Bile acid conjugates of a nonsteroidal glucocorticoid receptor modulator

Article abstract of DOI:10.1016/j.bmcl.2004.06.022

Bile acid conjugates of a selective nonsteroidal glucocorticoid receptor modulator were prepared and evaluated. Potent GR binding conjugates that showed improved metabolic stability were discovered. However, cellular potency and pharmacokinetics were not substantially improved.

Full text of DOI:10.1016/j.bmcl.2004.06.022

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4179–4183
Bile acid conjugates of a nonsteroidal glucocorticoid
receptor modulator
Noah Tu,a, J. T. Link,a,* Bryan K. Sorensen,a Maurice Emery,a,à
Marlena Grynfarb,b Annika Goos-Nilssonb and Bach Nguyena
aMetabolic Disease Research, Abbott Laboratories, Dept 4CB, Building 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 9 June 2004; accepted 9 June 2004
Available online 2 July 2004
Abstract—Bile acid conjugates of a selective nonsteroidal glucocorticoid receptor modulator were prepared and evaluated. Potent
GR binding conjugates that showed improved metabolic stability were discovered. However, cellular potency and pharmacokinetics
were not substantially improved.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The incidence of type 2 diabetes continues to rise
increasing the need for safer and more efficacious med-
ications for this debilitating disease.1 One strategy for
discovering anti-diabetics is to design compounds that
inhibit hepatic glucose production (HGP).2 One target
of these efforts is to identify glucocorticoid receptor
(GR) antagonists, or modulators, that inhibit the
expression of key enzymes involved in HGP.3 Due to the
undesired effect of systemic GR antagonism, such as
effects on the HPA axis or bone, researchers have sought
to target these agents to the liver.4 One strategy to
accomplish this goal is to discover antagonists that are
enterohepatically recirculated.5 A tactic, which has been
explored to increase the liver selectivity of small mole-
cules is to conjugate them to enterohepatically recircu-
antagonists to bile acids to determine if their profile
would improve.
lated bile acids.6–8 Recently, we discovered
a
nonsteroidal series of potent and selective GR modula-
tors that have modest functional activity and poor
metabolic stability.9 Typically, the most potent members
of the series contain the core sulfonamide structure 1
and are widely distributed (systemically available) fol-
lowing oral dosing (Fig. 1). We conjugated these
Keywords: Bile acid; Glucocorticoid receptor modulator; Conjugate.
* Corresponding author. Tel.: +1-847-935-6806; fax: +1-847-938-1674;
e-mail: james.link@abbott.com
Figure 1. Representative nonsteroidal glucocorticoid receptor antag-
onist 1, bile acids 2 and 3, and generic bile acid conjugate 4.
Present address: S*BIO, 1 Science Park Road, Singapore 117528.
Present address: Amgen, Thousand Oaks, CA 91320, USA.
à
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.022

4180
N. Tu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4179–4183
Our efforts centered on the bile acids cholic acid (2) and
taurocholic acid (3). Cholic acid (2) is partially con-
verted in vivo to taurocholic acid (3) and both com-
pounds are enterohepatically recirculated in rodents and
humans. Our initial plan was to discover GR modula-
tors attached via a linker to a carboxylic acid that would
reach outside of the GR ligand binding domain (LBD).
These compounds could be coupled to appropriately
functionalized bile acid pieces. This should lead to
conjugates whose potencies are nearly equivalent to the
unconjugated precursors. The GR modulator bile acid
conjugates 4 prepared to evaluate this strategy were
conjugated via the 3, 7, and 12 hydroxyl groups.
beled agonist dexamethasone. Binding selectivity against
other nuclear hormone receptors, including the human
progesterone receptor (h-PR), mineralocorticoid recep-
tor (h-MR), androgen receptor (h-AR), estrogen recep-
tor (h-ER), and thyroid-hormone receptor (h-TR a and
b), was checked for important compounds. A reporter
cell line (GRAF) expressing the glucocorticoid response
element linked to the alkaline phosphatase reporter gene
was utilized to assess the functional activity of the
compounds (Table 1).4
A set of GR modulator acids was first prepared to
determine the optimal site for conjugation. Due to the
lipophilic nature of the GR LBD, it was hypothesized
that if a potent modulator acid were discovered, the acid
would extend outside of the pocket into solvent. The
corresponding amide conjugates would then be expected
to retain the potency of the parent acid. 3-b-(2-Amino-
ethoxy)-cholic acid was selected as a bile acid amine for
conjugation. This was due to its equatorial linker chain
that would minimize interaction between the GR mod-
ulator and the bile acid. Dibenzylaniline modulators 17
and 18 were first examined for modification. Linkers
were incorporated at several positions. However, only
the para-position of the benzyl group tolerated large
substituents (data not shown). Addition of a three car-
bon acid chain produced acid 20 that has micromolar
activity. Its corresponding conjugate 19 has improved
(although modest) potency. This suggested the amide in
20 remained within the LBD, and that the para-position
could function as a linking site.
The GR modulators were prepared from 4-fluorobenz-
aldehyde 5 and a halophenol 6 (Scheme 1). Nucleophilic
aromatic substitution yielded biphenyl ether aldehyde 7.
Reductive amination with 2-methyl-3-nitroaniline 8
provided aniline 9. Alkylation with benzyl bromide gave
nitroarene 10 that was reduced, and sulfonylated, to
provide bromide 11. Installation of different linkers,
utilizing palladium mediated cross-coupling reactions,
allowed for the preparation of a variety of conjugates.
One example is the Negishi coupling10 of bromide 11
with 3-ethoxy-3-oxopropylzinc bromide that, after
hydrolysis, provided acid 12 suitable for coupling to an
appropriately modified bile acid. Similar chemistry
provided the GR modulators shown in Table 1.
Cholic acid 2 was converted to 3-b-hydroxyethyl ether
13 according to known procedures (Scheme 2).11 The
primary alcohol in triol 13 was then selectively con-
verted to the corresponding bromide. Displacement with
ammonia yielded amine 14; ready for coupling and de-
protection to provide the GR-modulator conjugates 4.
Due to the potency (h-GR binding IC50 ¼ 2.7 nM, r-GR
binding IC50 ¼ 1.3 nM) of diaryl ether 21 the modulator
core was enlarged.9b Acids 24 and 26 extended from the
smaller aniline benzyl substituent showed moderate
binding potency with the meta-isomer have better po-
tency. Conjugation yielded 23 and 25 that have roughly
twofold weaker binding potency than their parents. This
is surprising when compared with the modulator con-
jugate pair 19 and 20 suggesting different binding
modes.
O-(Benzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
tetrafluoroborate (TBTU)12 mediated coupling of GR
modulator acid 12 with cholic acid derived amine 14
yielded amide 15 (Scheme 3). Hydrolysis then gave the
deprotected diol–acid conjugate 16. Similar chemistry
provided the conjugates described herein.
Compounds were assayed in a h-GR competitive bind-
ing assay measuring the displacement of the radiola-
Extending the length of the linker was accomplished by
linking through the aryloxy group of 21 and 22. Modi-
Scheme 1. Reagents and conditions: (a) K2CO3, DMF, 100 ꢁC, 12 h, 79%; (b) 8, CH3CO2H, DCE, rt, 4 h; Na(OAc)3BH3, 12 h, 85%; (c) BnBr, i-
Pr2NEt, DMF, 95 ꢁC, 12 h, 87%; (d) (i) Fe, NH4Cl, 80 ꢁC, 1 h; (ii) MsCl, pyr, rt, 1 h, 81%; (e) (i) EtO2CCH2CH2ZnBr, Pd(PPh3)4, THF, 75 ꢁC, 12 h,
77%; (ii) NaOH, THF, EtOH, H2O, rt, 12 h, 98%.

N. Tu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4179–4183
4181
Table 1. GR binding and cellular GRAF activity for compounds 16–35. In the case of the conjugates, the structure shown is for the para-substituted
conjugate where the para refers to the aromatic ring to which the linker (between the bile acid and the GR modulator) is attached
Compds
Structure
GR Binding
IC50 (nM)a
GRAF IC50
(nM)a
R
17 (R ¼ H)
18 (R ¼ F)
28
15
250
100
R
N
O O
S
N
H
O
HO
H
OH
H
19 (20)
560 (>1000)
310 (>1000)
N
H
N
H
H
OH
O
O O
S
N
H
R
H
O
21 (R ¼ H)
22 (R ¼ F)
2.7
95
235
R
N
1800
O
O
O
S
N
H
OPh
O
HO
H
OH
H
para 23 (24)
meta 25 (26)
260 (170)
93 (56)
ND (ND)
ND (310)
N
H
H
H
OH
N
( )2
O
O O
S
H
O
N
H
O
HO
H
OH
H
para 27 (28)
meta 29 (30)
50 (8.8)
62 (270)
ND (670)
390 (ND)
N
H
H
H
OH
O
O O
N
O
S
H
N
H
O
F
O
HO
H
OH
F
H
31 (32)
270 (6.9)
ND (220)
N
H
H
OH
O
O
O O
S
NH
O
H
N
H
O
para 16 (12)
X ¼ CH2
17 (34)
57 (63)
450 (500)
HO
H
OH
H
N
X
O
meta 33 (34)
X ¼ O
590 (2300)
O O
H
H
OH
S
O
HN
N
H
O
H
The activity of the unconjugated modulator acid is shown in parenthesis. The activity of the corresponding meta-isomer is also shown in some cases.
a Values are geometric means of two experiments (NA ¼ not active, ND ¼ not determined).

4182
N. Tu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4179–4183
Scheme 2. Reagents and conditions: (a) Ref. 11; (b) (i) NBS, PPh3,
0 ꢁC fi rt, 3 h, 79%; (ii) NH3, MeOH, 100 ꢁC, 16 h, 65%.
Scheme 4. Metabolic stability of a nonsteroidal glucocorticoid recep-
tor modulator 17 and the related bile acid conjugate 19 in rat
hepatocytes.
important property for a liver targeting group to confer,
allowing for a diversity of GR modulator structures.
In a pharmacokinetic study in Sprague–Dawley rats
conjugate 16 (MW ¼ 978.3) had lower bioavailiability
(F <5%) than the nonconjugate acid 12 (F ¼ 10%). This
suggests that the conjugate is not passively absorbed, or
actively transported, into the bloodstream by the rat
ileal bile acid transporter (r-iBat). It was also hoped that
conjugate formation would potentially improve cellular
potency of the modulators. Initial GRAF results did not
result in improvement (Table 2). However, this cell line
does not contain bile acid transporters so poor activity
may be expected (and perhaps even desired). To explore
this possibility, compounds were analyzed in a second
assay measuring the blockade of dexamethasone-in-
duced activity of tyrosine amino transferase (TAT) in
freshly isolated rat hepatocytes. Improvement of con-
jugate 16 relative to acid 12 was not observed and did
not approach the potency of the steroidal antagonist
RU-486. Biochemical studies of conjugate 16 to block
the ability of rat hepatocytes to take up radiolabeled
taurocholate (TCA) indicated the conjugate strongly
interacted with the transporters. Interestingly, acid 12
also altered taurocholate uptake more potently than
taurocholate itself. However, the compound was not
enterohepatically recirculated as assessed by bile con-
centration measurements in bile cannulated rats.
Scheme 3. Reagents and conditions: (a) TBTU, i-Pr2NEt, DMF,
50 ꢁC, 5 h, 83%; (b) LiOH, dioxane, H2O, rt, 16 h, 95%.
fication of the phenoxy group gave the potent acid 28
and less potent meta-substituted isomer 30. Conjugate
27 has 5–6-fold weaker binding to GR compared to its
parent acid 28. Interestingly, conjugation of acid 30
improved binding potency, as seen in 29. These results
suggest the linker region still remains within the GR
pocket. Similar results were obtained with a one carbon
shorter linker in conjugate 31 and parent acid 32.
Exchanging the two fluorines for protons decreases the
potency of acid 12 relative to 32. However, the binding
potency of conjugate 16 is dramatically improved. The
conjugate 16 is 20-fold selective over AR, MR, and PR
(data not shown) retaining this important property of its
parent 2. The related meta-substituted ethers 33 and 34
show similar potency to one another. Comparison with
29 and 30 suggests this is coincidental rather than
reflective of extending outside of the pocket. Functional
activity, as assessed in the GRAF assay, was greater
than tenfold worse than binding for all modulators and
conjugates.
Bile acid conjugates of a series of metabolically labile
GR modulators were prepared. SAR analysis suggests
Table 2. Rat hepatocyte TAT and transport assay results for com-
pounds 12 and 16
Compds GRAF
IC50, nMa
Rat hepatocyte
3H-Taurocholate
TAT IC50 (lM)a uptake IC50 (lM)a
12
16
500
NA (>30)
NA (>30)
ND
1.3
4
450
ND
4.8
GR modulator-bile acid conjugate 19 was evaluated for
metabolic stability in rat hepatocytes (Scheme 4).
Conjugate 19 was remarkably stable relative to GR
modulator 17 indicating that conjugation to cholic acid
dramatically improved metabolic stability. This is an
TCA
RU-486
6
ND
0.27
a Values are geometric means of two experiments (NA ¼ not active,
ND ¼ not determined).

N. Tu et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4179–4183
4183
6. Enhsen, A.; Kramer, W.; Wess, G. DDT 1998, 3, 410–
418.
7. (a) Kramer, W.; Wess, G.; Schubert, G.; Bickel, M.;
Girbig, F.; Gutjahr, U.; Kowalewski, S.; Baringhaus,
that the linkage point between the modulator and the
bile acid was within the LBD. However, potent conju-
gates were discovered that have increased metabolic
stability. Unfortunately, conjugation did not greatly
improve cellular potency or the pharmacokinetic profile
of the modulators. These results indicate that cholic acid
does not function as a universal liver targeting group for
all compounds. Rather, bile acid conjugates need to be
evaluated on a case by case basis.
€
K.-H.; Enhsen, A.; Glombik, H.; Mullner, S.; Necker-
mann, G.; Schulz, S.; Petzinger, E. J. Biol. Chem. 1992,
267, 18598–18604; (b) Kim, D.-C.; Harrison, A. W.;
Ruwart, M. J.; Wilkinson, K. F.; Fisher, J. F.; Hidalgo, I.
J.; Borchardt, R. T. J. Drug Target. 1993, 1, 347–359; (c)
Holzinger, F.; Schteingart, C. D.; Ton-Nu, H.-T.; Eming,
S. A.; Monte, M. J.; Hagey, L. R.; Hofman, A. F.
Hepatology 1997, 26, 1263–1271; (d) Petzinger, E.; Wick-
boldt, A.; Pagels, P.; Starke, D.; Kramer, W. Hepatology
1999, 30, 1257–1268.
Acknowledgements
8. 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,
We thank Dr. Kennan Marsh for assistance with the
design of the pharmacokinetic studies. Jiahong Wang,
Masha Kalmanovich and Steven Fung are thanked for
the rat hepatocyte assays.
€
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.
ꢀ
€
9. (a) Link, J. T.; Sorensen, B. K.; Patel, J.; Emery, M.;
Grynfarb, M.; Goos-Nilsson, A. Bioorg. Med. Chem. Lett.
2004, 14, 2209–2212; (b) 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.
10. Negishi, E.; Liu, F. In Metal-catalyzed Cross-coupling
Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH:
Weinheim, 1998; pp 1–42.
References and notes
1. Zimmet, P.; Alberti, K. G. M. M.; Shaw, J. Nature 2001,
414, 782–787.
2. (a) Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.;
Rohde, J. J.; Richards, S. J.; Souers, A. J.; Szczepankie-
wicz, B. G. Curr. Med. Chem. 2003, 10, 99–121; (b)
Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.;
Richards, S. J.; Rohde, J. J.; Souers, A. J.; Szczepankie-
wicz, B. G. Curr. Med. Chem. 2003, 10, 123–153.
3. Coghlan, M. J.; Elmore, S. W.; Kym, P. R.; Kort, M. E.
Ann. Rep. Med. Chem. 2002, 37, 167–176.
4. Apelqvist, T.; Effendic, S. WO 99/63976, 1999.
5. Sorensen, B. K.; 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.
11. (a) Wess, G.; Kramer, W.; Bartmann, W.; Enhsen, A.;
€
Glombik, S.; Mullner, S.; Bock, K.; Dries, A.; Kleine, H.;
Schmitt, W. Tetrahedron Lett. 1992, 33, 195–198; (b)
Wess, G.; Kramer, W.; Enhsen, A.; Glombik, S.; Baring-
haus, K.-H.; Bock, K.; Kleine, H.; Schmitt, W. Tetrahe-
dron Lett. 1993, 34, 817–818.
€
12. Konig, W.; Geiger, R. Chem. Ber. 1970, 103, 788–
798.