Differentiation of in vitro transcriptional repression and activation profiles of selective glucocorticoid modulators

Article abstract of DOI:10.1016/j.bmcl.2004.01.044

The SAR at C-5 of the 10-methoxy-2,2,4-trimethylbenzopyrano[3,4-f] quinoline core leading to identification of (-) anti 1-methylcyclohexen-3-yl as the optimum substituent that imparts minimal GR mediated in vitro transcriptional activation while maintaining full transcriptional repression is described. The in vitro profile of these candidates in human cell assays relevant to the therapeutic window of glucocorticoid modulators is outlined.

Full text of DOI:10.1016/j.bmcl.2004.01.044

Bioorganic & Medicinal Chemistry Letters 14 (2004) 1721–1727
Differentiation of in vitro transcriptional repression and activation
profiles of selective glucocorticoid modulators
Steven W. Elmore,a,* John K. Pratt,a Michael J. Coghlan,a,y Yue Mao,a Brian E. Green,a
a
David D. Anderson,a Michael A. Stashko,a Chun W. Lin,a DouglasFalls,
Masaki Nakane,a Loan Miller,a CurtisM. Tyree, Jeffrey N. Minerb and Ben Lanea
b
aGlobal Pharmaceutical Research and Development, Abbott Laboratories, Abbott Park, IL 60064-3500, USA
bNew Leads Discovery and Transcription Research, Ligand Pharmaceuticals, Inc., San Diego, CA 92121, USA
Received 26 November 2003; accepted 19 January 2004
Abstract—The SAR at C-5 of the 10-methoxy-2,2,4-trimethylbenzopyrano[3,4-f]quinoline core leading to identification of (À) anti
1-methylcyclohexen-3-yl as the optimum substituent that imparts minimal GR mediated in vitro transcriptional activation while
maintaining full transcriptional repression is described. The in vitro profile of these candidates in human cell assays relevant to the
therapeutic window of glucocorticoid modulatorsisoutlined.
# 2004 Elsevier Ltd. All rights reserved.
Glucocorticoids (GCs) such as prednisolone (pred, 1)1
and dexamethasone2 have long been considered some of
the most effective antiinflammatory treatments for
maladies ranging from asthma to rheumatoid arthritis.
Chronic exposure to GCs, however produces a number
of common, undesired effects associated with altered
bone, glucose and lipid metabolism that manifests itself
into GC induced osteoporosis, glucose intolerance,
altered adipose differentiantion and fat redistribution
(moon face, hump back).3À6 Other hypertensive and
cardiovascular side effects of commonly used GCs such
as pred may be due at least in part to their cross reac-
tivity with other steroid receptors such as the miner-
alocorticoid receptor (MR).
activation function of the receptor complex hasbeen
associated with many of the undesired metabolic side
4,12À14
effectsdecsribed above.
The down-regulation of
specific genes occurs by an indirect mechanism in which
the activated GR complex bindsto and inactivatesother
transcription factors essentially ‘turning off’ the genes
they would normally regulate. The inhibitory action of
the GRC on pro-inflammatory transcription factors
such as AP-1 and NF-kB isbelieved to produce antiin-
flammatory effects by the repression of numerous cyto-
kines, adhesion molecules and enzymes associated with
synthesis of inflammatory mediators.15-20 Among these
are interleukin-1 (IL-1), TNF-a, GM-CSF, IL-3, IL-4,
IL-5, IL-6, IL-8, E-selectin, I-CAM, Cox-2, iNOS, and
PLA2. This repression mechanism is believed to be the
basis of the antiinflammatory effect of glucocorticoid
drugs.21, 22
When GCsbind the cytoplasmic glucocorticoid receptor
(GR), the resulting activated GR–ligand complex
(GRC) translocates to the nucleus where it can either
positively or negatively effect the expression of specific
genes.7À11 The up-regulation of genesrequiresa recep-
tor homodimer to act directly asan endogenoustran-
scription factor by binding to specific promoter regions
termed glucocorticoid response elements (GREs). This
* Corresponding author. Tel.: +1-847-937-7850; fax: +1-847-935-
5165; e-mail: steve.elmore@abbott.com
Present address: Eli Lilly and Co., Lilly Corporate Center, Indiana-
polis, IN 46285, USA.
y
We have been engaged in a research program dedicated
to finding small molecule ligands of the glucocorticoid
0960-894X/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.01.044

1722
S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
receptor with unique transcriptional repression/activa-
tion profilesthat would provide efficaciousantiinflam-
matory activity with diminished GC induced metabolic
side effects. The 2,5-dihydro-2,2,4-trimethyl-1H-[1]-
benzopyrano[3,4-f]quinoline core 2 isa non-tseroidal
tetracylic scaffold that has emerged as a general phar-
macophore for several members of the nuclear hormone
receptor family.23À28 We have described our initial find-
ings on the structure activity relationship of this core
related to GR and found that proper C-10 substitution
was critical for GR selectivity while C-5 substitution could
be used to modulate transcriptional activity.29,30 C-5
aromatic compoundssuch as 3 possess the full spectrum
of transcriptional activation and repression activity
equivalent to pred. C-5 non-aromatic compoundssuch
as 4a and b maintain efficacious repression activity in an
E-Selectin (E-Sel) cotransfection assay while showing
reduced levelsof GRE activation in an MMTV-GRE
cotransfection assay.31 Furthermore, 4a exhibited effi-
cacy similar to pred in several in vivo models of inflam-
mation, while displaying reduced side effects in at least
two critical parameters (glucose metabolism and
bone).32
Our strategy was to first rapidly survey the effects of
substitution at each position of the allyl side chain in 4a
(Fig. 1a). Compounds 6 and 13–15 were prepared in a
single step by Lewis acid promoted addition of known
or commercially available allyl silanes to C-5 methyl
acetal 5 (Fig. 2).31 Palladium catalyzed methylation and
carbonylation of vinyl bromide 6 and hydrolysis of allyl
acetate 15 gave 7, 8 and 16, respectively (Scheme 1).
Addition of 1-(t-butyldimethylsilyl)-1-methoxy ethene
to 5 yielded the methyl acetate that wastransformed to
aldehyde 9 via Dibal-H reduction of the corresponding
Weinreb amide. Terminally substituted allyl derivatives
10a–e were prepared by Wittig olefination of aldehyde
9. Treatment of 5 with methyl (triphenylphosphor-
anylidene)acetate gave the trans-methylbutenoate 11
that washydrolyzed to provide allyl alcohol 16. Hydride
reduction of the corresponding mesylate provided the
trans-butenyl analogue 12.
Table 1 outlinesreceptor binding and cotranfsection
data for substituted allyl analogues. All of these deriva-
tivesexhibited potent GR binding affinity and were
highly selective over PR. Small non-polar substituents
at all positions maintained transcriptional activation
and repression activity while more hydrophilic sub-
stituents at the 20 or 30 positions reduced both activation
and repression potency and efficacy. Simple alkyl term-
inal olefin substitution demonstrated a distinct advan-
tage by maintaining excellent E-selectin repression
activity equivalent to the parent allyl compound 4a
while exhibiting a diminished ability to activate GRE.
The 30,30-dimethyl analogue 10c proved optimal with
full E-sel repression function but significantly reduced
ability to activate GRE (16% dex efficacy vs68% for
4a).
Given the relationship between C-5 substitution and the
transcriptional activity of these compounds, the pro-
spect of further separating the antiinflammatory activity
from side effects prompted a continued refinement of
the C-5 position. Herein we describe the structure
activity relationship study that led to the identification
of 1- methylcyclohexen-3-yl asthe optimum C-5 usb-
stitutent that imparts an improved in vitro transcrip-
tional activation/repression profile. The biological
profile of similar 9-OH-10-OMe substituted analogues
haspreviously been reported. 33 We highlight here the in
vitro profilesof the 9-H-10-OMe analoguesin both
cotransfection and native protein assays which we
believe demonstrate potential for an improved thera-
peutic window.
We next examined the effectsof rigidification of the C-5
allyl group by its inclusion in a six-membered ring (Fig.
1b). Analogues 18–24 (Table 2) were prepared in a sin-
gle step from 5 and known allylsilanes using the method
described in Figure 2. An unattractive aspect of this
tactic was the creation of a second stereogenic center in
two of the three possible templates. Although these and
related compoundswere firts prepared asmixturesof
Compoundswere first evaluated in a competition bind-
ing assay to determine their intrinsic affinity for the a-
isoform of the human glucocorticoid receptor (hGR).
Mindful of progesterone receptor cross reactivity that
has previously been noted in this series, progesterone
receptor binding was also routinely monitored. Those
compoundsthat demontsrated high affinity for hGR
were then screened for their ability to functionally up-
or down-regulate gene transcription in the MMTV-
GRE and E-Sel co-transfection assays, respectively.31
We have primarily utilized these co-transfection assays
to guide our SAR studies to identify compounds worthy
of more in depth analysis. Compounds of particular
interest were then evaluated for their ability to activate
or repress the transcription of genes relevant to the anti-
inflammatory or metabolic side effects of GC in human
cell native protein assays. The repression of cytokine
stimulated IL-6, collagenase and PGE-2 expression was
used as a measure of antiinflammatory activity.32,33 The
activation of aromatase and tyrosine amino transferase
(TAT) was used as a measure of endocrine related side
effects while the repression of osteocalcin (OC) was used
Figure 1. (a) Definition C-5 allyl substitution postions; (b) Three
possible modes of cyclization C-5 allyl side chain.
Figure 2. General synthetic method for additions of allyl silanes to
ꢀ
32
.
asa measure of effect on bone.
methylacetal, 5: allyl silane: CH2Cl2, BF3 Et2O, À78 to 0 C.

S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
1723
diastereomers, our strategy was to first analyze these
mixtures in hope of identifying promising structural
variants. We then planned to focus on these candidates
and rigorously synthesize and characterize all stereo-
isomeric components.
eight-membered cycloalken-3-yl rings( 20–23) confirmed
the six-membered ring to have the best balance of
effective E-Sel repression and low GRE activation.
Superimposing the optimum terminal olefin substitution
(i.e., 10c) with the C-5 cycloalkenyl 20 yielded 1-
methylcyclohexen-3-yl analogue 24 that maintained
excellent E-Sel repression and showed a further decrease
in GRE activation (5% dex vs68% dex for 4a). We
chose to concentrate on the 2:1 diasteromeric mixture
24 to determine the contributionsmade by each of its
component isomers to its in vitro profile.
Of the possible modes of cyclization (18–20), only the
1030 analogue 20 maintained E-selectin repression activ-
ity equivalent to 4a. Interestingly, 20 also showed a
nine-fold decrease in GRE activation efficacy compared
to 4a. Subsequent comparison of the five, six, seven and
Table 1. In vitro receptor binding and cotransfection assay data for C-5 substituted allyl compounds. Compounds 6, 13–15 were prepared
according to the method described in Figure 2 utilizing the allylsilanes below. The phosphonium reagents used to prepare compounds 10a–e are also
provided
Compd
R1
Reagent
GR binding (nM) PR binding (nM)
GRE activation
meanꢁSEMa
E-selectin repressionb
meanꢁSEM
meanꢁSEM
2.4ꢁ0.3*
meanꢁSEM
IC50, (nM)c eff. (%dex) IC50, (nM)c eff. (%dex)
d
Pred
—
NA
NA
—
8.0ꢁ1.1*
33ꢁ8.0 *
89ꢁ19*
68ꢁ38*
2.1ꢁ0.2*
13ꢁ6.3*
99ꢁ1*
94ꢁ2*
4a
2.5ꢁ0.46*
2.2ꢁ1.1*
230ꢁ11
1800ꢁ480
13
14
7
—
110ꢁ70
—
73ꢁ13
—
40ꢁ17
—
89ꢁ3
—
—
NA
NA
3.1ꢁ0.7*
16ꢁ5.8*
32ꢁ8.3
—
66ꢁ19
240ꢁ68
—
81ꢁ22
54ꢁ6
—
42ꢁ3.0
110ꢁ37
—
87ꢁ7
74ꢁ2
5ꢁ1
6
—
8
—
15
16
10a
12
10b
10c
10e
10d
11
17
13ꢁ4.9*
12ꢁ4.7*
2.9ꢁ0.9
0.60ꢁ0.14
3.1ꢁ1.3
6.1ꢁ1.0*
4.5ꢁ0.9
2.1ꢁ0.4*
7.8ꢁ1.6*
26ꢁ5.6*
—
—
390ꢁ40
310ꢁ6.6
39ꢁ20
150ꢁ101
89ꢁ15
—
48ꢁ16
50ꢁ1
46ꢁ11
53ꢁ11
24ꢁ2
16ꢁ6
—
160ꢁ1.3
141ꢁ60
23ꢁ6.4
38ꢁ2.1
88ꢁ54
23ꢁ16
216.7
74ꢁ6
80ꢁ9
91ꢁ4
94ꢁ1
81ꢁ2
92ꢁ1
73ꢁ6
94ꢁ1
77ꢁ2
44ꢁ28
NA
EtPPh3Br
1200ꢁ983
2100ꢁ1300
—
NA
n-PrPPh3Br
i-PrPPh3Br
(cypent)PPh3Br
Ph2P(O)CHF2
Ph3P=CHCO2Me
NA
1600ꢁ420
—
—
2300ꢁ990
—
76ꢁ0.50
220ꢁ48
—
60ꢁ16
46ꢁ10
7ꢁ1
20ꢁ14
87ꢁ10
150ꢁ8.9
—
a Values with standard deviation represent the mean value of two experiments with triplicate determinations, values with an asterisk represent the
mean value of at least three separate experiments in triplicate with standard error (SEM) and values without standard deviation represent a single
experiment in triplicate.
b GRE activation efficacies are represented as the percentage of the maximal response of dexamethasone.
c All IC50 values were determined from full seven point, half-log concentration response curves in CV-1 cells and were calculated as the concen-
tration at half the maximal response.
d A hyphen indicatesa binding potency of >5000 nM, a functional potency that wasnot calculated due to low efficacy or not functionally effica-
cious. NA not applicable.

1724
S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
Unable to successfully separate the components of 24
directly by chiral HPLC, we resorted to independent
synthesis of each diastereomer in racemic form (Scheme
2) with the hope of resolving them individually. Addi-
tion of 3-(t-butyldimethylsiloxy-methoxy-methylene)-
cyclohexene to 5 followed by Dibal-H reduction of the
resulting mixture provided a (6:1) separable mixture of
allylic alcohol diastereomers (ꢁ) 26 and (ꢁ) 27. These
were each converted to methylcyclohexene analogues
(ꢁ) 28 and (ꢁ) 29, respectively, by reduction of their
corresponding mesylates. The relative stereochemistry
at the C-5 juncture of (ꢁ) 28 wasunambiguoulsy
determined by single crystal X-ray crystallography to be
anti. Both (ꢁ) 28 and (ꢁ) 29 were then easily resolved
by chiral HPLC to provide their component enantio-
mers( 30–33).34 Consistent with previous results, the (À)
enantiomerswere far more active than their (+) anti-
podes.24,31 The absolute stereochemistry of 30–33 was
Scheme 1. (a) (2-bromoallyl)trimethylsilane, BF3-OEt2, CH2Cl2, À78–
0 ꢀC, 90%; (b) Me4Sn, (PPh3)2PdCl2, HMPA, 85 ꢀC, 60 h, 83%; (c)
(PPh3)2(CO)2Ni, Et3N, MeOH, Á, 16 h, 40%; (d) 1-TBDMSO-1-
methoxyethene,
BF3-OEt2,
CH2Cl2,
À78–0 ꢀC,
90%;
(e)
NHMe(OMe)-HCl, Me3Al, tol, 40 ꢀC, 2 h, 62%; (f) Dibal-H, THF,
À78 ꢀC, 93%; (g) RPPh3Br, n-BuLi, THF/Et2O (5:2), À78 ꢀC to rt, 25–
90%; (h) Ph3PCHCO2CH3, THF, 45 ꢀC, 1 h, 80%; (i) LiAlH4, AlCl3,
Et2O, rt, 1 h, 53%; (j) EtN(i-Pr)2, MsCl, CH2Cl2, À10 ꢀC then
LiEt3BH, 38%.
Scheme 2. (a) BF3–OEt2, CH2Cl2, À25–5 ꢀC, 73%; Dibal-H, THF,
0 ꢀC, 86%; (c) EtN(i-Pr)2, MsCl, CH2Cl2, À10 ꢀC then LiEt3BH, 73%,
74%.
Table 2. In vitro receptor binding and cotransfection assay data for compounds containing a constrained C-5 allyl group. Compounds were pre-
pared according to the method in Figure 2 utilizing the allylsilanes below. Isolated reaction yields and diastereomeric product ratios based on 1H
NMR integration are provided where appropriate
Compd
R1
Allylsilane
Yield
GR bindinga (nM) PR binding (nM)
GRE activation
meanꢁSEM
E-selectin repression
meanꢁSEM
Ratio (syn:anti)
NA
meanꢁSEM
2.4ꢁ0.3*
meanꢁSEM
IC50, (nM) eff. (%dex) IC50, (nM) eff. (%dex)
Pred
—
NA
—
8.0ꢁ1.1*
33ꢁ8.0 *
89ꢁ19*
68ꢁ38*
2.1ꢁ0.2*
13ꢁ6.3*
99ꢁ1*
94ꢁ2*
4a
93%
2.5ꢁ0.46*
78ꢁ21
1800ꢁ480
18
19
20
21
22
23
24
85%
—
—
—
—
—
13ꢁ1
8ꢁ0.0*
34ꢁ1
15ꢁ6
—
220
320ꢁ29
21ꢁ1.2*
12ꢁ4.2
110ꢁ13
230
45ꢁ9
45% (4:1)
91% (1.1:1)
90% (1.5:1)
96% (1:1)
73% (1.4:1)
80% (2:1)
11ꢁ3.7
49ꢁ16
94ꢁ2*
96ꢁ1
5.5ꢁ1.1*
4.1ꢁ0.9*
0.50ꢁ0.30
65ꢁ15*
2400ꢁ157*
2200ꢁ130*
—
13ꢁ3.3*
12ꢁ4.9
—
85ꢁ1
—
—
21ꢁ12
97ꢁ1*
2.0ꢁ0.2*
980ꢁ86*
15ꢁ7.6*
5.0ꢁ1.0*
7.0ꢁ0.60*
a Values for data in this Table are represented in identical fashion to those in Table 1.

S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
1725
Table 3. In vitro receptor binding and cotransfection assay data for pure 5-(1-methylcyclohexen-3-yl) enantiomers 30–33
Compd
R1
GR bindinga (nM) PR binding (nM)
GRE activation meanꢁSEM
E-selectin repression meanꢁSEM
meanꢁSEM
2.4ꢁ0.3*
meanꢁSEM
IC50, (nM)
eff. (%dex)
IC50, (nM)
eff. (%dex)
Pred
NA
—
—
8.0ꢁ1.1*
89ꢁ19*
2.1ꢁ0.2*
99ꢁ1.0*
92ꢁ1.0
(À) 4a
(À)
1.4ꢁ0.62
0.7ꢁ0.1*
1300ꢁ430
710ꢁ155*
14ꢁ5.2
40ꢁ0.6
87ꢁ16
84ꢁ20
6.9
30
31
32
33
(À)-anti
(+)-anti
(À)-syn
(+)-syn
19ꢁ0.4
90ꢁ6
1700ꢁ752
1.5ꢁ0.4*
49ꢁ37
—
430ꢁ96*
—
—
—
—
—
—
2.0ꢁ2.0
93ꢁ4.0
49ꢁ18
3.0ꢁ1.0
16ꢁ4.0
5.8ꢁ0.0
440ꢁ91
a Values for data in this Table are represented in identical fashion to those in Table 1.
assigned by analogy to the parent allyl compound in
which the active (À) enantiomer bearsC-5 (S) configur-
ation.35
To determine the compoundstranscriptional profile in a
more physiologically relevant cellular context, we next
evaluated (À) 30 and (À) 32 in a variety of human
native cell assays evaluating gene products pertinent to
both GC metabolic side effects and antiinflammatory
properties. First, the repression of cytokine induced
inflammatory mediatorsIL-6, PGE-2 and collagenaes
were used as a measure of potential antiinflammatory
activity. While both diastereomers elicited nearly full
repression of all of the inflammatory mediators (Table
5), (À) 32 wasslightly more potent and efficaciousthan
itscounterpart ( À) 30.
Both (À)-anti 30 and (À)-syn 32 maintain high GR
affinity, but show a two to three fold increased affinity
for PR relative to other analoguesin the C-5 alkyl series
(Table 3). Given that the similar 9-hydroxy containing
analogues were reported to possess PR ‘superagonist’
activity,33 (À) 30 and (À) 32 were evaluated in the
MMTV PR-B activation assay to determine their ability
to functionally regulate PR. While (À) 30 wasunable to
activate PR in this assay, it exhibited weak, partial PR
antagonism (241 nM, 61% antagonist of progesterone).
(À) 32, On the other hand, elicited a weak, partial PR
agonism that is significantly lower than that seen with
progesterone or either 9-hydroxy-10-methoxy-5-
(methylcyclohexen-3-yl) stereoisomers reported pre-
viously (see Table 4 for comparison).
The promoter regionsof both aromatase and TAT are
known to contain GREsand their activation hasbeen
associated with certain endocrine side effects of GCs.
Consequently, the ability to up-regulate these proteins
was used as a measure of potential metabolic side
effects.36,37 OC repression is considered a marker of the
destructive osteoporotic effects of GCs and was eval-
uated asa measure of the compoundspotential effect on
bone.37 Both (À) 30 and (À) 32 exhibit significantly
reduced ability compared to pred to up-regulate aro-
matase and TAT (Table 5). In contrast to the GRE
cotransfection assay, (À) 30 induceslower levelsof
GRE mediated transcription than (À) 32 with only 26%
and 49% dex efficacy for aromatase and TAT, respec-
tively. In addition, compound (À) 30 shows a much
lower ability to repress OC compared to (À) 32 or pred.
The combination of potent, efficaciousrepreisosn of
inflammatory mediators, the significantly reduced ability
to negatively effect markersof GC induced endocrine and
bone side effects and lack of PR agonist cross reactivity
suggests (À) 30 to be the most attractive candidate to
date. The efficacy and side effect profiles of (À) 30 in in
vivo modelsof inflammation hasnot yet been studied.
The transcriptional activity profile of (À) 30 in both E-
sel and GRE cotransfection assays is quite similar to
allyl (À) 4a. Compound (À) 32, on the other hand, isa
full E-sel repressor, but shows a drastically reduced
capacity to induce GRE activation [(À) 32=3% dex
versus (À) 30=84% dex].
Table 4. PR binding and activation activity of (À) 30 and (À) 32
compared to their previously reported 9-OH counterparts33
PR bindinga (nM)
PR-B activationb
meanꢁSEM
IC50 (nM) eff. (% prog)
30 (À)-anti 9-H
(À)-anti 9-OH
710ꢁ155*
52ꢁ22
—
210
14
13
190
70
32 (À)-syn
9-H
(À)-syn 9-OH
430ꢁ96*
11.4ꢁ3.5
7.1
260
a Values for data in this Table are represented in identical fashion to
those in Table 1.
b PR-B activation efficaciesare repreesnted asthe percentage of the
maximal response of progesterone.
From these studies, it is especially apparent that small,
subtle structural changes in the GR ligand can produce
significant differences in transcriptional profiles. We

1726
S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
Table 5. Human cell native protein assays for transcriptional activation of aromatase, TAT and transcriptional repression of osteocalcin, IL-6,
PGE2 and collangenase for active enantiomers (À)-anti 30 and (À)-syn 32
Aromatasea
TAT
Osteocalcin
IL-6
PGE2
Collagenase
Compd
EC50
,
eff.
(%dex)
EC50
,
eff.
(%dex)
EC50
,
eff.
(%dex)
EC50
,
eff.
(%dex)
EC50
,
eff.
(%dex)
EC50
,
eff.
(%dex)
(nM)c
(nM)c
(nM)c
(nM)c
(nM)c
(nM)c
pred
45ꢁ4.1 96ꢁ3.5
48ꢁ24
40ꢁ23
95ꢁ5.3
71ꢁ17
49ꢁ3.6
32
n.d.
—
94
n.d.
38ꢁ7.0
3.6ꢁ2.0 98ꢁ2.2 2.0ꢁ1.1 100ꢁ0.60 0.40ꢁ0.08 98 ꢁ0.90
(À) 4a
(À) 30
(À) 32
160ꢁ36
640ꢁ35
59ꢁ20
82ꢁ8.5
30ꢁ17
62ꢁ31
45ꢁ27
83ꢁ1.5
82ꢁ1.3
n.d.
35ꢁ14
n.d.
95ꢁ3.4
95ꢁ3.2
37ꢁ7.8
32ꢁ7.6
3.4ꢁ1.9
77ꢁ6.2
89ꢁ4.0
97ꢁ1.6
26ꢁ6.8 110ꢁ31
43ꢁ2.6 290ꢁ140 67ꢁ17
54ꢁ14 71ꢁ19
90ꢁ4.3 5.4ꢁ3.8
a Values for data in this Table are represented in identical fashion to those in Table 1. n.d. not determined.
have been able to achieve a significant in vitro differ-
entiation of transcriptional repression/activation in
human cell native protein assays that may be pertinent
to the antiinflammatory and side effect profiles of GCs.
Given that the moderate transcriptional differentiation
seen with 4a has been shown to translate to measurable
in vivo improvements in at least two side effect para-
meters (glucose metabolism, bone), one would expect
the more dissociated profile of (À) 30 to yield an
improved in vivo therapeutic window.
7. Coghlan, M. J.; Elmore, S. W.; Kym, P. R.; Kort, M. E.
Current Topics in Med. Chem. 2003, 3, 1617.
8. Barnes, P. J.; Pedersen, S.; Busse, W. W. Am. J. Respir.
Crit. Care Med. 1998, 157, 1.
9. Adcock, I. M. Pulm. Pharmacol. Ther. 2000, 13, 115.
10. Beato, M.; Truss, M.; Chavez, S. Ann. N. Y. Acad. Sci.
1996, 784, 93.
11. Evans, R. M. Science 1988, 240, 889.
12. Baxter, J. D. Pharmacol. Ther. B 1976, 2, 605.
13. Sels, F.; Dequeker, J.; Verwilghen, J.; Mbuyi-Muamba,
J. M. Lupus 1996, 5, 89.
14. Anonymous Nutr. Rev. 1976, 34, 185.
15. Gottlicher, M.; Heck, S.; Herrlich, P. J. Mol. Med. 1998,
76, 480.
16. Gottlicher, M.; Heck, S.; Doucas, V.; Wade, E.; Kull-
mann, M.; Cato, A. C.; Evans, R. M.; Herrlich, P. Ster-
oids 1996, 61, 257.
17. Yang-Yen, H. F.; Chambard, J. C.; Sun, Y. L.; Smeal, T.;
Schmidt, T. J.; Drouin, J.; Karin, M. Cell 1990, 62, 1205.
18. Jonat, C.; Rahmsdorf, H. J.; Park, K. K.; Cato, A. C.;
Gebel, S.; Ponta, H.; Herrlich, P. Cell 1990, 62, 1189.
19. Van der Burg, B.; Liden, J.; Okret, S.; Delaunay, F.;
Wissink, S.; Van der Saag, P. T.; Gustafsson, J. A. Trends
Endocrinol. Metab. 1997, 8, 152.
Although the exact mechanism by which GR mediates
transcription is not fully elucidated, one possible expla-
nation for these differences are the varied abilities of the
GRC to properly associate with accessory proteins
necessary for normal function of the transcriptional
machinery.38 Ligand-dependent GC regulated tran-
scription appears to require the association of the GRC
with specific co-activator or co-repressor proteins.39
These accessory proteins are recruited through a pro-
tein-protein surface interaction between an amphipathic
a-helix on the coactivator/corepressor to a hydrophobic
groove on the surface of the activated GRC. This
hydrophobic binding groove is intimately associated
with the ligand binding domain (LBD). The X-ray
crystal structure of the ternary complex comprised of
the GR LBD in itsagonits form bound to dex and a
coactivator motif from TIF-2 elegantly depictsthisclose
association.40,41 Accessory protein recruitment should
therefore be highly dependent upon the ligand and the
intricate conformational changesit impartson GR
upon binding. The ability of GR when bound to (À) 30
to differentially recruit known coactivatorsand co-
repressors of GR has not yet been investigated.
20. Brattsand, R.; Linden, M. Aliment. Pharmacol. Ther.
1996, 10, 81.
21. Cato, A. C.; Wade, E. BioEssays 1996, 18, 371.
22. Heck, S. Forschungszent. Karlsruhe 1998, 1.
23. Edwards, J. P.; West, S. J.; Marschke, K. B.; Mais, D. E.;
Gottardis, M.; Jones, T. K. J. Med. Chem. 1998, 41,
303.
24. Edwards, J. P.; Zhi, L.; Pooley, C. L. F.; Tegley, C. M.;
West, S. J.; Wang, M. W.; Gottardis, M. M.; Pathirana,
C.; Schader, W. T.; Jones, T. K. J. Med. Chem. 1998, 41,
2779.
25. Zhi, L.; Tegley, C. M.; Edwards, J. P.; West, S. J.;
Marschke, K. B.; Gottardis, M. M.; Mais, D. E.; Jones,
T. K. Bioorg. Med. Chem. Lett. 1998, 8, 3365.
26. Zhi, L.; Tegley, C. M.; Pio, B.; Edwards, J. P.; Jones,
T. K.; Marschke, K. B.; Mais, D. E.; Risek, B.; Schrader,
W. T. Bioorg. Med. Chem. Lett. 2003, 13, 2071.
27. Zhi, L.; Ringgeberg, J. D.; Edwards, J. P.; Tegley, C. M.;
West, S. J.; Pio, B.; Motamedi, M.; Jones, T. K.;
Marschke, K. B.; Mais, D. E.; Schrader, W. T. Bioorg.
Med. Chem. Lett. 2003, 13, 2075.
28. Zhi, L.; Tegley, C. M.; Pio, B.; Edwards, J. P.; Motamedi,
M.; Jones, T. K.; Marschke, K. B.; Mais, D. E.; Risek, B.;
Schrader, W. T. J. Med. Chem. 2003, 46, 4104.
29. Coghlan, M. J.; Kym, P. R.; Elmore, S. W.; Wang, A. X.;
Luly, J. R.; Wilcox, D.; Stashko, M.; Lin, C. W.; Miner,
J.; Tyree, C.; Nakane, M.; Jacobson, P.; Lane, B. C. J.
Med. Chem. 2001, 44, 2879.
References and notes
1. Ali, S. L. Anal. Profiles Drug Subst. Excipients 1992, 21,
415.
2. Cohen, E. M. Anal. Profiles Drug Subst. 1973, 2, 163.
3. Schimmer, B. P., Parker, K. L. In Goodman and Gilman’s
The Pharmacological Basis of Therapeutics, 9th ed. Hard-
man, J. G., Limbird, L. E., Molinoff, P. B., Ruddon,
R. W., Gilman, A. G. Eds.; McGraw-Hill: New York,
1996, pp 1459-1485.
4. Gennari, C. Br. J. Rheumatol. 1993, 32, 11.
5. Andrews, R. C.; Walker, B. R. Clin. Sci. 1999, 513.
6. Remesar, X.; Fernandez-Lopez, J. A.; Alemany, M. Med.
Res. Rev. 1993, 13, 623.
30. Kym, P. R.; Kort, M. E.; Coghlan, M. J.; Moore, J. L.;
Tang, R.; Ratajczyk, J. D.; Larson, D. P.; Elmore, S. W.;

S. W. Elmore et al. / Bioorg. Med. Chem. Lett. 14 (2004) 1721–1727
1727
Pratt, J. K.; Stashko, M. A.; Falls, D.; Lin, C. W.;
Nakane, M.; Miller, L.; Tyree, C. M.; Miner, J. N.;
Jacobson, P. B.; Wilcox, D. M.; Nguyen, P.; Lane, B. C.
J. Med. Chem. 2003, 46, 1016.
(ꢁ) 29: 250ꢂ21.1 mm Regis(R,R) WELK- O-1 column,
isocratic 2% EtOH in Hexanes, 12.0 mL/min.
35. Ku, Y. Y.; Grieme, T.; Raje, P. S.; King, S. A.; Morton,
H. E. J. Am. Chem. Soc. 2002, 124, 4282.
31. Elmore, S. W.; Coghlan, M. J.; Anderson, D. D.; Pratt,
J. K.; Green, B. E.; Wang, A. W.; Stashko, M. A.; Lin,
C. W.; Tyree, C. M.; Miner, J. N.; Jacobson, P. B.; Wil-
cox, D. M.; Lane, B. C. J. Med. Chem. 2001, 44, 4481.
32. Coghlan, M. J.; Jacobson, P. B.; Lane, B.; Nakane, M.;
Lin, C. W.; Elmore, S. W.; Kym, P. R.; Luly, J. R.; Car-
ter, G. W.; Turner, R.; Tyree, C. M.; Hu, J.; Elgort, M.;
Rosen, J.; Miner, J. N. Mol. Endocrol. 2003, 17, 860.
33. Lin, C. W.; Makane, M.; Stashko, M.; Falls, D.; Kuk, J.;
Miller, L.; Huang, R.; Tyree, C.; Miner, J. N.; Rosen, J.;
Kym, P. R.; Coghlan, M. J.; Carter, G.; Lane, B. C. Mol.
Pharmacol. 2002, 62, 297.
36. Guido, E. C.; Delorme, E. O.; Clemm, D. L.; Stein, R. B.;
Rosen, J.; Miner, J. N. Mol. Endocrol. 1996, 10, 1178.
37. Crilly, R.; Cawood, M.; Marshall, D. H.; Nordin,
B. E. J. R. Soc. Med. 1978, 71, 733.
38. Meyer, T.; Gustafsson, J. A.; Carlstedt-Duke, J. DNA
Cell Biol. 1997, 16, 919.
39. Miner, J. Biochem. Pharmacol. 2002, 64, 355.
40. Jenkins, B. D.; Pullen, C. B.; Darimont, B. D. Trends
Endocrinol. Metab. 2001, 12, 122.
41. Bledsoe, R. K.; Montana, V. G.; Stanley, T. B.; Delves,
C. J.; Apolito, C. J.; McKee, D. D.; Consler, T. G.; Parks,
D. J.; Stewart, E. L.; Willson, T. M.; Lambert, M. H.;
Moore, J. T.; Pearce, K. H.; Xu, H. E. Cell 2002, 110, 93.
34. The chiral HPLC conditions used to separate (ꢁ) 28 and