Total synthesis of eryvarin H and its derivatives and their biological activity as ERRγ inverse agonist

Article abstract of DOI:10.1039/c3ob41264d

Total synthesis of eryvarin H and a biological investigation of its analogues as a potential inverse agonist of ERRγ are described here. Among the 13 analogues prepared by the modular synthetic route, eryvarin H and compound 12 showed meaningful ERRγ inverse agonistic activities along with moderate selectivity over ERα and other nuclear receptors in the cell-based reporter gene assay.

Full text of DOI:10.1039/c3ob41264d

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Total synthesis of eryvarin H and its derivatives and
their biological activity as ERRγ inverse agonist†
Cite this: Org. Biomol. Chem., 2013, 11,
5782
Ja Young Koo,a Sangmi Oh,b Seung-Rye Cho,c Minseob Koh,a Won-Keun Oh,d
Received 19th June 2013,
Accepted 12th July 2013
Hueng-Sik Choi*c and Seung Bum Park*a,e
DOI: 10.1039/c3ob41264d
www.rsc.org/obc
Total synthesis of eryvarin H and a biological investigation of its regulation of nuclear receptors, especially orphan nuclear
analogues as a potential inverse agonist of ERRγ are described receptors.
here. Among the 13 analogues prepared by the modular synthetic
We have been interested in estrogen-related receptors
route, eryvarin H and compound 12 showed meaningful ERRγ (ERRs) that are orphan NRs and closely related to estrogen
inverse agonistic activities along with moderate selectivity over receptors (ERs) in their sequence homology. But their differ-
ERα and other nuclear receptors in the cell-based reporter gene ence in the ligand binding domain (LBD) differentiates their
assay.
characteristics from those of ERs.5 ERRs have various func-
tional roles associated with diseases including cancer and
metabolic diseases as ER does.6 Among the members of the
ERR family, we paid particular attention to ERRγ due to its
crucial role in various biological events, such as hepatic
insulin signaling, gluconeogenesis, regulating oxidative metab-
olism, suppressing tumor growth of prostate cancer cells, and
modulating cell proliferation and estrogen signaling in breast
cancer.6–8 ERRγ is a third subtype receptor and is highly
expressed in various human adult tissues including brain,
kidney, skeletal muscle, heart, and placenta.9 Based on these
observations, ERRγ became an emerging target for cancers
and metabolic diseases, even though there are a limited
number of ERRγ agonists and inverse agonists that control its
downstream activity.10–12 In addition, most of the reported
ligands suffered non-specific interactions with various gene
products and caused functional cross-talks with other nuclear
receptors, especially with ERα, in the estrogenic signaling
pathway.10,11 Therefore, it is essential to discover novel bio-
active small molecules that selectively regulate ERRγ to eluci-
date its functional roles in our body. As a continuation of our
efforts7,10 in this field, we pursued the identification of novel
ERRγ inverse agonists to deliver new potential drug-like mole-
cules to the biomedical community.
Nuclear receptors (NRs) have been recognized as one of the
most important targets for the development of novel thera-
peutic agents via regulating various disease-relevant cellular
functions at the transcription level.1 NRs are described as tran-
scriptional factors mainly regulated by endogenous ligands in
biological systems. Among them, a series of NRs are categor-
ized as orphan nuclear receptors when they do not have any
identified endogenous ligands, and some of the orphan
nuclear receptors still possess constitutive transcriptional
activities without their endogenous ligands.2 Therefore, the
identification of bioactive small molecules that can specifically
control their activity can provide an important clue to the
development of novel therapeutic agents as well as a research
tool for deciphering complex events in biological systems.3
Unlike the genetic approach, bioactive small molecules can
perturb particular functions of wild-type gene products in
a temporal and reversible manner.4 Therefore, we aimed
to identify new synthetic ligands for the transcriptional
aDepartment of Chemistry, Seoul National University, Seoul, Korea.
E-mail: sbpark@snu.ac.kr
bMedicinal Chemistry Team, Institut Pasteur Korea, Seongnam-si,
Gyeonggi-do 463-400, Korea
cNational Creative Research Initiatives Center for Nuclear Receptor Signals, Hormone
Research Center, School of Biological Sciences and Technology, Chonnam National
University, Gwangju 500-757, Korea. E-mail: hsc@chonnam.ac.kr
dCollege of Pharmacy, Seoul National University, Seoul 151-742, Korea
eDepartment of Biophysics and Chemical Biology/Bio-MAX Institute, Seoul National
Under this mission statement, we searched for the new
chemical entity from the collection of natural products using
structure-based in silico analysis. With the X-ray co-crystal struc-
ture of ERRγ with GSK5182, a known ERRγ inverse agonist, we
performed a docking simulation study of the natural product
collection at the LBD of ERRγ, which allows the identification of
potential ligands for ERRγ. After computer-based screening of
4000 structurally diverse natural products from the Korea Bio-
active Natural Material Bank, we selected the natural product
University, Seoul 151-747, Korea. E-mail: sbpark@snu.ac.kr
†Electronic supplementary information (ESI) available: Detailed procedures for
synthetic and biological experiments, spectroscopic data of all new compounds.
See DOI: 10.1039/c3ob41264d
5782 | Org. Biomol. Chem., 2013, 11, 5782–5786
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Communication
eryvarin H (1) as a potential inverse agonist of ERRγ. Eryvarin linkage on eryvarin H (1) (Fig. 1c). However, even though there
H was first isolated from the roots of Erythrina species, is a series of potential specific interactions of eryvarin H with
Erythrina variegata13 or Erythrina abyssinica in 2003.14 In the ERRγ, it is essential to have a structural modularity to ensure
past, eryvarin H has been known to be an anti-microbial the specificity due to the high degree of structural similarity of
agent14 or a cellular radical scavenger.15 However, eryvarin ERRγ with ERα. In addition, the extraction from natural
H (1) used in previous reports was extracted from natural resources only provides the limited quantity of eryvarin H
resources with a limited quantity. Therefore, herein, we accom- without an access to its analogues. Therefore, we pursued the
plished the first total synthesis of eryvarin H (1) and its deriva- total synthesis of eryvarin H (1) and the subsequent structural
tives using a modular synthetic route. The resulting eryvarin H modification for structure–activity relationship study.
and its derivatives were evaluated for their function as selective
Eryvarin
H (1) is composed of two aromatic rings,
ligands for ERRγ.
2H-chromen-7-ol and electron-rich aryl ring, that can be con-
For the discovery of a novel ERRγ inverse agonist, we first nected via C–C bond formation using Pd-mediated cross-coup-
explored the in silico docking study of structurally diverse ling (Fig. 1e). Based on this retrosynthetic analysis of eryvarin
natural products with ERRγ co-crystallized with GSK5182 (PDB H (1), we prepared the two cross-coupling partners, vinyl
ID: 2GPU).12 GSK5182 is a confirmed ERRγ inverse agonist and bromide 2a and arylboronic ester 3. For the preparation of the
binds at its ligand binding site. From the collection of natural vinyl bromide (2a) part, resorcinol was first transformed to
products, we searched for potential ligands that can bind at 7-hydroxychroman-4-one (4) by the treatment of 3-chloro-
the LBD of ERRγ in a similar way to GSK5182. Based on the propionic acid via acid-catalyzed electrophilic aromatic acyla-
binding scores in computational docking analysis, we selected tion, followed by O-alkylation.16 The mono-bromination at the
a list of natural products. Among them, we selected eryvarin α position to carbonyl in compound 4 led to the formation of
H (1) as a potential inverse agonist of ERRγ. As shown in compound 5. The tert-butyldimethylsilyl (TBS) protection of
Fig. 1a and b, eryvarin H (1) nicely occupies the empty space at phenol in compound 5, carbonyl reduction using NaBH4,
the active site of ERRγ with dipole–dipole interaction as well and the subsequent acid-catalyzed dehydration allowed the
as hydrophobic interactions, which was clearly visualized in formation of desired vinyl bromide (2a) in good yields
the electrostatic potential clouds of eryvarin H at the LBD of (Scheme 1a). The synthesis of its cross-coupling counterpart,
ERRγ. The whole amino acid map within the range of 2 Å also arylboronic ester (3), was started by the treatment of 2,5-
reveals the specific interactions, such as hydrogen bonding of dimethoxybenzaldehyde with Br2 under mild acidic conditions
Asp273, Tyr326 and Asn346 with hydroxyl groups on eryvarin to provide its brominated compound 5.17 This aryl bromina-
H (1) and dipole–dipole interaction of Cys269 with ether tion was quite regioselective at the C-4 position due to the
presence of aldehyde moiety as a biasing element, which was
interconverted to phenol moiety through Baeyer–Villiger
Scheme 1 Synthetic route for total synthesis of eryvarin H via C–C bond for-
mation between two coupling partners (2a and 3). Reagents and conditions:
(A) Preparation of eryvarin H (1) through 2a. (a) 3-Chloropropionic acid, TfOH,
Fig. 1 Docking simulation of eryvarin H at the ligand binding pocket of ERRγ.
(a) Potential binding mode of eryvarin H; (b) possible hydrophobic and dipole–
dipole interactions of eryvarin H with residues at the ligand binding site of ERRγ
shown in a space filling model. The red color represents partial positive charge,
and the blue color represents partial negative charge. (c) Potential hydrogen
bonding interactions of eryvarin H with polar amino acids (Asp273, Tyr326, and
Asn346); (d) schematic diagram of amino acids within the 2 Å range from ery-
varin H; (e) chemical structure of eryvarin H and its retrosynthetic analysis.
80 °C, 2 h; (b) 2 N NaOH, 0 °C to r.t., 3 h, 62% (2-step yield); (c) CuBr2, EtOAc–
CHCl3–MeOH, 70 °C, 4 h, 67%; (d) TBSCl, imidazole, DCM, r.t., 1 h; (e) NaBH4,
EtOH, r.t., 1 h; (f ) TsOH–H2O, toluene, 80 °C, 120 W μW, 20 min, 61% (3-step
yield); (g) 3, Pd(PPh3)4, Na2CO3, toluene–EtOH–H2O, 80 °C, 2 h; (h) HF/pyridine,
THF, r.t., 3 h, 73% (2-step yield). (B) Preparation of 3. (i) Br2, acetic acid, r.t., 12 h,
62%; ( j) m-CPBA, DCM, r.t., 3 h; (k) NaOH, methanol, r.t., 3 h; (l) TBSCl, imid-
azole, DCM, r.t., 3 h, 95% (3-step yield); (m) pinacolborane, Pd(OAc)2, DPEphos,
TEA, 1,4-dioxane, 100 °C, 12 h, 63%.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 5782–5786 | 5783

View Article Online
Communication
Organic & Biomolecular Chemistry
oxidation of 6 and the direct hydrolysis of the resulting ester
After the preparation of eryvarin H (1) and its analogues
in a basic methanol solution. The resulting phenol was then (8–19), their biological activities on the transcriptional regu-
protected by TBSCl to prepare 7 in excellent 3-step yields. lation toward ERRγ were evaluated using the cell-based repor-
Finally, various conditions were tested for the introduction of ter gene assay to confirm their roles as inverse agonists. In
C–C cross-coupling moiety, but Pd-catalyzed boronic ester for- this assay, we measure the chemoluminescence induced by
mation with pinacolborane, DPEphos, triethylamine (TEA) in the expression changes of luciferase upon treatment of small
1,4-dioxane only yielded the desired product 3 in a reasonable molecules after the transient transfection of DNA plasmid
yield (Scheme 1b). After the preparation of both key partners, (Gal4-fused ERRγ-LBD construct) to HEK-293 T human kidney
we successfully finished up the first total synthesis of eryvarin cell line by calcium phosphate transfection protocol.19 The
H (1) via the formation of C–C covalent bonds between vinyl resulting luminescence signal was normalized with β-gal
bromide 2a and arylboronic ester 3 using Suzuki–Miyaura expression to minimize the false positives caused by cellular
coupling reaction, followed by TBS deprotection with HF/pyri- cytotoxicity.20 As shown in Fig. 2, we confirmed that eryvarin
dine with 8 linear steps in 18% overall yield (see ESI†).
H (1) is an ERRγ inverse agonist and showed a drastic
After the completion of total synthesis of eryvarin H, we reduction in the chemoluminescence signal of downstream
envisioned the preparation of its systematic analogues for the luciferase. Interestingly, all 13 analogues including eryvarin
molecular level understanding of its binding event with ERRγ. H showed some levels of inverse agonistic activities toward
On the basis of our in silico docking simulation of eryvarin H ERRγ, though their relative luminescence signals were varied
at the LBD of ERRγ, we hypothesized that the specific inter- on the basis of chemical structure (see ESI†). Among them,
action of 2′,5′-dimethoxy-4′-hydroxyphenyl on eryvarin H at the compounds 1, 11, and 12 showed more potent inverse
deep binding pocket of ERRγ is crucial for its functional agonism. Speaking of their structure–activity relationship,
modulation. We also revealed some flexible and empty regions either a hydroxy or methoxy group at the C-4′ position is essen-
at the ligand binding site in the docking structure of ERRγ tial for their activity and the steric or hydrophobic elements
with eryvarin H, which might provide an opportunity to are not necessary at the C-2 position in the case of analogues
induce some additional binding interactions for enhanced (16–19) containing 2H-chromenol with dimethyl substituents
potency and selectivity toward ERRγ over ERα. Therefore, we instead of two hydrogens. The additional bulky dimethyl sub-
prepared 12 derivatives (8–19) of eryvarin H using the modular stituents at the C-2 position might block the specific inter-
synthetic procedure (see ESI†). As shown in Fig. 2, we focused actions with the ligand binding pocket of ERRγ, which
on the diversification of aryl substituents to decipher the significantly reduced the inverse agonistic activity.
origin of specific hydrogen bondings at the ligand binding
Furthermore, transcriptional activities of these 13 ana-
pocket. Dimethyl-substituted vinyl bromide 2b, TIPS-protected logues towards other kinds of NRs (ERα, mCAR, HNF4, SF-1)21
3-bromo-2,2-dimethyl-2H-chromen-7-ol, was synthesized using were measured by reporter gene assay systems to investigate
a reported procedure.18
the selective ERRγ inverse agonistic activity over other NRs
(Fig. 4). Owing to the high degree of structural similarity and
DNA sequence homology of ERRγ with ERα, the ligand-
induced transcriptional activity of ERRγ is compared with that
of ERα. As shown in Fig. 3 and 4, eryvarin H and compound 12
showed the selective inverse agonistic activity toward ERRγ
Fig. 3 Biological evaluation of eryvarin H (1) and its derivatives (8–19) as
inverse agonists using the cell-based reporter gene assay in HEK-293 T cells after
transfection of Gal4-ERRγ, pFR-Luc, and β-gal plasmids. Compounds 1, 11 and
12 showed good inverse agonistic activities among all tested compounds. All
compounds including GSK5182 were treated at the final concentration of
10 μM. R.L.U. is relative luminescent unit.
Fig. 2 Derivatives of eryvarin
H containing isoflav-3-ene structural motif
(8–19). These derivatives were systematically prepared by Pd-mediated Suzuki
cross-coupling between aryl bromide structures (2a or 2b) and various boronic
esters/boronic acids.
5784 | Org. Biomol. Chem., 2013, 11, 5782–5786
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Communication
the follow-up in vivo biological study of eryvarin H and its ana-
logues due to their limited potential as a selective ERRγ
inverse agonist. Even though this study failed to produce a
selective and potent inverse agonist of ERRγ, this study clearly
exemplified the rational drug discovery procedure using in
silico docking simulation with a collection of natural products,
followed by total synthesis of a natural product and its ana-
logues as a potential perturbagen of proteins-of-interest. We
envision that this rational approach can be adapted to identify
a new small molecule that modulates the function of new
therapeutic targets and key signaling pathways.
This work was supported by the Bio & Medical Technology
Development Program (2012M3A9C4048780), a Global Frontier
Project Grant (2011-0032150), the WCU Program (R31-10032),
the Basic Research Laboratory (2010-0019766), and a National
Creative Research Initiatives Grant (20110018305 to H.-S.C.)
funded by the National Research Foundation of Korea (NRF).
J.Y.K. and M.K. are grateful for a BK21 Scholarship and a Seoul
Science Fellowship.
Fig. 4 Transcriptional activities of eryvarin H, 11, and 12 towards other NRs: (a)
ERα, (b) mCAR, (c) HNF4 and (d) SF-1. Normalized luciferase activities of each NR
were tested in HEK-293. T cells upon treatment of three compounds to test their
selective inverse agonism over that in ERRγ. E2 refers to estradiol.
over ERα. However, the efficacy and specificity of eryvarin H we
observed in this study are not comparable to those of GSK5182
that we have used for the biochemical studies of ERRγ.
References
In conclusion, natural products have been serving as essen-
tial perturbagens in various biological processes due to their
evolutional heritage and structural diversity. As a continuation
of our effort at the functional modification of ERRγ, we aimed
to develop a novel ERRγ inverse agonist via in silico docking
analysis with a collection of natural products. The computer-
aided ligand discovery allowed the identification of eryvarin H
(1) as a potential inverse agonist of ERRγ. Along with a limited
supply of eryvarin H via extraction from natural resources, the
inability for a structure–activity relationship of this natural
product against ERRγ led us to pursue the first total synthesis
of eryvarin H and its structural analogues. In addition, we
wanted to decipher the molecular basis of its interaction with
residues at the ligand binding pocket of ERRγ, which has high
structural similarity and sequence homology with ERα. There-
fore, we prepared two key intermediates, vinyl bromides (2a)
and arylboronic ester (3), and successfully completed the total
synthesis of eryvarin H through Pd-mediated Suzuki–Miyaura
cross-coupling of two intermediates in reasonable overall
yields. With this modular synthetic route, we also prepared 12
derivatives (8–19) of eryvarin H simply by changing the substi-
tuents on arylboronic esters or introducing a dimethyl substi-
tuent at the C-2 position of TBS-protected 3-bromo-2H-
chromen-7-ol. The resulting 13 analogues including eryvarin H
were subjected to the cell-based reporter gene assay using DNA
plasmid containing Gal4-fused ERRγ LBD to measure their
biological activity as ERRγ inverse agonists. Among these
derivatives, eryvarin H and compound 12 showed meaningful
ERRγ inverse agonistic activities along with moderate selec-
tivity over ERα and other NRs. However, the level of efficacy
and specificity of eryvarin H toward ERRγ over ERα is not com-
parable to that of GSK5182 that we have used for the biochemi-
cal studies of ERRγ. Therefore, we unfortunately discontinued
1 B. Horard and J.-M. Vanacker, J. Mol. Endocrinol., 2003, 31,
349–357.
2 Y. Shi, Drug Discovery Today, 2007, 12, 440–445.
3 J. T. Moore, J. L. Collins and K. H. Pearce, ChemMedChem,
2006, 1, 504–523.
4 S. L. Schreiber, Bioorg. Med. Chem., 1998, 6, 1127–1152.
5 V. Giguère, Trends Endocrinol. Metab., 2002, 13, 220–225.
6 E. A. Ariazi and V. C. Jordan, Curr. Top. Med. Chem., 2006,
6, 203–215.
7 (a) Y.-B. Xie, J.-H. Park, D.-K. Kim, J. H. Hwang, S. Oh,
S. B. Park, M. Shong, I.-K. Lee and H.-S. Choi, J. Biol.
Chem., 2009, 284, 28762–28774; (b) D.-K. Kim, J. R. Kim,
M. Koh, Y. D. Kim, J.-M. Lee, D. Chanda, S. B. Park,
J.-J. Min, C.-H. Lee, T.-S. Park and H.-S. Choi, J. Biol. Chem.,
2011, 286, 38035–38042; (c) D.-K. Kim, D. Ryu, M. Koh,
M.-W. Lee, D. Lim, M.-J. Kim, Y.-H. Kim, W.-J. Cho,
C.-H. Lee, S. B. Park, S.-H. Koo and H.-S. Choi, J. Biol.
Chem., 2012, 287, 21628–21639; (d) J. H. Lee, E.-J. Kim,
D.-K. Kim, J.-M. Lee, S. B. Park, I.-K. Lee, R. A. Harris,
M.-O. Lee and H.-S. Choi, PLoS One, 2012, 7, e46324;
(e) D.-K. Kim, G.-T. Gang, D. Ryu, M. Koh, Y.-N. Kim,
S. S. Kim, J. Park, Y.-H. Kim, T. Sim, I.-K. Lee, C. S. Choi,
S. B. Park, C.-H. Lee, S.-H. Koo and H.-S. Choi, Diabetes,
2013, DOI: 10.2337/db12-0946.
8 (a) V. Giguère, Endocr. Rev., 2008, 29, 677–696; (b) C. Tiraby,
B. C. Hazen, M. L. Gantner and A. Kralli, Cancer Res., 2011,
71, 2518–2528; (c) E. A. Ariazi, G. M. Clark and J. E. Mertz,
Cancer Res., 2002, 62, 6510–6518; (d) S. Yu, X. Wang,
C.-F. Ng, S. Chen and F. L. Chan, Cancer Res., 2007, 67,
4904–4914.
9 D. J. Heard, P. L. Norby, J. Holloway and H. Vissing, Mol.
Endocrinol., 2000, 14, 382–392.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 5782–5786 | 5785

View Article Online
Communication
Organic & Biomolecular Chemistry
10 (a) Y. Kim, M. Koh, D.-K. Kim, H.-S. Choi and S. B. Park, 13 H. Tanaka, M. Hirata, H. Etoh, M. Sako, M. Sato, J. Murata,
J. Comb. Chem., 2009, 11, 928–937; (b) M. Koh and
S. B. Park, Mol. Diversity, 2011, 15, 69–81.
H. Murata, D. Darnaedi and T. Fukai, Heterocycles, 2003,
60, 2767–2773.
11 (a) H. Greschik, R. Flaig, J.-P. Renaud and D. Moras, J. Biol. 14 A. Yenesew, S. Derese, B. Irungu, J. O. Midiwo,
Chem., 2004, 279, 33639–33649; (b) C. M. Klinge,
S. C. Jernigan and K. E. Risinger, Endocrinology, 2002, 143,
N. C. Waters, P. Liyala, H. Akala, M. Heydenreich and
M. G. Peter, Planta Med., 2003, 69, 658–661.
853–867; (c) P. Coward, D. Lee, M. V. Hull and 15 A. Yenesew, H. Twinomuhwezi, B. T. Kiremire,
J. M. Lehmann, Proc. Natl. Acad. Sci. U. S. A., 2001, 98,
8880–8884; (d) D. D. Yu and B. M. Forman, Bioorg. Med.
M. N. Mbugua, P. M. Gitu, M. Heydenreich and
M. G. Peter, Bull. Chem. Soc. Ethiop., 2009, 23, 205–210.
Chem. Lett., 2005, 15, 1311–1313; (e) S. Takayanagi, 16 K. Koch and M. S. Biggers, J. Org. Chem., 1994, 59, 1216–1218.
T. Tokunaga, X. Liu, H. Okada, A. Matsushima and 17 M. S. Sardessai and H. N. Abramson, Org. Prep. Proced. Int.,
Y. Shimohigashi, Toxicol. Lett., 2006, 167, 95–105;
(f) J. Wang, F. Fang, Z. Huang, Y. Wang and C. Wong, FEBS 18 S. K. Ko, H. J. Jang, E. Kim and S. B. Park, Chem. Commun.,
Lett., 2009, 583, 643–647; (g) Y. Wang, F. Fang and 2006, 2962–2964.
C.-W. Wong, Biochem. Pharmacol., 2010, 80, 80–85; 19 J. Sambrook and D. W. Russel, Molecular Cloning: A Labo-
1991, 23, 419–424.
(h) J. Hirvonen, A.-M. Rajalin, G. Wohlfahrt, H. Adlercreutz,
K. Wähälä and P. Aarnisalo, J. Steroid Biochem. Mol. Biol.,
2011, 123, 46–57.
ratory Manual, CSHL Press, New York, 3rd edn, 2001, vol. 3,
pp. 16.21–16.26.
20 J. Sambrook and D. W. Russel, Molecular Cloning: A Labo-
ratory Manual, CSHL Press, New York, 3rd edn, 2001, vol. 3,
pp. 17.48–17.51.
12 E. Y. H. Chao, J. L. Collins, S. Gaillard, A. B. Miller,
L. Wang, L. A. Orband-Miller, R. T. Nolte, D. P. McDonnell,
T. M. Willson and W. J. Zuercher, Bioorg. Med. Chem. Lett., 21 H. Gronemeyer, J.-A. Gustafsson and V. Laudet, Nat. Rev.
2006, 16, 821–824.
Drug Discov, 2004, 3, 950–964.
5786 | Org. Biomol. Chem., 2013, 11, 5782–5786
This journal is © The Royal Society of Chemistry 2013