Novel synthesis of the hexahydroimidazo[1,5b]isoquinoline scaffold: Application to the synthesis of glucocorticoid receptor modulators

Article abstract of DOI:10.1021/jm901551w

The first stereoselective synthesis of the hexahydroimidazo[1,5b] isoquinoline (HHII) scaffold as a surrogate for the steroidal A-B ring system is described. The structure-activity relationships of the analogs derived from this scaffold show that the basic imidazole moiety is tolerated by the glucocorticoid receptor (GR) in terms of binding affinity, although the partial agonist activity in the transrepressive assays depends on the substitution pattern on the B-ring.More importantly, most compounds in the HHII series bearing a tertiary alcohol moiety on the B-ring are either inactive or significantly less active in inducing GR-mediated transactivation, thus displaying a "dissociated" pharmacology in vitro. 2010 American Chemical Society.

Full text of DOI:10.1021/jm901551w

1270 J. Med. Chem. 2010, 53, 1270–1280
DOI: 10.1021/jm901551w
Novel Synthesis of the Hexahydroimidazo[1,5b]isoquinoline Scaffold: Application to the Synthesis of
Glucocorticoid Receptor Modulators†
Hai-Yun Xiao, Dauh-Rurng Wu, Mary F. Malley, Jack Z. Gougoutas, Sium F. Habte, Mark D. Cunningham,
John E. Somerville, John H. Dodd, Joel C. Barrish, Steven G. Nadler, and T. G. Murali Dhar*
Bristol-Myers Squibb Company, Research and Development, Princeton, New Jersey 08543-4000
Received October 19, 2009
The first stereoselective synthesis of the hexahydroimidazo[1,5b]isoquinoline (HHII) scaffold as a
surrogate for the steroidal A-B ring system is described. The structure-activity relationships of the
analogs derived from this scaffold show that the basic imidazole moiety is tolerated by the glucocorti-
coid receptor (GR) in terms of binding affinity, although the partial agonist activity in the transrepres-
sive assays depends on the substitution pattern on the B-ring. More importantly, most compounds in the
HHII series bearing a tertiary alcohol moiety on the B-ring are either inactive or significantly less active
in inducing GR-mediated transactivation, thus displaying a “dissociated” pharmacology in vitro.
Introduction
DNA binding, leads to suppression of inflammation.5,6 These
studies suggested the potential of designing GR ligands that
would maintain the anti-inflammatory properties associated
with transrepression and reduce the side effects associated
with transactivation, thereby improving the therapeutic index
of this important class of anti-inflammatory drugs.7
Based on this observation a number of reports have appea-
red in the open and patent literature8,9 describing the in vitro
and in vivo characterization of “dissociated glucocorticoids”,
also referred to as “selective glucocorticoid agonists.” In 2004,
Ali et al .10 and Scanlan et al.11 reported the synthesis and SAR
of pyrazolo[3,2-c] analogs (3) as partial agonists in various in
vitro functional assays of transactivation and transrepression
for hGR alpha (Figure 1). Both groups (i) removed the C and D
rings of fluorocortivazol (2) while maintaining the key hydroxyl
group meant to mimic the C-11 hydroxyl group of glucocorti-
coids and (ii) retained the 4-fluorophenylpyrazole moiety, first
identified by Hirschmann et al.12 as an excellent surrogate for
the 3-keto group of dexamethasone (1) or 16R-methylcortisol,
with enhanced anti-inflammatory activity.
Although a few other heterocycles have been prepared as a
replacement for the pyrazole moiety of fluorocortivazol,13 to
the best of our knowledge, the isosteric and isoelectronic
replacement of the pyrazole group with an imidazole has not
been reported in over four decades of steroid research. One
reason could be that most of the heterocyclic surrogates for
the 3-keto group of steroids are made from either commerci-
ally available Wieland-Mischer ketone or from functionalized
2-formyl-substituted steroids. The presence of a nitrogen atom
at the ring junction necessitated a completely new synthesis of
the hexahydroimidazo[1,5b]isoquinoline (hereafter, HHII, see
4, Figure 1) scaffold. This paper reports the first stereoselective
synthesis of the HHII ring system as a replacement for the A-B
ring of the glucocorticoid scaffold, wherein the pyrazole moiety
in the Scanlan/Merck chemotype is replaced by an imidazole
ring. We also report the consequences of replacing the neut-
ral pyrazole ring with a basic imidazole moiety (predicted
pKBHþ ∼ 6.4) on the SAR of these compounds and their poten-
tial to function as GR modulators.
The glucocorticoid receptor (GRa) is a member of the
nuclear hormone receptor superfamily and a member of the
class 1 steroid hormone family of transcription factors. Gluco-
corticoids that interact with GR have been used for over
50 years to treat inflammatory diseases. A number of studies
have shown that glucocorticoids exert their anti-inflammatory
activity via the inhibition of the transcription factors
NF-κB and AP-1.1 This inhibition is termed transrepression.
It has been shown that the primary mechanism for inhibition of
these transcription factors is via a direct physical interaction of
the GR monomer that alters the transcription factor complex
and inhibits the ability of NF-κB and activator protein-1 (AP-1)
to stimulate transcription.2 Other mechanisms such as seques-
tration of coactivators by GR have also been proposed.3,4
While transrepressive mechanisms contribute to the anti-
inflammatory activity of glucocorticoids, the interaction of a
glucocorticoid with GR can also lead to the induction and
transcription of certain genes such as tyrosine amino transfer-
ase (TAT). This induction of transcription is termed transac-
tivation. Unlike transrepression, transactivation appears to
require dimerization of GR and binding to consensus seq-
uences termed as glucocorticoid response element (GRE).
Studies employing transgenic GR dimerization defective
mice that cannot bind DNA have shown that the transactiva-
tion (DNA binding) activities of GR could be separated or
dissociated from the transrepressive (non-DNA binding)
effect of GR. These studies also suggest that many of the side
effects related to glucocorticoid therapy are due to the ability
of GR to induce transcription of various genes involved in
metabolism, whereas transrepression, which does not require
†
Dedicated to Dr. A. V. Rama Rao on the occasion of his 75th
birthday.
*To whom correspondence should be addressed. Phone: 609-252-
4158. Fax: 609-252-6804. E-mail: murali.dhar@bms.com.
a Abbreviations: HHII, hexahydroimidazo[1,5b]isoquinoline; GR,
glucocorticoid receptor; TAT, tyrosine amino transferase; GRE, gluco-
corticoid response element; AP-1, Activator Protein-1; ELAM,
Endothelial Leukocyte Adhesion Molecule.
r
pubs.acs.org/jmc
Published on Web 01/04/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3 1271
Figure 1. Evolution of the hexahydroimidazo[1,5b]isoquinoline scaffold.
Scheme 1a
a Reagents and conditions: (a) HCONH2, 170 °C, 43%; (b) HCHO, EtOH, RT, 96%; (c) SOCl2, CH2Cl2, 100%; (d) 2-methylcyclohexane-1,3-dione,
LiI, DBU, THF, RT to 60 °C, 60%; (e) (EtO)2P(O)CH2NC, n-BuLi, -50 °C to RT; (f), NaOAc, AcOH, H2O, 170 °C, microwave, 30 min, 55% over steps e
and f; (g) Chiralpak-AD column, CO2, MeOH; (h), RMgBr or RLi; (i) silica gel column chromatography for separation of diasteromers; (j) Dess-Martin
periodane, pyridine, CH2Cl2; (k) MeMgBr, THF.
Chemistry. Although a number of approaches were tried
to synthesize this novel ring system, the optimized protocol is
outlined in Scheme 1. Reaction of 2-bromo-1-(4-fluorophe-
nyl)ethanone (5) with formamide led to 4-(4-fluorophenyl)-
1H-imidazole (6) in good yield. Hydroxymethylation fol-
lowed by chlorination with thionyl chloride afforded the
chloromethyl derivative (7). After considerable amount of
experimentation, treatment of the chloromethyl derivative
(7) with 2-methyl-1,3-cyclohexanedione in the presence of
DBU and LiI14 gave us the desired C-alkylated product (8) in
60% isolated yield.
Attempts to convert the 1,3-diketone (8) to the 1,3-difor-
myl compound or its equivalent either failed entirely (e.g.,
MeOCH2SiMe3/t-BuLi or MeOCH2P(O)Ph2/LDA or N2CH-
SiMe3/LDA) or gave the desired product in very low yield
(e.g., Ph3PþCH2OMeCl-/t-BuOK), probably due to the
sterically hindered nature of the diketone. It is reported in
the literature that diethyl isocyanomethylphosphonate is a
useful reagent for the one carbon homologation of hindered
aldehydes and ketones.15,16 The reaction proceeds via an
intermediate isocyanide, which on acidic hydrolysis affords
the one carbon homologated aldehyde. In the event, reaction

1272 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3
Xiao et al.
Table 1. In Vitro GR Binding and Functional Activity for Secondary Alcohols
a See Experimental Section for retention times of isomers A and B. b Values are mean of at least two experiments done in triplicate. A value where the
standard deviation is not shown is from a single experiment. c Activator protein-1 (AP-1) assay in human A549 lung epithelial cell line. Experiments were
run in duplicate. d Endothelial leukocyte adhesion molecule (ELAM) assay in human A549 lung epithelial cell line. Experiments were run in duplicate.
e Efficacy represented as a percentage of the maximal response of dexamethasone. f See ref 11. g See ref 10.
of the 1,3-diketone (8) with diethyl isocyanomethylphospho-
nate gave the intermediate R,β-unsaturated isocyanide (9),
which was subjected to acidic hydrolysis employing sodium
acetate/acetic acid/water under microwave irradiation con-
ditions, at 170 °C. This led to the formation of the desired tri-
cyclic hexahydroimidazo[1,5b]isoquinoline (HHII) compound
(13) in 55% overall yield from the diketone (8) as a single
diastereomer. As outlined in Scheme 1, we believe that hydra-
tion of the isocyanide group leads to the formation of the
intermediate N-formylenamine (10, characterized by LCMS),
which isomerizes to the N-formylimine (11) via an enamine-
imine tautomeric equilibrium that places the bulky 4-(4-fluoro-
phenyl)-1H-imidazolylmethylene and N-formylimine groups in
an equatorial position around the cyclohexane ring. Subsequent
hydrolysis leads to the formation of the dialdehyde intermedi-
ate (12), which undergoes intramolecular ring closure at the
more electron rich 5-position of the imidazole ring, to yield
compound (13) as a single diastereomer with the desired
relative stereochemistry at the quaternary carbon bearing the
methyl group and the adjacent carbon atom anchoring the
formyl moiety (vide infra). Compound (13) was resolved into
its individual enantiomers using a Chiralpak-AD column and
each enantiomer was individually converted into the corres-
ponding secondary alcohols using Grignard or organolithium
reagents. The secondary alcohols were obtained as a mixture of
diastereomers (approximate ratio of 1:1 in most cases) that
could easily be separated by silica gel flash column chromatog-
raphy. The slower eluting enantiomeric aldehyde (isomer B)

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3 1273
Table 2. In Vitro GR Binding and Functional Activity for Tertiary Alcohol Derivatives
a Values are mean of at least two experiments done in triplicate. A value where the standard deviation is not shown is from a single experiment.
b Activator protein-1 (AP-1) assay in human A549 lung epithelial cell line. Experiments were run in duplicate. c Endothelial leukocyte adhesion molecule
(ELAM) assay in human A549 lung epithelial cell line. Experiments were run in duplicate. d GR tranactivation assay (agonist mode) in the NP-1 HeLa
cell line. e Efficacy represented as a percentage of the maximal response of dexamethasone. f See ref 10.
was used for further synthetic transformations, because the
secondary alcohols obtained from this enantiomer were more
potent than the corresponding analogs from the faster eluting
aldehyde (isomer A). Oxidation of the secondary alcohols to
the ketone, followed by addition of methylmagnesium bro-
mide, afforded the tertiary alcohols (28-51), mostly as single
diastereomers. The absolute stereochemistry of the tertiary
alcohols was established by single crystal X-ray analysis of
compounds 24 and 31, employing the anomalous scattering
method.17
repression. In the presence of test compounds, the ability of
the compounds to block activation of the AP-1 reporter by
phorbol myristate acetate (PMA) for 6 h is measured. In a
second assay (ELAM, endothelial leukocyte adhesion
molecule) that also utilizes a stable version of the A549 cell
line, a 400 bp portion of the E-selectin gene promoter region
is used to control a luciferase reporter. In the presence of test
compounds, the ability of the compounds to block activation
of the E-selectin reporter by IL-1β for 6 h is measured. This
assay is primarily NFκB dependent. The direct transcrip-
tional activation by GR in response to bound ligand was
measured using a cell based reporter assay in the agonist
mode. The NP-1 HeLa cell line contains a GR chimera that
activates the GAL4-luciferase reporter.18 In the agonist
mode, the ability of the bound compounds to directly acti-
vate the GR chimera is measured. Data for transrepression
(AP-1 and ELAM) and transactivation (NP-1) for represen-
tative compounds is shown in Tables 1-3.
Results and Discussion
Biological Assays. Compounds were initially measured for
their binding affinity to GRR in a competition based fluore-
scence polarization assay. Partially purified GRR protein
was mixed with a fluorescently tagged dexamethasone as a
ligand in the presence of test compounds. Compounds that
displayed potent GR binding were then evaluated in the
functional transrepression and transactivation assays. Com-
pounds underwent an assessment of their ability to repress tran-
scription using two cell-based reporter assays. In one assay,
a stable clone of the human lung epithelial cell line A549,
which contains a concatemer of five AP-1 response elements
controlling a luciferase reporter, was used to measure trans-
Structure-activity relationships for the hexahydroimidazo-
[1,5b]isoquinoline (HHII) based GR modulators are outlined
in Tables 1-3. Table 1 shows the GR binding and functional
data for a series of secondary alcohols derived from aldehydes
13A and 13B. Greater GR binding potency was demonstrated
with compounds derived from enantiomer 13B in comparison
to 13A (compare 14 and 15 against 16 and 17). However, there
was less of a stereochemical bias in terms of GR binding for
compounds within a diastereomeric set (compare 18 with 19
and compare 20 with 21). Although, the GR binding affinities
for the HHII compounds in the secondary alcohol series

1274 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3
Xiao et al.
Table 3. In Vitro GR Binding and Functional Activity for Tertiary Alcohols
a Values are mean of at least two experiments done in triplicate. A value where the standard deviation is not shown is from a single experiment. ND =
not determined. b Activator protein-1 (AP-1) assay in human A549 lung epithelial cell line. Experiments were run in duplicate. c Endothelial leukocyte
adhesion molecule (ELAM) assay in human A549 lung epithelial cell line. Experiments were run in duplicate. d GR tranactivation assay (agonist mode)
in the NP-1 HeLa cell line. e Efficacy represented as a percentage of the maximal response of dexamethasone. f See ref 10.
compared favorably with those reported by the Merck
and Scanlan groups, the lack of functional activity in the
transrepression assays was disappointing, because the corre-
sponding compounds in the pyrazole series are reporte-
dly partial agonists in the AP-1 and IL-6 transrepression
assays (compare compounds 16 with 22 and 20 with 23 in
Table 1).
We then decided to explore tertiary alcohols at the C-1
position of the HHII scaffold, since the corresponding
pyrazolo[3,2-c] analogs in the Merck series of GR modulators
displayed a significant boost in activity in the in vitro transre-
pression assays. We were mindful, however, that this increase
in transrepressive activity was reportedly accompanied by an
increase in transactivation, thereby decreasing the potential
of these compounds to function as dissociated GR agonists
(see compound 27, Table 2).
binding potency relative to the corresponding secondary alco-
hols as well as displaying partial agonist profiles in the AP-1
and ELAM transrepression assays (compare compound 19,
Table 1 with compound 26 in Table 2). More importantly,
compounds 24 and 26 (Table 2) did not show any signifi-
cant functional activity in the NP-1 transactivation assay.
A significant loss of GR binding affinity was recorded in
the case of methyl ether 25, along with a complete loss
of cellular functional activity (transrepression and trans-
activation). This was not unanticipated, because the C-11
OH group of glucocorticoids (including deacylcortivazol),
which corresponds to the C-1 hydroxyl of the HHII chemo-
types, forms an important H-bond with Asn564 on helix 3 of
the GR ligand binding domain.19 The dissociated nature of
compounds 24 and 26, in the in vitro transactivation and
transrepression assays prompted us to extensively explore
the SAR around the tertiary alcohol series of the HHII
chemotype.
The initial compounds made in the tertiary alcohol series of
the HHII chemotype showed significant improvement in GR

Article
Table 3 outlines the GR binding and functional activities
in the AP-1, ELAM, and NP-1 assays for a set of tertiary
alcohols.
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3 1275
the secondary alcohols. More importantly, these analogs were
either inactive or displayed significantly lower activity in the
NP-1 transactivation assay and therefore by definition
are “dissociated” in vitro. However, improvements in in vitro
and in vivo pharmacokinetic properties are needed to evaluate
these compounds in preclinical models for anti-inflammatory
activity (for example, mouse or rat adjuvant-induced arthritis
models) and side effects (for example, impact on glucose
metabolism) before making the conclusion about the ability
of these compounds to function as “dissociated” glucocorti-
coids in vivo.
Most of the analogs in the phenyl series (compounds
28-33) were partial agonists in the transrepression assays
and, as noted earlier, were inactive or displayed significantly
lower activity in the NP-1 transactivation assay. The pyridyl
regioisomers although of similar potency in the GR binding
assay to the phenyl analogs were either inactive or signifi-
cantly less active in the transrepression AP-1 and ELAM
assays (compare compound 28 with 34-36). The lack of
functional activity with the pyridyl regioisomers in relation
to the phenylanalog 28 may suggestthe inabilityof the pyridyl
regioisomers to engage helix-12 of the GR ligand binding
domain to attain the agonist conformation.20,21 The 2-thienyl
analogs 37-39 showed GR binding and functional activity
similar to the phenyl analogs. It is of interest to note that
subtle changes in functionality lead to significant changes in
potency in the functional assays for transrepression and
transactivation (compare 28 and 31 in the phenyl series and
24 and 39 in the 2-thienyl series). The 3-thienyl analog 40 had
significant potency in the GR binding and transrepression
assays and like the 2-thienyl analog 24 was inactive in the
NP-1 transactivation assay. As expected, the functional pot-
ency of the 2-thiazolyl analog 42 in transrepression assays was
better than the pyridyl regioisomers but significantly lower
than that of the thienyl analogs. A number of alkyl and
cycloaklyl analogs were also synthesized (analogs 43-51),
and most of them displayed partial agonist profiles in the
transrepression AP-1 and ELAM assays without significant
activity in the NP-1 transactivation assay.
Although a head-to-head comparison of the tertiary alco-
hol analogs in the HHII and pyrazolo[3,2-c] series in the
transrepression and transactivation assays is not possible,
because, of the different functional assays employed, it is clear
that the compounds display similar binding affinity and either
show a partial agonist profile or are inactive in the transacti-
vation functional assays (for example, compare compound 40
with 41 in Table 3).
The compounds listed in Tables 2 and 3 were tested for
selectivity against the androgen receptor (AR), progesterone
receptor (PR), and, in some cases, against the mineralocorti-
coid(MR) receptor. Almost allthe analogsdisplayed excellent
selectivity against the AR and MR receptors (IC50s>4000 nM
for AR in the binding assay and 30-50% inhibition at 10 μM
against MR in an antagonist mode functional assay, and no
activity at 10 μM in an agonist mode assay). The PR binding
activity varied between 100 and >1000 nM, depending on the
analog tested. For example, compound 31 has a Ki of 133 nM,
while 43 has a Ki of 1810 nM in the PR binding assay.22
A number of analogs listed in Tables 2 and 3 were also tested
in in vitro profiling assays for metabolic stability and their
ability toinhibit cytochromeP450. Unfortunately, most ofthe
compounds exhibited moderate to poor stability when incu-
bated in vitro with human and mouse liver microsomes23 and
so were not advanced for detailed in vivo evaluation.
Experimental Section
All procedures were carried out under a nitrogen atmosphere
unless otherwise indicated using anhydrous solvents purchased
from commercial sources without further purification. Reactions
requiring anhydrous solvents were performed in glassware, which
was dried using a heat gun and placed under a nitrogen atmo-
sphere. Column chromatography was performed on silica gel
using the solvent systems indicated. Solvent systems are reported
as v/v percent ratios. All reactions were monitored by TLC using
EM science, 0.25 mm, precoated silica gel plates or by LC/MS.
Yields refer to chromatographically and spectroscopically pure
compounds, except as otherwise noted. Proton NMR spectra
were recorded on a Bruker 400 MHz or a Joel 500 MHz NMR
spectrometer. Chemical shifts are reported in δ values relative to
tetramethylsilane. Electrospray ionization (ESI) mass spectra
were obtained on a Waters Micromass LC/MS single quadrupole
mass spectrometer. The purity of the final products was deter-
mined by HPLC, using the following methodfor each compound:
column YMC S5 CombiScreen ODS column 4.6 ꢀ 50 mm;
solvent A = water 90%/MeOH 10%/H3PO4 0.2%; solvent B =
MeOH 90%/water 10%/H3PO4 0.2%; flow rate 4 mL/min, λ =
220 nm; run time = 5 min. Most final products had a purity
of g95%. For final products whose purity is <95%, the percent
purity is specified is at the end of the description of its synthesis.
General Procedure for the Synthesis of Hexahydroimidazo-
[1,5b]isoquinolineCoreandSecondaryAlcoholsOutlinedinTable1
(4-Fluorophenyl)((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,
8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol (18, 19).
(a) A mixture of 2-bromo-1-(4-fluorophenyl)ethanone (10.9 g, 50
mmol) and formamide (14 mL, 350 mmol) was stirred at 170 °C
under nitrogen for 4 h. The reaction mixture was cooled and
diluted with ethyl acetate (40 mL). Saturated aqueous sodium
bicarbonate solution (50 mL) was added with caution to the
reaction mixture at 0 °C. The aqueous layer was separated and
extracted with ethyl acetate (2 ꢀ 40 mL). The combined organic
solutions were washed with water (30 mL) and brine (30 mL),
dried (Na2SO4), and concentrated. Silica gel flash chromatogra-
phy (10-100% ethyl acetate in hexanes) gave 4-(4-fluorophenyl)-
1H-imidazole (6; 3.46 g, 21 mmol, 43% yield, 95% purity) as a
solid. LC/MS = 163.1 (M þ H)þ; 1H NMR (400 MHz, chloro-
form-d) δ ppm 8.07 (1H, s), 7.72 (2H, dd, J=8.8 and 5.3 Hz), 7.32
(1H, s), 7.10 (2H, t, J=8.8 Hz).
(b) To a stirred solution of 4-(4-fluorophenyl)-1H-imidazole
(10 g, 62 mmol) in 95% ethanol (25 mL) was added aqueous
formaldehyde solution (37%, 9 mL) slowly. The reaction mix-
ture was stirred at RT for 1 h before water (25 mL) was added.
The reaction mixture was stirred at RT for an additional 3 h. The
solid that separated from solution was filtered, washed with
aqueous ethanol solution, and dried to give (4-(4-fluorophenyl)-
1H-imidazol-1-yl)methanol (11.4 g, 59 mmol, 96% yield) as a
solid. 1H NMR (400 MHz, MeOD) δ ppm 7.68-7.74 (3H, m),
7.40 (1H, s), 7.10 (2H, t, J=8.9 Hz), 4.65 (2H, s).
Conclusion
The first synthesis of the hexahydroimidazo[1,5b]isoquino-
line ring system as an abbreviated steroid scaffold is described.
The replacement of the neutral pyrazole ring with a basic imida-
zole moiety is tolerated by the glucocorticoid receptor in terms
of binding and the functional activity of tertiary alcohols in
the transrepression assays is significantly superior to that of
(c) To a stirred solution of thionyl chloride (18 mL) in
anhydrous methylene chloride (180 mL) was added (4-(4-fluo-
rophenyl)-1H-imidazol-1-yl)methanol (11.4 g, 59 mmol) por-
tionwise at RT under nitrogen. The reaction mixture was stirred
at RT for 3.5 h before anhydrous toluene (90 mL) was added.

1276 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3
Xiao et al.
The mixture was stirred at RT for 30 min. Concentration under
reduced pressure gave 1-(chloromethyl)-4-(4-fluorophenyl)-1H-
imidazole hydrochloride (7; 14.5 g, 59 mmol, 100% yield) as a
solid. MS found: (M þ H)þ = 211.3. The solid was sufficiently
pure to be used for the next step without further purification.
(d) To a stirred solution of 2-methylcyclohexane-1,3-dione
(10 g, 79 mmol), lithium iodide (26 g, 194 mmol) in anhydrous
THF (300 mL) cooled in a water bath was added 1,8-dia-
zabicyclo[5.4.0]-7-undecene (14 mL, 94 mmol) dropwise under
nitrogen. The reaction mixture was stirred at RT for 40 min
before 1-(chloromethyl)-4-(4-fluorophenyl)-1H-imidazole hydro-
chloride (9.8 g, 40 mmol) was added in one portion. The reaction
mixture was stirred at RT for 3 h and at 60 °C for 2 h. After the
mixture was cooled, water (250 mL) and heptane (250 mL) were
added. The aqueous layer was separated and extracted with
ethyl acetate (2ꢀ250 mL). The combined organic solutions were
washed with 16% aqueous potassium carbonate solution (3 ꢀ
50 mL), dried (Na2SO4), and concentrated. Silica gel flash chro-
matography (20-100% ethyl acetate in hexanes) gave 2-((4-(4-
fluorophenyl)-1H-imidazol-1-yl)methyl)-2-methylcyclohexane-
1,3-dione (8; 7.1 g, 24 mmol, 60% yield, 93% purity) as a solid.
LC/MS=301.4 (MþH)þ; 1H NMR (400 MHz, chloroform-d) δ
ppm 7.70 (2H, dd, J=8.6 and 5.1 Hz), 7.47 (1H, s), 7.06 (1H, s),
7.05 (2H, t, J = 8.6 Hz), 4.44 (2H, s), 2.67-2.78 (2H, m), 2.46
(2H, ddd, J = 17.04, 6.4, and 4.58 Hz), 1.96-2.07 (1H, m),
1.48-1.62 (1H, m), 1.37 (3H, s).
(e) To a stirred solution of diethyl isocyanomethylphospho-
nate (6.0 mL, 37 mmol) in anhydrous THF (160 mL) was added
butyl lithium solution (1.6 M in hexanes, 23 mL, 37 mmol)
dropwise at -50 to -60 °C under argon. The mixture was stirred
at the same temperature for 30 min before a solution of 2-((4-(4-
fluorophenyl)-1H-imidazol-1-yl)methyl)-2-methylcyclohexane-
1,3-dione (4.5 g in 30 mL of anhydrous THF, 15 mmol) was
added dropwise at -50 to -60 °C. The reaction mixture was
stirred at the same temperature for 30 min and at RT for 2 h,
concentrated under reduced pressure, dissolved in some ethyl
acetate, and filtered through a silica gel pad, which was then
rinsed with ethyl acetate. The filtrate was concentrated under
reduced pressure to give a solid that was used in the next step
without further purification.
Enantiomers 13A and 13B were used to synthesize the secondary
alcohols shown in Table 1. Enantiomer 13B was used to synthesize
the tertiary alcohol analogues shown in Tables 2 and 3 using the
following representative procedures.
(h) (4-Fluorophenyl)((5aR,6S)-1-(4-fluorophenyl)-5a-meth-
yl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol
(18, 19): To a stirred solution of (5aR,6S)-1-(4-fluorophenyl)-
5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinoline-
6-carbaldehyde (35 mg, 0.11 mmol) in anhydrous THF (1 mL)
was added 4-fluorophenymagnesium bromide solution (1 M in
THF, 0.6 mL, 0.6 mmol) dropwise at 0 °C under nitrogen.
The reaction mixture was stirred at 0 °C for 40 min and at RT for
40 min before being quenched with saturated aqueous ammo-
nium chloride solution (1 mL). The reaction mixture was extrac-
ted with ethyl acetate (2 ꢀ 2 mL), dried (Na2SO4), and concen-
trated. Prep HPLC purification (YMC S5 20ꢀ100 mm, 10 min
run; solvent A: 10% MeOH/90% H2O/0.1% TFA; solvent B:
90% MeOH/10% H2O/0.1% TFA) gave 18 as the faster eluting
isomer followed by 19, which were characterized as their TFA
salts. Analytical data for 18 (7 mg, 0.013 mmol, 12% yield, 97%
purity). LCMS=407.4 (Mþ1)þ. 1H NMR (400 MHz, MeOD)
δ ppm 8.88 (1H, s), 7.66 (2H, dd, J=8.9 and 5.3 Hz), 7.42 (2H,
dd, J=8.6 and 5.6 Hz), 7.31 (2H, t, J=8.6 Hz), 7.08 (2H, t, J=
8.6 Hz), 6.45 (1H, s), 5.06 (1H, s), 4.85 (1H, d, J=13.2 Hz), 4.11
(1H, d, J = 12.7 Hz), 2.46-2.52 (2H, m), 1.85-1.92 (1H, m),
1.79 (1H, td, J=12.5 and 3.1 Hz), 1.69-1.75 (1H, m), 1.63 (1H,
dd, J = 13.0 and 2.8 Hz), 1.37 (3H, s), 1.16-1.29 (1H, m);
Analytical data for 19 (6 mg, 0.012 mmol, 11% yield, 94%
purity). LCMS=407.4 (Mþ1)þ. 1H NMR (400 MHz, MeOD)
δ ppm 8.98 (1H, s), 7.67 (2H, dd, J=8.6 and 5.1 Hz), 7.40 (2H,
dd, J=8.6 and 5.6 Hz), 7.32 (2H, t, J=8.6 Hz), 7.09 (2H, t, J=
8.9 Hz), 6.51 (1H, d, J=2.0 Hz), 5.54 (1H, d, J=13.7 Hz), 4.58
(1H, d, J=10.2 Hz), 4.08 (1H, d, J=13.7 Hz), 2.35-2.55 (2H,
m), 1.83-1.93 (1H, m), 1.70-1.79 (1H, m), 1.34 (3H, s),
1.19-1.30 (2H, m), 1.10-1.17 (1H, m).
A mixture of the secondary alcohols 18 and 19 (41 mg, 0.79 mmol,
72% yield) was also obtained during prep HPLC.
The following compounds were prepared in the similar fashion.
Benzo[b]thiophen-3-yl(1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,
8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol (14): 7 mg,
0.016 mmol, 12% yield, 90% purity. LCMS=445.3 (Mþ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 8.01-8.08 (1H, m),
7.88-7.94 (1H, m), 7.69 (2H, dd, J = 8.9 and 5.3 Hz), 7.50 (1H,
s), 7.38-7.42 (2H, m), 7.37 (1H, s), 7.09 (2H, t, J= 8.9 Hz), 6.46
(1H, d, J=2.0 Hz), 5.15 (1H, d, J=12.7 Hz), 5.06 (1H, d, J=10.2
Hz), 3.88 (1H, d, J=13.2 Hz), 2.17-2.42 (3H, m), 1.64-1.72 (1H,
m), 1.29 (3H, s), 1.17-1.23 (3H, m).
Benzo[b]thiophen-3-yl(1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,
8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol (15): 4 mg,
0.009 mmol, 7% yield, 90% purity. LCMS =445.3 (M þ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 7.90 (1H, d, J=7.6 Hz),
7.64-7.78 (4H, m), 7.48 (1H, s), 7.35-7.46 (2H, m), 7.10 (2H, t,
J=8.9 Hz), 6.39 (1H, d, J=1.0 Hz), 5.47 (1H, s), 4.53 (1H, d, J=
11.7 Hz), 3.90 (1H, d, J=11.7 Hz), 2.33-2.49 (2H, m), 1.73-1.93
(4H, m), 1.34 (3H, s), 1.19-1.32 (1H, m).
Benzo[b]thiophen-3-yl(1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,
8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol (16): 8 mg,
0.018 mmol, 14% yield, 90% purity. LCMS=445.3 (Mþ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 8.02-8.07 (1H, m),
7.88-7.93 (1H, m), 7.69 (2H, dd, J = 8.9 and 5.3 Hz), 7.46 (1H,
s), 7.38-7.43 (2H, m), 7.36 (1H, s), 7.08 (2H, t, J= 8.6 Hz), 6.46
(1H, d, J = 1.5 Hz), 5.14 (1H, d, J = 13.2 Hz), 5.06 (1H, d, J =
10.2 Hz), 3.87 (1H, d, J=13.2 Hz), 2.17-2.41 (3H, m), 1.64-1.71
(1H, m), 1.28 (3H, s), 1.17-1.23 (3H, m).
Benzo[b]thiophen-3-yl(1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,
8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)methanol (17): 4 mg,
0.009 mmol, 7% yield, 90% purity. LCMS =445.3 (M þ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 7.90 (1H, d, J=8.1 Hz),
7.64-7.75 (4H, m), 7.48 (1H, s), 7.35-7.45 (2H, m), 7.10 (2H, t,
J= 8.6 Hz), 6.38 (1H, s), 5.46 (1H, s), 4.51 (1H, d, J= 12.2 Hz),
(f) One quarter of the above solid was mixed with sodium
acetate (1.5 g), acetic acid (4 mL), and water (4 mL). The mixture
was heated at 170 °C in a microwave reactor (Emrys Optimizer,
Personal Chemistry) under nitrogen for 30 min. The above
microwave reaction was repeated three times. The four micro-
wave reaction mixtures were combined and concentrated under
reduced pressure to remove solvents. The residue was diluted
with water (40 mL) and ethyl acetate (80 mL) and made basic using
potassium carbonate and sodium bicarbonate solid at 0 °C. The
aqueous layer was separated and extracted with ethyl acetate
(3 ꢀ 30 mL). The combined organic solutions were dried (Na2-
SO4) and concentrated. Silica gel flash chromatography (30-
100% ethyl acetate in hexanes) gave 1-(4-fluorophenyl)-5a-
methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinoline-6-car-
baldehyde 13A and 13B (2.55 g, 8.2 mmol, 55% yield over two
1
steps, 95% purity) as a solid. LCMS = 311.2 (M þ H)þ. H
NMR (400 MHz, chloroform-d) δ ppm 9.77 (1H, d, J=1.5 Hz),
7.68 (2H, dd, J=9.2 and 5.6 Hz), 7.52 (1H, s), 7.10 (2H, t, J=
8.9 Hz), 6.47 (1H, d, J=2.0 Hz), 4.49 (1H, d, J=12.2 Hz), 3.80
(1H, d, J=12.7 Hz), 2.36-2.51 (2H, m), 2.31 (1H, ddd, J=13.1,
3.4, and 1.3 Hz), 2.13 (1H, ddd, J = 13.4, 3.4, and 1.5 Hz),
2.01-2.09 (1H, m), 1.82 (1H, qd, J = 13.1 and 3.3 Hz), 1.42-
1.57 (1H, m), 1.17 (3H, s).
(g) 1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimi-
dazo[1,5-b]isoquinoline-6-carbaldehyde 13A and 13B were re-
solved into its enantiomers using a Chiralpak-AD column
[conditions: ChiralpakAD-H (3 ꢀ 25 cm) CO2/MeOH 68/32,
70 mL/min, at 35 °C and 100 bars at 220 nm; sample: 2.5 g in
22.5 mL of chloroform and IPA (1/1), loading 1.5 cm3/16.2 min];
retention time for 13A: 5.40 min; retention time for 13B: 8.18 min.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3 1277
3.89 (1H, d, J=11.7 Hz), 2.34-2.45 (2H, m), 1.76-1.93 (4H, m),
1.34 (3H, s), 1.19-1.32 (1H, m).
Step A: To a stirred solution of (5aR,6S)-1-(4-fluorophenyl)-5a-
methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinoline-6-car-
baldehyde (isomer 13B, 90 mg, 0.29 mmol) in anhydrous THF
(2.5 mL) was added thiophen-2-ylmagnesium bromide solution
(1 M in THF, 1.5 mL, 1.5 mmol) dropwise at 0 °C under
nitrogen. The reaction mixture was stirred at 0 °C for 30 min
and at RT for 30 min before being quenched with saturated
aqueous ammonium chloride solution (3 mL). The reaction
mixture was extracted with ethyl acetate (3 ꢀ 2 mL), dried
(Na2SO4), filtered through a silica gel pad, and concentrated.
The crude secondary alcohol mixture was used for the next step
without further purification.
1-(1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo-
[1,5-b]isoquinolin-6-yl)but-3-en-1-ol as a TFA salt (20): 6 mg,
0.013 mmol, 13% yield, 98% purity. LCMS = 353.4 (M þ 1)þ.
1H NMR (400 MHz, MeOD) δ ppm 8.88 (1H, s), 7.65 (2H, dd,
J = 8.6 and 5.1 Hz), 7.31 (2H, t, J = 8.9 Hz), 6.44 (1H, d, J =
1.5 Hz), 5.81-5.93 (1H, m), 5.08-5.17 (2H, m), 4.65 (1H, d, J=
12.7 Hz), 3.90 (1H, t, J = 7.1 Hz), 3.80 (1H, d, J = 13.2 Hz),
2.45-2.57 (2H, m), 2.40 (1H, dt, J=13.7 and 6.9 Hz), 2.21-2.31
(1H, m), 1.95-2.03 (1H, m), 1.68-1.85 (2H, m), 1.58-1.64 (1H,
m), 1.36-1.50 (1H, m), 1.24 (3H, s).
1-(1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo-
[1,5-b]isoquinolin-6-yl)but-3-en-1-ol as a TFA salt (21): 6 mg,
0.013 mmol, 13% yield, 93% purity. LCMS = 353.4 (M þ 1)þ.
1H NMR (400 MHz, MeOD) δ ppm 8.93 (1H, s), 7.65 (2H, dd,
J = 8.6 and 5.1 Hz), 7.31 (2H, t, J = 8.6 Hz), 6.48 (1H, d, J =
2.0 Hz), 5.90-6.02 (1H, m), 5.29 (1H, d, J=13.7 Hz), 5.09-5.18
(2H, m), 3.89 (1H, d, J = 13.7 Hz), 3.70-3.77 (1H, m), 2.37-
2.56 (3H, m), 2.27 (1H, ddd, J = 14.5, 7.6, and 7.4 Hz), 1.83-
1.97 (2H, m), 1.59-1.67 (1H, m), 1.31-1.45 (2H, m), 1.21 (3H, s).
Representative Examples for the Synthesis of Tertiary Alcohols
Outlined in Tables 2 and 3. (S)-1-(4-Fluorophenyl)-1-((5aR,6S)-
1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]-
isoquinolin-6-yl)ethanol (26).Step A: A TFA salt of (4-fluorophenyl)-
((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydro-
imidazo[1,5-b]isoquinolin-6-yl)methanol (18 and 19; 40 mg, 0.79
mmol) was neutralized with saturated aqueous sodium bicarbonate
solution and extracted with ethyl acetate. The ethyl acetate extract
was dried (Na2SO4) and concentrated. A total of 32 mg (0.079
mmol) of the secondary alcohol was dissolved in anhydrous
methylene chloride (2 mL). N-Methylmorpholine oxide (37 mg,
0.32 mmol) was added and the contents stirred at 0 °C for 5 min
before tetrapropylammonium perruthenate (6 mg, 0.016 mmol)
was added. The reaction mixture was stirred at 0 °C for 30 min and
at RT for 2 h. Silica gel flash chromatography of the reaction
mixture (10-100% ethyl acetate in hexanes) gave (4-fluorophenyl)-
((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimi-
dazo[1,5-b]isoquinolin-6-yl)methanone (24 mg, 0.059 mmol, 75%
yield) as a glassy solid. LCMS = 405.4 (M þ 1)þ. 1H NMR (400
MHz, chloroform-d) δ ppm 7.97 (2H, dd, J=8.9 and 5.3 Hz), 7.66
(2H, dd, J=8.9 and 5.3 Hz), 7.28 (1H, s), 7.18 (2H, t, J=8.6 Hz),
7.09 (2H, t, J=8.6 Hz), 6.46 (1H, d, J=1.5 Hz), 3.82 (1H, d, J=
12.3 Hz), 3.69 (1H, d, J=12.3 Hz), 3.50 (1H, dd, J=12.5 and 3.3
Hz), 2.41-2.57 (2H, m), 1.96-2.09 (2H, m), 1.82-1.90 (1H, m),
1.51-1.64 (1H, m), 1.31 (3H, s).
Step B: The secondary alcohol obtained in step A was
dissolved in anhydrous methylene chloride (2 mL). Pyridine
(0.07 mL, 0.87 mmol) and Dess-Martin periodinane (185 mg,
0.44 mmol) were added sequentially at RT. The reaction mixture
was stirred at RT for 1 h before being quenched by the slow
addition of saturated aqueous sodium bicarbonate solution
(5 mL) and saturated aqueous sodium thiosulphate solution
(3 mL). The reaction mixture was stirred at RT for 30 min and
the aqueous layer was separated and extracted with methylene
chloride (3 ꢀ 2 mL). The combined organic extracts were dried
(Na2SO4) and concentrated. Silica gel flash chromatography
(20-100% ethyl acetate in hexanes) gave ((5aR,6S)-1-(4-fluo-
rophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoqui-
nolin-6-yl)(thiophen-2-yl)methanone, which was used as such
for the subsequent step without further purification.
Step C: To a stirred solution of the above ((5aR,6S)-1-(4-
fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]-
isoquinolin-6-yl)(thiophen-2-yl)methanone (step B) in anhy-
drous THF (4 mL) was added methylmagnesium bromide
solution (3 M in diethyl ether, 2 mL, 6 mmol) dropwise at
-78 °C under argon. The reaction mixture was then stirred at
RT for 2.5 h before saturated aqueous ammonium chloride
solution (6 mL) was added slowly with water bath cooling. The
reaction mixture was extracted with ethyl acetate (3 ꢀ 5 mL),
dried (Na2SO4), and concentrated. Silica gel flash chromatog-
raphy (10-100% ethyl acetate in hexanes) gave (S)-1-((5aR,6S)-
1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo-
[1,5-b]isoquinolin-6-yl)-1-(thiophen-2-yl)ethanol (58 mg, 0.14 mmol,
49% yield over three steps, 99% purity) as a solid. MS found:
(MþH)þ=409.2; 1H NMR (400 MHz, CDCl3) δ ppm 7.73 (2H,
dd, J = 8.90 and 5.34 Hz), 7.58 (1H, s), 7.29 (m, under solvent
peak), 7.12 (2H, t, J=8.90 Hz), 7.00-7.02 (1H, m), 6.96-6.99
(1H, m), 6.42 (1H, s), 4.99 (1H, d, J=13.22 Hz), 3.88 (1H, d, J=
13.23 Hz), 2.41 (2H, m), 2.16 (1H, s), 2.12 (1H, dd, J=12.21 and
3.56 Hz), 1.82 (3H, s), 1.63-1.69 (1H, m), 1.59 (1H, dt, J=13.23
and 3.05 Hz), 1.34 (3H, s), 1.22-1.31 (1H, m).
Step B: To a stirred solution of (4-fluorophenyl)((5aR,6S)-
1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo-
[1,5-b]isoquinolin-6-yl)methanone (step A, 24 mg, 0.059 mmol)
in anhydrous THF (1 mL) was added methylmagnesium bro-
mide (3 M in diethyl ether, 0.15 mL, 0.45 mmol) dropwise at -78 °C
under argon. The reaction mixture was then stirred at RT for 4 h
before being quenched with saturated aqueous ammonium
chloride solution (2 mL) at 0 °C. The reaction mixture was
extracted with ethyl acetate (3 ꢀ 2 mL), dried (Na2SO4), and
concentrated. The residue that was obtained was purified by
prep HPLC (YMC S5 20ꢀ100 mm, 10 min run, solvent A: 10%
MeOH/90% H2O/0.1% TFA; solvent B: 90% MeOH/10%
H2O/0.1% TFA) to yield (S)-1-(4-fluorophenyl)-1-((5aR,6S)-
1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo-
[1,5-b]isoquinolin-6-yl)ethanol (26; 15 mg 0.028 mmol, 95%
purity, 47% yield) as a TFA salt. LCMS = 421.4 (M þ 1)þ. 1H
NMR (400 MHz, MeOD) δ ppm 8.84 (1H, s), 7.61-7.67 (2H,
m), 7.55-7.61 (2H, m), 7.31 (2H, t, J=8.6 Hz), 7.06 (2H, t, J=
8.9 Hz), 6.43 (1H, s), 5.27 (1H, d, J=13.7 Hz), 4.13 (1H, d, J=
13.7 Hz), 2.43-2.50 (2H, m), 1.96 (1H, dd, J=12.2 and 3.6 Hz),
1.83 (1H, dt, J=12.2 and 3.6 Hz), 1.69 (3H, s), 1.55-1.67 (2H,
m), 1.42 (3H, s), 1.11-1.28 (1H, m).
The following compounds were prepared in a similar fashion.
In some cases, the compounds were further purified by crystal-
lization from ethyl acetate and heptanes.
(S)-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-phenylethanol (28): 5 mg,
0.012 mmol, 25% yield, 95% purity. LCMS = 403.2 (M þ 1)þ.
1H NMR 1H NMR (500 MHz, chloroform-d) δ ppm 7.69 (2H,
dd, J = 8.8 and 5.5 Hz), 7.50-7.54 (2H, m), 7.36-7.41 (3H,
m), 7.29-7.34 (1H, m), 7.09 (2H, t, J = 8.7 Hz), 6.37 (1H, s),
4.62 (1H, d, J = 13.2 Hz), 3.75 (1H, d, J = 13.2 Hz), 2.33-
2.42 (2H, m), 1.97 (1H, dd, J = 11.7 and 4.0 Hz), 1.76-1.83
(1H, m), 1.73 (3H, s), 1.53-1.66 (2H, m), 1.32 (3H, s), 1.13-1.30
(1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-o-tolylethanol (29): 14 mg,
0.034 mmol, 76% yield, 90% purity. LCMS=417.2 (Mþ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 7.68 (2H, dd, J=8.9 and
5.3 Hz), 7.40-7.45 (1H, m), 7.33 (1H, s), 7.14-7.26 (3H, m), 7.09
(2H, t, J=8.6 Hz), 6.39(1H, s), 4.73 (1H, d, J=13.2Hz), 3.87 (1H,
d, J= 12.7 Hz), 2.63 (3H, s), 2.33-2.41 (2H, m), 2.17-2.27 (2H,
m), 1.82 (3H, s), 1.74-1.87 (1H, m), 1.55-1.64 (2H, m), 1.34 (3H,
s), 1.13-1.25 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(thiophen-2-yl)ethanol (24).

1278 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3
Xiao et al.
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-m-tolylethanol (30): 11 mg,
0.026 mmol, 58% yield, 96% purity. LCMS=417.2 (Mþ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 7.74 (2H, dd, J=8.9 and
5.3 Hz), 7.23-7.33 (4H, m), 7.10-7.17 (3H, m), 6.36 (1H, s), 4.68
(1H, d, J = 13.2 Hz), 3.84 (1H, d, J = 13.2 Hz), 2.40 (3H, s),
2.35-2.42 (4H, m), 1.96 (1H, dd, J=11.7and4.1Hz), 1.71(3H, s),
1.52-1.86 (2H, m), 1.34 (3H, s), 1.17-1.30 (1H, m).
(S)-1-(4-Chlorophenyl)-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-
5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)ethanol (31):
49 mg, 0.11 mmol, 43% yield, 99% purity. LCMS=437.1 (Mþ
1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.71 (2H, dd, J=
8.9 and 5.3 Hz), 7.54 (1H, br s), 7.45 (2H, d, J = 8.7 Hz), 7.34
(2H, d, J = 8.6 Hz), 7.11 (2H, t, J = 8.9 Hz), 6.38 (1H, s), 4.84
(1H, d, J=13.2 Hz), 3.83 (1H, d, J=12.7 Hz), 2.34-2.40 (2H,
m), 1.88-1.94 (1H, m), 1.79 (1H, dt, J=12.7 and 3.6 Hz), 1.76
(1H, s), 1.72 (3H, s), 1.49-1.58 (2H, m), 1.33 (3H, s), 1.10-1.21
(1H, m).
3-((S)-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-
hexahydroimidazo[1,5-b]isoquinolin-6-yl)-1-hydroxyethyl)benzo-
nitrile (32): 9 mg, 0.017 mmol, 37% yield, 95% purity. LCMS=
428.3 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.87 (1H, s),
7.95 (1H, s), 7.90 (1H, d, J = 8.1 Hz), 7.61-7.68 (3H, m), 7.54
(1H, t, J = 7.8 Hz), 7.31 (2H, t, J = 8.8 Hz), 6.44 (1H, s), 5.33
(1H, d, J=13.6 Hz), 4.16 (1H, d, J=13.6 Hz), 2.44-2.51 (2H,
m), 1.95 (1H, dd, J=12.3 and 3.3 Hz), 1.83 (1H, dt, J=12.7 and
3.7 Hz), 1.72 (3H, s), 1.63-1.76 (1H, m), 1.54 (1H, dd, J=13.8
and 3.3 Hz), 1.44 (3H, s), 1.10-1.26 (1H, m).
4-((S)-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-5,5a,6,7,8,9-
hexahydroimidazo[1,5-b]isoquinolin-6-yl)-1-hydroxyethyl)benzo-
nitrile (33): 4 mg, 0.007 mmol, 16% yield, 94% purity. LCMS=
428.1 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.84 (1H, s),
7.78 (2H, d, J=8.6 Hz), 7.72 (2H, d, J=8.6 Hz), 7.64 (2H, dd,
J = 8.8 and 5.0 Hz), 7.31 (2H, t, J = 8.7 Hz), 6.44 (1H, s), 5.26
(1H, d, J=13.8 Hz), 4.12 (1H, d, J=13.8 Hz), 2.48 (2H, dd, J=
8.2 and 3.4 Hz), 1.97 (1H, dd, J=12.6 and 3.3 Hz), 1.83 (1H, dt,
J=12.6 and 3.7 Hz), 1.72 (3H, s), 1.65-1.77 (1H, m), 1.52-1.60
(1H, m), 1.43 (3H, s), 1.11-1.22 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(pyridin-2-yl)ethanol, 2 TFA
salt (34): 13 mg, 0.021 mmol, 46% yield, 90% purity. LCMS=
404.1 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.91 (1H, s),
8.79 (1H, dd, J = 5.7 and 1.1 Hz), 8.52 (1H, td, J = 8.0 and
1.6 Hz), 8.17 (1H, d, J=8.3 Hz), 7.92-7.97 (1H, m), 7.65 (2H,
dd, J=8.9 and 5.2 Hz), 7.31 (2H, t, J=8.8 Hz), 6.47 (1H, s), 5.11
(1H, d, J=13.6 Hz), 4.15 (1H, d, J=13.6 Hz), 2.48-2.54 (2H,
m), 2.08 (1H, dd, J =12.5 and 3.1 Hz), 1.86 (3H, s), 1.76-1.91
(2H, m), 1.56-1.64 (1H, m), 1.45 (3H, s), 1.21-1.33 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(pyridin-3-yl)ethanol, 2 TFA
salt (35): 22 mg, 0.035 mmol, 77% yield, 95% purity. LCMS=
404.1 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 9.06 (1H, d,
J=1.8 Hz), 9.00 (1H, s), 8.81 (1H, dt, J=8.3 and 1.6 Hz), 8.79
(1H, d, J=5.5 Hz), 8.07 (1H, dd, J=8.3 and 5.8 Hz), 7.66 (2H,
dd, J=8.9 and 5.2 Hz), 7.31 (2H, t, J=8.8 Hz), 6.47 (1H, s), 5.52
(1H, d, J=13.6 Hz), 4.24 (1H, d, J=13.8 Hz), 2.46-2.54 (2H,
m), 1.98 (1H, dd, J =12.5 and 2.9 Hz), 1.83 (3H, s), 1.66-1.90
(2H, m), 1.46 (3H, s), 1.40-1.49 (1H, m), 1.12-1.26 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(pyridin-4-yl)ethanol, 2 TFA
salt (36): 20 mg, 0.032 mmol, 70% yield, 90% purity. LCMS=
404.1 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.97 (1H, s),
8.83 (2H, d, J=6.8 Hz), 8.28 (2H, d, J=6.8 Hz), 7.65 (2H, dd,
J = 8.8 and 5.0 Hz), 7.31 (2H, t, J = 8.8 Hz), 6.47 (1H, s), 5.39
(1H, d, J=13.8 Hz), 4.17 (1H, d, J=13.8 Hz), 2.47-2.54 (2H,
m), 2.00 (1H, dd, J =12.3 and 3.3 Hz), 1.81 (3H, s), 1.71-1.90
(2H, m), 1.50 (1H, dd, J = 13.3 and 3.3 Hz), 1.46 (3H, s),
1.15-1.27 (1H, m).
ethanol (37): 11 mg, 0.026 mmol, 58% yield, 95% purity. LCMS=
423.2 (M þ 1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.68
(2H, dd, J=8.9 and 5.3 Hz), 7.34 (1H, s), 7.05-7.12 (3H, m), 6.82
(1H, d, J =5.1 Hz), 6.40 (1H, s), 4.73 (1H, d, J = 13.2 Hz), 3.81
(1H, d, J=12.7 Hz), 2.38 (3H, s), 2.33-2.44 (2H, m), 2.28 (1H, s),
2.21 (1H, dd, J=12.5 and 3.3 Hz), 1.81 (3H, s), 1.72-1.88 (2H, m),
1.61 (1H, qd, J=12.8 and 3.3 Hz), 1.31 (3H, s), 1.18-1.28 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(5-methylthiophen-2-yl)-
ethanol (38): 13 mg, 0.031 mmol, 69% yield, 93% purity. LCMS=
423.1 (M þ 1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.70
(2H, dd, J=8.6 and 5.6 Hz), 7.46 (1H, s), 7.09 (2H, t, J=8.6 Hz),
6.76 (1H, d, J = 3.6 Hz), 6.58 (1H, dd, J = 3.3 and 1.3 Hz), 6.40
(1H, s), 5.01 (1H, d, J=13.2 Hz), 3.83 (1H, d, J=13.2 Hz), 2.47
(3H, s), 2.34-2.42 (2H, m), 2.05-2.11 (2H, m), 1.80 (1H, dt, J=
12.5 and 3.7 Hz), 1.75 (3H, s), 1.59-1.68 (1H, m), 1.51 (1H, qd, J=
12.8 and 3.3 Hz), 1.30 (3H, s), 1.21-1.28 (1H, m).
(S)-1-(5-Chlorothiophen-2-yl)-1-((5aR,6S)-1-(4-fluorophenyl)-
5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)-
ethanol (39): 12 mg, 0.027 mmol, 60% yield, 94% purity. LCMS=
443.1 (M þ 1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.67
(2H, dd, J=8.9 and 5.3 Hz), 7.41 (1H, s), 7.08 (2H, t, J=8.6 Hz),
6.70-6.77 (2H, m), 6.40 (1H, s), 5.00 (1H, d, J = 13.2 Hz), 3.80
(1H, d, J=13.2 Hz), 2.72 (1H, br s), 2.30-2.41 (2H, m), 1.88-2.03
(1H, m), 1.81 (1H, dt, J=12.6 and 3.4 Hz), 1.72 (3H, s), 1.43-1.68
(2H, m), 1.29 (3H, s), 1.16-1.27 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(thiophen-3-yl)ethanol (40):
8 mg, 0.02 mmol, 56% yield, 98% purity. LCMS = 409.3 (M þ
1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.70 (2H, dd, J=
8.9 and 5.3 Hz), 7.45 (1H, s), 7.33 (1H, dd, J=5.1 and 3.1 Hz),
7.20 (1H, dd, J = 3.1 and 1.5 Hz), 7.15 (1H, dd, J = 5.1 and
1.5 Hz), 7.10 (2H, t, J = 8.6 Hz), 6.39 (1H, s), 4.73 (1H, d, J =
13.2 Hz), 3.77 (1H, d, J=13.2 Hz), 2.35-2.41 (2H, m), 1.98 (1H,
dd, J = 11.7 and 4.1 Hz), 1.77-1.85 (2H, m), 1.73 (3H, s),
1.50-1.66 (2H, m), 1.31 (3H, s), 1.15-1.29 (1H, m).
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(thiazol-2-yl)ethanol (42):
12 mg, 0.029 mmol, 50% yield, 90% purity. LCMS=410.1 (Mþ
1)þ. 1H NMR (400 MHz, chloroform-d) δ ppm 7.72 (1H, d, J=
3.3 Hz), 7.63-7.69 (2H, m), 7.35 (1H, s), 7.34 (1H, d, J =
3.0 Hz), 7.08 (2H, t, J = 8.8 Hz), 6.38 (1H, s), 4.52 (1H, d, J =
13.1 Hz), 3.72 (1H, d, J=13.1 Hz), 2.36-2.42 (2H, m), 2.15 (1H,
dd, J = 12.5 and 2.9 Hz), 1.82-1.94 (2H, m), 1.80 (3H, s),
1.58-1.75 (2H, m), 1.32 (3H, s), 1.19-1.30 (1H, m).
(R)-2-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)pent-4-en-2-ol (43): 1.4 mg,
0.0038 mmol, 6% yield, 96% purity. LCMS = 367.4 (M þ 1)þ.
1H NMR (400 MHz, chloroform-d) δ ppm 7.70 (2H, dd, J =
8.6 and 5.6 Hz), 7.52 (1H, s), 7.10 (2H, t, J=8.6 Hz), 6.40 (1H, s),
5.84-5.96 (1H, m), 5.31 (1H, dd, J=10.2 and 2.0 Hz), 5.21 (1H,
d, J=16.3 Hz), 5.10 (1H, d, J=13.2 Hz), 3.71 (1H, d, J=13.7
Hz), 2.56 (1H, dd, J=13.2 and 8.6 Hz), 2.35-2.45 (2H, m), 2.31
(1H, dd, J = 13.2 and 6.6 Hz), 1.88-1.98 (1H, m), 1.72-1.82
(2H, m), 1.50-1.63 (1H, m), 1.33-1.43 (1H, m), 1.31 (3H, s),
1.25 (3H, s), 1.17-1.26 (1H, m).
(R)-2-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)hexan-2-ol (44): 3 mg, 0.0078
1
mmol, 17% yield, 90% purity. LCMS = 383.2 (M þ 1)þ. H
NMR (400 MHz, chloroform-d) δ ppm 7.69 (2H, dd, J=8.6 and
5.6 Hz), 7.47 (1H, s), 7.09 (2H, t, J=8.6 Hz), 6.39 (1H, s), 4.99 (1H,
d, J=13.2 Hz), 3.76 (1H, d, J=13.2 Hz), 2.38 (2H, d, J=6.1 Hz),
1.31 (3H, s), 1.23 (3H, s), 1.18-1.96 (12H, m), 0.95 (3H, t, J =
6.9 Hz).
(R)-2-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-4-phenylbutan-2-ol (45):3.8 mg,
0.0088 mmol, 20% yield, 96% purity. LCMS=431.2 (Mþ1)þ. 1H
NMR (400 MHz, chloroform-d) δ ppm 7.70 (2H, dd, J=8.9 and
5.3 Hz), 7.50 (1H, s), 7.29-7.35 (2H, m), 7.19-7.25 (3H, m), 7.09
(2H, t, J=8.6 Hz), 6.39 (1H, s), 5.03 (1H, d, J=13.2 Hz), 3.79 (1H,
(S)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(3-methylthiophen-2-yl)-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3 1279
d, J=13.2 Hz), 2.80 (1H, td, J=12.6 and 4.8 Hz), 2.64 (1H, td, J=
12.6 and 4.8 Hz), 2.27-2.44 (2H, m), 2.15 (1H, td, J = 13.0 and
5.1 Hz), 1.85-1.95 (1H, m), 1.67-1.84 (3H, m), 1.49-1.62 (2H, m),
1.44 (3H, s), 1.32-1.47 (1H, m), 1.23 (3H, s).
(R)-1-(4-Fluorophenyl)-2-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-
5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)propan-2-ol
(46): 4.7 mg, 0.0086 mmol, 33% yield, 95% purity. LCMS =
435.4 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.93 (1H, s),
7.62-7.69 (2H, m), 7.23-7.36 (4H, m), 7.00 (2H, t, J=8.9 Hz),
6.47 (1H, s), 5.52 (1H, d, J=14.2 Hz), 4.10 (1H, d, J=13.7 Hz),
3.12 (1H, d, J=13.2 Hz), 2.74 (1H, d, J=13.2 Hz), 2.47-2.59
(2H, m), 1.75-2.06 (4H, m), 1.46 (3H, s), 1.28-1.43 (1H, m),
1.16 (3H, s).
LCMS =423.2 (M þ 1)þ. 1H NMR (400 MHz, MeOD) δ ppm
9.00 (1H, s), 7.64-7.71 (2H, m), 7.41 (1H, dd, J = 5.16 and
1.13 Hz), 7.33 (2H, t, J = 8.69 Hz), 7.07 (1H, dd, J = 3.53 and
1.26 Hz), 6.98 (1H, dd, J=5.04 and 3.53 Hz), 6.48 (1H, s), 5.31
(1H, d, J =13.85 Hz), 4.23 (1H, d, J=14.10 Hz), 3.13 (3H, s),
2.47-2.54 (2H, m), 2.20 (1H, dd, J = 12.46 and 2.90 Hz),
1.79-1.86 (1H, m), 1.78 (3H, s), 1.60-1.74 (1H, m, J = 12.97,
12.97, 12.84, and 3.27 Hz), 1.49 (1H, d, J=13.35 Hz), 1.40 (3H,
s), 1.17-1.28 (1H, m).
Receptor Binding Assay. The GR ligand binding assay was
conducted in fluorescence polarization format which measures
the competition between a test compound and a fluorescently
labeled ligand (GS-red) for binding to the full length or ligand
binding domain of GRR (Invitrogen, cat#P2893). Compounds
were tested in concentrations ranging from 5 μM to 85 pM. IC50
values were determined by fitting the fluorescence polarization
signal data using a four parameter logistic equation. The Ki
values were determined by application of the Cheng-Prusoff
equation to the IC50 values, where Ki = IC50/(1 þ ligand con-
centration/Kd) (Cheng, Y.; Prusoff, W. H. Relationship between
the inhibition constant (Ki) and the concentration of inhibitor
which causes 50% inhibition (IC50) of an enzymatic reaction.
Biochem. Pharmacol. 1973, 22, 3099-3108). The Kd value used
for GR was 0.3 nM, as supplied by the assay manufacturer
(Invitrogen, cat#P2893). Data shown represent the mean of two
or more experiments.
Transrepression Assays. AP-1 activity was measured using an
AP-1 response element (5 copies) cloned into a luciferase
reporter vector. This reporter is stably transfected into the
human A549 lung epithelial cell line. AP-1 activity is induced
by addition of phorbol myristate acetate (PMA; 15 ng/mL), and
inhibition of induction by compounds is quantitated by measur-
ing decreased luciferase activity. NFκB is measured using a
truncated, NFκB dependent, E-selectin promoter (∼400 bp)
cloned into a luciferase reporter vector. This reporter is stably
transfected into the human A549 lung epithelial cell line. NFκB
activity is induced using IL-1β (0.5 ng/mL) and inhibition of
induction by compounds is quantitated by measuring decreased
luciferase activity.
Transactivation Assay. NP-1 activity is measured using a GR
ligand binding domain (GR-LBD) chimera cloned into a GAL4
Luciferase reporter system. This reporter system is stably trans-
fected into a HeLa cell line (Webster, N. J. G.; Green, S.; Jin,
J. R.; Chambon, P. The hormone-binding domains of the estrogen
and glucocorticoid receptors contain an inducible transcription
activation function. Cell 1988, 54 (2), 199-207). Response to
ligand/compound induced binding is quantitated by measuring
luciferase activity. Direct activation of the GR-LBD by com-
pounds can be measured (agonist) as increased luciferase acti-
vity, or the ability of compounds to inhibit 10 nM dexamethasone
induction can be measured (antagonist) as decreased luciferase
activity.
(R)-1-Cyclopropyl-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-
5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)ethanol (47):
4 mg, 0.008 mmol, 24% yield, 90% purity. LCMS=367.4 (Mþ
1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.91 (1H, s), 7.62-7.67
(2H, m), 7.31 (2H, t, J= 8.6 Hz), 6.44 (1H, s), 5.42 (1H, d, J=
13.7 Hz), 3.93 (1H, d, J = 14.2 Hz), 2.43-2.59 (2H, m), 1.92-
2.08 (2H, m), 1.89 (1H, dd, J=12.7 and 3.1 Hz), 1.59-1.71 (1H,
m, J = 12.8, 12.8, 12.7, and 2.8 Hz), 1.38-1.47 (1H, m), 1.33
(3H, s), 1.26 (3H, s), 1.10-1.23 (1H, m), 0.26-0.50 (4H, m).
(R)-1-Cyclohexyl-1-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-
5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)ethanol (48):
2.6 mg, 0.005 mmol, 11% yield, 90% purity. LCMS=409.2 (Mþ
1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.93 (1H, s), 7.61-7.67
(2H, m), 7.31 (2H, t, J= 8.9 Hz), 6.43 (1H, s), 5.58 (1H, d, J=
13.7 Hz), 3.87 (1H, d, J = 14.2 Hz), 2.41-2.56 (2H, m),
1.53-2.00 (9H, m), 1.35 (3H, s), 1.27 (3H, s), 1.04-1.41 (7 H, m).
(R)-1-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-1-(tetrahydro-2H-pyran-4-yl)-
ethanol (49): 6 mg, 0.01 mmol, 10% yield, 90% purity. LCMS=
411.3 (Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.94 (1H, s),
7.61-7.67 (2H, m), 7.31 (2H, t, J = 8.8 Hz), 6.45 (1H, s), 5.57
(1H, d, J = 13.8 Hz), 3.96-4.06 (2H, m), 3.88 (1H, d, J = 13.8
Hz), 3.36-3.45 (2H, m), 2.43-2.55 (2H, m), 1.93 (2H, dd, J =
11.7 and 3.4 Hz), 1.36 (3H, s), 1.30-1.75 (8H, m), 1.29 (3H, s).
(S)-1,1,1-Trifluoro-2-((5aR,6S)-1-(4-fluorophenyl)-5a-methyl-
5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquinolin-6-yl)propan-2-ol
(50): 1 mg, 0.002 mmol, 12% yield, 98% purity. LCMS=395.2
(Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.96 (1H, s), 7.64
(2H, dd, J=8.9 and 5.2 Hz), 7.28-7.34 (2H, m), 6.48 (1H, d, J=
2.0 Hz), 5.56 (1H, d, J = 13.8 Hz), 3.93 (1H, d, J = 13.8 Hz),
2.42-2.58 (2H, m), 2.18 (1H, dd, J = 12.7 and 3.4 Hz), 1.86-
1.98 (2H, m), 1.66-1.79 (1H, m), 1.51 (3H, s), 1.37-1.46 (1H,
m), 1.36 (3H, s).
(R)-2-((5aR,6S)-1-(4-Fluorophenyl)-5a-methyl-5,5a,6,7,8,9-hexa-
hydroimidazo[1,5-b]isoquinolin-6-yl)-4-methylpentan-2-ol (51):
7.8 mg, 0.016 mmol, 23% yield, 98% purity. LCMS = 383.3
(Mþ1)þ. 1H NMR (400 MHz, MeOD) δ ppm 8.91 (1H, s), 7.64
(2H, dd, J=8.6 and 5.1 Hz), 7.31 (2H, t, J=8.9 Hz), 6.43 (1H,
s), 5.39 (1H, d, J=13.7 Hz), 4.08 (1H, d, J=13.7 Hz), 2.43-2.53
(2H, m), 1.83-1.99 (2H, m), 1.56-1.80 (4H, m), 1.39 (2H, dd,
J=14.0 and 3.8 Hz), 1.34 (3H, s), 1.32 (3H, s), 1.05 (3H, d, J=
6.6 Hz), 0.97 (3H, d, J = 6.6 Hz).
(5aR,6S)-1-(4-Fluorophenyl)-6-((S)-1-methoxy-1-(thiophen-2-yl)-
ethyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquino-
line (25). To a stirred solution of (S)-1-((5aR,6S)-1-(4-fluoro-
phenyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]isoquino-
lin-6-yl)-1-(thiophen-2-yl)ethanol (24; 13 mg, 0.032 mmol) in
anhydrous dimethylformamide (0.3 mL) was added sodium
hydride (60% oil dispersion, 9 mg, 0.23 mmol) at 0 °C. The reac-
tion mixture was stirred at RT for 40 min before methyl iodide
(1 drop) was added. The reaction mixture was stirred at RT for a
further 30 min and purified using reverse phase HPLC (YMC S5
20 ꢀ 100 mm, 10 min run, solvent A: 10% MeOH/90% H2O/
0.1% TFA; solvent B: 90% MeOH/10% H2O/0.1% TFA) to
yield (5aR,6S)-1-(4-fluorophenyl)-6-((S)-1-methoxy-1-(thiophen-
2-yl)ethyl)-5a-methyl-5,5a,6,7,8,9-hexahydroimidazo[1,5-b]iso-
Acknowledgment. The authors would like to thank Dr.
David S. Weinstein for useful discussions regarding the manu-
script and Mary Ellen Cvijic, Ding Ren Shen, and Melissa
Yarde for their support to the project.
References
(1) Hayashi, R.; Wada, H.; Ito, K.; Adcock, I. M. Effects of glucocorti-
coids on gene transcription. Eur. J. Pharmacol. 2004, 500, 51–62.
(2) Caldenhoven, E.; Liden, J.; Wissink, S.; Van de Stolpe, A.; Raaijmakers,
J.; Koenderman, L.; Okret, S.; Gustafsson, J. A.; Van der Saag,
P. T. Negative cross-talk between RelA and the glucocorticoid
receptor: A possible mechanism for the antiinflammatory action of
glucocorticoids. Mol. Endocrinol. 1995, 4, 401–412.
(3) Kamei, Y.; Xu, L.; Heinzel, T.; Torchia, J.; Kurokawa, R.; Gloss,
B.; Lin, S.; Heyman, R. A.; Rose, D. W.; Glass, C. K.; Rosenfield,
M. G. A CBP integrator complex mediates transcriptional acti-
vation and AP-1 inhibition by nuclear receptors. Cell 1996, 85,
403–414.
quinoline TFA salt (15 mg, 0.028 mmol, 87% yield, 96% purity).
3

1280 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 3
Xiao et al.
(4) Chakravarti, D.; LaMorte, V. J.; Nelson, M. C.; Nakajima, T.;
Schulman, I. G.; Juguilon, H.; Montminy, M.; Evans, R. M. Role
of CBP/P300 in nuclear receptor signaling. Nature 1996, 383,
99–103.
(5) Reichardt, H. M.; Kaestner, K. H.; Tuckerman, J.; Kretz, O.;
Wessely, O.; Bock, R.; Gass, P.; Schmid, W.; Herrlich, P.; Angel,
P.; Schutz, G. DNA binding of the glucocorticoid receptor is not
essential for survival. Cell 1998, 93, 531–541.
(6) Reichardt, H. M.; Tuckerman, J. P.; Gottlicher, M.; Vujic, M.;
Weih, F; Angel, P.; Herrlich, P.; Schutz, G. Repression of inflam-
matory responses in the absence of DNA binding by the glucocor-
ticoid receptor. EMBO J. 2001, 20, 7168–7173.
(7) Newtown, R.; Holden, N. S. Separating transrepression and
transactivation. A distressing divorce for the glucocorticoid recep-
tor? Mol. Pharmacol. 2007, 72, 799–809.
Harris, G.; Kelly, T. M.; Mazur, P.; Pandit, S.; Santoro, J.; Sitlani,
A.; Wang, C.; Williamson, J.; Miller, D. K.; Thompson, C. M.;
Zaller, D. M.; Forrest, M. J.; Carballo-Jane, E.; Luell, S. Novel
N-arylpyrazolo[3,2-c]-based ligands for the glucocorticoid recep-
tor. Receptor binding and in vivo activity. J. Med. Chem. 2004, 47,
2441–2452.
(11) Shah, N.; Scanlan, T. S. Design and evaluation of novel nonster-
oidal dissociating glucocorticoid receptor ligands. Bioorg. Med.
Chem. Lett. 2004, 14, 5199–5203.
(12) Hirschmann, R.; Steinberg, N. G.; Buchschacher, P.; Fried, J. H.;
Kent, G. J.; Tishler, M. Syntheses and structures of steroidal
4-pregneno[2,3-c]pyrazoles. A novel class of potent antiinflamma-
tory steroids. J. Am. Chem. Soc. 1963, 85, 120–122.
(13) Mrozik, H.; Buchschacher, P.; Hannah, J.; Fried, J. H. Hetero-
cyclic steroids in the antiinflammatory series. J. Med. Chem. 1964,
7, 584–589.
(8) Some examples from the literature: (a) Vayssiere, B. M.; Dupont,
S.; Choquart, A.; Petit, F.; Garcia, T.; Marchandeau, C.; Gronemeyer,
H.; Resche-Rigon, M. Synthetic glucocorticoids that dissociate
transactivation and AP-1 transrepression exhibit anti-inflamma-
tory in vivo. Mol. Endocrinol. 1997, 11, 1245–1255. (b) Schacke, H.;
Schottelius, A.; Docke, W.-D.; Strehlke, P.; Jaroch, S.; Schmees, N.;
Rehwinkel, H.; Hennekes, H.; Asadullah, K. Dissociation of transacti-
vation from transrepression by a selective glucocorticoid receptor
agonist leads to separation of therapeutic effects from side effects.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 227–32. (c) Thompson, C. F.;
Quraishi, N.; Ali, A.; Tata, J. R.; Hammond, M. L.; Balkovec, J. M.;
Einstein, M.; Ge, L.; Harris, G.; Kelly, T. M.; Mazur, P.; Pandit, S.;
Santoro, J.; Sitlani, A.; Wang, C.; Williamson, J.; Miller, D. K.; Yamin,
T. D.; Thompson, C. M.; O'Neill, E. A.; Zaller, D.; Forrest, M. J.;
Carballo-Jane, E.; Luell, S. Novel heterocyclic glucocorticoids in vitro
profile and in vivo efficacy. Bioorg. Med. Chem. Lett. 2005, 15, 2163–
2167. (d) Barker, M.; Clackers, M.; Copley, R.; Demaine, D. A.;
Humphreys, D.; Inglis, G. G. A.; Johnston, M. J.; Jones, H. T.; Haase,
M. V.; House, D.; Loiseau, R.; Nisbet, L.; Pacquet, F.; Skone, P. A.;
Shanahan, S. E.; Tape, D.; Vinader, V. M.; Washington, M.; Uings, I.;
Upton, R.; McLay, I. M.; Macdonald, S. J. F. Dissociated nonsteroidal
glucocorticoid receptor modulators: Discovery of the agonist trigger in
a tetrahydronaphthalene-benzoxazine series. J. Med. Chem. 2006, 49,
4216–4231. (e) Kuzmich, D.; Kirrane, T.; Proudfoot, J.; Bekkali, Y.;
Zindell, R.; Beck, L.; Nelson, R.; Shih, C.-K.; Kukulka, A. J.; Paw, Z.;
Reilly, P.; Deleon, R.; Cardozo, M.; Nabozny, G.; Thomson, D.
Identification of dissociated non-steroidal glucocorticoidreceptor ago-
nists. Bioorg. Med. Chem. Lett. 2007, 17, 5025–5031. (f) Ray, N. C.;
Clark, R. D.; Clark, D. E.; Williams, K.; Hickin, H. G.; Crackett, P. H.;
Dyke, H. J.; Lockey, P. M.; Wong, M.; Devos, R.; White, A.; Belanoff,
J. K. Discovery and optimization of novel, non-steroidal glucocorticoid
receptor modulators. Bioorg. Med. Chem. Lett. 2007, 17, 4901–4905.
(g) Robinson, R. P.; Buckbinder, L.; Haugeto, A. I.; McNiff, P. A.;
Millham, M. L.; Reese, M. R.; Schaefer, J. F.; Abramov, Y. A.; Bordner,
J.; Chantigny, Y. A.; Kleinman, E. F.; Laird, E. R.; Morgan, B. P.;
Murray, J. C.; Salter, E. D.; Wessel, M. D.; Yocum, S. A. J. Med. Chem.
2009, 52, 1731–1743. (h) Yang, B. V.; Vacarro, W.; Doweyko, A.;
Doweyko, L.; Huynh, T.; Tortolani, D.; Nadler, S.; McKay, L.; Somerville,
J.; Holloway, D. A.; Habte, S.; Weinstein, D. S.; Barrish, J. Discovery of
novel dihyro-9,10-ethano-anthracene carboxamides as glucocorticoid
receptor modulators. Bioorg. Med. Chem. Lett. 2009, 19, 2139–2143.
(9) Schacke, H.; Berger, M.; Hansson, T. G.; McKerrecher, D.;
Rehwinkel, H. Dissociated non-steroidal glucocorticoid receptor
modulators: An update on new compounds. Expert Opin. Ther.
Pat. 2008, 18, 339 and references cited therein.
(14) Bedekar, A. V.; Watanabe, T.; Tanaka, K.; Fuji, K. A convenient
method for selective C-alkylation of 2-methyl-1,3-diketones.
Synthesis 1995, 9, 1069–1070.
(15) Schollkopf, U.; Stafforst, D.; Jentch, R. Liebigs Ann. Chem. 1977,
1167–1173.
(16) For a recent use of diethyl isocyanomethylphosphonate, see:
Etayo, P.; Badorrey, R.; Diaz-de-Villegas, M. D.; Galvez, J. A.
Asymmetric homologation of ketones: A new entry to orthogon-
ally protected (2R,4R)-piperidine-2,4-dicarboxylic acid. J. Org.
Chem. 2008, 73, 8594–8597.
(17) Crystallographic data (excluding structure factors) for compounds
24 and 31 have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication numbers
CCDC 751588 and 751589. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.; fax: þ44 1223
336033; e-mail: deposit@ccdc.cam.ac.uk). Compound 24 was
crystallized from ethyl acetate and heptane. Compound 31 was
crystallized from a solvent mixture comprising ethyl acetate,
heptane, methanol, and dichloromethane.
(18) Webster, N. J. G.; Green, S.; Jin, J. R.; Chambon, P. The hormone-
binding domains of the estrogen and glucocorticoid receptors
contain an inducible transcription activation function. Cell 1988,
54, 199–207.
(19) Suino-Powell, K.; Xu, Y.; Zhang, C.; Tao, Y.-G.; Tolbert, D. W.;
Simons, S. S., Jr.; Xu, E. H. Doubling the size of the glucocorticoid
receptor ligand binding pocket by deacylcortivazol. Mol. Cell. Biol.
2008, 28, 1915–1923.
(20) 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. The three-dimensional
structures of antagonistic and agonistic forms of the glucocorticoid
receptor ligand-binding domain. J. Biol. Chem. 2003, 278, 22748–
22754.
(21) Zhang, J.; Simisky, J.; Tsai, F. T. F.; Geller, D. S. A critical role of
helix 3-helix 5 interaction in steroid hormone receptor function.
Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 2707–2712.
(22) PR binding data for selected compounds: 19, 451 ( 75 nM; 24, 230
( 161 nM; 31, 238 ( 100 nM; 37, 132 ( 16 nM; 39, 409 ( 158 nM.
(23) For example, the in vitro metabolic stability of compounds 31 and
37 was evaluated at 3 μM in the presence of human, rat, and mouse
liver microsomes. After a 10 min incubation, the % compound
remaining for compound 31 in humans, rats, and mice was 56, 51,
and 73%, respectively; the corresponding values for compound 37
was 3, 2, and 21%, respectively.
(10) Ali, A.; Thompson, C. F.; Balkovec, J. M.; Graham, D. W.;
Hammond, M. L.; Quraishi, N.; Tata, J. R.; Einstein, M.; Ge, L.;