Viewpoint pubs.acs.org/jmc
Cite This: J. Med. Chem. 2018, 61, 10412−10414
A Drug with Lipophilicity-Dependent Potency Can Be Metabolically Stable: Discovery of a Potent and Selective Retinoic Acid ReceptorRelated Orphan Receptor C2 (RORC2) Inverse Agonist as an Orally Bioavailable Anti-Inflammatory Agent Yibing Wang and Hong Liu*
Downloaded via 95.85.70.57 on December 19, 2018 at 18:31:25 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, China ABSTRACT: IL-17 drives an amplification mechanism in inflammatory diseases such as psoriasis, psoriatic arthritis, and ankylosing spondylitis. The production of IL-17 depends on the activity of RORC2 in immune cells, which suggests that RORC2 inverse agonists are capable of anti-inflammatory therapy by reducing IL-17 levels. However, oral delivery of inverse agonists remains a challenge since the binding pocket of RORC2 prefers to accommodate lipophilic ligands, which tend to have poor metabolic stability. This Viewpoint discusses recent results published in this journal that identified a potent, selective, and orally bioavailable RORC2 inverse agonist as an anti-inflammatory agent, optimized from a high-throughput screening (HTS) hit through a combination of de novo and structure-guided design.
luciferase reporter assay in which the intrinsic transcriptional activity of RORC2 is the functional readout, including the possibility of allosteric binding mechanisms beyond the usual competitive binding mechanism with the endogenous ligand; (2) a further GAL4-65 reporter assay to modify the positive results, excluding the transcriptional inhibition through nonspecificity for RORC2, cytotoxicity, or specific interference with the luminescence readout; (3) a comparable reporter assay of GAL4-RORC2, GAL4-RORA, and GAL4-RORB in murine cells for characterization of isoform selectivity. After the high-throughput screening of about 150 000 compounds and application of the above strategies, compound 3 (Figure 1) was selected as a potent and selective RORC2 inverse agonist that, however, showed poor metabolic stability.
IL-17 participates in the pathogenesis and progression of various inflammatory diseases such as psoriasis, psoriatic arthritis, and ankylosing spondylitis. It mediates the amplification mechanism in inflammation since it triggers the secretion of proinflammatory mediators at the site of inflammation, and these further recruit additional inflammatory cells to the tissue. The production of IL-17 occurs in immune cells, especially in T helper 17 (Th17) cells, in which the nuclear hormone receptor RORC2 is indispensable for the IL-17 expression and differentiation of Th17 cells from naive CD4+ T cells. It is assumed that with the action of an inverse agonist against RORC2, the production of IL-17 will decrease either in the presence of agonists or if the receptor has intrinsic activity. However, the clinical approach to modulate IL-17 levels is still dominated by monoclonal antibody therapies. The discovery of orally bioavailable small molecules remains a challenge mainly because the inverse agonists need sufficiently high lipophilicity to display their activities in the lipophilic pocket of RORC2, but lipophilic ligands usually have poor metabolic stability and are not suitable for oral delivery. In a recent issue of the Journal of Medicinal Chemistry, Schnute et al.1 report their work on the identification of a potent, selective, and orally bioavailable RORC2 inverse agonist optimized from a HTS hit through de novo design and optimization focusing on lipophilicity and structure-guided conformational restriction. Although diverse chemotypes of the small molecular modulators of RORC receptors have been reported,2−5 most of them displayed their potencies by depending on the binding in a hydrophobic pocket in the ligand binding domain and displayed poor to moderate metabolic stability. In this work, Schnute et al.1 initiated their research with high-throughput screening to find a promising hit. First, the researchers established an efficient and comprehensive in vitro screening protocol for the identification of potent and selective hits: (1) a cell-based GAL4-RORC2 © 2018 American Chemical Society
Figure 1. Structures of compounds 3, 8b, and 66.
The optimization of hit compound 3 started with the introduction of a methyl group to the indole nitrogen, which significantly improved the potency and afforded an increase of 1.4 in lipophilicity efficiency (LIPE). Modification of the other two vectors, the benzamide and the piperidine amide, failed to improve potency but provided insight into the binding mode between the inverse agonist (8b) and RORC2 ligand binding domain (LBD) via the cocrystal structure of 8b with the RORC2 LBD. Received: October 4, 2018 Published: December 3, 2018 10412
DOI: 10.1021/acs.jmedchem.8b01545 J. Med. Chem. 2018, 61, 10412−10414
Journal of Medicinal Chemistry
Viewpoint
the N-Me substituent that afforded a hydrophobic interaction with RORC2, increased the lipophilicity efficiency (ΔLIPE = 1.4), and introduced another methyl in the 4-position of the indole ring, minimizing the calculated energetic penalty paid from ground state conformation to the bound state torsion (113° dihedral angle of C4−C5−N−C). This required conformational restriction of the ligand, which was verified by the cocrystal structure of this new compound with RORC2 LBD. Upon subsequent replacement of the indole ring with the pyrrolo[2,3-b]pyridine scaffold, the inhibition of IL-17 production was remarkably strengthened with another improvement in LIPE (ΔLIPE = 2.1) of the compound. Interestingly, the 4-methyl substituted compounds exhibited a pseudoirreversible behavior. Their effective dissociation halflife from the recombinant receptor were much longer (10−20 h) than the half-life of the RORC2 receptor in human Th17 cells (5 h), which provided their potential for a sustained effect dependent on the dissociation rate of these compounds from the binding pocket. Although there was indeed a significant improvement in LIPE and potency, the goal of reducing metabolic clearance had not been accomplished by the above modifications. Further studies revealed that the lipophilicity of the piperidine amide plays a critical role in metabolic stability as well as in the potency. Substituents of larger size and higher lipophilicity were much more favorable for improving potency but less beneficial to a satisfying metabolic stability. As a result, the isopropyl analog held the best balance of potency, metabolic stability, and LIPE. On the basis of the breakthrough on metabolic stability and the analyses of two cocrystal structures, a fine-tuning of the conformational restriction, replacing the 4-methyl with 4trifluoromethyl substituent, afforded a highly potent, metabolically stable and selective lead compound. The metabolically stable trifluoromethyl group not only contributes to the conformational restriction but also attenuates the polarization of the pyrrolopyridine nitrogen so as to be more effectively accommodated in the hydrophobic binding pocket. Indeed, the trifluoromethyl group is more lipophilic than its methyl counterpart and causes an increase of the molecular lipophilicity back to the high levels similar to those observed during early structural modifications. Nonetheless, the metabolic stability was well preserved, which might have resulted from the specific modification avoiding the main metabolic site. The optimized lead 66 displays high selectivity over 48 nuclear receptors and excellent in vitro and in vivo potency. Compound 66 significantly inhibits the IL-17 production in murine Th17 cells with 92% maximum inhibition. The IC50 was 32 nM, supporting the in vivo efficacy evaluation in imiquimod-induced skin inflammation mouse model. As expected, compound 66, orally administered once daily over 5 days, demonstrated a dose-dependent inhibition of ear swelling in the model, and the maximum inhibition (46%) was comparable to that of the anti-IL-17 monoclonal antibody (45%) that was administered by intraperitoneal injection. The level of IL-17A protein in the mouse ear was also significantly reduced by the oral administration of compound 66 at all doses. It is notable that the oral bioavailability of compound 66 in spray-dried dispersion form was up to 83% in rats and 70% in dogs. Overall, this work demonstrates an excellent example of lead discovery that balances lipophilicity-dependent potency and
On the basis of an in-depth analysis of the cocrystal structure, the authors took advantage of their insight into the structural basis for the efficacy induced by their newly designed inverse agonist (8b) to use structure-guided design to achieve optimal lipophilicity efficiency and receptor residence time. As shown in Figure 2a, the RORC2 active conformation requires
Figure 2. RORC2 active conformation and the binding mode with compound 8b shown in the cocrystal structure (6CN5). (a) Key triplet latch for stabilizing the RORC2 active conformation (3KYT). (b) Conformational changes in the piperidine amide moiety binding site. Residues in gray depict cocrystal structure with compound 8b (6CN5), and residues in blue white depict the RORC2 active conformation (3KYT). (c) Binding mode of RORC2 with the benzamide indole moiety of ligand 8b. (d) The N-methyl moiety of indole 8b extends into a small hydrophobic pocket.
the close association of helix-12 (H12) with the surrounding structure to form a cleft for coactivator peptide binding. This conformation is stabilized by the interactions among His479 of H11, Tyr502 of H12, and Phe506 of H12. A key hydrogen bond formed by H479 and Tyr502 bridges the two helices, and the benzyl of Phe506 adopts a favorable orientation to face the aromatic stacking arrangement with both His479 and Tyr502. Therefore, the charged His479 tends to hold H12 in an active conformation via a cation−π interaction. In stark contrast, the original active conformation is disrupted when the endogenous ligand-binding pocket is occupied by inverse agonist 8b. His479, the key residue for stabilizing the active conformation of RORC2 LBD, is now engaged in hydrogen bonding with the piperidine amide carbonyl of compound 8b (Figure 2b), which leads to a highly disordered state of H12, and disappearance of the coactivator binding cleft. As a result of the torsional change in Trp317, the cyclohexylamide substituent (Figure 2b) and the indole ring located in a hydrophobic region between Met365 and Val376 (Figure 2c) can assume hydrophobic interactions. The cyano group on the benzamide moiety forms a bifurcated hydrogen bond with Arg367 and the backbone amide NH of Leu287, and the benzamide NH and the backbone carbonyl of Phe377 form another hydrogen bonding interaction (Figure 2c). The introduction of the N-Me group induces hydrophobic interactions between the methyl group and hydrophobic residues of Ile400, Phe401, and Tyr369 within a small pocket (Figure 2d). After obtaining the cocrystal structure, Schnute et al. changed their strategy from empirical medicinal chemical modification to structure-guided optimization. They retained 10413
DOI: 10.1021/acs.jmedchem.8b01545 J. Med. Chem. 2018, 61, 10412−10414
Journal of Medicinal Chemistry
Viewpoint
quinazolinedione derivatives as ROR gamma t inverse agonist. Bioorg. Med. Chem. 2018, 26, 721−736.
metabolic stability by de novo design and structure-guided optimization focusing on the improvement of lipophilicity efficiency and conformational restriction for optimal ground state energetics and receptor residence time. The optimized lead (66) displayed satisfactory in vitro and in vivo potency after oral administration and has remarkable oral bioavailability in preclinical models, which may support the potential of small molecular modulators of RORC2 as oral therapeutics for the treatment of inflammatory diseases. Nonetheless, the oral bioavailability in mice, the model used for evaluation of the in vivo efficacy, was not as promising as that in rats or dogs. It may still take a long time before obtaining an orally bioavailable small molecule for clinical use.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Hong Liu: 0000-0002-0863-7853 Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Schnute, M. E.; Wennerstal, M.; Alley, J.; Bengtsson, M.; Blinn, J. R.; Bolten, C. W.; Braden, T.; Bonn, T.; Carlsson, B.; Caspers, N.; Chen, M.; Choi, C.; Collis, L. P.; Crouse, K.; Farnegardh, M.; Fennell, K. F.; Fish, S.; Flick, A. C.; Goos-Nilsson, A.; Gullberg, H.; Harris, P. K.; Heasley, S. E.; Hegen, M.; Hromockyj, A. E.; Hu, X.; Husman, B.; Janosik, T.; Jones, P.; Kaila, N.; Kallin, E.; Kauppi, B.; Kiefer, J. R.; Knafels, J.; Koehler, K.; Kruger, L.; Kurumbail, R. G.; Kyne, R. E., Jr.; Li, W.; Lofstedt, J.; Long, S. A.; Menard, C. A.; Mente, S.; Messing, D.; Meyers, M. J.; Napierata, L.; Noteberg, D.; Nuhant, P.; Pelc, M. J.; Prinsen, M. J.; Rhonnstad, P.; Backstrom-Rydin, E.; Sandberg, J.; Sandstrom, M.; Shah, F.; Sjoberg, M.; Sundell, A.; Taylor, A. P.; Thorarensen, A.; Trujillo, J. I.; Trzupek, J. D.; Unwalla, R.; Vajdos, F. F.; Weinberg, R. A.; Wood, D. C.; Xing, L.; Zamaratski, E.; Zapf, C. W.; Zhao, Y.; Wilhelmsson, A.; Berstein, G. Discovery of 3-Cyano-N(3-(1-isobutyrylpiperidin-4-yl)-1-methyl-4-(trifluoromethyl)-1Hpyrrolo[2,3-b]pyridin-5-yl)benzamide: A Potent, Selective, and Orally Bioavailable Retinoic Acid Receptor-Related Orphan Receptor C2 Inverse Agonist. J. Med. Chem. 2018, DOI: 10.1021/acs.jmedchem.8b00392. (2) Fauber, B. P.; Magnuson, S. Modulators of the Nuclear Receptor Retinoic Acid Receptor-Related Orphan Receptor-gamma (ROR gamma or RORc). J. Med. Chem. 2014, 57, 5871−5892. (3) Fauber, B. P.; Rene, O.; Deng, Y. Z.; DeVoss, J.; Eidenschenk, C.; Everett, C.; Ganguli, A.; Gobbi, A.; Hawkins, J.; Johnson, A. R.; La, H.; Lesch, J.; Lockey, P.; Norman, M.; Ouyang, W. J.; Summerhill, S.; Wong, H. Discovery of 1-{4-[3-Fluoro-4-((3S,6R)-3-methyl-1,1dioxo-6-phenyl-[1,2]thiazinan-2-ylmethyl)-phenyl]-piperazin-1-yl}ethanone (GNE-3500): a Potent, Selective, and Orally Bioavailable Retinoic Acid Receptor-Related Orphan Receptor C (RORc or ROR gamma) Inverse Agonist. J. Med. Chem. 2015, 58, 5308−5322. (4) Hirata, K.; Kotoku, M.; Seki, N.; Maeba, T.; Maeda, K.; Hirashima, S.; Sakai, T.; Obika, S.; Hori, A.; Hase, Y.; Yamaguchi, T.; Katsuda, Y.; Hata, T.; Miyagawa, N.; Arita, K.; Nomura, Y.; Asahina, K.; Aratsu, Y.; Kamada, M.; Adachi, T.; Noguchi, M.; Doi, S.; Crowe, P.; Bradley, E.; Steensma, R.; Tao, H. Y.; Fenn, M.; Babine, R.; Li, X. L.; Thacher, S.; Hashimoto, H.; Shiozaki, M. SAR Exploration Guided by LE and Fsp(3): Discovery of a Selective and Orally Efficacious ROR gamma Inhibitor. ACS Med. Chem. Lett. 2016, 7, 23−27. (5) Fukase, Y.; Sato, A.; Tomata, Y.; Ochida, A.; Kono, M.; Yonemori, K.; Koga, K.; Okui, T.; Yamasaki, M.; Fujitani, Y.; Nakagawa, H.; Koyama, R.; Nakayama, M.; Skene, R.; Sang, B. C.; Hoffman, I.; Shirai, J.; Yamamoto, S. Identification of novel 10414
DOI: 10.1021/acs.jmedchem.8b01545 J. Med. Chem. 2018, 61, 10412−10414
