Steroidal C-21 heteroaryl thioethers. Part 3: Pregn-4-eno-[3,2-c]pyrazole fused A ring modified steroids as selective glucocorticoid receptor modulators (dissociated steroids)

Article abstract of DOI:10.1016/j.bmcl.2012.03.015

The introduction of A ring pyrazole modification to the hydrocortisone C-21 heteroaryl thioethers generated compounds with excellent transrepression potency (IL-8 inhibition) compared to their hydrocortisone analogs. However, the transcriptional transacti

Full text of DOI:10.1016/j.bmcl.2012.03.015

Bioorganic & Medicinal Chemistry Letters 22 (2012) 3291–3295
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Steroidal C-21 heteroaryl thioethers. Part 3: Pregn-4-eno-[3,2-c]pyrazole
fused A ring modified steroids as selective glucocorticoid receptor modulators
(dissociated steroids)
⇑
Purakkattle Biju , Hongwu Wang, John Anthes, Kevin McCormick, Robert Aslanian, Michael Berlin,
Rema Bitar, Yeon-Hee Lim, Yoon Joo Lee, Daniel Prelusky, Robbie Mcleod, Yanlin Jia, Xiomora Fernandez,
Gissela Lieber, Johanna Jimenez, Steve Eckel, Aileen House, Richard Chapman, Jonathan Phillips
Merck Research Laboratories, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The introduction of A ring pyrazole modification to the hydrocortisone C-21 heteroaryl thioethers generated
compounds with excellent transrepression potency (IL-8 inhibition) compared to their hydrocortisone analogs.
However, the transcriptional transactivation activity of these compounds were considerably higher than the
corresponding hydrocortisone analogs. Among all the compounds evaluated, a quinoxaline thioether modifi-
cation demonstrated the best overall in vitro separation.
Received 22 January 2012
Revised 1 March 2012
Accepted 5 March 2012
Available online 11 March 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Dedicated to Professor Gilbert Stork on the
occasion of his 90th birthday.
Keywords:
Selective glucocorticoid receptor
modulators
Dissociated steroids
Prednisolone
Transactivation
Transrepression
Pyrazole
We recently reported1 compounds of structural types 1, 3, and 4
as both inhaled and oral dissociated steroids for the treatment of
various inflammatory conditions. The key structural feature critical
for the dissociated profile is the C-21 heteroaryl thioether modifi-
cation. Interestingly the majority of these compounds demon-
strated significantly reduced transactivational activity while
keeping potent anti-inflammatory transrepressional activity in
both in vitro and in vivo assays. Although we were not successful
in getting a crystal structure of any of these compounds within
the glucocorticoid receptor, molecular modeling was carried out
to explore the binding mode of these steroids to the glucocorticoid
receptor (GR) and to further understand the mechanisms of the
dissociation profile. The protein structure used in this study was
retrieved from the PDB (PDB ID: 1P93, GR-LBD in complex with
dexamethasone2). Due to the large size of the C-21 group, these
compounds could not fit into the steroid binding site of the crystal
structure. Hence we thought it would be reasonable to use a
truncated version of the crystal structure (with residues after
R
R
S
S
O
O
O
OH
R1
O
HO
HO
O
O
O
R1 = H, SR= OH; 2
R1 = H, 3; R1 = Me 4
1
R
1 = H; R = N-methyl-2-
R
benzimidazole; 5 OH
X
O
OH
O
OH
HO
HO
N
N
N
N
R1
6; X = O and S
7
Figure 1. Dissociated steroids: steroidal[3,2-c]pyrazole fused steroid
A ring
modification.
T739 removed) to model these compounds. A previously reported
⇑
Corresponding author.
E-mail address: purakkattle.biju@merck.com (P. Biju).
compound1b 5 (Fig. 1) was docked to this truncated GR crystal
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.03.015

3292
P. Biju et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3291–3295
Table 1
Transrepression and transactivation screening of C-21 thioether derivatives of various
steroidal[3,2-c]pyrazoles
N
S
S
N564
O
OH
HO
N
Q570
R611
N
R
Q642
Compds
R
IL-8 inh. IC50
,
Rat TAT. IC50
,
h-TAT IC50,
nMa efficacy (%) nMa efficacy (%) nMa efficacy (%)
11
0.4 (99)
1.7 (97)
3.2 (107)
24.1 (83)
40.9 (135)
101 (110)
13a
F
14
15
7.5 (95)
(20)
57.3 (67)
NT
165 (45)
NT
Figure 2. The binding model of compound 5. GR protein is shown as cartoon with
helix 12 colored blue, the loop between helix 11 and helix 12 colored orange and
the rest of the protein pale green. The carbon atoms of compound 5 are colored
magenta. The C21 group fits under this loop, disturbing its conformation from the
agonist bound state. This may induce conformational changes to the helix 12 region
and leads to the dissociation profile possibly due to the lack of dimerization of the
GR–GC monomer complex, which is a key event required for the transcriptional
transactivation.
Me
16
107 (66)
NT
NT
OMe
17
18
19
179 (66)
86.4 (78)
7.1 (98)
NT
NT
N
structure. Docking of compound 5 was carried out using glide XP
methodology3 in Schrödinger modeling package. Chain A of 1P93,
prepared as described above, was used to generate the energy grids
for docking.
In the docking model of compound 5 (Fig. 2), the polar interac-
tions with GR resembled to that of dexamethasone, for example,
the C3-carbonyl group with Gln570 and Arg611, the 11b-hydroxyl
NT
NT
O
54.7 (63)
400 (53)
20
21
(27)
(8)
NT
NT
NT
NT
CN
Me
Me
group with Asn564, and the 17a-hydroxyl group with Gln642. The
steroid core maintained tight packing with the hydrophobic resi-
dues of the binding pocket. These interactions contributed to the
high binding affinity of compound 5 (hGR IC50 = 7.3 nM). When
compound 5 overlaid on the whole crystal structure, the C-21
Me
Note: The top concentration for the Emax when there are no IC50 available is 1
lM.
a
Values are means of two experiments.
group oriented directly under the loop between the helix 11 and
helix 12 occupying the space of Ile747. The binding interaction of
compound 5 in this region of GR could disrupt the conformation
of this loop, which may subsequently induce conformational
changes to the downstream region of helix 12. Induced-fit docking
was employed to further elucidate protein conformation around
the loop region and we observed the proposed conformational
changes in the GR structure. Based on these docking studies, we
hypothesize that the helix-12 conformational changes, could prob-
ably prevent the dimerization of the glucocorticoid receptor–glu-
cocorticoid (steroid) monomer complex, which is a key event
required for the transcriptional transactivation.4
As an ongoing efforts to further optimize the anti-inflammatory
potency and dissociation, we were interested in introducing a pyr-
azole modification in the A ring of compound 1 like deacylcortivaz-
ole5 (DAC) 7, a well studied glucocorticoid receptor agonist with
interesting pharmacological properties. Historically, the androst-
4-eno and androstano-[3,2-c]pyrazole modification was first
introduced by Clinton et al.6 in 1959 and later by de Ruggieri
et al.7 Later in 1963, Merck researchers8 Max Tishler and Ralph
Hirschman extended this modification to pregn-4-eno-[3,2-c]pyra-
zole and developed a new series of potent anti-inflammatory
hybrid steroids. Herein, we report the structure activity studies
of a variety of N-substituted steroidal-pyrazoles with various C-
21 heteroaryl ether and thioether derivatives towards the optimi-
zation of both dissociation and transrepression potency.
OH
O
OH
O
O
O
HO
O
O
O
a
b
O
O
8
9
OH
OH
O
O
O
O
HO
O
O
OH
O
O
c, d
N
N
11
10
e
N
F
OMs
OH
S
S
O
O
OH
HO
HO
f
N
N
N
N
13a
12
F
F
Scheme 1. Reagents and conditions: (a) formaldehyde, CHCl3, concd HCl (b)
sodium methoxide, THF, 60 °C, 45 min, then HCOOEt, 24 h (c) p-fluorophenylhydr-
azine, NaOAc, CH3COOH, rt, 60% (d) 50% aq. HCOOH, 70 °C, 65% (e) CH3SO2Cl,
Hunig’s base, CH2Cl2, 0 °C–rt, 90% (f) 2-mercaptobenzothiazole, K2CO3, acetone,
reflux, 92%.

P. Biju et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3291–3295
3293
Table 2
Table 3
Transrepression and transactivation screening of C-21 ether and thioether derivatives
in 4-fluorophenyl pyrazole series
Transrepression and transactivation screening of C-21 tether derivatives in hydro-
cortisone series
R
R
S
X
O
O
OH
HO
OH
HO
N
N
O
Compds
R
IL-8 inh. IC50, nMa
efficacy (%)
hGRE. IC50, nMa
efficacy (%)
h-GR
IC50, nM
F
N
Compds
R
X
O
IL-8 inh. IC50
,
r. TAT. IC50
,
h-TAT IC50
,
31
32
36 (97)
(30)
4.8
8.9
S
N
nMa efficacy (%) nMa efficacy
(%)
nMa efficacy
(%)
56.1 (84)
(8)
N
11
R=H
0.4 (99)
3.2 (107)
40.9 (135)
Me
N
S
13a
13b
S
1.7 (97)
2.1 (96)
24.1 (83)
84 (71)
101 (110)
(31)
N
33
34
35
36
37
26.7 (98)
21.9 (95)
89.3 (83)
22.1 (83)
61.3 (87)
(35)
(57)
(0)
3.9
4.3
6.6
NT
NT
O
N
O
Me
Me
N
N
22a
22b
S
1.1 (99)
2.9 (97)
174 (89)
78 (74)
88.7 (115)
(34)
S
Me
O
N
N
23a
23b
S
0.72 (99)
1.06 (99)
52.5 (98)
58 (106)
30.5 (68)
29.2 (67)
(1)
O
N
N
24a
24b
S
5.3 (97)
1.8 (97)
24.3 (115)
44.7 (59)
18.3 (116)
(38)
N
N
N
(37)
O
N
O
N
N
38
39
40
59.1 (75)
29.2 (80)
23.8 (65)
(16)
4.7
25a
25b
S
1.99 (98)
1.89 (97)
(25)
(41)
(2)
S
N
Ph
O
98.7 (52)
58.5 (54)
(14)
NT
N
O
26
27
28
29
S
S
S
S
1.6 (99)
0.32 (99)
0.5 (99)
0.6 (99)
65.4 (54)
13.3 (80)
4.5 (111)
14.3 (66)
37.3 (63)
38.6 (63)
4.7 (116)
37.7 (72)
N
14.9
N
Me
N
N
N
O
41
42
127 (59)
27.2 (90)
(18)
(28)
NT
NT
N
Cl
Cl
N
N
Note: The top concentration for the Emax when there are no IC50 available is 1
lM.
N
O
a
Values are means of two experiments.
NH
30
S
3.0 (99)
86.5 (59)
93.5 (51)
studied in two different cell based assays. Transrepression was
measured9 in a TNF stimulated IL-8 cytokine synthesis assay in
human bronchial epithelial cells (H292 cells). Functional transacti-
vation was measured by induction of tyrosineaminotransferase
(TAT), an enzyme with well characterized endogenous glucocorti-
coid response elements that catalyzes the degradation of tyrosine
to p-hydroxy phenyl pyruvate. Human HepG2 liver cells and rat
H4II-E cells were used for the TAT functional assay. A recombinant
hGRE luciferase assay was also used for measuring the transcrip-
tional transactivation.
The parent hydroxy compound 11 exhibited a classical steroid
like profile with a very potent IL-8 inhibition and a robust tran-
scriptional transactivation in both rat and human TAT induction
assays. A limited SAR was generated (Table 1) with N-substituted
pyrazoles ranging from aromatic and aliphatic hydrophobic groups
(e.g., phenyl, 4-fluorophenyl, cyclohexyl and t-butyl) to aromatic
and aliphatic polar groups (e.g., pyridyl, methoxy-phenyl, tetrahy-
dropyranyl, and propionitrile). Pyrazoles with 4-fluorophenyl 13a,
phenyl 14 and cyclohexyl 19 demonstrated promising anti-inflam-
matory profile, with a relatively weaker TAT induction compared
to the parent steroid 11. The pyrazole substitution with polar
groups such as methoxy-phenyl, pyridyl, tetrahydropyran, and
Note: The top concentration for the Emax when there are no IC50 available is 1
l
M.
a
Values are means of two experiments.
A
general procedure for the synthesis of fused pyrazole
compounds is outlined in Scheme 1. The commercially available
hydrocortisone 8, was converted to the ketal 9 by treating with
formaldehyde in chloroform. The formylation of the A ring was
carried out with sodium methoxide and ethyl formate in anhy-
drous THF to afford the keto-aldehyde 10. Subsequent treatment
of the keto-aldehyde 10 with 4-fluorophenyl hydrazine and so-
dium acetate in glacial acetic acid generated the corresponding
pyrazole, which was deprotected to the pyrazole-steroid 11 by
the treatment with 50% aqueous formic acid at 70 °C in moderate
yield. The alcohol 11 was converted to the corresponding mesylate
12 and subsequent treatment with 2-mercaptobenzothiazole and
potassium carbonate in acetone generated the target compound
13 in excellent yield.
By following the sequence of reactions outlined in Scheme 1,
the keto-aldehyde 10 was treated with a number of hydrazines
and generated the corresponding 2-mercaptobenzothiazole com-
pounds (Table 1). The dissociated profile of these compounds were

3294
P. Biju et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3291–3295
Table 4
Transrepression and transactivation screening of C-21 ether and thioether derivatives in phenyl and cyclohexyl pyrazole series
R
X
O
OH
HO
N
N
R1
Compds
R
X
R1
IL-8 inh.
r. TAT.
h-TAT
hGRE luciferase
5.57 (96)
IC50, nMa efficacy (%)
IC50, nMa efficacy (%)
IC50, nMa efficacy (%)
43a
43b
R=H
R=H
O
O
Ph
0.63 (100)
2.37 (99)
3.23 (96)
NT
143 (70)
NT
Cyclohexyl
N
S
14
19
S
Ph
7.5 (95)
7.1 (98)
57.3 (67)
54.7 (63)
165 (45)
400 (53)
36.9 (89)
130 (78)
Cyclohexyl
N
N
44a
S
Ph
4.7 (97)
78.0 (75)
66.7 (49)
19.1 (64)
Me
44b
45
Cyclohexyl
Cyclohexyl
6.6 (96)
1000 (44)
31.6 (65)
(19)
108 (75)
30.1 (72)
N
S
S
S
3.12 (99)
82.1 (66)
O
N
N
N
46
Ph
0.89 (99)
33.8 (81)
225 (82)
37.7 (64)
596 (64)
6.9 (83)
O
47a
47b
Ph
1.5 (100)
4.4 (96)
8.4 (102)
43.3 (87)
Cyclohexyl
N
48
S
S
Cyclohexyl
7.3 (98)
(36)
1000 (46)
60.1 (43)
137 (76)
N
49a
49b
Ph
14.8 (95)
44.7 (92)
82.3 (95)
101 (69)
262 (71)
N
Cyclohexyl
N
N
50a
S
Ph
5.36 (97)
113 (67)
226 (41)
28.8 (68)
50b
50c
S
O
Cyclohexyl
Ph
12.4 (97)
3.7 (95)
(40)
(33)
(27)
(43)
158.0 (84)
31.0 (56)
Me
Me
N
51
52
S
S
Cyclohexyl
Ph
2.8 (99)
135 (96)
258 (62)
168 (59)
278 (50)
31.8 (88)
28.6 (86)
S
N
3.43 (97)
N
Note: The top concentration for the Emax when there are no IC50 available is 1
lM.
a
Values are means of two experiments
propionitrile produced compounds with significantly weaker
inhibitory activity towards IL-8.
activation. Again a noticeable difference in the dissociation profile
was observed for the quinazoline compound 25b like the corre-
sponding thioether analog 25a. Table 4 lists some of the C-21
modifications in the phenyl and cyclohexyl pyrazole series. These
compounds showed a similar overall profile to fluorophenyl series.
Interestingly, the best compounds identified in the phenyl series
were the thio and oxo quinoxaline compounds 50a and 50c. The
benzimidazole compound 44b and the quinoxaline compound 50b
were identified as the best compounds in the cyclohexyl series.
In order to evaluate some of these compounds anti-inflamma-
tory activity in vivo, the blood levels in rats following oral admin-
istration were determined (Table 5). Compounds 11, 27, 43a, 46
showed good blood levels, with a comparable pharmacokinetic
profile to that of prednisolone. We were interested in the quinox-
aline compound 25a and 50a due to the negligible blood levels as
a possible inhaled steroid. However, these two compounds did not
demonstrate significant inhibition of total cells when dosed intrat-
racheally in Brown-Norway rat allergic lung inflammation model.
The compounds with better blood levels were evaluated in a three
day Brown-Norway rat oral allergic lung inflammation model. The
Owing to the interesting anti-inflammatory profile of 4-fluoro-
phenyl, phenyl and cyclohexyl substituted pyrazoles, we decided
to investigate the C-21 oxo and thio SAR in detail (Table 2) to opti-
mize the pharmacological properties. As evident from the Table 2,
the compounds with thioether modification showed very potent
IL-8 inhibition. Compounds with azabenzoxazole 27, pyridyl 28,
and pyrimidinyl 29 demonstrated excellent potency towards IL-8
inhibition, a several fold improvement compared to their hydro-
cortisone analogs (Table 3). However, their transactivation potency
and efficacy were significantly higher than the corresponding
hydrocortisone analogs (Table 3). Generally most of the hydrocor-
tisone analogs demonstrated potent binding affinity to GR and a
highly desirable dissociation profile with minimal transactivation-
al activity. The quinazoline compound 25a notably showed a very
potent IL-8 inhibition, well separated from TAT activation in both
rat and human cell types. At this stage we prepared some of the
C-21 oxygen analogs. Interestingly most of them displayed a
parallel SAR with that of the thioethers with a lesser hTAT

P. Biju et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3291–3295
3295
Table 5
Rat PK and in vivo transrepression and transactivation screening
Acknowledgments
Compds BNR lung inflammation
(3 d, 10 mg/kg po) %
BNR thymolysis (3 d,
10 mg/kg po) %
thymus wt decrease
Rat PK AUC
(hr ng/mL)/
Cmax (ng/mL)
We thank Dr. John C. Hunter, Dr. Malcom Maccoss, Dr. John
Piwinski and Dr. William Greenlee from the Merck Research
Laboratory for their support and guidance.
total cell inhibition
2
11
25a
27
32
36
43a
46
56
NT
9*
48
39
39
NT
62
28*
53
NT
NT
70
5
1252/682
1044/240
0/0
1631/348
150/115
274/255
2484/628
2034/467
64/21
References and notes
1. (a) Biju, P.; McCormick, K.; Aslanian, R.; Berlin, R.; Solomon, D.; Chapman, R.;
McLeod, R.; Prelusky, D.; Eckel, S.; Kelly, G.; Natiello, M.; House, A.; Fernandez,
X.; Bitar, R.; Phillips, J.; Anthes, J. Bioorg. Med. Chem. Lett. 2011, 21, 6343; (b)
Biju, P.; McCormick, K.; Aslanian, R.; Berlin, M.; Solomon, D.; Wang, H.; Lee, Y.
J.; Bitar, R.; Prelusky, D.; McLeod, R.; Jia, Y.; Fernandez, X.; Eckel, S.; House, A.;
Lieber, G.; Jimenez, J.; Kelly, G.; Chapman, R.; Phillips, J.; Anthes, J. Bioorg. Med.
Chem. Lett. 2012, 22, 1086.
2. Kauppi, B.; Jakob, C.; Farnegardh, M.; Yang, J.; Ahola, H.; Alarcon, M.; Calles, K.;
Engstrom, O.; Harlan, J.; Muchmore, S.; Ramqvist, A. K.; Thorell, S.; Ohman, L.;
Greer, J.; Gustafsson, J. A.; Carlstedt-Duke, J.; Carlquist, M. J. Biol. Chem. 2003,
278, 22748.
3. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739.
4. (a) Barnes, P. J. Clin. Sci. (London, England: 1979) 1998, 94, 557; (b) Rhen, T.;
Cidlowski, J. A. N. Engl. J. Med. 2005, 353, 1711.
5. (a) Harmon, J. M.; Schmidt, T. J.; Thompson, E. B. J. Steroid Biochem. 1981, 14,
273; (b) Madauss, K. P.; Bledsoe, R. K.; Mclay, I.; Stewart, E. L.; Uings, I. J.;
Weingarten, G.; Williams, S. P. Bioorg. Med. Chem. Lett. 2008, 18, 6097.
6. (a) Clinton, R. O.; Manson, A. J.; Stonner, F. W.; Neumann, H. C.; Christiansen, R.
G.; Clarke, R. L.; Ackerman, J. H.; Page, D. F.; Dean, J. W.; Dickinson, W. B.;
Carabateas, Clarissa J. Am. Chem. Soc. 1961, 83, 1478; (b) Clinton, R. O.; Manson,
A. J.; Stonner, F. W.; Beyler, A. L.; Potts, G. O.; Arnold, Aaron J. Am. Chem. Soc.
1959, 81, 1513.
7. DeRuggieri, P.; Gandolfi, C.; Chiaramonti, D. Gazz. Chim. Ital. 1962, 92, 768.
8. (a) Tishler, M.; Steinberg, N. G.; Hirschmann, R. F. 1962, 24 pp US 3067193.; (b)
Hirschmann, R.; Buchschacher, P.; Steinberg, N. G.; Fried, J. H.; Ellis, R.; Kent, G.
J.; Tishler, M. J. Am. Chem. Soc. 1964, 86, 1520.
0
NT
67
NT
50a
*
Meaning compound is tested in intratracheal assay (it) at 2 mg/kg and not po.
anti-inflammatory activity was measured as the percent inhibition
of the total number of cells, while transactivation was measured
by the percentage decrease of the thymus10 weight after dosing
the compound at 10 mg/kg orally for three days. Although com-
pounds 27 and 46 showed significant inhibition of total cells, they
showed significant thymolysis compared to the hydrocortisone
analogs 32 and 36 and the previously reported1b prednisolone
compounds.
In summary, the A ring pyrazole modification of hydrocortisone
combined with various C-21 thio and oxo heteroaryl modification
generated compounds with potent anti-inflammatory activity and
moderate dissociation profile in in vitro functional assays. Although
the key compounds failed to differentiate efficacy and thymolysis,
we have continued interest in these types of compounds. The dis-
covery of a selective glucocorticoid receptor modulator may offer
potential therapeutic benefits over the existing oral anti-inflamma-
tory steroids for treatment of various inflammatory conditions.
9. For biological assay details, refer WO 2009085879.
10. Reichardt, H. M.; Kaestner, K. H.; Tuckermann, J.; Kretz, O.; Wessely, O.; Bock,
R.; Gass, P.; Schmid, W.; Herrlich, P.; Angel, P.; Schutz, G. Cell 1998, 93, 531.