A designed antagonist of the thyroid hormone receptor.

Article abstract of DOI:10.1016/S0960-894X(01)00521-2

We synthesized an analogue of the thyromimetic GC-1 bearing the same hydrophobic appendage as the estrogen receptor antagonist ICI-164,384. While having reduced affinity for the thyroid hormone receptors compared to GC-1, it behaves in a manner consistent

Full text of DOI:10.1016/S0960-894X(01)00521-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2821–2825
A Designed Antagonist of the Thyroid Hormone Receptor
Hikari A. I. Yoshihara,a James W. Apriletti,b John D. Baxterb and Thomas S. Scanlana,*
aDepartments of Pharmaceutical Chemistry and Cellular & Molecular Pharmacology, University of California,
San Francisco, CA 94143-0446, USA
bMetabolic Research Unit, University of California, San Francisco, CA 94143-0540, USA
Received 12 June 2001; accepted 27 July 2001
Abstract—We synthesized an analogue of the thyromimetic GC-1 bearing the same hydrophobic appendage as the estrogen recep-
tor antagonist ICI-164,384. While having reduced affinity for the thyroid hormone receptors compared to GC-1, it behaves in a
manner consistent with a competitive antagonist in a transactivation assay. # 2001 Elsevier Science Ltd. All rights reserved.
Thyroid hormone, 3,30,5-triiodo-l-thyronine (T3, Fig. 1),
plays important roles in development and homeostasis.1
This hormone is synthesized in peripheral tissues by
conversion of its precursor, thyroxine (3,5,30,50-tetra-
iodo-l-thyronine), derived from the thyroid gland, and
to a lesser extent by the thyroid gland. T3 is recognized
in target tissues by thyroid hormone receptors (TRs)2
which are members of the nuclear receptor superfamily
of transcription regulators.3 These receptors mediate the
effects of thyroid hormone by positively and negatively
regulating the transcription of target genes.2
Previously we reported the potent halogen-free thyro-
mimetic GC-1 (Fig. 1) with a modest selectivity for TRb
over TRa.11 In this study we report the design and
synthesis of GC-1 analogues bearing substituents on the
carbon atom that bridges the two aromatic rings.12 The
design of these compounds was based on using a
Excess activity of the thyroid gland leads to hyperthy-
roidism that can result in cardiac arrhythmias, heart
failure, weakness, and nervousness.1 Current approa-
ches to treating this disorder include surgical or
radiological ablation of the thyroid gland, or the
administration of drugs such as propylthiouracil and
methimazole that inhibit the biosynthesis of T3.1 The
cardiac antiarrhythmic drug amiodarone has antithy-
roid effects,4,5 and its major metabolite desethylamio-
darone (Fig. 1) antagonizes TR under certain in vitro
conditions.6À8 Desethylamiodarone, a non-competitive
antagonist of TRb, has been shown to interfere with the
interaction of TRb with p160 co-activator GRIP1.9
However, desethylamiodarone has not been shown to
be an antagonist in cells in culture or in animals.
Amiodarone has a low affinity for the TR,6 a slow onset
of action, and results in serious side effects.10 Thus,
these compounds may not be true TR antagonists, and to
our knowledge, no compound with clear TR antagonist
activity has been reported.
Figure 1. Chemical structures of thyroid hormone (T3), thyromimetic
GC-1, the antithyroid and cardiac antiarrhythmic drug amiodarone,
and estrogen antagonist ICI-164,384.
*Corresponding author. Tel.: +1-415-476-3620; fax: +1-415-502-
7220; e-mail: scanlan@cgl.ucsf.edu
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00521-2

2822
H. A. I. Yoshihara et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2821–2825
chemical structure that should bind to TR, but contain
a side group to perturb the normal agonist-induced
folding of the receptor as based on the TR X-ray crystal
structure.19 One of these analogues was found to
antagonize T3 in transactivation assays using cultured
cells.
arylcerium reagent16,17 instead of its aryllithium pre-
cursor gave much higher yields with this sterically hin-
dered electrophile. Crude carbinol 6 was reacted with
methanol under acidic conditions, substituting the
hydroxyl with a methoxyl group. The resulting product
was more chromatographically mobile and easier to
purify than 6. Selective removal of the TBMPS protect-
ing group with triethylamine trihydrofluoride produced
phenol 7, which was alkylated with either ethyl- or
methylbromoacetate, yielding oxyacetic esters 8 and 9,
respectively.
In this and previous13 reports, we differentially pro-
tected the phenolic hydroxyls to improve a low-yielding
step in the original synthesis of GC-1. Starting with 2-
isopropylphenol (1, Scheme 1), bromination and treat-
ment with triisopropylsilyl chloride yielded bromo-
arylsilyl ether 2. 4-Bromo-3,5-dimethylphenol (3) was
protected as the tert-butylmethoxyphenylsilyl (TBMPS)
acetal, which can be selectively cleaved in the presence
of silyl ethers.14,15 Formylation of 4 to benzaldehyde 5
was followed by the coupling of the two aryl rings via
the arylcerium(III) reagent prepared in situ from the
aryllithium derived from arylbromide 2. Using the
Attempts at preparing GC-1 analogues with alkoxy
substituents at the bridging carbon were unsuccessful,
presumably due to the lability of the ether. The pro-
ducts of allylation (10, Scheme 2) or arylation (12)
reactions of ethyl ester 8 via acid solvolysis were suffi-
ciently stable to be deprotected, yielding GC-1 analo-
gues 13 and 15. Hydroboration and oxidation of butene
Scheme 1.
Scheme 2.

H. A. I. Yoshihara et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2821–2825
2823
10 followed by the removal of TIPS and methyl ester
protecting groups produced GC-1 analogue 14.
tected with and without first subjecting it to catalytic
hydrogenation, yielding GC-1 analogues 20 and 21.
TR binding data of analogues 13–15 are summarized in
Table 1. Analogue 13, having the highest affinity, was
tested for its ability to activate TR-mediated transacti-
vation; it exhibited low-potency agonistic activity with
EC50 values of approximately 1 mM for TRb and 10 mM
for TRa (data not shown).
The tighter binding analogue 21 (Table 1) was tested for
its ability to activate TR-mediated transcription. The
assay consisted of JEG-3 cells transiently transfected via
electroporation with a TR-expression plasmid and a
DR-4-driven luciferase reporter construct.22 Having
only the slightest ability to activate transcription, its
ability to block agonist-induced transactivation was
examined. The competition curve against 300 pM T3
(Fig. 2a) indicates antagonism, with increasing con-
centrations of 21 resulting in reduced transactivation.
To assess whether the decrease in transactivation was
due to competition with T3 for TR rather than the
inhibition of another factor required for transactivation
or reporter activity, dose–response curves of T3 in the
presence and absence of 10 mM 21 were obtained
(Fig. 2b). The similarity of the curves, differing only in
potency (EC50: 0.64Æ0.14!4.7Æ1.3 nM), is indicative
of competition between T3 and 21. Similar results were
observed with TRa1 (data not shown), with the EC50 of
T3 shifting from 0.17Æ0.05 nM to 2.2Æ1.1 nM in the
presence of 10 mM 21. Antagonism was observed when
the method of transfection was either electroporation or
calcium phosphate but not when using lipofection, pre-
sumably due to the sequestration of 21 by residual
lipofection reagent.
Inspection of the X-ray crystal structures of the ligand
binding domains of estrogen receptor a18 and rat
19
TRa1 suggested that the estrogen antagonist ICI-
164,38420,21 (Fig. 1) projects its long alkylamide appen-
dage in a direction similar to the direction a substituent
on the bridging carbon of GC-1 would project. This led
us to synthesize a GC-1 analogue with the same alkyl-
amide extension. As shown in Scheme 3, methyl ether 9
was allylated, hydroborated and oxidized to butanal 18.
8-Bromooctanoic acid (16) was coupled with N-
methylbutylamine using HBTU, then reacted neat with
molten triphenylphosphine at 125 ꢀC. The crude triphenyl-
phosphonium bromide was deprotonated with lithium
bis(trimethylsilyl)amide to generate the ylid and coupled
with butanal 18. The resulting olefin (19) was depro-
Table 1. Binding affinities of GC-1 bridge-substituted analogues for
TR
With its structural similarity to thyroid hormone and
competitive binding to the receptor, 21 likely binds in
the same ligand binding pocket as agonist ligands. The
Compd
Kd (hTRa1)ÆSE (nM)
Kd (hTRb1)ÆSE (nM)
T3
13
14
15
20
21
0.17Æ0.02
430Æ60
0.14Æ0.01
90Æ10
.
recently published X-ray crystal structure of ERb ICI-
164,38423 shows the steroid core in the ligand binding
pocket with the alkylamide side chain protruding from
the pocket in a manner similar to that observed for the
hydrophobic extensions of other estrogen receptor
1400Æ800
1400Æ700
240Æ50
3600Æ140
4200Æ200
720Æ100
148Æ13
112Æ18
Scheme 3.

2824
H. A. I. Yoshihara et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2821–2825
estrogen antagonist ICI-164,384 and its analogues. For
the ICI series, only the a-epimers are active estrogen
antagonists,21 indicating that the direction in which the
alkylamide extension protrudes is critical for activity.
Since the one stereocenter of 21 is the carbon which
bears the alkylamide appendage, each enantiomer will
project the appendage in a different direction and interact
differently with TR.
Unlike GC-1, analogues 14, 15, 20, and 21 do not show
a preference in binding to the b subtype of TR, although
they share many structural features, including an oxy-
acetic acid side chain, which may be responsible for
conferring GC-1’s b-selectivity.25 Indeed, they have
varying degrees of preference for TRa (Table 1). Thus
in the context of non-isosteric thyroid hormone analo-
gues, additional factors may influence the preferential
binding to one TR subtype.
These results support the notion that the design princi-
ples of other nuclear receptor antagonists26 can also be
successfully applied to the thyroid hormone receptor.
Thyroid hormone analogue 21 represents a prototype
for a new class of antithyroid compounds that will serve
as useful pharmacological probes of thyroid hormone
action.
Acknowledgements
This work was supported by the National Institutes of
Health (DK52798, T.S.S.). H.A.I.Y. is grateful for
financial support from the National Cancer Institute,
Institutional Training Grant T32 CA09270, with sup-
plemental support from the UCSF Dept. of Biochemis-
try’s Herbert W. Boyer Fund. We thank Martin
Privalsky and Keith Yamamoto for the gifts of plas-
mids, used in the transfection assay. J.D.B. has pro-
prietary interests in, and serves as a consultant and
Deputy director to, Karo Bio AB that has commercial
interests in this area of research.
Figure 2. Antagonism by 21 of TR-mediated transactivation of luci-
ferase under the control of a DR4-type thyroid hormone response
element in transiently transfected JEG-3 cells: (a) dose–response of 21
alone (triangles) or in competition with 300 pM T3 (squares); (b) dose–
response of T3 in presence (triangles) and absence (squares) of 10 mM
21.
modulators complexed with ER,18,24 resulting in struc-
tural perturbations to a surface of the ligand binding
domain that interacts with co-regulatory factors. This
binding mode requires the steroid core of ICI-164,384 to
rotate by approximately 180ꢀ from the orientation
.
found in the ERa estradiol structure. For 21 to bind TR
in such a fashion would require a similar change in the
orientation of its thyronine core, which, with its bent
profile incurred by the sp3-hybridized carbon linking the
two aromatic rings, may have a higher energetic cost
than the comparatively planar steroid estradiol. A sig-
nificant penalty with regard to binding affinity is incur-
red with substitution at the bridging carbon (Table 1).
However, comparing analogue 13 to 21, increasing the
side chain length from 3 to 16 carbons has no additional
cost in binding affinity and changes the analogue from
an agonist to an antagonist.
References and Notes
1. Braverman, L. E.; Utiger, R. D.; Werner, S. C.; Ingbar, S.
H. Werner and Ingbar’s the Thyroid: a Fundamental and Clin-
ical Text, 6th ed.; Lippincott: Philadelphia, 1991.
2. Zhang, J.; Lazar, M. A. Annu. Rev. Physiol. 2000, 62, 439.
3. Mangelsdorf, D. J.; Thummel, C.; Beato, M.; Herrlich, P.;
Schutz, G.; Umesono, K.; Blumberg, B.; Kastner, P.; Mark,
M.; Chambon, P. Cell 1995, 83, 835.
4. Singh, B. N.; Vaughan Williams, E. M. Br. J. Pharmacol.
1970, 39, 657.
5. Burger, A.; Dinichert, D.; Nicod, P.; Jenny, M.; Lemarc-
hand-Beraud, T.; Vallotton, M. B. J. Clin. Invest. 1976, 58,
255.
6. Latham, K. R.; Sellitti, D. F.; Goldstein, R. E. J. Am. Coll.
Cardiol. 1987, 9, 872.
7. Bakker, O.; Vanbeeren, H. C.; Wiersinga, W. M. Endocri-
nology 1994, 134, 1665.
8. Vanbeeren, H. C.; Bakker, O.; Wiersinga, W. M. Mol. Cell.
Endocrinol. 1995, 112, 15.
All analogues were synthesized as racemic mixtures with
the substitution reactions of 8 and 9 proceeding under
solvolytic conditions through a presumed carbocation
intermediate. Attempts to prepare enantiopure versions
of 21, via resolution of diastereomeric derivatives of 21
or intermediates to 21, were unsuccessful. The enantio-
mers may have differing activities, as in the case of the

H. A. I. Yoshihara et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2821–2825
2825
9. van Beeren, H. C.; Bakker, O.; Wiersinga, W. M. FEBS
Lett. 2000, 481, 213.
20. Bowler, J.; Lilley, T. J.; Pittam, J. D.; Wakeling, A. E.
Steroids 1989, 54, 71.
10. Morse, R. M.; Valenzuela, G. A.; Greenwald, T. P.; Eulie,
P. J.; Wesley, R. C.; McCallum, R. W. Ann. Intern. Med. 1988,
109, 838.
11. Chiellini, G.; Apriletti, J. W.; Yoshihara, H. A. I.; Baxter,
J. D.; Ribeiro, R. C. J.; Scanlan, T. S. Chem. Biol. 1998, 5,
299.
12. Yoshihara, H. A. I.; Chiellini, G.; Mitchison, T. J.; Scan-
lan, T. S. Bioorg. Med. Chem. 1998, 6, 1179.
13. Chiellini, G.; Nguyen, N. H.; Yoshihara, H. A. I.; Scan-
lan, T. S. Bioorg. Med. Chem. Lett. 2000, 10, 2607.
14. Guindon, Y.; Fortin, R.; Yoakim, C.; Gillard, J. W. Tet-
rahedron Lett. 1984, 25, 4717.
15. Gillard, J. W.; Fortin, R.; Morton, H. E.; Yoakim, C.;
Quesnelle, C. A.; Daignault, S.; Guindon, Y. J. Org. Chem.
1988, 53, 2602.
16. Imamoto, T.; Sugiura, Y.; Takiyama, N. Tetrahedron
Lett. 1984, 25, 4233.
17. Imamoto, T.; Kusumoto, T.; Tawarayama, Y.; Sugiura,
Y.; Mita, T.; Hatanaka, Y.; Yokoyama, M. J. Org. Chem.
1984, 49, 3904.
18. Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard,
R. E.; Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G. L.;
Gustafsson, J.A.; Carlquist, M. Nature (London) 1997, 389,
753.
19. Wagner, R. L.; Apriletti, J. W.; McGrath, M. E.; West,
B. L.; Baxter, J. D.; Fletterick, R. J. Nature (London) 1995,
378, 690.
21. Wakeling, A. E.; Bowler, J. J. Endocrinol. 1987, 112, R7.
22. Transfection assay: JEG-3 cells were grown to ꢁ80%
confluency in MEM Eagle’s with Earle’s BSS with 5% FBS,
penicillin and streptomycin. Approximately 1.5ꢂ106 cells were
transfected by electroporation (BioRad Gene Pulser, 0.22 kV,
960 mF, 0.4 cm cuvettes) in 500 mL electroporation buffer with
5 mg of a DR4-pGL3 reporter construct, 1 mg of pRL-TK
(Promega) and 1 mg of hTR in a pSG5 vector, per transfection.
Cells were plated in 12-well plates in growth medium. After 6
h, medium was changed to a version containing charcoal-
stripped FBS and drugs or vehicle (EtOH) were added in tri-
plicate. After 24 h, cells were harvested and assayed for luci-
ferase activity using the Promega Dual Luciferase kit and
Analytical Luminescence Laboratory Monolight1 3010
luminometer. Binding assay: the TR-binding assay was
performed as previously described.11
23. Pike, A. C. W.; Brzozowski, A. M.; Walton, J.; Hubbard,
R. E.; Thorsell, A. G.; Li, Y. L.; Gustafsson, J. A.; Carlquist,
M. Structure 2001, 9, 145.
24. Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.;
Kushner, P. J.; Agard, D. A.; Greene, G. L. Cell 1998, 95, 927.
25. Wagner, R. L.; Huber, B. R.; Shiau, A. K.; Kelly, A.;
Cunha Lima, S. T.; Scanlan, T. S.; Apriletti, J. W.; Baxter,
J. D.; West, B. L.; Fletterick, R. J. Mol. Endocrinol. 2001, 15,
398.
26. Weatherman, R. V.; Fletterick, R. J.; Scanlan, T. S. Annu.
Rev. Biochem. 1999, 68, 559.