5450
J. L. Gilmore et al. / Bioorg. Med. Chem. Lett. 23 (2013) 5448–5451
Table 2
also shows increased transactivation as measured in the NP-1 as-
In vitro NHR selectivity profiles of selected indazoles
say with an EC50 of 126 nM (86% of dex). The most promising com-
pound in this series is 3l which has moderate functional activity in
AP-1 (57 nM, 74% efficiency) and E-selectin (58 nM; 65% efficiency)
and poor activity in the NP-1 assay (429 nM; 12% efficiency).
For the sulfonamide derivatives (4a–h), we examined a set of
small alkyl groups ranging from methyl to propyl in length since
larger alkyl substituents showed significantly lower activity (data
not shown). These compounds all had good GR binding activity.
The methyl (4a) and ethyl (4b) sulfonamide derivatives had very
similar activities in both the transrepression and transactivation
functional assays with activities of 20.5 nM (86% efficiency) and
28.8 nM (84% efficiency) in the AP-1 assay, 19.1 nM (97% effi-
ciency) and 15.6 nM (87% efficiency) in the E-selectin assay, and
148 nM (32% efficiency) and 214 nM (19% efficiency) in the NP-1
assay, respectively. The increased size of the propyl compound 4c
led to the some loss of the transrepression functional activities,
however the cyclopropyl derivative restored both the AP-1 and
E-selectin activity to that of the methyl and ethyl analogs. The tri-
fluoroethyl sulfonamide 4e gave a compound with similar AP-1
activity (20.4 nM; 74% efficiency and slightly better E-selectin
activity (7.6 nM; 86% efficiency) while retaining poor activity in
the NP-1 assay (202 nM; 24% efficiency). When the R1/R2 groups
in the sulfonamide series are changed from dimethyl to mono-
methyl (4f–h), activity in the AP-1 and E-selectin assays increased
for all the compounds tested analogous to the regular amides de-
scribed in the previous Letter. However, these compounds also
showed a significant increase in NP-1 activity ranging from 12 to
26 nM with >100% efficiencies.
The selectivities against other nuclear hormone receptors for
one representative compound from each series are shown in
Table 2.12,13 The compounds from the amide and urea series show
excellent selectivity’s against progesterone receptor (PR), androgen
receptor (AR) and mineralcorticoid receptor (MR) whereas com-
pounds from the sulfonamide series show excellent selectivies
over AR and MR but only modest selectivity versus PR.
In an effort to better understand the binding and functional
activity of this series of indazoles, we used molecular modeling
to dock selected compounds into the GR X-ray crystal structure
with dexamethasone.14 Figure 2 shows the docking and minimiza-
tion studies done with compound 3c (amide series), compound 3l
(urea series), and compound 4e (sulfonamide series). All three
compounds show a similar binding poses in the GR active site.
Their respective carbonyl or sulfonyl groups form a hydrogen bond
to Gln 642 on helix 11 and the NH of these moieties is hydrogen
bonded to Asn 564 on helix 3. Additionally, the N-1 nitrogen of
the indazole forms a hydrogen bond to Gln 570 in all three poses.
The 4-fluorophenyl moiety on the indazole is positioned in a
hydrophobic pocket that appears to mimic the same group in the
2.5A co-crystal structure of fluorocortivazol with the GR LBD from
Suino-Powell et al.15
Compds GR binding Ki AR binding Ki PR binding Ki MR agonism
(nM)
(nM)
(nM)
IC50 (nM)
3b
3j
4e
0.81
5.14
0.64
>5000
>5000
>5000
>5000
>5000
406
>5000
>5000
>5000
the putative hydrogen bond to Asn 564 observed in the GR ligand
binding domain crystal structure (vide infra). As shown in Table 1,
the acetamide 3a (Z@Me) shows good GR binding but very little
activity in any of the functional assays. This led us to investigate
other groups that could lower the pKa of the NH even further. Com-
pound 3b, which has a CF3 substitution has good GR binding and
modest functional activity in the AP-1 and E-selectin assays. Plac-
ing a methylene spacer to give a trifluoroethyl derivative (3c) led to
a compound with good activity in both the AP-1 (12.4 nM, 91% effi-
ciency) and E-selectin assays (6.85 nM, 90% efficiency). This trifluo-
roethyl derivative also shows modest activity in the NP-1 assay
(129 nM, 65% efficiency). Adding another methylene spacer leads
to a loss of activity in the AP-1 and E-selectin assays (data not
shown). These data suggest that the GR binding and functional
activity in the transrepression assays is independent of the pKa
of the amide nitrogen. Other larger groups such as pivaloyl (3d)
and benzyl (3f) also lead to diminished activity especially in the
case of benzyl. When the CF3 group of the trifluoroethyl moiety
is replaced with a t-butyl group (3e), the resulting compound has
modest functional activity in the AP-1 (251 nM, 62% efficiency)
and E-selectin (138 nM, 46% efficiency) assays and no activity in
the NP-1 assay (>10,000 nM). The trifluoroethyl series was further
explored by exchanging the Y group from phenyl to isobutyl (3g)
which resulted in diminished functional activity. Next, the effect
of monomethyl substitution at the R1/R2 positions was exam-
ined—a change in the forward amide series9 that conferred signif-
icant increase in potency. The two more active stereoisomers from
the forward amide series were prepared in the reversed amide ser-
ies (3h and 3i) and are shown in Table 1. The monomethyl substi-
tution led to a modest boost in potency in the AP-1 assay for one
stereoisomer while the other was found to be less potent. Unfortu-
nately, the more active stereoisomer 3i also exhibited a significant
increase in the NP-1 activity (20.7 nM, 102% efficiency).
For the urea analogs (3j–m), two sets of compounds are dis-
closed in Table 1. One set is a trifluoroethyl urea derivative with
phenyl and isobutyl groups at the Y position and the other is a t-
butyl urea derivative with the phenyl and isobutyl groups at the
Y position. The trifluoroethyl urea series has moderate activity in
the AP-1 and ELAM assays whereas the t-butylurea series has much
better activity in these functional assays. The most active com-
pound in this series is 3m which has the isobutyl group at the Y po-
sition. This compound has an AP-1 EC50 of 25 nM with 100%
efficiency and an ELAM EC50 of 11.3 nM with 94% efficiency but
Figure 2. Molecular models of compound 3c (amide series), compound 3l (urea series), and compound 4e (sulfonamide series) docked into the GR LBD.
Gilmore, John L.
