Synthesis, activity, metabolic stability, and pharmacokinetics of glucocorticoid receptor modulator-statin hybrids

Article abstract of DOI:10.1016/j.bmcl.2004.06.021

The synthesis, activity, metabolic stability, and pharmacokinetics of steroidal and nonsteroidal glucocorticoid receptor modulator-statin hybrids is reported. Potent steroidal antagonist-statin hybrids like 22 (h-GR binding IC₅₀=7nM) and nonsteroidal modulator hybrids like 16 (h-GR binding IC₅₀=2nM) were discovered. Appending a 'statin'-like diol-acid group to the modulators dramatically improved metabolic stability (and in some cases hepatocyte activity), but did not impart hepatoselectivity.

Full text of DOI:10.1016/j.bmcl.2004.06.021

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4173–4178
Synthesis, activity, metabolic stability, and pharmacokinetics of
glucocorticoid receptor modulator–statin hybrids
J. T. Link,a,* Bryan K. Sorensen,a Chunqiu Lai,a Jiahong Wang,a Steven Fung,a
Daisy Deng,a Maurice Emery,a, Sherry Carroll,a Marlena Grynfarb,b
Annika Goos-Nilssonb and Thomas von Gelderna
aMetabolic Disease Research, Abbott Laboratories, Dept 4 CB, Bldg. AP-10, Rm. L-14, 100 Abbott Park Road,
Abbott Park, IL 60064-6098, USA
bKaro Bio AB, Novum, SE-141 57 Huddinge, Sweden
Received 9 February 2004; revised 7 June 2004; accepted 9 June 2004
Available online 2 July 2004
Abstract—The synthesis, activity, metabolic stability, and pharmacokinetics of steroidal and nonsteroidal glucocorticoid receptor
modulator–statin hybrids is reported. Potent steroidal antagonist–statin hybrids like 22 (h-GR binding IC50 ¼ 7 nM) and nonste-
roidal modulator hybrids like 16 (h-GR binding IC50 ¼ 2 nM) were discovered. Appending a ‘statin’-like diol-acid group to the
modulators dramatically improved metabolic stability (and in some cases hepatocyte activity), but did not impart hepatoselectivity.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Metformin has become a cornerstone of antidiabetic
treatment both as a mono-therapy and in combination
with other agents.1 Although its precise mechanism of
action is still under investigation, its predominant
activity is inhibition of hepatic glucose production
(HGP). Its therapeutic and commercial success has
stimulated research on multiple targets whose regulation
may moderate HGP.2 Antagonists of the glucocorticoid
receptor (GR) have been explored as potential antidia-
betic agents that inhibit HGP.3 Because of the possi-
bility of undesired effects due to extrahepatic GR
antagonist exposure, attempts have been made to target
them to the liver.4 Steroidal GR antagonists, like mife-
pristone (RU-486) (1), are well known members of this
class (Fig. 1). However, they are unselective and show
significant progesterone receptor (PR) activity leading to
their abortive properties. Recently, several classes of
nonsteroidal GR ligands, including a series of GR
selective modulators from our laboratory have been
identified.5;6 Both the steroidal and nonsteroidal GR
antagonists are typically widely distributed after oral
Figure 1. GR ligands and HMG CoA reductase inhibitors.
Keywords: Glucocorticoid receptor modulator; Statin.
* Corresponding author. Tel.: +1-847-935-6806; fax: +1-847-938-1674;
dosing potentially leading to both hepatic and systemic
effects. We have sought to alter the pharmacokinetics of
e-mail: james.link@abbott.com
Present address: Amgen, Thousand Oaks, CA 91320, USA.
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.021

4174
J. T. Link et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4173–4178
GR antagonist mifepristone (1) and the GR selective
nonsteroidal modulator 2 to increase their hepatic
selectivity.
One example of liver selective small molecule thera-
peutics is the statin class of 3-hydroxy-3-methylglutaryl
coenzyme A reductase inhibitors.7 These compounds,
which prevent coronary heart disease by lowering low-
density lipoprotein cholesterol levels, show preferential
distribution and action in the liver. However, some
potentially beneficial extrahepatic effects have also been
studied.8 Lovastatin (3) and pravastatin (4) are two
statins that have polar sidechains that structurally rep-
resent the class. Inactive hydroxylactones, like 3, upon
ingestion and absorption, undergo lactone hydrolysis to
the active diol-acids. Frequently the lipophilic lactones
are able to passively cross membranes more efficiently
than the polar diol-acids. Several diol-acids, like prav-
astatin (4), are passively and actively taken up by
transporters (frequently the organic ion transporter)
into hepatocytes.9 This increases their hepatoselectivity.
Pravastatin (4) is orally absorbed without the necessity
of administering the corresponding lactone.
Scheme 1. Reagents and conditions: (a) i. allyl iodide, K2CO3, ace-
tone, 40 ꢁC, 12 h, 41%, ii. 4-fluorobenzaldehyde, K2CO3, DMF,
100 ꢁC, 12 h, 81%; (b) 2-methyl-3-nitroaniline, AcOH, DCE, rt, 4 h,
Na(OAc)3BH, 12 h, 85%; (c) benzylbromide, i-Pr2NEt, DMF, 90 ꢁC,
12 h, 94%; (d) Pd(PPh3)4, PhSiH3, CH2Cl2, rt, 1 h, 95%.
We prepared hydroxylactone GR antagonist/modula-
tor-hybrids 5 and diol-acid GR antagonists/modulator-
hybrids 6 in order to liver target GR antagonists or
modulators (Fig. 2). In the case of hybrids with the
aniline bearing site of mifepristone (1), the GR binding
activity was expected to be similar to 1 because the ap-
pended groups would project outside of the ligand-
binding domain of the GR.10 Reported structure activity
studies also indicated large groups could be linked to
this site without large reductions in potency.11 The
activity of the hybrids of the nonsteroidal GR modula-
tor 2 was less predictable. The site of attachment and
linker would need to be experimentally determined. If
these hybrids target the GR modulators to the liver they
could be optimized for dual lipid-lowering activity.
Alternatively, HMG CoA activity could be potentially
removed by altering the diol stereochemistry.
fluorobenzaldehyde provided aldehyde 8. Reductive
amination with 2-methyl-3-nitroaniline followed by
alkylation with benzyl bromide gave allyl ether 10. De-
protection using tetrakis(triphenylphosphine)palla-
dium[0] in the presence of phenylsilane yielded phenol
11.12
The key portion of the statins was appended from a
known precursor via precedented alkylation chemistry
(Scheme 2).13 Phenol 11 and mesylate 12 reacted to give
ether 13 in 62% yield. Conversion of nitroarene 13 into
methylsulfonamide 14, followed by deprotection, gave
the diol-acid 16 in good yield. Lactone 17 was prepared
from t-butyl ester 15 by treatment with trifluoroacetic
acid. Steroidal GR antagonist–statin hybrids 21–24 were
prepared from a known intermediate using a similar
strategy.14
Two sites in the nonsteroidal modulators tolerated large
polar substituents. At these positions, GR modulator-
statin hybrids were prepared using an aryl ether linker.
The core of one modulator was prepared from hydro-
quinone 7 (Scheme 1). Allylation followed by a
nucleophilic aromatic substitution reaction with 4-
The antagonists and modulators were assayed in a h-GR
binding assay, monitoring for the displacement of
radiolabeled dexamethasone. Functional activity was
assessed in a GR alkaline phosphatase reporter whole
cell assay (GRAF).6a A second assay measuring the
blockade of dexamethasone-induced activity of tyrosine
amino transferase (TAT) in freshly isolated rat hepato-
cytes was also employed. Since rat hepatocytes have
transporters (the organic ion transporter among others),
it was hoped diol-acid hybrids would show improved
activity in this assay. Weaker activity might also be
observed in the GRAF assay if the statin substituents
lowered cell permeability.
The nonsteroidal GR modulator 2 binds potently to h-
GR (IC50 ¼ 2.7 nM) (Table 1). It also has excellent
selectivity over other nuclear hormone receptors
(>100·), including human progesterone receptor (h-
PR), mineralocorticoid (h-MR), androgen receptor (h-
Figure 2. GR ligand–statin hybrids.

J. T. Link et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4173–4178
4175
Scheme 2. Reagents and conditions: (a) K2CO3, 18-crown-6, DMSO, 80 ꢁC, 18 h, 62%; (b) Fe, NH4Cl, EtOH, H2O, 80 ꢁC, 1 h; MsCl, pyr, rt, 1 h,
84%; (c) 3 N HCl, THF, EtOH, rt, 12 h, 97%; (d) NaOH, EtOH, rt, 2.5 h, 72%; (e) TFA, CH2Cl2, 0 ꢁC fi rt, 2 h, 82%.
Table 1. GR binding and functional assay results for nonsteroidal modulators 2 and 16–20
R3
R2
R4
R1
N
O
O
O
S
N
H
Compds
R1
R2
H
R3
H
R4
h-GR Binding
IC50, nMa
GRAF
IC50, nMa
Rat TAT
IC50, lMa
2
H
H
H
2.7
240
300
>30
>30
O
OH OH
16
H
H
2.1
O
HO
HO
O
17
H
H
H
H
2.1
440
>30
O
O
18
19
H
H
F
F
F
F
21
15
385
760
>30
>30
O
OH OH
O
HO
HO
O
20
H
F
F
25
670
>30
O
O
a Values are the geometric means of two experiments (na ¼ not active, nd ¼ not determined).
AR), and estrogen receptor (ERa and b) (data not
shown). However, sulfonamide 2 has modest functional
activity in the GRAF assay and is inactive in the
rat hepatocyte TAT assay. para-Substituted hybrids

4176
J. T. Link et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4173–4178
diol-acid 16 and lactone 17 have similar activity to the
parent structure and are equally selective for h-GR.
Diol-acid 16 shows little species selectivity (r-GR bind-
ing IC50 ¼ 2.6 nM). meta-Substituted hybrids of the h-
GR selective difluoride 18 were also prepared. Although
less potent, these hybrids (19 and 20) also maintained
the activity of the parent 18. Interestingly, the diol-acid
16 and lactone 17 show similar activity, as do the other
pair 19 and 20. Although potent binding statin hybrids
were identified their hepatocyte functional activity did
not improve.
The metabolic stability of representative GR modulator
and antagonist–statin hybrids were evaluated in rat and
human microsomes and hepatocytes. For the lactone 17,
the appearance of the corresponding diol-acid 16 was
also monitored. Diol-acid 16 was metabolically stable
when exposed to rat, or human, liver microsomes
(incubation period ¼ 60 min). The parent dibenzylaniline
2 is rapidly metabolized in microsomal incubations
indicating the diol-acid group stabilizes the core struc-
ture. This is a desirable property for a liver targeting
group. Lactone 17 under identical conditions converted
to the stable diol-acid 16 during the incubation period.
Diol-acid 16 underwent metabolism by both rat and
human hepatocytes. In rat hepatocytes, metabolism
occurred more extensively consuming all but 5%,
whereas 60% of the parent remained at the end of the in
human hepatocytes (6 h studies). The identity of the
metabolites was not determined but they are less polar
than the diol-acid 16. This led us to speculate that they
are dehydration products. During these experiments it
was also noted that lactone 17 converts to acid 16 in
buffer (half life not determined). This may also be the
cause of their similar activity in the in vitro assays.
Steroidal GR antagonists–statin hybrids were also
evaluated (Table 2). The steroidal core is less selective
due to its potent antagonism of h-PR. A shorter
monohydroxyacid 21 showed diminished activity rela-
tive to mifepristone (1). However, the longer chain diol-
acid 22 and corresponding lactone 23 retain similar
h-GR binding potency to the parent 1. Aniline hybrids
22 and 23 also have potent functional antagonist activity
(GRAF IC50 ¼ 18 and 23 nM, respectively). Ether
hybrid diol-acid 24 has potent binding activity but has
significantly weaker GRAF activity, relative to mife-
pristone (1) and closely related aniline 21, indicating it is
probably not cell permeable. Interestingly, all the com-
pounds show good potency in the rat hepatocyte TAT
assay. This suggests diol-acid 24 and the other steroid
hybrids are actively transported into hepatocytes.
Pharmacokinetic studies at 5 mpk in Sprague–Dawley
rats of open chain acid 16 showed the compound was
cleared rapidly (Clp ¼ 5.2 L/h kg), had a high volume of
distribution (Vb ¼ 6.2 L/kg), modest AUC (1.0 lg h/mL)
Table 2. GR binding and functional assay results for steroidal antagonists 1 and 21–25
R
OH
H
H
O
Compds
R
h-GR binding
IC50, nMa
h-PR Binding
IC50, nM
GRAF
IC50, nMa
Rat TAT
IC50, lMa
1
NMe2
1.1
2.9
4.8
0.27
O
O
OH
21
22
23
24
37
200
76
0.22
N
HO
HO
OH OH
6.8
2.3
9.2
32
3.8
47
18
23
0.13
0.14
0.56
N
O
O
N
HO
O
OH OH
2900
O
HO
a Values are the geometric means of two experiments (na ¼ not active, nd ¼ not determined).

J. T. Link et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4173–4178
4177
Acknowledgements
We thank Dr. Richard Granneman for his insight and
thoughtful suggestions of potential modifications to in-
crease the liver selectivity of GR ligands. We also
acknowledge Dr. Alan E. Walker and Kaneka Corpora-
tion for a generous gift of t-butyl-(3R,5S)-6-hydroxy-3,5-
O-isopropylidene-3,5-dihydroxyhexanoate. Dr. Kennan
Marsh is thanked for her assistance with the design of the
pharmacokinetic experiments.
References and notes
1. Cusi, K.; DeFronzo, R. A. Diabetes Rev. 1998, 6, 89–131.
2. (a) Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.;
Rohde, J. J.; Richards, S. J.; Souers, A. J.; Szczepankiew-
ica, B. G. Curr. Med. Chem. 2003, 10, 99–122; (b)
Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.;
Richards, S. J.; Rohde, J. J.; Souers, A. J.; Szczepankiew-
ica, B. G. Curr. Med. Chem. 2003, 10, 123–154.
Figure 3. Rat pharmacokinetics of GR modulator–statin hybrid 16.
3. Friedman, J. E.; Sun, Y.; Ishizuka, T.; Farrell, C. J.;
McCormack, S. E.; Herron, L. M.; Hakimi, P.; Lechner,
P.; Yun, J. S. J. Biol. Chem. 1997, 272, 31475–31481.
4. (a) Apelqvist, T.; Wu, J.; Koehler, K. F. WO00058337,
2000; (b) Sorensen, B.; Link, J. T.; von Geldern, T.;
Emery, M.; Wang, J.; Hickman, B.; Grynfarb, M.; Goos-
Nilsson, A.; Carroll, S. Bioorg. Med. Chem. Lett. 2003, 13,
2307–2310.
5. (a) Link, J. T.; Sorensen, B. K.; Patel, J. R.; Arendsen, D.
L.; Li, G. WO 064550, 2002; (b) Link, J. T.; Sorensen, B.
K.; Patel, J.; Emery, M.; Grynfarb, M.; Goos-Nilsson, A.
Bioorg. Med. Chem. Lett. 2004, 14, 2209–2212; (c) Link, J.
T.; Sorensen, B. K.; Patel, J.; Arendsen, D.; Li, G.;
Swanson, S.; Nguyen, B.; Emery, M.; Grynfarb, M.;
Goos-Nilsson, A. Bioorg. Med. Chem. Lett. 2004, 14, in
press. doi:10.1016/j.bmcl.2004.06.023.
and a short half life (t1=2 ¼ 48 min) after iv dosing. Bio-
availability was moderate (F ¼ 54%) (Fig. 3). Although
the high clearance may be consistent with liver uptake,
liver levels at t ¼ 7 h were low (0.024 lg/g) in the iv
animals and the orally treated animals were not appre-
ciably higher (0.1 lg/mg). Diol-acid 19 gave a similar
profile. Oral administration of the corresponding lac-
tone 20 provided nearly identical blood and liver levels
of diol-acid 19 when compared with oral dosing of 19
itself. The data is consistent with lactone hydrolysis
prior to absorption. No lactone 20 was detected in either
plasma or liver.
Steroidal hybrid 24 had a similar profile to the nonste-
roid 16. Steroidal diol-acid 24 was stable in rat and
human liver microsomes (>90% remaining after 60 min).
Good stability was also observed in rat (75% remaining)
and human (>95% remaining) hepatocytes after a 6 h
incubation. As with the nonsteroidal diol-acids, this
profile represents a substantial improvement over the
rapid metabolism of mifepristone (1). Oral/iv pharma-
cokinetic studies at 5 mg/kg in Sprague–Dawley rats
indicated steroidal antagonist 24 had a high iv clearance
consistent with liver uptake. However, low bioavail-
ability (<5%) and liver levels (<5 lg/g) after iv or oral
dosing were observed.
6. (a) Apelqvist, T. WO 63976, 1999; (b) Kym, P.; Lane, B.;
Pratt, J.; von Geldern, T.; Winn, M.; Brenneman, J.; Patel,
J.; Arendsen, D.; Akritopoulou-Zanze, I. U.S. Patent
6,329,534 B1, 2001; (c) Miner, J. N.; Tyree, C.; Hu, J.;
Berger, E.; Marschke, K.; Nakane, M.; Coghlan, M. J.;
Clemm, D.; Lane, B.; Rosen, J. Mol. Endocrinol. 2002, 17,
117–127; (d) Morgan, B. P.; Swick, A. G.; Hargrove, D.
M.; LaFlamme, J. A.; Moynihan, M. S.; Carroll, R. S.;
Martin, K. A.; Lee, E.; Decosta, D.; Bordner, J. J. Med.
Chem. 2002, 45, 2417–2424; (e) Grese, T. A.; Jadhav, P.
K.; Neel, D. A.; Steinberg, M. I.; Lander, P. A. WO
078394, 2003; (f) Amjad, A.; Balkovec, J. M.; Graham,
D. W.; Thompson, C. F.; Nazia, Q. WO 086294, 2003.
7. (a) Hamelin, B. A.; Turgeon, J. Trends Pharm. Sci. 1998,
19, 26–37; (b) Corsini, A.; Bellosta, S.; Baetta, R.;
Fumagalli, R.; Paoletti, R.; Bernini, F. Pharm. Ther.
1999, 84, 413–428; (c) Hatanaka, T. Clin. Pharmacokinet.
2000, 39, 397–412; (d) McTaggart, F.; Buckett, L.;
Davidson, R.; Holdgate, G.; McCormick, A.; Schneck,
D.; Smith, G.; Warwick, M. Am. J. Cardiol. 2001, 87,
28D–32B; (e) Nezasa, K.; Higaki, K.; Matsumura, T.;
Inazawa, K.; Hasegawa, H.; Nakano, M.; Koike, M. Drug
Metab. Dispos. 2002, 30, 1158–1163.
Nonsteroidal GR selective modulator–statin hybrids
and steroidal GR antagonist–statin hybrids were pre-
pared that have similar nuclear hormone activity as
their potent parent compounds. The dihydroxyhexa-
noic acid side chains improved the metabolic stability
of both the steroidal antagonist and nonsteroidal
modulator. It also improved the cellular activity of the
steroid hybrids in hepatocytes. These are valuable
properties for a polar low molecular weight group to
confer. However, neither class of hybrids achieved the
necessary liver levels (or liver selectivity) to provide
sufficient GR blockade to lower blood glucose levels.
The successful realization of liver targeting for these
series with other structural motifs will be reported in
future publications.11
8. Mundy, G.; Garrett, R.; Harris, S.; Chan, J.; Chen, D.;
Rossini, G.; Boyce, B.; Zhao, M.; Gutierrez, G. Science
1999, 286, 1946–1949.
9. Nakai, D.; Nakagomi, R.; Furuta, Y.; Tokui, T.; Abe, T.;
Ikeda, T.; Nishimura, K. J. Pharm. Exp. Ther. 2001, 297,
861–897.
€
10. Kauppi, B.; Jakob, C.; Farnegardh, M.; Yang, J.; Ahola,
H.; Alarcon, M.; Calles, K.; Engstrom, O.; Harlan, J.;
ꢀ
€

4178
J. T. Link et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4173–4178
€
Muchmore, S.; Ramqvist, A.-K.; Thorell, S.; Ohman, L.;
ꢁ
12. Dessolin, M.; Guillerez, M.-G.; Thieriet, N.; Guibe, F.;
Loffet, A. Tetrahedron Lett. 1995, 36, 5741–5744.
13. Jendralla, H.; Granzer, E.; Kerekjarto, B. V.; Krause, R.;
Schacht, U.; Baader, E.; Bartmann, W.; Beck, G.; Berg-
mann, A.; Kesseler, K.; Wess, G.; Chen, L.-J.; Granata,
ꢀ
Greer, J.; Gustaffson, J.-A.; Carlstedt-Duke, J.; Carlquist,
M. J. Biol. Chem. 2003, 278, 22748–22754.
11. von Geldern, T.; Tu, N.; Kym, P. R.; Link, J. T.; Jae, H.;
Lai, C.; Apelqvist, T.; Rhonnstadt, P.; Hagberg, L.;
Koehler, K.; Grynfarb, M.; Goos-Nilsson, A.; Sandberg,
€
S.; Herchen, J.; Kleine, H.; Schussler, H.; Wagner, K. J.
Med. Chem. 1991, 34, 2962–2983.
€
J.; Osterlund, M.; Wang, J.; Fung, S.; Wilcox, D.; Nguyen,
€
P.; Jakob, C.; Hutchins, C.; Farnegardh, M.; Kauppi,
B.; Ohman, L.; Jacobsen, P. B. J. Med. Chem., in
press.
ꢀ
ꢁ
14. (a) Belanger, A.; Philibert, D.; Teutsch, G. Steroids 1981,
€
37, 361–382; (b) Acosta, K.; Cessac, J. W.; Rao, N.; Kim,
H. K. J. Chem. Soc., Chem. Commun. 1994, 1985–1986.