfonylnitrobenzoates, a new class of irreversible inhibitors of the interaction of the thyroid hormone receptor and its obligate coactivators that functionally antagonizes thyroid hormone. J Biol Chem 2011;286(14):11895 11908. 83. Johnson RL, Hwang JY, Arnold LA, Huang R, Wichterman J, Augustinaite I, Austin CP, Inglese J, Guy RK, Huang W. A quantitative high-throughput screen identies novel inhibitors of the interaction of thyroid receptor beta with a peptide of steroid receptor coactivator 2. J Biomol Screen 2011;16(6):618627. 84. Huang W, Johnson RL, Huang R, Wichterman J, Hwang JY, Guy RK. qHTS for Inhibitors of the Interaction of Thyroid Hormone Receptor and Steroid Receptor Coregulator 2. Probe Reports from the NIH Molecular Libraries Program, 2011/07/08 ed. Bethesda, MD: National Center for Biotechnology Information (US); 2010. 85. Hwang JY, Attia RR, Zhu F, Yang L, Lemoff A, Jeffries C, Connelly MC, Guy RK. Synthesis and evaluation of sulfonylnitrophenylthiazoles (SNPTs) as thyroid hormone receptor-coactivator interaction inhibitors. J Med Chem 2012;55(5):23012310. Medicinal Research Reviews DOI 10.1002/med
34
r CABONI AND LLOYD
86. Mita Y, Dodo K, Noguchi-Yachide T, Miyachi H, Makishima M, Hashimoto Y, Ishikawa M. LXXLL peptide mimetics as inhibitors of the interaction of vitamin D receptor with coactivators. Bioorg Med Chem Lett 2010;20(5):17121717. 87. Cummings MD, Schubert C, Parks DJ, Calvo RR, LaFrance LV, Lattanze J, Milkiewicz KL, Lu T. Substituted 1,4-benzodiazepine-2,5-diones as alpha-helix mimetic antagonists of the HDM2-p53 protein-protein interaction. Chem Biol Drug Des 2006;67(3):201205. 88. Nandhikonda P, Lynt WZ, McCallum MM, Ara T, Baranowski AM, Yuan NY, Pearson D, Bikle DD, Guy RK, Arnold LA. Discovery of the rst irreversible small molecule inhibitors of the interaction between the vitamin D receptor and coactivators. J Med Chem 2012;55(10):46404651. 89. Willson TM, Kliewer SA. PXR, CAR and drug metabolism. Nat Rev Drug Discov 2002;1(4):259 266. 90. Biswas A, Mani S, Redinbo MR, Krasowski MD, Li H, Ekins S. Elucidating the Jekyll and Hyde nature of PXR: The case for discovering antagonists or allosteric antagonists. Pharm Res 2009;26(8):18071815. 91. Xue Y, Chao E, Zuercher WJ, Willson TM, Collins JL, Redinbo MR. Crystal structure of the PXR-T1317 complex provides a scaffold to examine the potential for receptor antagonism. Bioorg Med Chem 2007;15(5):21562166. 92. Takeshita A, Taguchi M, Koibuchi N, Ozawa Y. Putative role of the orphan nuclear receptor SXR (steroid and xenobiotic receptor) in the mechanism of CYP3A4 inhibition by xenobiotics. J Biol Chem 2002;277(36):3245332458. 93. Huang H, Wang H, Sinz M, Zoeckler M, Staudinger J, Redinbo MR, Teotico DG, Locker J, Kalpana GV, Mani S. Inhibition of drug metabolism by blocking the activation of nuclear receptors by ketoconazole. Oncogene 2007;26(2):258268. 94. Wang H, Huang H, Li H, Teotico DG, Sinz M, Baker SD, Staudinger J, Kalpana G, Redinbo MR, Mani S. Activated pregnenolone X-receptor is a target for ketoconazole and its analogs. Clin Cancer Res 2007;13(8):24882495. 95. Reinhardt TA, Horst RL. Ketoconazole inhibits self-induced metabolism of 1,25-dihydroxyvitamin D3 and amplies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 1989;272(2):459465. 96. Loose DS, Kan PB, Hirst MA, Marcus RA, Feldman D. Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 1983;71(5):14951499. 97. Loose DS, Stover EP, Feldman D. Ketoconazole binds to glucocorticoid receptors and exhibits glucocorticoid antagonist activity in cultured cells. J Clin Invest 1983;72(1):404408. 98. Duret C, Daujat-Chavanieu M, Pascussi JM, Pichard-Garcia L, Balaguer P, Fabre JM, Vilarem MJ, Maurel P, Gerbal-Chaloin S. Ketoconazole and miconazole are antagonists of the human glucocorticoid receptor: Consequences on the expression and function of the constitutive androstane receptor and the pregnane X receptor. Mol Pharmacol 2006;70(1):329339. 99. Eil C. Ketoconazole binds to the human androgen receptor. Horm Metab Res 1992;24(8):367 370. 100. Jones JO, An WF, Diamond MI. AR inhibitors identied by high-throughput microscopy detection of conformational change and subcellular localization. ACS Chem Biol 2009;4(3):199208. 101. Wang H, Li H, Moore LB, Johnson MD, Maglich JM, Goodwin B, Ittoop OR, Wisely B, Creech K, Parks DJ, Collins JL, Willson TM, Kalpana GV, Venkatesh M, Xie W, Cho SY, Roboz J, Redinbo M, Moore JT, Mani S. The phytoestrogen coumestrol is a naturally occurring antagonist of the human pregnane X receptor. Mol Endocrinol 2008;22(4):838857. 102. Ekins S, Chang C, Mani S, Krasowski MD, Reschly EJ, Iyer M, Kholodovych V, Ai N, Welsh WJ, Sinz M, Swaan PW, Patel R, Bachmann K. Human pregnane X receptor antagonists and agonists dene molecular requirements for different binding sites. Mol Pharmacol 2007;72(3):592603. 103. Ekins S, Kholodovych V, Ai N, Sinz M, Gal J, Gera L, Welsh WJ, Bachmann K, Mani S. Computational discovery of novel low micromolar human pregnane X receptor antagonists. Mol Pharmacol 2008;74(3):662672. Medicinal Research Reviews DOI 10.1002/med
r 35
104. Rodriguez AL, Tamrazi A, Collins ML, Katzenellenbogen JA. Design, synthesis, and in vitro biological evaluation of small molecule inhibitors of estrogen receptor alpha coactivator binding. J Med Chem 2004;47(3):600611. 105. Parent AA, Gunther JR, Katzenellenbogen JA. Blocking estrogen signaling after the hormone: Pyrimidine-core inhibitors of estrogen receptor-coactivator binding. J Med Chem 2008;51(20):6512 6530. 106. Gunther JR, Moore TW, Collins ML, Katzenellenbogen JA. Amphipathic benzenes are designed inhibitors of the estrogen receptor alpha/steroid receptor coactivator interaction. ACS Chem Biol 2008;3(5):282286. 107. Shao D, Berrodin TJ, Manas E, Hauze D, Powers R, Bapat A, Gonder D, Winneker RC, Frail DE. Identication of novel estrogen receptor alpha antagonists. J Steroid Biochem Mol Biol 2004;88(45):351360. 108. LaFrate AL, Gunther JR, Carlson KE, Katzenellenbogen JA. Synthesis and biological evaluation of guanylhydrazone coactivator binding inhibitors for the estrogen receptor. Bioorg Med Chem 2008;16(23):1007510084. 109. Zhou HB, Collins ML, Gunther JR, Comninos JS, Katzenellenbogen JA. Bicyclo[2.2.2]octanes: Close structural mimics of the nuclear receptor-binding motif of steroid receptor coactivators. Bioorg Med Chem Lett 2007;17(15):41184122. 110. Becerril J, Hamilton AD. Helix mimetics as inhibitors of the interaction of the estrogen receptor with coactivator peptides. Angew Chem Int Ed Engl 2007;46(24):44714473. 111. Fletcher S, Hamilton AD. Targeting protein-protein interactions by rational design: Mimicry of protein surfaces. J R Soc Interface 2006;3(7):215233. 112. Williams AB, Weiser PT, Hanson RN, Gunther JR, Katzenellenbogen JA. Synthesis of biphenyl proteomimetics as estrogen receptor-alpha coactivator binding inhibitors. Org Lett 2009;11(23):5370 5373. 113. Sun A, Moore TW, Gunther JR, Kim MS, Rhoden E, Du Y, Fu H, Snyder JP, Katzenellenbogen JA. Discovering small-molecule estrogen receptor alpha/coactivator binding inhibitors: Highthroughput screening, ligand development, and models for enhanced potency. ChemMedChem 2011;6(4):654666. 114. Wang Y, Chirgadze NY, Briggs SL, Khan S, Jensen EV, Burris TP. A second binding site for hydroxytamoxifen within the coactivator-binding groove of estrogen receptor beta. Proc Natl Acad Sci USA 2006;103(26):99089911. 115. Kojetin DJ, Burris TP, Jensen EV, Khan SA. Implications of the binding of tamoxifen to the coactivator recognition site of the estrogen receptor. Endocr Relat Cancer 2008;15(4):851 870. 116. Mizwicki MT, Keidel D, Bula CM, Bishop JE, Zanello LP, Wurtz JM, Moras D, Norman AW. Identication of an alternative ligand-binding pocket in the nuclear vitamin D receptor and its functional importance in 1alpha,25(OH)2-vitamin D3 signaling. Proc Natl Acad Sci USA 2004;101(35):12876 12881. 117. Gunther JR, Parent AA, Katzenellenbogen JA. Alternative inhibition of androgen receptor signaling: Peptidomimetic pyrimidines as direct androgen receptor/coactivator disruptors. ACS Chem Biol 2009;4(6):435440. 118. Estebanez-Perpina E, Arnold LA, Nguyen P, Rodrigues ED, Mar E, Bateman R, Pallai P, Shokat KM, Baxter JD, Guy RK, Webb P, Fletterick RJ. A surface on the androgen receptor that allosterically regulates coactivator binding. Proc Natl Acad Sci USA 2007;104(41):1607416079. 119. Buzon V, Carbo LR, Estruch SB, Fletterick RJ, Estebanez-Perpina E. A conserved surface on the ligand binding domain of nuclear receptors for allosteric control. Mol Cell Endocrinol 2011;348(2):394402. 120. Lack NA, Axerio-Cilies P, Tavassoli P, Han FQ, Chan KH, Feau C, LeBlanc E, Guns ET, Guy RK, Rennie PS, Cherkasov A. Targeting the binding function 3 (BF3) site of the human androgen receptor through virtual screening. J Med Chem 2011;54(24):85638573. Medicinal Research Reviews DOI 10.1002/med
36
r CABONI AND LLOYD
121. De Leon JT, Iwai A, Feau C, Garcia Y, Balsiger HA, Storer CL, Suro RM, Garza KM, Lee S, Kim YS, Chen Y, Ning YM, Riggs DL, Fletterick RJ, Guy RK, Trepel JB, Neckers LM, Cox MB. Targeting the regulation of androgen receptor signaling by the heat shock protein 90 cochaperone FKBP52 in prostate cancer cells. Proc Natl Acad Sci USA 2011;108(29):11878 11883. 122. Grosdidier S, Carbo LR, Buzon V, Brooke G, Nguyen P, Baxter JD, Bevan C, Webb P, EstebanezPerpina E, Fernandez-Recio J. Allosteric conversation in the androgen receptor ligand-binding domain surfaces. Mol Endocrinol 2012;26(7):10781090. 123. Caboni L, Kinsella GK, Blanco F, Fayne D, Jagoe WN, Carr M, Williams DC, Meegan MJ, Lloyd DG. True antiandrogens-selective non-ligand-binding pocket disruptors of androgen receptor-coactivator interactions: Novel tools for prostate cancer. J Med Chem 2012;55(4):1635 1644. 124. Axerio-Cilies P, Lack NA, Nayana MR, Chan KH, Yeung A, Leblanc E, Guns ES, Rennie PS, Cherkasov A. Inhibitors of androgen receptor activation function-2 (AF2) site identied through virtual screening. J Med Chem 2011;54(18):61976205. 125. Jones JO, Bolton EC, Huang Y, Feau C, Guy RK, Yamamoto KR, Hann B, Diamond MI. Non-competitive androgen receptor inhibition in vitro and in vivo. Proc Natl Acad Sci USA 2009;106(17):72337238. 126. Jones JO, Diamond MI. A cellular conformation-based screen for androgen receptor inhibitors. ACS Chem Biol 2008;3(7):412418. 127. Dodou K, Anderson RJ, Small DA, Groundwater PW. Investigations on gossypol: Past and present developments. Expert Opin Investig Drugs 2005;14(11):14191434. 128. Van Poznak C, Seidman AD, Reidenberg MM, Moasser MM, Sklarin N, Van Zee K, Borgen P, Gollub M, Bacotti D, Yao TJ, Bloch R, Ligueros M, Sonenberg M, Norton L, Hudis C. Oral gossypol in the treatment of patients with refractory metastatic breast cancer: A phase I/II clinical trial. Breast Cancer Res Treat 2001;66(3):239248. 129. Hiipakka RA, Zhang HZ, Dai W, Dai Q, Liao S. Structure-activity relationships for inhibition of human 5alpha-reductases by polyphenols. Biochem Pharmacol 2002;63(6):11651176. 130. Song D, Lorenzo B, Reidenberg MM. Inhibition of 11 beta-hydroxysteroid dehydrogenase by gossypol and bioavonoids. J Lab Clin Med 1992;120(5):792797. 131. Kitada S, Leone M, Sareth S, Zhai D, Reed JC, Pellecchia M. Discovery, characterization, and structure-activity relationships studies of proapoptotic polyphenols targeting B-cell lymphocyte/leukemia-2 proteins. J Med Chem 2003;46(20):42594264. 132. York B, OMalley BW. Steroid receptor coactivator (SRC) family: Masters of systems biology. J Biol Chem 2010;285(50):3874338750. 133. Walsh CA, Qin L, Tien JC, Young LS, Xu J. The function of steroid receptor coactivator-1 in normal tissues and cancer. Int J Biol Sci 2012;8(4):470485. 134. Gearhart MD, Dickinson L, Ehley J, Melander C, Dervan PB, Wright PE, Gottesfeld JM. Inhibition of DNA binding by human estrogen-related receptor 2 and estrogen receptor alpha with minor groove binding polyamides. Biochemistry 2005;44(11):41964203. 135. Chenoweth DM, Dervan PB. Structural basis for cyclic Py-Im polyamide allosteric inhibition of nuclear receptor binding. J Am Chem Soc 2010;132(41):1452114529. 136. Nickols NG, Dervan PB. Suppression of androgen receptor-mediated gene expression by a sequencespecic DNA-binding polyamide. Proc Natl Acad Sci USA 2007;104(25):1041810423. 137. Chenoweth DM, Harki DA, Phillips JW, Dose C, Dervan PB. Cyclic pyrrole-imidazole polyamides targeted to the androgen response element. J Am Chem Soc 2009;131(20):71827188. 138. Punchihewa C, De Alba A, Sidell N, Yang D. XR5944: A potent inhibitor of estrogen receptors. Mol Cancer Ther 2007;6(1):213219. 139. Whittal RM, Benz CC, Scott G, Semyonov J, Burlingame AL, Baldwin MA. Preferential oxidation of zinc nger 2 in estrogen receptor DNA-binding domain prevents dimerization and, hence, DNA binding. Biochemistry 2000;39(29):84068417. Medicinal Research Reviews DOI 10.1002/med
r 37
140. Schwabe JW, Chapman L, Finch JT, Rhodes D. The crystal structure of the estrogen receptor DNAbinding domain bound to DNA: How receptors discriminate between their response elements. Cell 1993;75(3):567578. 141. Maynard AT, Covell DG. Reactivity of zinc nger cores: Analysis of protein packing and electrostatic screening. J Am Chem Soc 2001;123(6):10471058. 142. Wang LH, Yang XY, Zhang X, Mihalic K, Fan YX, Xiao W, Howard OM, Appella E, Maynard AT, Farrar WL. Suppression of breast cancer by chemical modulation of vulnerable zinc ngers in estrogen receptor. Nat Med 2004;10(1):4047. 143. Wang LH, Yang XY, Zhang X, An P, Kim HJ, Huang J, Clarke R, Osborne CK, Inman JK, Appella E, Farrar WL. Disruption of estrogen receptor DNA-binding domain and related intramolecular communication restores tamoxifen sensitivity in resistant breast cancer. Cancer Cell 2006;10(6):487 499. 144. Mao C, Patterson NM, Cherian MT, Aninye IO, Zhang C, Montoya JB, Cheng J, Putt KS, Hergenrother PJ, Wilson EM, Nardulli AM, Nordeen SK, Shapiro DJ. A new small molecule inhibitor of estrogen receptor alpha binding to estrogen response elements blocks estrogen-dependent growth of cancer cells. J Biol Chem 2008;283(19):1281912830. 145. Kretzer NM, Cherian MT, Mao C, Aninye IO, Reynolds PD, Schiff R, Hergenrother PJ, Nordeen SK, Wilson EM, Shapiro DJ. A noncompetitive small molecule inhibitor of estrogen-regulated gene expression and breast cancer cell growth that enhances proteasome-dependent degradation of estrogen receptor {alpha}. J Biol Chem 2010;285(53):4186341873. 146. Balasubramanyam K, Swaminathan V, Ranganathan A, Kundu TK. Small molecule modulators of histone acetyltransferase p300. J Biol Chem 2003;278(21):1913419140. 147. Schultz DJ, Wickramasinghe NS, Ivanova MM, Isaacs SM, Dougherty SM, Imbert-Fernandez Y, Cunningham AR, Chen C, Klinge CM. Anacardic acid inhibits estrogen receptor alpha-DNA binding and reduces target gene transcription and breast cancer cell proliferation. Mol Cancer Ther 2010;9(3):594605. 148. Hemshekhar M, Sebastin Santhosh M, Kemparaju K, Girish KS. Emerging roles of anacardic acid and its derivatives: A pharmacological overview. Basic Clin Pharmacol Toxicol 2011. 149. Andersen RJ, Mawji NR, Wang J, Wang G, Haile S, Myung JK, Watt K, Tam T, Yang YC, Banuelos CA, Williams DE, McEwan IJ, Wang Y, Sadar MD. Regression of castrate-recurrent prostate cancer by a small-molecule inhibitor of the amino-terminus domain of the androgen receptor. Cancer Cell 2010;17(6):535546. 150. Sadar MD. Androgen-independent induction of prostate-specic antigen gene expression via crosstalk between the androgen receptor and protein kinase A signal transduction pathways. J Biol Chem 1999;274(12):77777783. 151. Ueda T, Bruchovsky N, Sadar MD. Activation of the androgen receptor N-terminal domain by interleukin-6 via MAPK and STAT3 signal transduction pathways. J Biol Chem 2002;277(9):7076 7085. 152. Ueda T, Mawji NR, Bruchovsky N, Sadar MD. Ligand-independent activation of the androgen receptor by interleukin-6 and the role of steroid receptor coactivator-1 in prostate cancer cells. J Biol Chem 2002;277(41):3808738094. 153. Quayle SN, Mawji NR, Wang J, Sadar MD. Androgen receptor decoy molecules block the growth of prostate cancer. Proc Natl Acad Sci USA 2007;104(4):13311336. 154. Sadar MD, Williams DE, Mawji NR, Patrick BO, Wikanta T, Chasanah E, Irianto HE, Soest RV, Andersen RJ. Sintokamides A to E, chlorinated peptides from the sponge Dysidea sp. that inhibit transactivation of the N-terminus of the androgen receptor in prostate cancer cells. Org Lett 2008;10(21):49474950. 155. Meimetis LG, Williams DE, Mawji NR, Banuelos CA, Lal AA, Park JJ, Tien AH, Fernandez JG, de Voogd NJ, Sadar MD, Andersen RJ. Niphatenones, glycerol ethers from the sponge Niphates digitalis block androgen receptor transcriptional activity in prostate cancer cells: Structure elucidation, synthesis, and biological activity. J Med Chem 2012;55(1):503514. Medicinal Research Reviews DOI 10.1002/med
38
r CABONI AND LLOYD
156. Dehm SM, Tindall DJ. Alternatively spliced androgen receptor variants. Endocr Relat Cancer 2011;18(5):R183R196. 157. Dehm SM, Tindall DJ. Ligand-independent androgen receptor activity is activation function-2independent and resistant to antiandrogens in androgen refractory prostate cancer cells. J Biol Chem 2006;281(38):2788227893. 158. Ottow E, Weinmann, H. Nuclear receptors as drug targets: A historical perspective of modern drug discovery. In: Ottow E, Weinmann, H, Eds. Nuclear Receptors as Drug Targets, Methods and Principles in Medicinal Chemistry. Wiley-VCH; 2008. p 123. 159. Norris JD, Joseph JD, Sherk AB, Juzumiene D, Turnbull PS, Rafferty SW, Cui H, Anderson E, Fan D, Dye DA, Deng X, Kazmin D, Chang CY, Willson TM, McDonnell DP. Differential presentation of protein interaction surfaces on the androgen receptor denes the pharmacological actions of bound ligands. Chem Biol 2009;16(4):452460. 160. Potashman MH, Duggan ME. Covalent modiers: An orthogonal approach to drug design. J Med Chem 2009;52(5):12311246.
Laura Caboni, is a PhD candidate in the School of Biochemistry and Immunology, based in the Trinity Biomedical Sciences Institute of Trinity College Dublin. She holds an honours degree in Pharmaceutical Chemistry and Technology from the University of Cagliari (Italy). Thereafter she moved to Ireland where she initially worked as a community pharmacist. In 2008 she was appointed to the four-year PhD programme in Molecular Medicine sponsored by the Health Research Board (HRB). David G. Lloyd PhD, is an Associate Professor in the School of Biochemistry & Immunology, based in the Trinity Biomedical Sciences Institute of Trinity College Dublin. He holds both a BSc and PhD from the School of Chemical Sciences in Dublin City University and worked as post-doctoral fellow in the School of Pharmacy, Trinity College Dublin. Thereafter, he took up a post as Senior Scientist in De Novo Pharmaceuticals in Cambridge, UK. He returned to Ireland and Trinity College Dublin in 2004 as the Hitachi Lecturer in Advanced Computing. In 2005 Prof Lloyd was appointed as Trinitys rst Associate Dean of Research, overseeing interactions between Trinity and its industry partners. In 2007 he became the Dean (Vice-President) of Research for TCD with responsibility for the overall research and innovation strategies within the institution. Since September 2011, Prof Lloyd has worked as Trinity College Dublins Bursar & Director of Strategic Innovation. In March 2012, Prof Lloyd was announced as the Inaugural Chairman of the Irish Research Council.

Med 21275

Uploaded by

Copyright:

Available Formats

Med 21275

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Med 21275

Uploaded by

Copyright:

Available Formats

Beyond the Ligand-Binding Pocket: Targeting Alternate Sites in Nuclear Receptors

Laura Caboni and David G. Lloyd

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

Medicinal Research Reviews DOI 10.1002/med

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

Medicinal Research Reviews DOI 10.1002/med

BEYOND THE LIGAND-BINDING POCKET

Medicinal Research Reviews DOI 10.1002/med

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

Medicinal Research Reviews DOI 10.1002/med

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

Medicinal Research Reviews DOI 10.1002/med

BEYOND THE LIGAND-BINDING POCKET

Medicinal Research Reviews DOI 10.1002/med

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

Medicinal Research Reviews DOI 10.1002/med

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET ACKNOWLEDGMENTS

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

62. 63. 64.

Medicinal Research Reviews DOI 10.1002/med

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

BEYOND THE LIGAND-BINDING POCKET

r CABONI AND LLOYD

Medicinal Research Reviews DOI 10.1002/med

You might also like