Identification of C-2 hydroxyethyl imidazopyrrolopyridines as potent JAK1 inhibitors with favorable physicochemical properties and high selectivity over JAK2

Article abstract of DOI:10.1021/jm4004895

Herein we report on the structure-based discovery of a C-2 hydroxyethyl moiety which provided consistently high levels of selectivity for JAK1 over JAK2 to the imidazopyrrolopyridine series of JAK1 inhibitors. X-ray structures of a C-2 hydroxyethyl analogue in complex with both JAK1 and JAK2 revealed differential ligand/protein interactions between the two isoforms and offered an explanation for the observed selectivity. Analysis of historical data from related molecules was used to develop a set of physicochemical compound design parameters to impart desirable properties such as acceptable membrane permeability, potent whole blood activity, and a high degree of metabolic stability. This work culminated in the identification of a highly JAK1 selective compound (31) exhibiting favorable oral bioavailability across a range of preclinical species and robust efficacy in a rat CIA model.

Full text of DOI:10.1021/jm4004895

Article
pubs.acs.org/jmc
Identification of C‑2 Hydroxyethyl Imidazopyrrolopyridines as Potent
JAK1 Inhibitors with Favorable Physicochemical Properties and High
Selectivity over JAK2
Mark Zak,*,† Christopher A. Hurley,¶ Stuart I. Ward,¶ Philippe Bergeron,† Kathy Barrett,‡
Mercedesz Balazs,◆ Wade S. Blair,‡ Richard Bull,¶ Paroma Chakravarty,# Christine Chang,‡
Peter Crackett,¶ Gauri Deshmukh,§ Jason DeVoss,◆ Peter S. Dragovich,† Charles Eigenbrot,○
Charles Ellwood,¶ Simon Gaines,¶ Nico Ghilardi,∥ Paul Gibbons,† Stefan Gradl,† Peter Gribling,◆
Chris Hamman,† Eric Harstad,⊥ Peter Hewitt,¶ Adam Johnson,‡ Tony Johnson,¶ Jane R. Kenny,§
Michael F. T. Koehler,† Pawan Bir Kohli,‡ Sharada Labadie,† Wyne P. Lee,◆ Jiangpeng Liao,+
Marya Liimatta,‡ Rohan Mendonca,† Raman Narukulla,¶ Rebecca Pulk,† Austin Reeve,¶ Scott Savage,▽
Steven Shia,○ Micah Steffek,○ Savita Ubhayakar,§ Anne van Abbema,‡ Ignacio Aliagas,†
Barbara Avitabile-Woo,¶ Yisong Xiao,+ Jing Yang,+ and Janusz J. Kulagowski¶
‡
§
†Departments of Discovery Chemistry, Biochemical and Cellular Pharmacology, Drug Metabolism and Pharmacokinetics,
∥Immunology, Safety Assessment, Small Molecule Pharmaceutical Sciences, ▽Small Molecule Process Chemistry, ○Structural
⊥
#
Biology, ◆Translational Immunology, Genentech, Inc., 1 DNA Way, South San Francisco, California 94080, United States
¶Argenta, 8/9 Spire Green Centre, Flex Meadow, Harlow, Essex CM19 5TR, United Kingdom
+WuXi AppTec Co., Ltd., 288 Fute Zhong Road, Waigaoqiao Free Trade Zone, Shanghai 200131, P. R. China
S
* Supporting Information
ABSTRACT: Herein we report on the structure-based
discovery of a C-2 hydroxyethyl moiety which provided
consistently high levels of selectivity for JAK1 over JAK2 to the
imidazopyrrolopyridine series of JAK1 inhibitors. X-ray
structures of a C-2 hydroxyethyl analogue in complex with
both JAK1 and JAK2 revealed differential ligand/protein
interactions between the two isoforms and offered an
explanation for the observed selectivity. Analysis of historical
data from related molecules was used to develop a set of
physicochemical compound design parameters to impart
desirable properties such as acceptable membrane perme-
ability, potent whole blood activity, and a high degree of metabolic stability. This work culminated in the identification of a highly
JAK1 selective compound (31) exhibiting favorable oral bioavailability across a range of preclinical species and robust efficacy in a
rat CIA model.
to treat RA while minimizing JAK2 inhibition could improve
the therapeutic index by avoiding anemia.7 As such, we wished
to identify potent and orally bioavailable small molecule
inhibitors of JAK1, with selectivity over JAK2, as potential
therapies for RA and other immunologic disorders.8,9
Our group has previously reported on the discovery and
optimization of the imidazopyrrolopyridine series of JAK1
inhibitors, including the observation that a methyl group at the
C-2 position may lead to improvements in JAK1 selectivity over
JAK2.7 One particularly striking example of the C-2 methyl
effect is exemplified by the matched pair comparison of
compounds 2 and 3 (Figure 1a). As depicted in Figure 1b, the
INTRODUCTION
■
The Janus kinases (JAK1, JAK2, JAK3, and TYK2) are
intracellular protein tyrosine kinases with essential roles in
immune function,1 inflammation,2 and hematopoiesis.3 Tar-
geted inhibition of the JAKs has shown promise against
associated hematopoietic and immunologic diseases, as
evidenced by recent approvals of the multi-JAK inhibitors
(R)-3-(4-(7H-pyrrolo[2,3-d]pyrimidin-4-yl)-1H-pyrazol-1-yl)-
3-cyclopentylpropanenitrile (INCB018424)4 and 1 (CP-
690,550)5 (see Table 1) for treatment of myelofibrosis and
rheumatoid arthritis (RA), respectively.6 Despite the clinical
success of these agents, there may be benefit in identifying
inhibitors with greater specificity for a single JAK isoform,
particularly for immunologic diseases such as RA. As previously
described, we believe selective JAK1 inhibition will be sufficient
Received: April 4, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. (a) Matched pair analysis of C-2 H (2) and C-2 Me (3) imidazopyrrolopyridines. (b) X-ray structure of 3 in complex with JAK1 (gray,
PDB 4EHZ) overlaid with a key residue difference with JAK2 (green, D939 from structure 4F09). Superscript letters indicate the following: (a)
arithmetic mean of at least three separate runs (n ≥ 3); (b) the average coefficient of variation was less than 0.3 times the mean for biochemical
assays;13 (c) JAK2 Ki/JAK1 Ki; (d) apparent permeability in MDCK transwell culture. A:B, apical-to-basolateral.
amplification of JAK1 selectivity may potentially be explained
by the differential proximity of the methyl group in compound
3 to a key residue difference between JAK1 (E966, 3.4 Å) and
JAK2 (D939, 5.1 Å). Despite this intriguing observation, the C-
2 methyl group did not lead to consistently high levels of JAK1
selectivity. The nature of the group extending beneath the P-
loop also played an important, yet poorly understood role, and
groups other than the 4-piperidine present in 3 generally
reduced JAK1 selectivity. Unfortunately, molecules containing
the selectivity-enhancing 4-piperidine group possessed charac-
teristics not compatible with orally administered therapeutics,
including relatively weak target potency and poor membrane
permeability. As part of our continued investigation of the
imidazopyrrolopyridine series, we wished to discover molecules
with JAK1 selectivity equal to or greater than 3 with improved
MDCK permeability and JAK1 potency.
however, offered a dramatic breakthrough in that it exhibited
greater JAK1 selectivity than compound 3 while maintaining
biochemical and cell-based potency equivalent to 1. Further
branching to give tertiary alcohol 9 was poorly tolerated from a
potency perspective, while homologation of the secondary
alcohol to produce compounds 10 and 11 led to no
improvement relative to parent compounds 7 and 8. Finally,
chiral diols 12 and 13 were found to maintain potency but lose
selectivity compared to 7 and 8. Excited by the highly ligand
efficient and predictably selective nature of the C-2 substituted
hydroxyls, we evaluated the most promising analogues in
additional assays.
The effect of plasma proteins and other blood components
on cell-based JAK1 inhibition was assessed by comparing the
potencies of multiple compounds in the standard low serum
cell-based assay (IL-6 pSTAT3) to the potencies in human
whole blood.13 As shown in Table 2, compound 1 maintained
potent JAK1 cell-based inhibition in human whole blood, with
no potency shift between the two assay formats. By contrast,
several of the other compounds tested were much less potent in
human whole blood, with compounds 6 and 11 notably shifting
by greater than 10-fold. Examination of the limited data set in
Table 2 indicated that there was a relationship between
cLogD7.414 and the magnitude of potency loss in human whole
blood for the imidazopyrrolopyridine series. The more
lipophilic analogues 6, 8, and 11 showed larger shifts than
the less lipophilic analogues 5, and 13.
Additional relationships between the physical or structural
properties of compounds and experimental outcomes in
relevant assays are also noted in Table 2. MDCK permeability
was very good for the monohydroxylated compounds but
dropped slightly for the diol 13, consistent with that
compound’s significant increase in TPSA.15 All compounds
were found to show good to moderate in vitro stability in the
presence of human liver microsomes (HLMs), however
stability in human hepatocytes varied widely.16 A relationship
was noted between the degree of branching surrounding the
alcohol and human hepatocyte stability. The primary alcohols 5
and 6 were found to be highly unstable in hepatocytes, while
the secondary and tertiary alcohols 8 and 9 exhibited moderate
and very good stability, respectively. We believe the contrasting
hepatocyte stability was related to the alcohols’ abilities to form
phase two conjugates. Indeed, hepatocyte metabolism identi-
fication studies on analogues related to primary alcohol 5 and
secondary alcohol 8 revealed significant glucuronidation of the
RESULTS AND DISCUSSION
■
With the goal of achieving greater and more consistent
selectivity for JAK1 over JAK2, we prepared a series of C-2
substituted alcohols (5−13, Table 1) designed to engage in
differential H-bonding interactions between E966 (JAK1) and
D939 (JAK2). Alcohols were chosen due to their modest
impact on parameters such as molecular weight, TPSA, and
basicity relative to other potential H-bonding groups. We
believed these attributes would allow for molecules to contain a
broad spectrum of moieties extending beneath the P-loop while
maintaining physicochemical parameters consistent with orally
dosed therapeutics.10 The initial SAR exploration of C-2
alcohols was carried out in combination with a cyclohexyl P-
loop group due to improved potency and MDCK permeability
relative to corresponding 4-piperidine analogues. The marketed
drug 1 and the C-2 methyl imidazopyrrolopyridine 47 are also
included in Table 1 for comparison. Compound 4 exhibited
levels of JAK1 potency and ligand efficiency11,12 comparable to
1, yet only modest improvement to JAK1 vs JAK2 selectivity in
both biochemical and cell-based assays. Gratifyingly, the
hydroxylated analogue 5 maintained equivalent potency with
improved JAK1 selectivity, supporting the hypothesis that a
hydrogen bond donor at the C-2 position could better exploit
the E966−D939 residue difference. The straight chain
homologated analogue 6 was slightly less potent and selective,
while the (S)-isomer (7) of the branched homologue subtly
improved JAK1 selectivity relative to 5 but suffered from
notably reduced potency. The corresponding (R)-isomer (8),
B
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. JAK1 Potency and Selectivity of 1 and C-2 Substituted Imidazopyrrolopyridines 4−13
a
b
c
Ligand efficiency11 = (−1.4log Ki)/(n heavy atoms). Ligand-lipophilicity efficiency12 = (−log Ki) − (cLogP). Arithmetic mean of at least three
separate runs (n ≥ 3).13 The average coefficient of variation was less than 0.3 times the mean for biochemical assays. JAK2 Ki/JAK1 Ki. JAK1 cell-
d
e
f
based assay.13 The average coefficient of variation was less than 0.5 times the mean for cell-based assays. JAK2 cell-based assay.13 (EPO pSTAT5
g
h
i
j
k
l
EC50)/(IL-6 pSTAT3 EC50). Not applicable. Single enantiomers. Known stereochemistry. Single enantiomers. Stereochemistry inferred from
potency relationships between 7 and 8.
Table 2. Additional Data for 1 and C-2 Substituted Imidazopyrrolopyridines
IL-6 pSTAT3 EC50
whole blood EC50
whole blood
TPSA
f
MDCK Papp A:B
g
HLM, HHep CLH
h
a,
b
a,
c
d
e
ex
(nM)
(nM)
shift
cLogD7.4
(Å2)
(× 10−6 cm/s)
(mL min−1 kg−1
)
1
53
64
43
0.8
4.9
12
1.5
1.7
2.1
2.1
2.6
2.6
0.9
88
66
66
66
66
66
86
5
313
1500
330
24
23
23
19
22
16
5.6, 19
5.3, 17
4.5, 11
5.8, 1.7
7.6, 12
1.8, 6.8
6
130
47
8
7.0
9
330
84
11
13
1600
220
19
69
3.2
a
b
Arithmetic mean of at least three separate runs (n ≥ 3). JAK1 cell-based assay.13 The average coefficient of variation was less than 0.5 times the
mean for cell-based assays. JAK1 cell-based assay (IL-6 pSTAT3) run in the presence of whole blood.13 (whole blood EC50)/(IL-6 pSTAT3 EC50).
c
d
e
f
g
See ref 14. See ref 15. Apparent permeability in MDCK transwell culture. A:B, apical-to-basolateral. Mean of at least two separate runs (n ≥ 2).
h
Hepatic clearance in the presence of human liver microsomes and human hepatocytes.16 Mean of at least two separate runs (n ≥ 2).
C
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Defining the ideal calculated property space of C-2 H and Me imidazopyrrolopyridines.a All compounds analyzed were synthesized prior to
the imidazopyrrolopyridines in Tables 1, 2, and 4. Superscript letters indicate the following: (a) compound structures and tabulated data are listed in
the Supporting Information; (b) (whole blood EC50)/(IL-6 pSTAT3 EC50);13 (c) see ref 14; (d) oral bioavailability in the rat; (e) apparent
permeability in MDCK transwell culture; A:B, apical-to-basolateral; (f) in vivo clearance in the rat after IV administration; (g) see ref 15; (h)
experimental hepatic clearance in the presence of human liver microsomes;16 (i) prospectively calculated probability of being stable in the presence
of human liver microsomes.17
former but minimal glucuronidation of the latter (data not
shown). Although greater branching led to better agreement
between the two in vitro metabolic stability assay formats, no
compound became significantly more stable in the presence of
human hepatocytes than in the presence of liver microsomes.
Therefore, a high degree of HLM stability was desired as a
prerequisite to acceptable human hepatocyte stability.
The emerging trends noted in the limited data set presented
in Table 2 prompted us to perform more extensive properties
analyses on existing compounds to ascertain whether an ideal
range of calculated properties could be identified to balance low
whole blood shift with acceptable membrane permeability and
HLM stability. We have previously reported on C-2 H and
methyl substituted imidazopyrrolopyridine compounds (Figure
2, R = H or Me),7,9 many of which were synthesized prior to
initiation of the current work. As shown in Figure 2a, analysis of
this historical data set revealed that compounds with cLogD7.4
greater than 2 possessed a high probability of exhibiting large
whole blood shifts (>5×), whereas compounds with cLogD7.4
<2 were more likely to fall into the low (<3×) or moderate
(3×−5×) shift ranges.
Figure 2b illustrates the relationships between rat oral
bioavailability and MDCK permeability in the C-2 H and
methyl substituted imidazopyrrolopyridine series. It appeared
that MDCK Papp (A:B) was an important determinant of oral
bioavailability. MDCK Papp values of >3 × 10−6 cm/s (A:B)
often produced compounds falling into the high (>50%) or
moderate (30−50%) bioavailability categories, while com-
pounds possessing MDCK Papp (A:B) below than that value
were often found to cluster in the low bioavailability class (%F
< 30%). TPSA and cLogD7.4 were found to be useful
computational predictors of acceptable MDCK permeability.
Figure 2c shows that compounds with TPSA values of >90 Å2
or cLogD7.4 <0.5 were less likely to demonstrate acceptable
MDCK permeability (Papp A:B >3 × 10−6 cm/s) than those
with TPSA <90 Å2 and cLogD7.4 >0.5.
As expected, high in vivo clearance could also erode oral
bioavailability. Indeed, the compound in the bottom right
quadrant of Figure 2b possessed extremely good MDCK
D
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
permeability yet exhibited low bioavailability (presumably due
to very high in vivo CL). In an effort to control in vivo CL, we
wished to identify compounds with a high degree of metabolic
stability in the presence of HLMs. We evaluated the ability of a
computational model trained on a set of experimental HLM
data to prospectively predict the probability of HLM stability of
new compounds in the imidazopyrrolopyridine series.17 Figure
2d shows the correlation between the model’s predicted
probability of HLM stability and the experimental CLH values
obtained by incubating compounds with HLMs. Indeed, the
model indicated that imidazopyrrolopyridine compounds with
calculated probabilities <0.5 were likely to possess poor
metabolic stability, i.e., high (>15 mL min−1 kg−1) or high−
moderate (10−15 mL min−1 kg−1) HLM CLH. By contrast,
calculated probabilities of >0.5 resulted in an enriched
proportion of molecules with good metabolic stability, i.e.,
low (<6.2 mL min−1 kg−1) or low−moderate (6.2−10 mL
min−1 kg−1) experimental HLM CLH values.
unlikely. As we have previously reported,7,9 many N-substituted
4-piperidine P-loop groups exhibited high levels of JAK1
potency in the imidazopyrrolopyridine series and, given the
ease of synthesis, appeared to offer an attractive means to tune
physicochemical properties. Unfortunately, we found it difficult
to identify substituted 4-piperidines with acceptable profiles
when combined with the C-2 (R)-hydroxyethyl selectivity
group. For example, compound 14 (Table 4) fell within the
desired calculated properties ranges but suffered from a large
whole blood shift. A related analogue (15) possessed
reasonable JAK1 potency in the biochemical and whole blood
assays but fell outside the ideal cLogD7.4 and TPSA ranges and
suffered from low MDCK permeability. Although multiple
other N-substituted 4-piperidine compounds were synthesized
(data not shown), none balanced acceptable JAK1 potency with
low whole blood shift and acceptable in vitro and in vivo
DMPK properties. The much more synthetically complex gem-
difluoro piperidine 16 was somewhat more promising in
balancing reasonable JAK1 potency and selectivity with the
other experimental parameters listed in Table 4. N-substituted
3-piperidine groups were not successful (e.g., 17 and 18), and
similar to the 4-piperidine isomers, it was not possible to
balance potency with the various other calculated and
experimental parameters. Tetrahydropyran (THP) P-loop
groups possessed ideal calculated properties, however, the 4-
regioisomer 19 and the (R)-epimer of the 3-regioisomer (20)
were disappointingly weak biochemical JAK1 inhibitors. In
contrast, the (S)-epimer of the 3-THP regioisomer (21) was far
more biochemically active, as well as being highly selective for
JAK1 over JAK2 (>25× and >50× in biochemical and cell-
based assays, respectively) and possessing good MDCK
permeability and excellent metabolic stability. In addition, a
very low whole blood shift allowed 21 to be the first C-2
hydroxyethyl compound to exhibit a whole blood JAK1 cellular
EC50 of <200 nM. A series of substituted cyclohexane and
cyclopentane P-loop groups was then examined. The cis and
trans 4-cyclohexanols 22 and 23 exhibited modest biochemical
and cell-based JAK1 potency, while the optimal isomer7 of the
3-cyclohexanol (24) showed improved potency, excellent JAK
selectivity, and a low whole blood shift. Unfortunately, despite
possessing cLogD7.4 and TPSA values within the optimal
calculated range, the MDCK permeability of 24 was low,
potentially due to the presence of a third hydrogen-bond donor
(HBD). Indeed, removal of the third HBD in the trans 4-
methoxy analogue 25 led to much improved MDCK
permeability. We previously found in the C-2 Me series that
a 3-methoxy substituted cyclohexane P-loop group was not
tolerated (data not shown), therefore a similar analogue was
not synthesized in the C-2 (R)-hydroxyethyl series. Instead, we
prepared the shielded 4-amino cyclohexanes 26−28. The
strongly electron withdrawing and lipophilic 2,2,2-trifluoroethyl
group allowed compound 26 to achieve excellent MDCK
permeability despite the presence of the third HBD. Gratify-
ingly, 26 also balanced robust JAK1 selectivity (>15×
biochemical, >25× cell-based) and excellent metabolic stability
with very potent whole blood JAK1 inhibition, comparable to
that of 1 (Table 2). The 2,2-difluoroethyl analogue 27 was also
highly potent and even more selective than 26, but it suffered
from reduced MDCK permeability. The cis 1,4-diamino
cyclohexane 28 suffered from reduced potency and decreased
metabolic stability relative to the trans version 26. Unlike the
matched pair of cis and trans cyclohexanes 26 and 28, the
corresponding cyclopentanes 29 and 30 exhibited similar
The correlations identified in Figure 2 for the C-2 H and Me
imidazopyrrolopyridine series are summarized in Table 3. A
Table 3. Compound Design Parameters for C-2
Hydroxyethyl Imidazopyrrolopyridines
target
value
calcd property
rationale
a
d
e
cLogD7.4
0.5−2.0
≤90 Å2
>0.5
balance MDCK Papp with whole blood shift
b
d
TPSA
MDCK Papp
f
HLM
human liver microsome stability
c
probability
a
b
c
See ref 14. See ref 15. Prospectively calculated probability of being
stable in the presence of human liver microsomes.17 Apparent
d
permeability in MDCK transwell culture; A:B, apical-to-basolateral.
e
f
(whole blood EC50)/(IL-6 pSTAT3 EC50).13 Experimental in vivo
stability in the presence of human liver microsomes.
cLogD7.4 range of 0.5−2.0 was desirable. Compounds with
lower values often suffered from poor MDCK permeability,
whereas analogues with higher values typically exhibited large
whole blood potency shifts. An additional determinant of
MDCK permeability was TPSA, with values of >90 Å2 often
associated with unacceptable permeability. Finally, compounds
with calculated HLM stability probability <0.5 were found to be
unlikely to possess satisfactory experimental metabolic stability
in the presence of HLMs. These correlations in the C-2 H and
Me series were used to define a set of compound design
parameters for further optimization of the consistently JAK1
selective C-2 hydroxyl series. Among the C-2 hydroxyls, the
(R)-hydroxyethyl analogue 8 appeared to offer the most
promise for optimization. In addition to possessing the highest
levels of JAK1 potency and selectivity of the monohydroxylated
analogues (Table 1), its TPSA was substantially lower than that
of diol 13, allowing for many more acceptable combinations
with P-loop groups to be designed. As such, the C-2 (R)-
hydroxyethyl moiety was used for all further optimization work.
The results of the P-loop optimization campaign are shown
in Table 4. All compounds in Table 4 possessed calculated
probabilities of HLM stability of ≥0.6 and, as expected from the
analysis of the C-2 H and Me imidazopyrrolopyridines (Figure
2d), the majority of them exhibited either low or moderate
experimental HLM CLH. Human hepatocyte CLH values were
generally in good agreement with the corresponding HLM
values, indicating that significant metabolism by processes such
as phase 2 conjugation or aldehyde oxidase activity was
E
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 4. SAR of C-2 (R)-Hydroxyethyl Imidazopyrrolopyridines
F
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 4. continued
a
b
c
Ligand efficiency11 = (−1.4log Ki)/(n heavy atoms). Ligand-lipophilicity efficiency12 = (−log Ki) − (cLogP). Arithmetic mean of at least three
d
e
f
separate runs (n ≥ 3). The average coefficient of variation was less than 0.3 times the mean for biochemical assays. JAK2 Ki/JAK1 Ki. JAK1 cell-
based assay.13 The average coefficient of variation was less than 0.5 times the mean for cell-based assays. (EPO pSTAT5 EC50)/(IL-6 pSTAT3
g
EC50). JAK1 cell-based assay (IL-6 pSTAT3) run in the presence of whole blood.13 (whole blood EC50)/(IL-6 pSTAT3 EC50). Hepatic clearance
h
i
j
in the presence of human liver microsomes and human hepatocytes.16 Mean of at least two separate runs (n ≥ 2). See ref 14. See ref 15. Apparent
k
l
m
n
permeability in MDCK transwell culture. A:B, apical-to-basolateral. Mean of at least two separate runs (n ≥ 2). Single, unknown 4-piperidine
epimer.
Figure 3. (a) X-ray structure of 31 in complex with JAK1 (PDB 4IVB, resolution = 1.9 Å). (b) X-ray structure of 31 in complex with JAK2 (PDB
4IVA, resolution = 1.5 Å). A key residue difference between JAK1 (E966) and JAK2 (D939) is highlighted in green. Notable crystallographic waters
are highlighted as red spheres. Hydrogen-bonding interactions are denoted as dashed lines. The electron density of the JAK1 structure was consistent
with two main conformations of the P-loop. For the sake of clarity, only one of these is depicted.
attributes. Both were promising, although the cis analogue 29
possessed better JAK1 selectivity. The final analogues examined
in combination with the C-2 (R)-hydroxyethyl selectivity group
were the cyano substituted cyclohexanes 31−34. While residing
at the upper limit of acceptable TPSA, these compounds
contained only two HBDs, and were thus considered worthy of
preparation. Although possessing borderline MDCK perme-
ability and somewhat reduced potency in the cell-based JAK1
assay relative to the optimal trifluoroethyl substituted amines
(e.g., 26 and 29), compound 31 exhibited a low whole blood
shift and, thus, maintained good whole blood potency. Notably,
31 also exhibited very high levels of JAK1 selectivity over JAK2
(>35× biochemical, >60× cell-based). The corresponding cis
isomer 32 was slightly more potent, although at the cost of
reduced JAK1 selectivity. The singly and doubly homologated
trans isomers 33 and 34 were extremely potent in all
biochemical and cell-based measures of the JAK1 inhibition
(including in human whole blood), although JAK1 selectivity
was reduced relative to 31. One final note on the inhibitors in
Table 4 is that a high degree of ligand efficiency was maintained
throughout the optimization effort, with the majority of sub-5
nM Ki compounds maintaining LE ≥ 0.45 and LLE ≥ 7.0.
As shown in Figure 3, X-ray crystal structures of the highly
JAK1 selective compound 31 were generated in complex with
both JAK1 and JAK2. Compound 31 adopts broadly similar
binding modes in the presence of each JAK isoform, with the
pyrrolopyridine moiety contacting the hinge residues as
expected, and the free lone pair of the imidazole ring interacting
with a crystallographic water. Also similar between the two
structures, the cyano group of the ligand does not interact
directly with the P-loop but rather extends beneath it and likely
participates in polar interactions with nearby crystallographic
waters. An important difference between the two structures is
observed in the region surrounding the key JAK1 (E966) vs
JAK2 (D939) residue difference targeted by the C-2 hydroxy
moiety. In the case of JAK1, the C-2 hydroxyethyl group forms
a favorable direct H-bonding interaction with E966 at an
optimal distance of 2.8 Å. The distance between the ligand and
the corresponding residue in JAK2 (D939) is 4.3 Å, too far for
a direct interaction. To satisfy the H-bonding partners of the
ligand and D939, a bridging, and presumably entropically
disfavored, water molecule is incorporated into the inhibitor−
protein complex. We believe the suboptimal interaction
between the C-2 hydroxyethyl group of the ligand and D939
of the protein produces the observed reduction in JAK2
potency and, hence, the excellent JAK1 selectivity.
Although the physicochemical properties of compound 31
allowed it to exhibit greater whole blood potency than the
G
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. (a−c) X-ray structures of compounds 33, 34, and 1, respectively, in complex with JAK1. PDB codes and resolutions are as follows: 4IVC,
2.4 Å (a); 4IVD, 1.9 Å (b); 3EYG,18 1.9 Å (c). A key residue difference between JAK1 (E966) and JAK2 (D939) is highlighted in green. Notable
crystallographic waters are highlighted as red spheres. Hydrogen-bonding interactions are denoted as dashed lines. (d) Superposition of the ligands
from the X-ray structures of 1,18 31, 33, and 34 in complex with JAK1.
unsubstituted cyclohexane 8, the cyano group was found to be
detrimental to intrinsic JAK1 biochemical potency (JAK1 Ki: 31
= 1.9 nM, 8 = 0.8 nM). By comparison, both the singly and
doubly homologated nitriles 33 and 34 exhibited improved
JAK1 biochemical inhibition relative to the unsubstituted
cyclohexane 8. Crystal structures of 33 and 34 in complex
with JAK1 were generated (Figure 4a,b) and compared to those
of both 31 (Figure 3a) and the published structure of 1 (Figure
4c).18 All structures revealed very similar interactions with the
hinge region of JAK1, with the imidazopyrrolopyridine (31, 33,
34) or pyrrolopyrimidine (1) cores contacting E957 and L959
and interacting with a crystallographic water molecule. The
interaction between the hydroxyethyl moieties of 31, 33, and
34 and E966 were likewise found to be very similar. However,
the overlay of the four JAK1 structures revealed a notable
difference between the highly potent biochemical inhibitors (1,
33, 34) and the less potent inhibitor 31 (Figure 4d). The
favorable interactions between the nitrile of 1 and the P-loop
have previously been described.18,19 Nitriles 33 and 34 adopt
conformations comparable to 1, extending up toward the P-
loop and likely participating in similar favorable interactions. In
contrast, and as noted above, the nitrile of 31 does not interact
directly with the P-loop, but instead extends in an
approximately perpendicular direction below it. This difference
in nitrile placement may explain the reduced biochemical
potency of 31 relative to 1, 33, and 34.
In vivo DMPK and associated data for notable C-2
hydroxyethyl compounds are depicted in Table 5. All
compounds possessed uniformly high levels of pH 2 solubility,
while pH 7.4 solubility was more variable and appeared to be
influenced by lipophilicity. Indeed, the compounds with the
lowest cLogD7.4 values (21 and 31) possessed the best pH 7.4
solubility. Additionally, all compounds tested possessed low to
moderate plasma protein binding across species. Similar to
human hepatocytes, the in vitro rat and dog hepatocyte CLH
values were found to be in the low to moderate range. The
compounds were generally less stable in the presence of
monkey hepatocytes, with typical in vitro CLH values in the
moderate to high range. The in vivo PK profiles of the tricyclic
compounds generally compared favorably with 1. Moderate to
high IV clearance was observed in the rat, yet oral exposure was
better than expected, with the ratio of experimental
bioavailability (F) to maximum theoretical bioavailability
(Fmax),20 typically larger than one. In vivo dog clearance was
typically moderate, although compounds 21 and 34 fell into the
low and high ranges, respectively. Similar to the rat, dog oral
exposure was generally greater than would be expected on the
basis of IV clearance (i.e., the ratio of F to Fmax was typically
>1). A definitive explanation for the unexpectedly good oral
exposure in rat and dog has not been established to date. The
compounds profiled in monkey PK exhibited moderate IV
clearance and lower F/Fmax ratios than in other species. Indeed,
the oral bioavailabilities of compounds 21 and 31 were very
H
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 5. In Vivo PK Data for 1 and Selected C-2 (R)-Hydroxyethyl Imidazopyrrolopyridines
MDCK Papp
A:B, B:A
a
d
e
f
f
g
g
h
i
aq sol, pH 2,
(× 10−6
PPB
Hep CLH
CL
T1/2
(h)
AUC
CMax
F
Fmax
b
c
Ex
pH 7.4
cm/s)
species
(%)
(mL min−1 kg−1
)
(mL min−1 kg−1
)
(μM·h)
(μM)
(%)
(%)
F/Fmax
j
1
R
62
19
18
0.6
1.2
2.1
0.58
1.9
0.28
1.3
27
78
48
11
39
59
2.5
2.0
0.8
D
M
2.0
1.0
21
26
29
31
33
34
>4000, >4000
3900, 130
7.6, 8.6
6.7, 10
10, 18
R
39
61
54
2.3
4.9
21
22
2.1
19
1.5
5.7
1.1
10.7
65.7
4.3
5.2
9.8
1.8
197
111
70
69
93
56
2.9
1.2
1.3
D
M
R
56
55
65
11
2.3
26
36
13
0.5
1.7
1.4
9.9
0.96
2.1
54
49
58
1.1
2.4
D
M
139
>4000, 70
R
49
59
56
14
9.8
27
55
14
1.1
1.2
1.0
9.0
0.98
3.0
61
29
55
2.1
2.4
D
M
131
>4000, 1900
>4000, 130
>4000, 16
2.7, 4.7
3.5, 4.9
4.7, 9.3
R
38
47
60
3.7
3
60
16
18
0.5
2.2
1.7
0.94
8.3
0.42
2.1
52
119
65
14
48
59
3.7
2.5
1.1
D
M
2.4
3.9
0.63
R
63
63
70
15
12
25
38
17
0.5
1.8
1.3
0.93
2.5
49
54
45
0.9
4.1
D
M
11.2
183
R
68
66
76
28
11
36
126
29
0.6
1.3
0.8
0.42
4.3
0.22
1.3
50
125
8
0
6
D
M
20
29
0.26
0.07
33
0.2
a
b
Thermodynamic solubility (reported in μM) of solid powder in aqueous buffer. Apparent permeability in MDCK transwell culture. A:B = apical-
c
to-basolateral, B:A = basolateral-to-apical. Mean of at least two separate runs (n ≥ 2). Abbreviations as follows: R = rat, D = dog, M = monkey.
d
e
f
Plasma protein binding. In vitro hepatic clearance in the presence of hepatocytes.16 Mean of at least two separate runs (n ≥ 2). In vivo parameters
after IV dosing. CL = clearance; T1/2 = half-life. Compounds 21, 26, 29, 31, 33, and 34 were formulated as solutions in PEG400/citrate buffer (pH
5) and dosed at 0.4 mg/kg. Data are the average from three separate animals. In vivo parameters after PO dosing. AUC = area under the curve, CMax
g
= maximum concentration. Compounds 21, 26, 29, 31, 33, and 34 were formulated as suspensions in MCT and dosed at 2 mg/kg. Data are the
h
i
j
average from three separate animals. Experimental oral bioavailability. Theoretical maximum achievable oral bioavailability.20 All values obtained or
calculated from ref 5. PO parameters scaled to 2 mg/kg.
close to what would be predicted based on IV clearance (F/Fmax
∼ 1). Compound 21 possessed the best rat and dog in vivo PK
of all compounds tested, with IV clearance and oral exposure
values superior to those reported for 1. Monkey PK, however,
was distinctly different and all IV and oral parameters were less
favorable than in the rat or dog. It is notable that of all species
tested, 21 was least stable in monkey hepatocytes and further
that monkey hepatocytes were very predictive of in vivo
clearance. It is possible that the moderate monkey PK
parameters may be due at least in part to the intrinsically
poorer metabolic stability of 21 in the monkey relative to other
species. The most JAK1 selective compound in this work (31)
exhibited moderate, yet consistent PK across all species tested,
and compared favorably to the preclinical PK reported for 1.
The off target kinase activities of multiple C-2 hydroxyethyl-
containing compounds were assessed in a panel of 285 kinases
(Table 6). Each compound was tested at a low (0.01−0.15 μM)
and high (0.1−1.0 μM) concentration. The concentrations
were chosen based on biochemical potency against JAK1, with
more potent compounds tested at lower concentrations. The
JAK family kinases were included as controls in every panel
tested and, as expected, each compound demonstrated virtually
complete (>90%) inhibition of JAK1 at both concentrations
tested. All compounds appeared to be relatively selective for the
JAK family of kinases over the broader kinome at the lower
concentration. However, at the higher concentration, the
compounds began to differentiate significantly, with 29 and
31 showing the highest levels of selectivity. Indeed, at a 1 μM
test concentration (≥500 × JAK1 Ki), the only non-JAK
kinases inhibited by >50% were LRRK2 (72%) by compound
29 and CSF1R (75%) and the V561D mutant form of PDGFR-
α (55%) by compound 31.
Other off-target activities of these compounds are
summarized in Table 7. In biochemical assays, all compounds
other than 1 were much less potent against JAK3 than JAK1,
while the differential between TYK2 and JAK1 inhibition was
smaller. No cellular measure of JAK3 inhibition independent
from JAK1 was available, however, cellular TYK2 inhibition was
measured in the IL-12 pSTAT4 assay. There was a consistently
larger cell shift between the biochemical and cell-based
measures of TYK2 inhibition relative to JAK1, thus compounds
exhibited greater cellular selectivity for JAK1 vs TYK2 than
would be expected based on their biochemical potencies. hERG
inhibition was not a concern for the majority of the compounds
I
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 6. Inhibition of a 285-Member Kinase Panel by Imidazopyrrolopyridine JAK1 Inhibitors
a
Off-target kinase inhibition assessed by SelectScreen Kinase Profiling Services (Invitrogen-Life Technologies, Madison, WI, USA). Each column
contains 285 cells (one kinase per cell). The cells are colored as follows: green, < 50% inhibition; yellow, 50−75% inhibition; red, >75% inhibition;
white, not tested. Identities of all kinases tested and corresponding % inhibition data may be found in the Supporting Information.
tested, with typical IC50 values of >100 μM being observed in a
functional patch clamp assay. The nitrile-containing analogues
31, 33, and 34 showed consistent, yet weak reversible inhibition
of CYP3A4 using a midazolam probe (IC50 = 6.7−8.8 μM),
however, the remaining isoforms were not inhibited up to the
highest concentration tested (10 μM). Finally, none of the
compounds tested in Table 7 showed time dependent
inhibition of any CYP isoform tested (3A4, 2C9, 2D6, 2C19,
1A2).
Because of multiple favorable features including high
selectivity for JAK1 over the other JAK family members and
the broader kinome, potent inhibition of cell-based JAK1
activity in human whole blood, high aqueous solubility, and
consistent cross-species preclinical PK comparing favorably
with 1, compound 31 was further progressed to a collagen
induced arthritis (CIA) efficacy study in the rat. As shown in
Figure 5a, compound 31 exhibited approximately dose
proportional increases in plasma exposure when dosed orally
in rats over a range of 20−500 mg/kg. These oral exposures
were also close to dose proportional with the low dose rat PK
study (2 mg/kg PO, Table 5). The oral exposures reported in
Figure 5a did not change appreciably between day 10 of the
CIA efficacy study, when the first daily dose of 31 was
administered, and day 20 of the study, when the eleventh and
final daily dose of 31 was delivered.
The results from the CIA efficacy model are shown in Figure
5b. Collagen-immunized animals receiving only the corn oil
vehicle exhibited marked increases in ankle thicknesses due to
arthritic disease (gray line), while robust suppression of
inflammation was observed even in the group treated with
the lowest dose of 31 (20 mg/kg). The mid-dose of 31 (100
mg/kg) fully suppressed inflammation and maintained ankle
̈
thicknesses equivalent to those of naive animals untreated with
collagen. Full suppression of inflammation was likewise
J
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 7. Other Off-Target Inhibition of Selected Analogues
Scheme 1. Preparation of Dianiline Intermediate 38
reversible CYP
hERG
(IC50, μM)
d
JAK3 K TYK2 K
IL-12 pSTAT4
b
IC
3A4(mid) ,
ai
a i
50 c
e
ex
(nM)
(nM)
EC50 (nM)
(μM)
other
1
0.4
110
76
7.3
31
710
>5800
310
21
26
29
31
33
34
>100
>100
>100
>100
>100
20
>10, >10
>10, >10
>10, >10
7.9, >10
8.8, >10
6.7, >10
3.4
20
a
53
2200
5600
100
Reagents and conditions: (i) DIPEA, i-PrOH, 80 °C; (ii) H2, Pd/C,
EtOH, THF, 50 °C.
280
5.9
7.8
12
0.4
1.7
980
further reduced with sodium borohydride to produce
homologated alcohol 6b. Enantiopure secondary alcohols 7a
and 8a were accessed by engaging 38 with the imidates derived
from (R)-(+)-lactamide or (S)-(−)-lactamide, respectively.
Oxidation of the secondary alcohol 7a and reaction of the
resulting ketone (9a) with methylmagnesium bromide afforded
tertiary alcohol 9b. Removal of the phenylsulfonyl protecting
groups present in 5a, 6b, 7a, 8a, and 9b under basic hydrolysis
conditions afforded the corresponding final products 5−9.
Treatment of dianiline 38 with triethyl orthoformate led to C-2
unsubsituted imidazole 39, which could be selectively
deprotonated with (tmp)2Mg·2LiCl24 and trapped by either
propionaldehyde or 2-(benzyloxy)acetaldehyde to provide
racemic secondary alcohols 10a and 12a, respectively. Removal
of the phenylsulfonamide protecting groups afforded free
azaindoles 10b and 12b, which were further separated into their
corresponding enantiopure components (10b → 10 + 11, 12b
→ 12c + 13c) via chiral preparative SFC. Removal of the
benzyl groups present in 12c and 13c under standard
conditions afforded diols 12 and 13.
As shown in Scheme 3, the aryl chloride present in 35 was
similarly displaced with additional amines25 14a−17a, 19a−
24a, 26a, and 28a−34a, then the nitro groups in the resulting
products were reduced to the corresponding anilines to provide
the c series of dianiline intermediates. Reaction of the dianilines
with the imidate derived from (R)-(+)-lactamide provided the
d series of (R)-alcohols. The phenylsulfonamide moieties
present in 15d, 16d, 19d−24d, and 31d−34d were removed
under hydrolytic conditions to provide final products 15, 16,
19−24, and 31−34, whereas intermediates 14d, 17d, 26d, and
28d−30d were further manipulated as described in Scheme 4.
As illustrated in Scheme 4, the Boc groups of 14d, 17d, 26d,
and 28d−30d were removed to liberate the corresponding free
a
Arithmetic mean of at least three separate runs (n ≥ 3). The average
coefficient of variation was less than 0.3 times the mean for
biochemical assays.13 The biochemical JAK3 and TYK2 Ki values of
the remaining compounds in this work are reported in the Supporting
Information. TYK2 cell-based assay.13 The average coefficient of
b
c
variation was less than 0.5 times the mean for cell-based assays. Patch
clamp assay conducted by ChanTest. Reversible CYP3A4 inhibition
measured in HLMs with midazolam as LCMS/MS probe. Other CYP
isoforms tested (LCMS/MS probe used): 3A4 (testosterone), 2C9
(warfarin), 2D6 (dextromethorphan), 2C19 (mephenytoin), 1A2
(phenacetin).
d
e
observed in the high dose group (500 mg/kg of 31). Given its
selectivity profile against the JAK family and the broader
kinome, it is likely that the robust CIA efficacy demonstrated by
31 is driven predominantly by JAK1 inhibition. As such, these
CIA results appear to support the hypothesis that selective
JAK1 inhibitors could provide treatments for patients suffering
from rheumatoid arthritis and other immunologic disorders.
CHEMISTRY
■
The preparation of analogues 5−13 commenced with synthesis
of dianiline intermediate 38 (Scheme 1).21 SNAr displacement
of the activated aryl chloride 359 by cyclohexylamine 36,
followed by reduction of the resulting aryl nitro 37, produced
dianiline 38.
As shown in Scheme 2, dianiline 38 was cyclized under
multiple conditions to form various C-2 substituted imidazoles.
Treatment of the dianiline 38 with the imidate formed by the
action of Meerwein’s reagent22 on 2-hydroxyacetamide
produced the primary alcohol 5a.23 Coupling of 38 to ethyl
3-chloro-3-oxopropanoate, followed by dehydrative cyclization,
yielded the ethyl ester-substituted imidazole 6a, which was
Figure 5. (a) Rat plasma exposures of 31 on days 10 and 20 of the CIA efficacy study.a (b) Mean ankle thicknesses of rats either treated or not
treated with 31 in the CIA efficacy study.a Superscript letter indicates the following: (a) 31 was formulated in corn oil and dosed orally once daily at
the indicated doses for 11 days. The first and last doses of 31 were administered on study days 10 and 20, respectively.
K
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Preparation of C-2 Hydroxyl Substituted Imidazopyrrolopyridine Analogues 5−13
a
23
−
Reagents and conditions: (i) 2-hydroxyacetamide, Et3(O+)BF4 , THF, EtOH, 25−70 °C; (ii) NaOH, H2O, EtOH or EtOH/THF, 25−50 °C;
−
(iii) TEA, ethyl 3-chloro-3-oxopropanoate, DCM, 25 °C, then AcOH, reflux; (iv) NaBH4, EtOH, 50 °C; (v) (R)-(+)-lactamide, Et3(O+)BF4 , THF,
23
−
EtOH, 25−70 °C; (vi) (S)-(−)-lactamide, Et3(O+)BF4 , THF, EtOH, 25−70 °C; (vii) Dess−Martin periodinane, DCM, 25 °C; (viii) MeMgBr,
THF, −10 to 25 °C; (ix) triethyl orthoformate, 82 °C; (x) (tmp)2Mg·2LiCl, propionaldehyde, THF, 0−25 °C; (xi) preparative chiral SFC; (xii)
(tmp)2Mg·2LiCl, 2-(benzyloxy)acetaldehyde, THF, 0−25 °C; (xiii) Pd(OH)2/C, MeOH, EtOH, H2, 50 psi, 25 °C.
amino groups (e series of intermediates). Alkylation or
acylation of the exposed amines and phenylsulfonamide
hydrolysis delivered final compounds 14, 17, 18, and 26−30.
potency was observed in human whole blood. Analysis of
historical data generated in the C-2 H9 and methyl7
imidazopyrrolopyridine series indicated that maintaining
cLogD7.4 <2 was important to minimizing whole blood shift.
Additional analysis of the historical data provided further
correlations of calculated physicochemical properties with
acceptable membrane permeability and HLM stability. These
analyses were used to formulate a set of physicochemical design
parameters (Table 3) that were employed to optimize the
highly JAK1 selective C-2 (R)-hydroxyethyl imidazopyrrolopyr-
idine series. Multiple compounds with favorable features
including excellent JAK1 vs JAK2 selectivity, potent cellular
JAK1 inhibition in human whole blood, and promising oral
pharmacokinetic parameters across multiple preclinical species
were, thereby, identified (e.g., 21, 26, 29, 31). These
compounds were found to show good levels of JAK family
CONCLUSION
■
Structure-based design targeting a key residue difference
between JAK1 (E966) and JAK2 (D939) was employed to
identify a C-2 (R)-hydroxyethyl group imparting consistent
JAK1 selectivity to the imidazopyrrolopyridine series. X-ray
structures of the highly selective inhibitor 31 in complex with
both JAK1 and JAK2 (Figure 3a,b) showed the C-2 (R)-
hydroxyethyl group participating in differential interactions
with the E966 and D939 residues and offered an explanation
for the observed selectivity. Although early analogues (such as
8) exhibited excellent potency in low serum measures of cell-
based JAK1 inhibition (IL-6 pSTAT3 assay), a large shift in
L
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 3. Preparation of C-2 (R)-Hydroxyethyl Imidazopyrrolopyridine Analogues 15, 16, 19−24, and 31−34
a
Reagents and conditions: (i) DIPEA, i-PrOH, 75−83 °C; (ii) H2, Pd/C, EtOH or EtOH/THF or EtOAc, 23−50 °C; (ii′) Fe, NH4Cl, EtOH, H2O,
reflux; (iii) (R)-(+)-lactamide, Et3(O+)BF4 , THF,23 EtOH, 23−70 °C; (iv) NaOH, H2O, EtOH or EtOH/THF or MeOH/THF, 25−60 °C. (b) 4-
Piperidine diastereomers separated by preparative chiral SFC. (c) Racemic mixture. (d) Diastereomeric mixture of 4-piperidine epimers.
−
an internal standard; tetramethylsilane (ppm = 0.00). The following
abbreviations are used: br = broad signal, s = singlet, d = doublet, dd =
doublet of doublets, t = triplet, q = quartet, m = multiplet. Purification
by silica gel chromatography was carried out using either hand-packed
glass columns or CombiFlash by Teledyne Isco systems with
prepacked cartridges. Some basic compounds were purified by capture
on SCX (strong cation exchange) sulfonic acid cartridges, followed by
elution with methanolic ammonia. Chemical purities were >95% for all
final compounds as assessed by LCMS. Further details on the LCMS
conditions used for individual compounds may be found in the
Supporting Information.
Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo[4,5-d]-
pyrrolo[2,3-b]pyridin-2-yl)methanol (5a). Triethyloxonium tetra-
fluoroborate (3.81 g, 201 mmol) was added to a solution of 2-
hydroxyacetamide (1.51 g, 201 mmol) in THF (100 mL) at 25 °C.23
The reaction mixture was stirred at 25 °C for 2 h. After concentrating
selectivity relative to the broader kinome, with 29 and 31 being
particularly discriminating (Table 6). Finally, compound 31 was
found to be highly efficacious in a rat CIA efficacy model
(Figure 5), supporting the hypothesis that selective JAK1
inhibition may be a viable method to treat RA and other
immunologic disorders.
EXPERIMENTAL SECTION
■
General. Unless otherwise indicated, all reagents and solvents were
purchased from commercial sources and used without further
purification. Moisture or oxygen sensitive reactions were conducted
under an atmosphere of argon or nitrogen gas. Unless otherwise
1
stated, H NMR spectra were recorded at 298−300 K using Varian
Unity Inova or Bruker Avance DRX400 instruments operating at the
indicated frequencies. Chemical shifts are expressed in ppm relative to
M
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 4. Preparation of C-2 (R)-Hydroxyethyl Imidazopyrrolopyridine Analogues 14, 17, 18, and 26−30
a
Reagents and conditions: (i) HCl/1,4-dioxane or TFA/H2O, 23 °C; (ii) 2,2,2-trifluoroethyl trifluoromethanesulfonate, TEA, DCM, DMF, 23 °C;
(iii) NaOH, H2O, EtOH or EtOH/THF or MeOH/THF, 25−60 °C; (iv) EDC, 4-DMAP, DIPEA, 2-cyanoacetic acid (for 17) or 3,3,3-
trifluoropropanoic acid (for 18), 23 °C; (v) 2,2,-difluoroethyl trifluoromethanesulfonate, TEA, DCM, DMF, 23 °C.
under vacuum, the resulting oil was dissolved in EtOH (50 mL), then
added to a mixture of 387 (2.92 g, 78.8 mmol) in EtOH (100 mL) and
stirred at 75 °C for 80 min. The reaction mixture was cooled to 25 °C
and allowed to stand for 2 h. The resulting precipitate was collected by
filtration and purified by column chromatography on silica gel, eluting
with EtOAc/heptane to afford 2.93 g (91%) of the title compound as
were dried over MgSO4, filtered, and concentrated under vacuum. The
resulting solid was triturated with diethyl ether (2 × 15 mL), then
dried under vacuum to afford the title product as an off-white solid
1
(26.2 mg, 18%). H NMR (400 MHz, DMSO-d6) δ 11.83 (s, 1H),
8.53 (s, 1H), 7.47 (s, 1H), 6.74 (s, 1H), 5.67 (s, 1H), 4.78 (d, J = 4.4
Hz, 2H), 4.75−4.61 (m, 1H), 2.39−2.20 (m, 2H), 2.02−1.87 (m, 4H),
1.86−1.74 (m, 1H), 1.62−1.37 (m, 3H). LCMS (method LCMS1,
ESI): RT = 5.81 min, m/z = 271.0 [M + H]+.
Ethyl 2-(1-Cyclohexyl-6-(phenylsulfonyl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)acetate (6a).
TEA (450 mg, 2.97 mmol) and ethyl 3-chloro-3-oxopropanoate
(327 mg, 3.24 mmol) were added to a solution of 387 (1.0 g, 2.7
mmol) in DCM (15 mL). The reaction mixture was stirred at 25 °C
for 1 h, then concentrated under vacuum. Acetic acid (10 mL) was
added to the residue, and the mixture was heated at reflux for 16 h.
The reaction mixture was then cooled to 25 °C and concentrated
under vacuum. The residue was purified by column chromatography
on silica gel, eluting with EtOAc/heptane to afford 210 mg (16%) of
1
an off-white solid. H NMR (400 MHz, DMSO-d6) δ 8.67 (s, 1H),
8.14 (d, J = 8.0 Hz, 2H), 7.98 (s, 1H), 7.72−7.68 (m, 1H), 7.65−7.57
(m, 2H), 7.12 (s, 1H), 5.74−5.68 (m, 1H), 4.82−4.77 (m, 2H), 4.74−
4.64 (m, 1H), 2.19−2.07 (m, 2H), 1.96−1.88 (m, 4H), 1.79−1.74 (m,
1H), 1.56−1.44 (m, 3H).
(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)methanol (5). First, 1.0 mL of a 1 N aqueous solution
of NaOH was added to a mixture of 5a (224 mg, 0.54 mmol) in EtOH
and THF (5 mL each). The reaction mixture was stirred at 50 °C for 1
h, then was cooled to 25 °C and poured into 100 mL of a half
saturated aqueous solution of NaHCO3. The aqueous layer was
extracted with EtOAc (2 × 100 mL), then the combined organic layers
N
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
the title compound as a yellow solid. This material was used in the
next step without further purification or characterization.
layer was washed with saturated aqueous NaHCO3 solution, dried over
MgSO4, and concentrated under vacuum to give the title compound
(199 mg, 66%) as an off-white solid.
2-(1-Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (6b). Sodium borohydride
was added to a solution of 6a (210 mg, 0.45 mmol) in EtOH (5 mL).
The reaction mixture was stirred at 50 °C for 30 min, then cooled to
25 °C and concentrated under vacuum. The residue was purified by
column chromatography on silica gel, eluting with EtOAc/heptane to
give the title compound (151 mg, 79%) as a gray solid. LCMS
(method LCMS2, ESI): RT = 1.37 min, m/z = 425.1 [M + H]+.
2-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)ethanol (6). Sodium hydroxide (8 mL of a 2 N aqueous
solution) was added to a solution of 6b (151 mg, 0.35 mmol) in EtOH
(4 mL). The reaction mixture was stirred at 25 °C for 16 h, then
adjusted to pH ∼7 with HCl (2 N aqueous solution) and extracted
with EtOAc (2 × 30 mL). The combined organic extracts were washed
with water and brine (10 mL each), dried over MgSO4, filtered, and
concentrated. The residue was purified by column chromatography on
silica gel, eluting with EtOAc/heptane to give the title compound (42
mg, 39%) as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 11.80 (s,
1H), 8.49 (s, 1H), 7.46 (s, 1H), 6.72 (s, 1H), 4.88 (t, J = 5.2 Hz, 1H),
4.56−4.43 (m, 1H), 3.87−3.81 (br, 2H), 3.13 (t, J = 6.7 Hz, 2H),
2.36−2.24 (m, 2H), 2.03−1.83 (m, 4H), 1.81−1.76 (m, 1H), 1.64−
1.39 (m, 3H). LCMS (method LCMS3, ESI): RT = 3.17 min, m/z =
285.1 [M + H]+.
2-(1-Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)propan-2-ol (9b). To a solution of
compound 9a (198 mg, 0.47 mmol) in THF (5 mL) was added
methylmagnesium bromide (1 mL of 1.0 M solution in dibutyl ether)
at −10 °C. The reaction mixture was warmed to 25 °C and stirred at
that temperature for 30 min. Water (10 mL) was added, and the
mixture was extracted with EtOAc (2 × 10 mL). The combined
extracts were washed with water and brine (10 mL each), dried over
MgSO4, and concentrated under vacuum. The residue was purified by
column chromatography on silica gel, eluting with EtOAc/heptane to
afford 200 mg (97%) of the title compound as a gray solid. LCMS
(method LCMS2, ESI): RT = 1.43 min, m/z = 439.2 [M + H]+.
2-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)propan-2-ol (9). NaOH (2 N aqueous solution, 10
mL) was added to a solution of 9b (200 mg, 0.46 mmol) in EtOH (5
mL). The reaction mixture was stirred at 25 °C for 16 h, then the pH
was adjusted to ∼7 with 2 N aqueous HCl solution. The mixture was
extracted with EtOAc (2 × 15 mL). The combined organic layers were
washed with water and brine (10 mL each), dried over MgSO4,
filtered, and concentrated under vacuum. The residue was purified by
column chromatography on silica gel, eluting with EtOAc/heptane to
1
afford 90 mg of the title compound as a white solid (69%). H NMR
(S)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)ethanol (7). Synthesized by methods similar to those
used for 8, substituting (S)-(−)-lactamide for (R)-(+)-lactamide.
(400 MHz, DMSO-d6) δ 11.83 (s, 1H), 8.53 (s, 1H), 7.47 (s, 1H),
6.74 (s, 1H), 5.74 (s, 1H), 5.43 (t, J = 12.3 Hz, 1H), 2.50 (s, 6H), 2.37
(dd, J = 23.4, 11.2 Hz, 2H), 2.04−1.85 (m, 4H), 1.84−1.76 (m, 1H),
1.57−1.37 (m, 3H). LCMS (method LCMS3, ESI): RT = 3.47 min,
m/z = 299.1 [M + H]+.
1
Analytical chiral SFC (method SFC1): RT = 0.57 min. H NMR and
LCMS match 8.
(R)-1-(1-Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo-
[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (8a). Triethyloxonium
tetrafluoroborate (0.569 g, 3.00 mmol) was added to a solution of (R)-
(+)-lactamide (267 mg, 3.00 mmol) in THF (25 mL) at 25 °C.23 The
resulting suspension was stirred at 25 °C for 2 h, then was
concentrated under reduced pressure to afford an oil. This material
was dissolved in EtOH (10 mL), and the resulting solution was added
to to a solution of 387 (370 mg, 1.00 mmol) in EtOH (4 mL) at 25
°C. The reaction mixture was heated at 75 °C for 1 h, then was cooled
to 25 °C and was partitioned between half-saturated aqueous
NaHCO3 (100 mL) and EtOAc (2 × 100 mL). The organic layers
were dried over MgSO4 and filtered, and the filtrate was concentrated
under vacuum. Purification of the resulting solid by column
chromatography on silica gel (gradient: 0−8% CH3OH in DCM)
afforded the title compound (202 mg, 95%) as a gray solid. LCMS
(method LCMS4, ESI): RT = 0.99 min, m/z = 425.3 [M + H]+.
(R)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-
b]pyridin-2-yl)ethanol (8). Sodium hydroxide (1 mL of a 1 N
aqueous solution) was added to a solution of 8a (203 mg, 0.48 mmol)
in a 1:1 mixture of THF and EtOH (10 mL) at 25 °C. The reaction
mixture was stirred at 50 °C for 1 h, then was cooled to 25 °C and was
partitioned between half-saturated aqueous NaHCO3 (100 mL) and
EtOAc (2 × 125 mL). The combined organic layers were dried over
MgSO4 and filtered, and the filtrate was concentrated under reduced
pressure. Purification of the residue by preparative HPLC (column:
Gemini-NX, 5 cm × 10 cm, 10 μm; detection:, UV 254 nm; mobile
phase A, water with 0.1% NH4OH; mobile phase B, CH3CN; flow
rate, 120 mL/min; gradient 5−95% B over 15 min) afforded the title
compound (46 mg, 34%) as an off-white solid. Analytical chiral SFC
1-Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo[4,5-d]-
pyrrolo[2,3-b]pyridine (39). A suspension of 387 (7.80 g, 21.1
mmol) in triethyl orthoformate (48 mL) was heated at 82 °C for 16 h.
The reaction mixture was cooled to 25 °C and concentrated under
vacuum. The residue was purified by column chromatography on silica
gel, eluting with EtOAc/heptane to afford 5.34 g (68%) of the title
compound as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 8.72 (s,
1H), 8.46 (s, 1H), 8.15−8.10 (m, 2H), 7.97 (d, J = 4.0 Hz, 1H), 7.73−
7.66 (m, 1H), 7.64−7.55 (m, 2H), 7.18 (d, J = 4.0 Hz, 1H), 4.64−4.49
(m, 1H), 2.15−2.03 (m, 2H), 1.91−1.68 (m, 5H), 1.67−1.51 (m, 2H),
1.36−1.21 (m, 1H). LCMS (method LCMS6, ESI): RT = 2.27 min,
m/z = 381.3 [M + H]+.
(
)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-
b]pyridin-2-yl)propan-1-ol (10b). 2,2,6,6-Tetramethylpiperidine-
magnesium chloride lithium chloride complex (1 M solution in
THF, 677 μL) was added to a solution of 39 (200 mg, 0.50 mmol) in
THF (4 mL) at 0 °C. After stirring for 5 min at 0 °C, the reaction
mixture was removed from the ice bath and stirred for 1 h at 25 °C.
The reaction mixture was then added to a stirred solution of
propionaldehyde (97 μL, 1.35 mmol) in THF (1 mL) and left to stir
for 16 h. EtOH (5 mL) and sodium hydroxide (1 N aqueous solution,
5 mL) were then added, and the reaction mixture was stirred at 50 °C
for 2 h. The reaction mixture was concentrated under vacuum and
purified by column chromatography eluting with EtOAc/heptane to
afford the racemic title compound (10b) as a white solid (18.1 mg,
12%).
(S)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)propan-1-ol (10) and (R)-1-(1-Cyclohexyl-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)propan-1-ol
(11). 10b was separated into its individual enantiomers by preparative
chiral SFC (Lux Cellulose-1 column, 21.2 mm × 150 mm, 5 μm;
detection, UV 220 nm; mobile phase A, CO2; mobile phase B, MeOH
containing 0.1% NH4OH; flow rate, 70 mL/min; gradient, isocratic;
A:B = 80:20). Isolation and concentration of the appropriate fractions
afforded two products with the following characteristics. 10: white
solid. Analytical chiral SFC (method SFC2): RT = 0.48 min. 1H NMR
and LCMS match 11. 11: white solid. Analytical chiral SFC (method
1
(method SFC1): RT = 0.54 min. H NMR (400 MHz, DMSO-d6) δ
11.84 (s, 1H), 8.55 (s, 1H), 7.47 (s, 1H), 6.74 (s, 1H), 5.66 (d, J = 6.5
Hz, 1H), 5.12 (p, J = 6.4 Hz, 1H), 4.91−4.76 (m, 1H), 2.41−2.22 (m,
2H), 2.02−1.84 (m, 4H), 1.80 (s, 1H), 1.64 (d, J = 6.4 Hz, 3H), 1.58−
1.42 (m, 3H). LCMS (method LCMS5, ESI): RT = 6.56 min, m/z =
285.3 [M + H]+.
1-(1-Cyclohexyl-6-(phenylsulfonyl)-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)ethanone (9a). Dess−Martin period-
inane (1.2 g, 2.8 mmol) was added to a solution of 7a (300 mg, 0.71
mmol) in DCM (5 mL). The reaction mixture was stirred for 30 min
at 25 °C, then diluted with an additional 50 mL of DCM. The organic
1
SFC2) RT = 0.45 min. H NMR (400 MHz, DMSO-d6) δ 11.85 (s,
1H), 8.54 (s, 1H), 7.47 (t, J = 3.0 Hz, 1H), 6.73 (s, 1H), 5.73−5.66
(m, 1H), 4.98−4.63 (m, 2H), 2.38−2.26 (m, 2H), 2.07−2.00 (m, 1H),
O
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
2.00−1.92 (m, 3H), 1.88 (d, J = 11.2 Hz, 2H), 1.83−1.78 (m, 1H),
1.56−1.43 (m, 3H), 0.98 (t, J = 7.4 Hz, 3H). LCMS (method LCMS3,
ESI): RT = 3.46 min, m/z = 299.1 [M + H]+.
DMSO-d6) δ 11.82 (s, 1H), 8.55 (s, 1H), 7.49 (t, J = 3.0 Hz, 1H), 6.90
(s, 1H), 5.68 (d, J = 6.6 Hz, 1H), 5.13 (t, J = 6.5 Hz, 1H), 4.93−4.81
(m, 1H), 3.38−3.30 (m, 2H), 3.21−3.11 (s, 2H), 2.66−2.55 (m, 4H),
1.93−1.81 (m, 2H), 1.63 (d, J = 6.4 Hz, 3H). LCMS (method
LCMS7, ESI): RT = 3.02 min, m/z = 368.3 [M + H]+.
(R)-3-(4-(2-(1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)piperidin-1-yl)propanenitrile (15). 15 was
synthesized by methods similar to those used for 8, substituting 3-
(4-aminopiperidin-1-yl)propanenitrile (15a)26 for cyclohexylamine
(36). 1H NMR (400 MHz, DMSO-d6) δ 11.81 (s, 1H), 8.55 (s,
1H), 7.40 (t, J = 3.0 Hz, 1H), 7.00 (br s, 1H), 5.69 (d, J = 6.7 Hz, 1H),
5.13 (p, J = 6.5 Hz, 1H), 4.94−4.81 (m, 1H), 3.18−3.11 (m, 2H),
2.82−2.67 (m, 4H), 2.67−2.55 (m, 2H), 2.25 (t, J = 11.9 Hz, 2H),
1.94−1.84 (m, 2H), 1.63 (d, J = 6.4 Hz, 3H). LCMS (method
LCMS7, ESI): RT = 1.38 min, m/z = 339.2 [M + H]+.
(
)-2-(Benzyloxy)-1-(1-cyclohexyl-6-(phenylsulfonyl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (12a).
A 1 M solution of 2,2,6,6-tetramethylpiperidinemagnesium chloride
lithium chloride complex in THF (16.8 mL, 16.8 mmol) was added to
a solution of 39 (5.34 g, 14.0 mmol) in anhydrous THF (100 mL) at 0
°C. The reaction mixture was stirred for 1 h at 0 °C, then added to a
solution of 2-(benzyloxy)acetaldehyde (3.95 mL, 28.1 mmol) in THF
(35 mL) at 0 °C. The mixture was stirred for 1 h at 0 °C, then slowly
warmed to 25 °C. Stirring was continued for an additional 14 h, then a
saturated aqueous solution of NH4Cl (50 mL) was added. The mixture
was extracted with EtOAc (3 × 75 mL), then the combined organic
layers were dried over MgSO4, filtered, and concentrated under
vacuum. The residue was purified by column chromatography eluting
with EtOAc/heptane to afford 6.56 g (88%) of the title compound as a
foamy off-white solid. LCMS (method LCMS6, ESI): RT = 2.71 min,
m/z = 531.3 [M + H]+.
(1R)-1-(1-(1-Ethyl-3,3-difluoropiperidin-4-yl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (16).
A diastereomeric mixture of 16 and its 4-piperidine epimer was
synthesized by methods similar to those used for 8, substituting ( )-1-
ethyl-3,3-difluoropiperidin-4-amine (16a)27 for cyclohexylamine (36).
The diastereomers were separated by preparative chiral SFC (Lux
Cellulose-1 column: 21.2 mm × 250 mm, 5 μm; detection, UV 220
nm; mobile phase A, CO2; mobile phase B, MeOH containing 0.1%
NH4OH; flow rate, 70 mL/min; gradient, isocratic; A:B = 85:15) to
(
)-2-(Benzyloxy)-1-(1-cyclohexyl-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (12b). To a solution of 12a
(6.56 g, 12.4 mmol) in EtOH (50 mL) was added a 1 M solution of
aqueous sodium hydroxide (24.7 mL, 24.7 mmol). The reaction
mixture was stirred at 23 °C for 16 h, then a 1 M aqueous solution of
HCl (25 mL) was added. The mixture was concentrated to dryness
under vacuum, then an additional 100 mL of EtOH was added. The
mixture was filtered, then the filtrate was concentrated under vacuum
to afford the crude title product as an off-white solid.
1
provide 16 (later eluting peak) as a white solid. H NMR (400 MHz,
DMSO-d6) δ 11.74 (s, 1H), 8.58 (s, 1H), 7.41 (t, J = 3.1 Hz, 1H),
6.74−6.62 (m, 1H), 5.75 (d, J = 7.4 Hz, 1H), 5.46−5.28 (m, 1H), 5.04
(p, J = 6.6 Hz, 1H), 3.38 (t, J = 11.8 Hz, 1H), 3.22−3.13 (m, 1H), 3.06
(qd, J = 12.4, 4.6 Hz, 1H), 2.69−2.53 (m, 3H), 2.33 (t, J = 11.7 Hz,
1H), 2.00−1.89 (m, 1H), 1.68 (d, J = 6.3 Hz, 3H), 1.11 (t, J = 7.1 Hz,
3H). LCMS (method LCMS1, ESI): RT = 2.96 min, m/z = 350.4 [M
+ H]+.
(R)-2-(Benzyloxy)-1-(1-cyclohexyl-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (12c) and (S)-2-(Benzyl-
oxy)-1-(1-cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)ethanol (13c). The crude residue obtained in the
previous step was purified and separated into its individual
enantiomers by preparative chiral SFC (column: Lux Cellulose-1, 3
cm × 25 cm, 5 μm; detection, UV 220 nm; mobile phase A, CO2;
mobile phase B, MeOH containing 0.1% dimethylamine; flow rate, 200
mL/min; gradient, isocratic; A:B = 70:30). First-eluting peak (13c):
off-white solid (2.12 g, 44%). 1H NMR (400 MHz, DMSO-d6) δ 11.82
(s, 1H), 8.53 (s, 1H), 7.46 (t, J = 2.9 Hz, 1H), 7.38−7.25 (m, 5H),
6.73 (s, 1H), 5.92 (d, J = 6.6 Hz, 1H), 5.17 (q, J = 6.3 Hz, 1H), 4.91−
4.69 (m, 1H), 4.61 (s, 2H), 4.05 (dd, J = 10.1, 4.6 Hz, 1H), 3.93 (dd, J
= 10.1, 7.4 Hz, 1H), 2.39−2.21 (m, 2H), 1.98−1.74 (m, 5H), 1.59−
1.38 (m, 3H). LCMS (method LCMS5, ESI): RT = 10.7 min, m/z =
391.1 [M + H]+. Second-eluting peak (12c): off-white solid (1.53 g,
(R)-tert-Butyl 3-(2-((R)-1-hydroxyethyl)-6-(phenylsulfonyl)-
imidazo[4,5-d]pyrrolo[2,3-b]pyridin-1(6H)-yl)piperidine-1-car-
boxylate (17d). 17d was synthesized by methods similar to those
used for 8a, substituting (R)-tert-butyl 3-aminopiperidine-1-carbox-
1
ylate (17a) for cyclohexylamine (36). H NMR (400 MHz, DMSO-
d6) δ 8.72 (s, 1H), 8.14 (d, J = 7.6 Hz, 2H), 7.96 (d, J = 3.8 Hz, 1H),
7.75−7.58 (m, 3H), 7.21 (d, J = 3.7 Hz, 1H), 5.85−5.75 (m, 1H),
5.19−5.08 (m, 1H), 4.99−4.86 (m, 1H), 4.27−4.17 (m, 1H), 4.15−
3.98 (m, 1H), 2.36−2.21 (m, 1H), 2.06−1.97 (m, 1H), 1.90 (br d, J =
12.1 Hz, 1H), 1.62 (d, J = 6.2 Hz, 3H), 1.51 (s, 3H), 1.38 (br s, 9H).
(R)-1-(1-((R)-Piperidin-3-yl)-1,6-dihydroimidazo[4,5-d]-
pyrrolo[2,3-b]pyridin-2-yl)ethanol (17f). 17d (8.09 g, 15.4 mmol)
was stirred in a 4.0 M solution of hydrogen chloride in 1,4-dioxane
(150 mL) for 3 h. The volatiles were removed under vacuum and the
solids were washed with Et2O (3 × 100 mL) and isolated by filtration.
The resulting hygroscopic white powder was dissolved in a 1.0 M
solution of sodium hydroxide in water (100 mL) and EtOH (100 mL)
and heated to 60 °C for 8 h. The crude reaction mixture was then
concentrated onto silica gel and purified by column chromatography,
eluting with a gradient of 0% to 20% 2 M NH3 in MeOH in DCM, to
1
32%). H NMR and LCMS match 13c.
(R)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-
b]pyridin-2-yl)ethane-1,2-diol (12). 12c was converted to 12 by
the same method used to convert 13c to 13. The 1H NMR and LCMS
of 12 match those of 13.
(S)-1-(1-Cyclohexyl-1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-2-yl)ethane-1,2-diol (13). To a solution of 13c (2.1 g, 5.3
mmol) in a 1:1 mixture of MeOH:EtOH (50 mL) was added 20%
palladium hydroxide on carbon (371 mg). The reaction mixture was
pressurized to 50 psi with hydrogen gas, then shaken on a Parr
apparatus for 48 h. The reaction mixture was filtered through Celite,
then the filtrate was concentrated under vacuum. The residue was
purified by reverse phase HPLC (Gemini-NX column, 30 mm × 100
mm, 5 μm; detection, UV 230 nm; mobile phase A, water with 0.1%
NH4OH; mobile phase B, CH3CN; flow rate, 60 mL/min; gradient 5−
50% B over 10 min) to afford 253 mg (17%) the title product as a
white solid. 1H NMR (400 MHz, DMSO-d6) δ 11.82 (s, 1H), 8.54 (s,
1H), 7.46 (t, J = 3.0 Hz, 1H), 6.74 (dd, J = 3.5, 1.9 Hz, 1H), 5.76 (d, J
= 6.4 Hz, 1H), 4.94 (q, J = 6.2 Hz, 1H), 4.82 (t, J = 6.0 Hz, 2H),
3.99−3.91 (m, 1H), 3.89−3.82 (m, 1H), 2.41−2.19 (m, 2H), 2.02−
1.68 (m, 5H), 1.60−1.39 (m, 3H). LCMS (method LCMS5, ESI): RT
1
provide the title compound as an off-white solid (2.70 g, 61%). H
NMR (400 MHz, DMSO-d6) δ 11.83 (s, 1H), 8.54 (s, 1H), 7.48 (s,
1H), 6.84 (s, 1H), 5.66 (d, J = 6.5 Hz, 1H), 5.17−5.04 (m, 1H), 4.86
(s, 1H), 3.43−3.36 (m, 1H), 3.17 (d, J = 5.1 Hz, 1H), 3.10−2.94 (m,
2H), 2.75 (t, J = 11.9 Hz, 1H), 2.46−2.38 (m, 1H), 2.03−1.80 (m,
2H), 1.70−1.57 (m, 4H).
3-((R)-3-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)piperidin-1-yl)-3-oxopropanenitrile (17). EDC
(77 mg, 0.40 mmol) was added to a solution of 17f (60 mg, 0.21
mmol), 4-dimethylaminopyridine (10 mg, 0.08 mmol), DIPEA (80
μL, 0.46 mmol), and 2-cyanoacetic acid (34 mg, 0.40 mmol) in DMF
(1 mL). The reaction mixture was stirred at 23 °C for 16 h, then
diluted with EtOAc (10 mL). The organic layer was washed
successively with 1 M aqueous HCl solution, saturated aqueous
NaHCO3 solution, and brine (1 × 3 mL each), then dried over
Na2SO4, filtered, and concentrated under vacuum. The residue was
purified by preparative HPLC (column: Gemini-NX, 5 cm × 10 cm,
10 μm; detection, UV 254 nm; mobile phase A, water with 0.1%
+
= 10.7 min, m/z = 301.1 [M + H] .
(R)-1-(1-(1-(2,2,2-Trifluoroethyl)piperidin-4-yl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (14).
14 was synthesized by methods similar to those used for 26,
substituting tert-butyl 4-aminopiperidine-1-carboxylate (14a) for tert-
butyl (trans-4-aminocyclohexyl)carbamate (26a). 1H NMR (400 MHz,
P
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
NH4OH; mobile phase B, CH3CN; flow rate, 120 mL/min; gradient
5−95% B over 15 min) to afford 50 mg (68%) of the title compound
as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 11.91 (s, 1H), 8.57
(d, J = 6.4 Hz, 1H), 8.32 (s, 1H), 7.49 (s, 1H), 6.80 (d, J = 13.8 Hz,
1H), 5.25−5.06 (m, 1H), 5.00−4.80 (m, 1H), 4.68−4.45 (m, 1H),
4.25−4.04 (m, 2H), 3.94−3.75 (m, 2H), 3.03−2.89 (m, 1H), 2.12−
1.89 (m, 2H), 1.86−1.71 (m, 1H), 1.71−1.59 (m, 4H). LCMS
(method LCMS8, ESI): RT = 2.70 min, m/z = 353.1 [M + H]+.
3,3,3-Trifluoro-1-((R)-3-(2-((R)-1-hydroxyethyl)imidazo[4,5-
d]pyrrolo[2,3-b]pyridin-1(6H)-yl)piperidin-1-yl)propan-1-one
(18). 18 was synthesized by methods similar to those used for 17,
substituting 3,3,3-trifluoropropanoic acid for 2-cyanoacetic acid. 1H
NMR (400 MHz, DMSO-d6) δ 11.92 (s, 1H), 8.58 (d, J = 8.4 Hz,
1H), 7.50 (s, 1H), 6.81 (d, J = 15.8 Hz, 1H), 5.85−5.70 (m, 1H),
5.24−5.06 (m, 1H), 4.98−4.79 (m, 1H), 4.73−4.51 (m, 1H), 4.04−
3.91 (m, 1H), 3.86−3.66 (m, 2H), 2.93 (t, J = 14.1 Hz, 1H), 2.70−
2.56 (m, 1H), 2.12−1.92 (m, 3H), 1.72−1.57 (m, 4H). LCMS
(method LCMS3, ESI): RT = 3.05 min, m/z = 396.1 [M + H]+.
(R)-1-(1-(Tetrahydro-2H-pyran-4-yl)-1,6-dihydroimidazo[4,5-
d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (19). 19 was synthesized by
methods similar to those used for 21, substituting tetrahydro-2H-
precipitate was collected by filtration, washed with cold EtOAc (50
mL), and dried under vacuum for 16 h to afford 60.9 g (80%) of the
title compound as a yellow solid. H NMR (400 MHz, DMSO-d6) δ
1
8.71 (s, 1H), 8.18−8.09 (m, 2H), 7.99 (d, J = 4.0 Hz, 1H), 7.74−7.67
(m, 1H), 7.65−7.57 (m, 2H), 7.26 (d, J = 4.1 Hz, 1H), 5.82 (d, J = 6.4
Hz, 1H), 5.14 (p, J = 6.3 Hz, 1H), 5.08−4.94 (m, 1H), 3.95 (d, J = 8.3
Hz, 3H), 3.68 (t, J = 11.2 Hz, 1H), 2.46−2.30 (m, 1H), 2.07 (d, J =
11.7 Hz, 1H), 1.95−1.75 (m, 2H), 1.62 (d, J = 6.3 Hz, 3H). LCMS
1
(method LCMS9, ESI): RT = 1.34 min, m/z = 427.2 [M + H]+. H
(R)-1-(1-((S)-Tetrahydro-2H-pyran-3-yl)-1,6-dihydroimidazo-
[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (21). To a suspension of
21d (40.0 g, 93.8 mmol) in EtOH (50 mL) was added an aqueous
solution of sodium hydroxide (9 M, 30 mL, 270 mmol). The reaction
mixture was stirred for 1 h at 60 °C, then cooled to room temperature.
The majority of the EtOH was removed under vacuum, then the
remaining aqueous solution was cooled to 0 °C. The pH was adjusted
to ∼7 with 37% aqueous hydrochloric acid solution. The resulting
solid was removed by filtration. The filtrate was concentrated under
vacuum, the residue was taken up in a 1:1 mixture of EtOH:EtOAc
(200 mL), and the solid was removed with another filtration. The
filtrate was concentrated, and the residue was dissolved in MeOH (150
mL) and treated with macroporous triethylammonium methylpolys-
tyrene carbonate resin to remove the remaining phenylsulfonic acid.
The resin was removed by filtration and washed with MeOH (30 mL).
The filtrate was concentrated and the residue crystallized from hot
CH3CN. The resulting solid was collected by filtration, washed with
cold CH3CN, and then taken up in water (15 mL). The water was
removed under vacuum to afford 10.6 g (40%) of the title compound
as a white solid. 1H NMR (400 MHz, DMSO-d6) δ 11.87 (s, 1H), 8.55
(s, 1H), 7.49 (d, J = 3.3 Hz, 1H), 6.85 (d, J = 3.4 Hz, 1H), 5.89−5.56
(m, 1H), 5.14 (q, J = 6.4 Hz, 1H), 5.08−4.94 (m, 1H), 4.11 (t, J = 11.1
Hz, 1H), 4.02−3.92 (m, 2H), 3.69 (td, J = 11.9, 2.5 Hz, 1H), 2.63−
2.51 (m, 1H), 2.14−2.03 (m, 1H), 1.97−1.74 (m, 2H), 1.63 (d, J = 6.4
Hz, 3H). LCMS (method LCMS1, ESI): RT = 3.60 min, m/z = 287.0
[M + H]+.
trans-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)cyclohexanol (22). 22 was synthesized by
methods similar to those used for 21, substituting trans-4-amino-
cyclohexanol (22a) for (S)-tetrahydro-2H-pyran-3-amine (21a). 1H
NMR (400 MHz, DMSO-d6) δ 11.81 (s, 1H), 8.54 (s, 1H), 7.46 (d, J
= 3.4 Hz, 1H), 6.66 (d, J = 3.5 Hz, 1H), 5.65 (br s, 1H), 5.12 (q, J =
6.4 Hz, 1H), 4.88−4.69 (m, 2H), 3.86−3.74 (m, 1H), 2.46−2.32 (m,
2H), 2.10−2.01 (m, 2H), 1.95−1.83 (m, 2H), 1.63 (d, J = 6.4 Hz,
3H), 1.53−1.39 (m, 2H). LCMS (method LCMS1, ESI): RT = 7.14
min, m/z = 301.2 [M + H]+.
cis-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)cyclohexanol (23). 23 was synthesized by
methods similar to those used for 21, substituting cis-4-amino-
cyclohexanol (23a) for (S)-tetrahydro-2H-pyran-3-amine (21a). 1H
NMR (400 MHz, DMSO-d6) δ 11.71 (s, 1H), 8.52 (s, 1H), 7.41 (s,
2H), 5.63 (br s, 1H), 5.11 (q, J = 6.5 Hz, 1H), 4.97−4.74 (m, 2H),
4.03 (s, 1H), 2.79−2.64 (m, 2H), 1.94−1.83 (m, 2H), 1.72−1.55 (m,
7H). LCMS (method LCMS1, ESI): RT = 7.18 min, m/z = 301.2 [M
+ H]+.
1
pyran-4-amine (19a) for (S)-tetrahydro-2H-pyran-3-amine (21a). H
NMR (400 MHz, DMSO-d6) δ 11.85 (s, 1H), 8.57 (s, 1H), 7.50 (s,
1H), 6.76 (s, 1H), 5.78−5.63 (m, 1H), 5.22−5.05 (m, 2H), 4.17−4.10
(m, 2H), 3.57 (t, J = 12.0 Hz, 2H), 2.70−2.53 (m, 2H), 1.92−1.80 (m,
2H), 1.65 (d, J = 6.4 Hz, 3H). LCMS (method LCMS3, ESI): RT =
2.65 min, m/z = 287.1 [M + H]+.
(R)-1-(1-((R)-Tetrahydro-2H-pyran-3-yl)-1,6-dihydroimidazo-
[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (20). 20 was synthe-
sized by methods similar to those used for 21, substituting (R)-
tetrahydro-2H-pyran-3-amine (20a) for (S)-tetrahydro-2H-pyran-3-
1
amine (21a). H NMR (400 MHz, DMSO-d6) δ 11.88 (s, 1H), 8.56
(s, 1H), 7.50 (s, 1H), 6.86 (s, 1H), 5.72−5.64 (m, 1H), 5.19−5.09 (m,
1H), 5.04−4.95 (m, 1H), 4.10 (t, J = 11.2 Hz, 1H), 4.02−3.91 (m,
2H), 3.70 (t, J = 11.8 Hz, 1H), 2.68−2.50 (m, 1H), 2.11−2.03 (m,
1H), 1.96−1.80 (m, 2H), 1.65 (br s, 3H). LCMS (method LCMS3,
ESI): RT = 2.72 min, m/z = 287.1 [M + H]+.
(S)-5-Nitro-1-(phenylsulfonyl)-N-(tetrahydro-2H-pyran-3-yl)-
1H-pyrrolo[2,3-b]pyridin-4-amine (21b). DIPEA (130 mL, 760
mmol) was added to a solution of (S)-tetrahydro-2H-pyran-3-amine
(21a,28 40.7 g, 296 mmol) and 359 (100 g, 296 mmol) in isopropyl
alcohol (1.0 L). The mixture was stirred at an internal temperature of
75 °C for 4 h. The mixture was then cooled slowly to 23 °C and
allowed to stand for 16 h. The precipitate was collected by filtration
and washed with cold isopropyl alcohol (100 mL). The solid was dried
in a vacuum oven at 50 °C for 24 h to afford 114 g (96%) of the title
compound as a yellow solid. 1H NMR (400 MHz, CDCl3) δ 9.21 (d, J
= 8.0 Hz, 1H), 9.11 (s, 1H), 8.19 (dd, J = 7.6, 1.6 Hz, 2H), 7.65−7.58
(m, 2H), 7.52 (t, J = 7.7 Hz, 2H), 6.75 (d, J = 4.1 Hz, 1H), 4.18−4.07
(m, 1H), 3.98 (dd, J = 11.5, 3.0 Hz, 1H), 3.81−3.65 (m, 2H), 3.57
(dd, J = 11.6, 6.5 Hz, 1H), 2.19−2.06 (m, 1H), 1.93−1.78 (m, 2H),
1.75−1.62 (m, 1H). LCMS (method LCMS9, ESI): RT = 1.63 min,
m/z = 403.2 [M + H]+.
(S)-1-(Phenylsulfonyl)-N4-(tetrahydro-2H-pyran-3-yl)-1H-
pyrrolo[2,3-b]pyridine-4,5-diamine (21c). Palladium on carbon
(10%, 30.0 g, 28.2 mmol) was added to a solution of 21b in EtOAc
(2.5 L). The mixture was stirred at 23 °C under an atmosphere of
hydrogen gas for 48 h, then filtered through Celite and washed with
10% MeOH in DCM (1 L). The filtrate was concentrated to afford
106 g (100%) of the title compound as an off-white solid. LCMS
(method LCMS9, ESI): RT = 1.13 min, m/z = 373.2 [M + H]+.
(R)-1-(6-(Phenylsulfonyl)-1-((S)-tetrahydro-2H-pyran-3-yl)-
1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol
(21d). (R)-(+)-lactamide (33.5 g, 376 mmol) was added to a
suspension of triethyloxonium tetrafluoroborate (73.6 g, 376 mmol) in
THF (800 mL).23 The reaction mixture was stirred for 1 h at 23 °C,
during which time it became a clear solution. The solution was then
added to a solution of 21c (70.0 g, 188 mmol) in EtOH (1.5 L), and
the reaction mixture was stirred at 67 °C for 2 h. The solution was
then cooled to room temperature, and EtOAc (500 mL) was added.
The mixture was allowed to stand for 16 h, then the resulting
(1R,3R)-3-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-
b]pyridin-1(6H)-yl)cyclohexanol (24). 24 was synthesized by
methods similar to those used for 21, substituting (1R,3R)-3-
aminocyclohexanol (24a)29 for (S)-tetrahydro-2H-pyran-3-amine
1
(21a). H NMR (400 MHz, DMSO-d6) δ 11.80 (s, 1H), 8.52 (s,
1H), 7.45 (t, J = 2.6 Hz, 1H), 6.66 (d, J = 3.0 Hz, 1H), 5.77−5.45 (m,
1H), 5.40−5.24 (m, 1H), 5.14−5.02 (q, J = 6.5 Hz, 1H), 4.68 (br s,
1H), 4.22 (br s, 1H), 2.48−2.40 (m, 1H), 2.35−2.19 (m, 1H), 2.01−
1.74 (m, 4H), 1.74−1.63 (m, 2H), 1.61 (d, J = 6.5 Hz, 3H). LCMS
(method LCMS1, ESI): RT = 3.87 min, m/z = 301.1 [M + H]+.
(R)-1-(1-(trans-4-Methoxycyclohexyl)-1,6-dihydroimidazo-
[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (25).27 1H NMR (400
MHz, DMSO-d6) δ 11.85 (s, 1H), 8.54 (s, 1H), 7.47 (t, J = 3.0 Hz,
1H), 6.69 (m, 1H), 5.66 (d, J = 6.7 Hz, 1H), 5.12 (m, 1H), 4.84 (m,
1H), 3.54 (m, 4H), 2.39 (m, 2H), 2.24 (m, 2H), 1.94 (m, 2H), 1.63
Q
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(d, J = 6.5 Hz, 3H), 1.40 (m, 2H). LCMS (method LCMS11, ESI):
RT = 2.34 min, m/z = 315.5 [M + H]+.
1.93−1.80 (m, 1H), 1.62 (d, J = 6.5 Hz, 3H). LCMS (method
LCMS1, ESI): RT = 7.54 min, m/z = 368.2 [M + H]+.
(R)-1-(1-((1S,3S)-3-((2,2,2-Trifluoroethyl)amino)cyclopentyl)-
1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol
(30). 30 was synthesized by methods similar to those used for 26,
substituting tert-butyl ((1S,3S)-3-aminocyclopentyl)carbamate (30a)32
for tert-butyl (trans-4-aminocyclohexyl)carbamate (26a). 1H NMR
(400 MHz, DMSO-d6) δ 11.83 (s, 1H), 8.55 (s, 1H), 7.45 (t, J = 3.0
Hz, 1H), 6.49 (dd, J = 3.5, 1.8 Hz, 1H), 5.60 (d, J = 6.6 Hz, 1H), 5.52
(p, J = 9.5 Hz, 1H), 5.10 (p, J = 6.5 Hz, 1H), 3.71−3.62 (m, 1H),
3.32−3.20 (m, 2H), 2.64−2.56 (m, 1H), 2.48−2.11 (m, 4H), 2.06−
1.96 (m, 1H), 1.71−1.62 (m, 1H), 1.64 (d, J = 6.4 Hz, 3H). LCMS
(method LCMS1, ESI): RT = 2.58 min, m/z = 368.1 [M + H]+.
trans-4-((5-Nitro-1-(phenylsulfonyl)-1H-pyrrolo[2,3-b]-
pyridin-4-yl)amino)cyclohexanecarbonitrile (31b). A mixture of
trans-4-aminocyclohexanecarbonitrile trifluoroacetic acid salt33 (31a,
6.36 g, 26.7 mmol), 359 (8.21 g, 24.3 mmol) and DIPEA (12.5 mL,
73.0 mmol) in isopropyl alcohol (200 mL) was heated at reflux for 3 h.
The cooled mixture was filtered and the precipitate washed with
isopropyl alcohol (100 mL) to afford 9.04 g (87%) of the title
(R)-1-(1-(trans-4-Aminocyclohexyl)-6-(phenylsulfonyl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (26e).
26d was synthesized by methods similar to those used for 31d,
substituting tert-butyl (trans-4-aminocyclohexyl)carbamate (26a) for
trans-4-aminocyclohexanecarbonitrile trifluoroacetic acid salt (31a).
26d (2.00 g, 3.71 mmol) was then added to a mixture of trifluoroacetic
acid (10 mL) and water (1 mL). The reaction mixture was stirred at 23
°C for 1 h, then concentrated under vacuum. The residue was taken up
in MeOH and purified by Isolute SCX column chromatography
(gradient: MeOH to 2 M NH3 in MeOH) to give 1.59 g (98%) of the
title compound as an off-white solid. LCMS (method LCMS10, ESI):
RT = 1.81 min, m/z = 440.3 [M + H]+.
(R)-1-(1-(trans-4-((2,2,2-Trifluoroethyl)amino)cyclohexyl)-
1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol
(26). 26e (200 mg, 0.46 mmol) was dissolved in DCM (2 mL) and
DMF (2 mL). TEA (254 μL, 1.84 mmol) was added, followed by
2,2,2-trifluoroethyl trifluoromethanesulfonate (131 μL, 0.91 mmol).
The reaction mixture was stirred for 16 h at 23 °C. Water (10 mL) was
added and the mixture was extracted with DCM (3 × 5 mL), and the
combined organic extracts were dried (Na2SO4), filtered, and
concentrated under vacuum. The residue was dissolved in MeOH (2
mL) and THF (2 mL), and 2 M aqueous NaOH (2 mL) was added.
The reaction mixture was stirred at 23 °C for 2 h, then concentrated
under vacuum. Then 1 M aqueous HCl solution (4 mL) was added to
the residue, and the mixture was concentrated under vacuum.
Purification by column chromatography on silica gel (gradient:
DCM to 8% (2 M NH3 in MeOH) in DCM) followed by trituration
with CH3CN gave 77 mg (44%) of the title compound as an off-white
1
compound as a bright-yellow solid. H NMR (400 MHz, CDCl3) δ
9.11 (s, 1H), 9.07 (d, J = 7.7 Hz, 1H), 8.20 (m, 2H), 7.63 (m, 2H),
7.53 (m, 2H), 6.67 (d, J = 4.2 Hz, 1H), 4.06 (m, 1H), 2.66 (m, 1H),
2.24 (m, 4H), 1.83 (m, 2H), 1.58 (m, 2H). LCMS (method LCMS11,
ESI): RT = 4.15 min, m/z = 426.0 [M + H]+.
trans-4-((5-Amino-1-(phenylsulfonyl)-1H-pyrrolo[2,3-b]-
pyridin-4-yl)amino)cyclohexanecarbonitrile (31c). A suspension
of 31b (5.00 g, 11.8 mmol), iron powder (−325 mesh, 2.59 g, 47.0
mmol), and ammonium chloride (3.74 g, 70.5 mmol) in a mixture of
EtOH and water (3:1, 400 mL) was mechanically stirred at reflux for 4
h. The cooled mixture was filtered through Celite. The filtrate was
concentrated under vacuum to remove the majority of the EtOH, then
the residual aqueous layer was extracted with DCM (3 × 100 mL).
The combined extracts were washed with water and brine (100 mL
each), dried over Na2SO4, filtered, and concentrated under vacuum.
Purification by diethyl ether trituration afforded 4.51 g (97%) of the
title compound as a light-gray solid. LCMS (method LCMS11, ESI):
RT = 2.69 min, m/z = 396.1 [M + H]+.
1
solid. H NMR (400 MHz, DMSO-d6) δ 11.82 (s, 1H), 8.54 (s, 1H),
7.47 (t, J = 3.0 Hz, 1H), 6.71−6.63 (dd, J = 3.5, 1.9 Hz, 1H), 5.64 (d, J
= 6.5 Hz, 1H), 5.12 (p, J = 6.5 Hz, 1H), 4.89−4.74 (m, 1H), 3.42−
3.31 (m, 2H), 2.92−2.74 (m, 1H), 2.46−2.29 (m, 3H), 2.07 (d, J =
12.2 Hz, 2H), 1.98−1.83 (m, 2H), 1.63 (d, J = 6.3 Hz, 3H), 1.40−1.22
(m, 2H). LCMS (method LCMS1, ESI): RT = 2.57 min, m/z = 382.4
[M + H]+.
(R)-1-(1-(trans-4-((2,2-Difluoroethyl)amino)cyclohexyl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (27).
27 was synthesized by methods similar to those used for 26,
substituting 2,2-difluoroethyl trifluoromethanesulfonate for 2,2,2-
trifluoroethyl trifluoromethanesulfonate. 1H NMR (400 MHz,
DMSO-d6) δ 11.81 (s, 1H), 8.54 (s, 1H), 7.46 (t, J = 2.9 Hz, 1H),
6.69 (t, J = 2.5 Hz, 1H), 6.01 (tt, J = 56.6, 4.3 Hz, 1H), 5.64 (d, J = 6.2
Hz, 1H), 5.12 (p, J = 6.5 Hz, 1H), 4.90−4.74 (m, 1H), 3.01 (t, J = 16.2
Hz, 2H), 2.89−2.72 (m, 1H), 2.38 (q, J = 12.8 Hz, 2H), 2.12 (d, J =
12.0 Hz, 2H), 2.01−1.83 (m, 3H), 1.63 (d, J = 6.4 Hz, 3H), 1.37−1.19
(m, 2H). LCMS (method LCMS1, ESI): RT = 2.20 min, m/z = 364.4
[M + H]+.
trans-4-(2-((R)-1-Hydroxyethyl)-6-(phenylsulfonyl)imidazo-
[4,5-d]pyrrolo[2,3-b]pyridin-1(6H)-yl)cyclohexanecarbonitrile
(31d). A suspension of triethyloxonium tetrafluoroborate (5.36 g, 28.2
mmol) and (R)-(+)-lactamide (2.85 g, 32.0 mmol) in dry THF (40
mL) was stirred at 23 °C under a nitrogen atmosphere for 2 h.23 The
mixture was concentrated under vacuum and stirred in absolute EtOH
(70 mL) for 10 min. 31c (4.00 g, 10.1 mmol) was added, and the
reaction mixture was heated at reflux for 2 h. The cooled mixture was
concentrated under vacuum, and the residue was triturated with
EtOAc to afford 4.50 g (99%) of the title compound as an off-white
1
solid. H NMR (400 MHz, DMSO-d6) δ 8.71 (s, 1H), 8.13 (m, 2H),
7.94 (d, J = 4.1 Hz, 1H), 7.69 (m, 1H), 7.61 (m, 2H), 7.25 (m, 1H),
5.74 (d, J = 6.7 Hz, 1H), 5.14 (m, 1H), 4.85 (m, 1H), 3.23 (m, 1H),
2.20 (m, 4H), 1.85 (m, 4H), 1.62 (d, J = 6.4 Hz, 3H). LCMS (method
LCMS11, ESI): RT = 3.33 min, m/z = 450.3 [M + H]+.
(R)-1-(1-(cis-4-((2,2,2-Trifluoroethyl)amino)cyclohexyl)-1,6-
dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol (28).
28 was synthesized by methods similar to those used for 26,
substituting tert-butyl (cis-4-aminocyclohexyl)carbamate (28a)30 for
trans-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)cyclohexanecarbonitrile (31). A solution of 31d
(1.20 g, 2.67 mmol) in MeOH:THF (1:1, 150 mL) was treated with a
1 M solution of aqeuous sodium hydroxide (20 mL) and stirred at 23
°C for 3 h. The organic solvent was removed under vacuum, and the
aqueous residue was extracted with EtOAc (3 × 20 mL). The
combined organic extracts were washed with water and brine (20 mL
each), dried over Na2SO4, filtered, and concentrated under vacuum.
Purification by column chromatography on silica gel (gradient: 0−10%
[2 M NH3 in MeOH] in EtOAc) with further purification by reverse
phase HPLC (XBridge Prep C18 OBD, 19 mm × 250 mm, 5 μm;
detection, UV 220 and 254 nm; mobile phase A, water with 0.1%
NH4OH; mobile phase B, CH3CN with 0.1% NH4OH; flow rate, 18
mL/min; gradient 5−75% B over 20 min) afforded 149 mg (18%) of
the title compound as a white solid. 1H NMR (400 MHz, DMSO-d6) δ
11.82 (s, 1H), 8.54 (s, 1H), 7.44 (t, J = 2.9 Hz, 1H), 6.79 (t, J = 2.5
Hz, 1H), 5.63 (d, J = 6.0 Hz, 1H), 5.13 (p, J = 6.3 Hz, 1H), 4.90−4.76
1
tert-butyl (trans-4-aminocyclohexyl)carbamate (26a). H NMR (400
MHz, DMSO-d6) δ 11.65 (s, 1H), 8.51 (s, 1H), 7.64 (s, 1H), 7.27 (t, J
= 3.0 Hz, 1H), 5.62 (d, J = 6.0 Hz, 1H), 5.11 (p, J = 6.4 Hz, 1H), 4.81
(tt, J = 12.3, 4.2 Hz, 1H), 3.43−3.32 (m, 2H), 3.00 (s, 1H), 2.79−2.58
(m, 3H), 1.99 (d, J = 13.2 Hz, 2H), 1.72−1.52 (m, 7H). LCMS
(method LCMS1, ESI): RT = 8.43 min, m/z = 382.2 [M + H]+.
(R)-1-(1-((1S,3R)-3-((2,2,2-Trifluoroethyl)amino)cyclopentyl)-
1,6-dihydroimidazo[4,5-d]pyrrolo[2,3-b]pyridin-2-yl)ethanol
(29). 29 was synthesized by methods similar to those used for 26,
substituting tert-butyl ((1R,3S)-3-aminocyclopentyl)carbamate
(29a)31 for tert-butyl (trans-4-aminocyclohexyl)carbamate (26a). H
1
NMR (400 MHz, DMSO-d6) δ 11.80 (s, 1H), 8.54 (s, 1H), 7.42 (t, J =
3.0 Hz, 1H), 6.96 (dd, J = 3.4, 1.9 Hz, 1H), 5.65 (d, J = 6.5 Hz, 1H),
5.38−5.23 (m, 1H), 5.12 (p, J = 6.5 Hz, 1H), 3.43−3.33 (m, 1H),
3.30−3.21 (m, 2H), 2.81 (td, J = 7.7, 5.0 Hz, 1H), 2.42 (td, J = 12.0,
11.3, 6.6 Hz, 1H), 2.32 (dt, J = 13.2, 6.8 Hz, 1H), 2.24−2.00 (m, 3H),
R
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(m, 1H), 3.21−3.05 (m, 1H), 2.41−2.18 (m, 4H), 2.01−1.90 (m, 2H),
1.90−1.71 (m, 2H), 1.63 (d, J = 6.4 Hz, 3H). LCMS (method
LCMS1, ESI): RT = 4.48 min, m/z = 310.1 [M + H]+.
ACKNOWLEDGMENTS
■
We thank Michael Hayes, Mengling Wong, Baiwei Lin, Deven
Wang, Yutao Jiang, and Yanzhou Liu for purification and
analytical support; Daniel Hascall, Grady Howes, Gigi Yuen,
Garima Porwal, and Mandy Yim for compound management
support; Hoa Le and Qin Yue for ADME support; Emile Plise
and Jonathan Cheng for MDCK data; Erlie Delarosa for
hepatocyte stability; Ning Liu for microsomal stability studies;
Suzanne Tay for TDI studies; Jasleen Sodhi for reversible CYP
inhibition measurement; Quynh Ho for plasma protein binding
measurement; Wei Jia, Po-Chang (Tom) Chiang, and Amy
Sambrone for formulations support; Zhiyu Huang for collation
of data from the rat CIA study.
cis-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-b]-
pyridin-1(6H)-yl)cyclohexanecarbonitrile (32). 32 was synthe-
sized by methods similar to those used for 31, substituting cis-4-
aminocyclohexanecarbonitrile trifluoroacetic acid salt (31a)33 for trans-
4-aminocyclohexanecarbonitrile trifluoroacetic acid salt (32a). 1H
NMR (400 MHz, DMSO-d6) δ 11.82 (s, 1H), 8.55 (s, 1H), 7.45 (t, J =
3.0 Hz, 1H), 7.18 (s, 1H), 5.65 (d, J = 6.8 Hz, 1H), 5.14 (p, J = 6.6 Hz,
1H), 4.89−4.76 (m, 1H), 3.34 (br s, 1H), 2.60−2.46 (m, 2H), 2.23−
2.10 (m, 2H), 2.01−1.81 (m, 4H), 1.64 (d, J = 6.4 Hz, 3H). LCMS
(method LCMS1, ESI): RT = 4.42 min, m/z = 310.1 [M + H]+.
2-(trans-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-
b]pyridin-1(6H)-yl)cyclohexyl)acetonitrile (33). 33 was synthe-
sized by methods similar to those used for 31, substituting 2-(trans-4-
aminocyclohexyl)acetonitrile (33a)27 for trans-4-aminocyclohexane-
carbonitrile trifluoroacetic acid salt (31a). 1H NMR (400 MHz,
DMSO-d6) δ 11.82 (s, 1H), 8.54 (s, 1H), 7.45 (t, J = 2.9 Hz, 1H), 6.72
(t, J = 2.4 Hz, 1H), 5.66 (d, J = 6.2 Hz, 1H), 5.13 (p, J = 6.5 Hz, 1H),
4.90−4.76 (m, 1H), 2.59 (d, J = 5.8 Hz, 2H), 2.45−2.30 (m, 2H),
2.14−1.88 (m, 5H), 1.63 (d, J = 6.3 Hz, 3H), 1.49−1.29 (m, 2H).
LCMS (method LCMS1, ESI): RT = 4.86 min, m/z = 324.1 [M +
H]+.
3-(trans-4-(2-((R)-1-Hydroxyethyl)imidazo[4,5-d]pyrrolo[2,3-
b]pyridin-1(6H)-yl)cyclohexyl)propanenitrile (34). 34 was syn-
thesized by methods similar to those used for 31, substituting 3-(trans-
4-aminocyclohexyl)propanenitrile trifluoroacetic acid salt (34a)27 for
trans-4-aminocyclohexanecarbonitrile trifluoroacetic acid salt (31a).
1H NMR (400 MHz, DMSO-d6) δ 11.82 (s, 1H), 8.54 (s, 1H), 7.47 (t,
J = 3.0 Hz, 1H), 6.68 (t, J = 2.5 Hz, 1H), 5.64 (d, J = 5.5 Hz, 1H), 5.12
(p, J = 6.5 Hz, 1H), 4.90−4.73 (m, 1H), 2.61 (t, J = 7.3 Hz, 2H),
2.43−2.29 (m, 2H), 2.05−1.86 (m, 4H), 1.76−1.56 (m, 6H), 1.31−
1.14 (m, 2H). LCMS (method LCMS1, ESI): RT = 6.13 min, m/z =
338.1 [M + H]+.
ABBREVIATIONS USED
■
ADME, absorption, distribution, metabolism, and excretion;
AUC, area under the curve; Bn, benzyl; Boc, tert-butyl
carbamate; brine, a saturated aqueous solution of sodium
chloride; CH3CN, acetonitrile; CIA, collagen induced arthritis;
CL, clearance; CLH, hepatic clearance, CLint, intrinsic clearance;
cLogD7.4, calculated log of partition coefficient between octanol
and pH7.4 aqueous buffer; Cmax, maximum concentration; CYP,
cytochrome P450; D, aspartic acid; DCM, dichloromethane;
DEA, diethylamine, DIPEA, N,N-diisopropylethylamine; DMF,
dimethylformamide; DMSO, dimethylsulfoxide; E, glutamic
acid; EC50, half-maximal effective concentration; EPO,
erythropoietin; EtOAc, ethyl acetate; EtOH, ethanol; ex,
example; F, oral bioavailability; Fmax, theoretical maximum
achievable oral bioavailability; h, hour; HBD, hydrogen-bond
donor; hERG, human ether-a-go-go-related gene; HLM, human
liver microsome; HPLC, high performance liquid chromatog-
raphy; IL-6, interleukin-6; i-PrOH, isopropyl alcohol; IV,
intravenous; JAK, Janus kinase; JAK1 biochemical selectivity
index, JAK2 Ki ÷ JAK1 Ki; JAK1 cellular selectivity index, EPO-
pSTAT5 EC50 ÷ IL6-pSTAT3 EC50; Ki, inhibition constant;
LCMS, liquid chromatography−mass spectrometry; LE, ligand
efficiency; LLE, ligand-lipophilicity efficiency; MCT, methyl
cellulose/Tween; MDCK, Madin−Darby canine kidney;
MeOH, methanol; min, minute; Papp, apparent permeability;
PEG400, polyethylene glycol 400; Ph, phenyl; PK, pharmaco-
kinetics; PO, by mouth; PPB, plasma protein binding; PSI,
pounds per square inch; RA, rheumatoid arthritis; RT,
retention time; SAR, structure−activity relationships; SFC,
supercritical fluid chromatography; STAT, signal transducer
and activator of transcription; (tmp)2Mg·2LiCl, 2,2,6,6-
tetramethylpiperidinemagnesium chloride lithium chloride
complex; TEA, triethylamine; TFA, trifluoroacetic acid; THF,
tetrahydrofuran; TPSA, topological polar surface area; UV,
ultraviolet
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic procedures for 16a, 33a, 34a, and 25; identities of
kinases in Table 6 and % inhibition by 21, 26, 29, 31, 33, and
34; biochemical JAK3 and TYK2 Ki values of 1−34; tabulated
compound structures and data used to produce the graphs in
Figure 2. Details of: in vitro and in vivo ADME experimental
procedures, JAK enzymatic and cellular assays (low serum and
whole blood), in vitro hERG assay, CIA efficacy study of 31 in
the rat, protein expression/purification, and crystallographic
methods and procedures for 31 (in complex with JAK1 and
JAK2) and 33 and 34 (in complex with JAK1), thermodynamic
solubility methods, and analytical LCMS and SFC conditions.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Accession Codes
PDB numbers: 4EHZ for 2 complexed with JAK1, 4F09 for 2
complexed with JAK2, 4IVB for 31 complexed with JAK1, 4IVA
for 31 complexed with JAK2, 4IVC for 33 complexed with
JAK1, 4IVD for 34 complexed with JAK1.
REFERENCES
■
(1) (a) O’Shea, J. J. Jaks, STATs, cytokine signal transduction, and
immunoregulation: are we there yet? Immunity 1997, 7, 1−11.
(b) Pesu, M.; Laurence, A.; Kishore, N.; Zwillich, S. H.; Chan, G.;
O’Shea, J. J. Therapeutic targeting of Janus kinases. Immunol. Rev.
2008, 223, 132−142.
(2) O’Shea, J. J.; Pesu, M.; Borie, D. C.; Changelian, P. S. A new
modality for immunosuppression: targeting the JAK/STAT pathway.
Nature Rev. Drug. Discovery 2004, 3, 555−564.
(3) (a) Clark, S. C.; Kamen, R. The human hematopoietic colony-
stimulating factors. Science 1987, 236, 1229−1237. (b) Metcalf, D. The
molecular control of cell division, differentiation commitment and
maturation in haemopoietic cells. Nature 1989, 339, 27−30.
AUTHOR INFORMATION
Corresponding Author
*Phone: 650-467-4533. E-mail: mzak@gene.com.
■
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
S
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(c) Witthuhn, B. A.; Quelle, F. W.; Silvennoinen, O.; Yi, T.; Tang, B.;
Miura, O.; Ihle, J. N. JAK2 associates with the erythropoietin receptor
and is tyrosine phosphorylated and activated following stimulation
with erythropoietin. Cell 1993, 74, 227−236.
(4) Lin, Q.; Meloni, D.; Pan, Y.; Xia, M.; Rodgers, J.; Shepard, S.; Li,
M.; Galya, L.; Metcalf, B.; Yue, T. Y.; Liu, P.; Zhou, J. Enantioselective
synthesis of Janus kinase inhibitor INCB018424 via an organocatalytic
aza-Michael reaction. Org. Lett. 2009, 11, 1999−2002.
(5) Flanagan, M. E.; Blumenkopf, T. A.; Brissette, W. H.; Brown, M.
F.; Casavant, J. M.; Shang-Poa, C.; Doty, J. L.; Elliott, E. A.; Fisher, M.
B.; Hines, M.; Kent, C.; Kudlacz, E. M.; Lillie, B. M.; Magnuson, K. S.;
McCurdy, S. P.; Munchhof, M. J.; Perry, B. D.; Sawyer, P. S.; Strelevitz,
T. J.; Subramanyam, C.; Sun, J.; Whipple, D. A.; Changelian, P. S.
Discovery of CP-690,550: a potent and selective Janus kinase (JAK)
inhibitor for the treatment of autoimmune diseases and organ
transplant rejection. J. Med. Chem. 2010, 53, 8468−8484.
(6) INCB018424/Ruxolitnib/Jakafi is FDA-approved for the treat-
ment of myelofibrosis. Incyte Press Release. http://investor.incyte.
com/phoenix.zhtmL?c=69764&p=irol-newsArticle&ID=
1631201&highlight (accessed March 27, 2013). CP-690,550/Tofaci-
tinib/Xeljanz is FDA-approved for the treatment of adults with
moderately to severely active rheumatoid arthritis (RA) who have had
an inadequate response or intolerance to methotrexate. Pfizer Press
Release. http://press.pfizer.com/press-release/multimedia-us-food-
and-drug-administration-approves-pfizers-xeljanz-tofacitinib-citrat (ac-
cessed March 27, 2013).
(7) Zak, M.; Mendonca, R.; Balazs, M.; Barrett, K.; Bergeron, P.;
Blair, W. S.; Chang, C.; Deshmukh, G.; DeVoss, J.; Dragovich, P. S.;
Eigenbrot, C.; Ghilardi, N.; Gibbons, P.; Gradl, S.; Hamman, C.;
Hanan, E. J.; Harstad, E.; Hewitt, P. R.; Hurley, C. A.; Jin, T.; Johnson,
A.; Johnson, T.; Kenny, J. R.; Koehler, M. F. T.; Bir Kohli, P.;
Kulagowski, J. J.; Labadie, S.; Liao, J.; Liimatta, M.; Lin, Z.; Lupardus,
P. J.; Maxey, R. J.; Murray, J. M.; Pulk, R.; Rodriguez, M.; Savage, S.;
Shia, S.; Steffek, M.; Ubhayakar, S.; Ultsch, M.; van Abbema, A.; Ward,
S. I.; Xiao, L.; Xiao, Y. Discovery and optimization of C-2 methyl
imidazopyrrolopyridines as potent and orally bioavailable JAK1
inhibitors with selectivity over JAK2. J. Med. Chem. 2012, 55, 6176−
6193.
(8) A summary of previous disclosures of compounds exhibiting
JAK1 vs JAK2 selectivity is presented in ref 9. Additionally, a number
of publications describing JAK2 inhibitors with varying JAK family and
other kinase selectivity profiles have appeared recently. These include:
(a) Schenkel, L. B.; Huang, X.; Cheng, A.; Deak, H. L.; Doherty, E.;
Emkey, R.; Gu, Y.; Gunaydin, H.; Kim, J. L.; Lee, J.; Loberg, R.;
Olivieri, P.; Pistillo, J.; Tang, J.; Wan, Q.; Wang, H. L.; Wang, S. W.;
Wells, M. C.; Wu, B.; Yu, V.; Liu, L.; Geuns-Meyer, S. Discovery of
potent and highly selective thienopyridine Janus kinase 2 inhibitors. J.
Med. Chem. 2011, 54, 8440−8450. (b) Lim, J.; Taoka, B.; Otte, R. D.;
Spencer, K.; Dinsmore, C. J.; Altman, M. D.; Chan, G.; Rosenstein, C.;
Sharma, S.; Su, H. P.; Szewczak, A. A.; Xu, L.; Yin, H.; Zugay-Murphy,
J.; Marshall, C. G.; Young, J. R. Discovery of 1-amino-5H-pyrido[4,3-
b]indol-4-carboxamide inhibitors of Janus kinase 2 (JAK2) for the
treatment of myeloproliferative disorder. J. Med. Chem. 2011, 54,
7334−7349. (c) Siu, T.; Kozina, E. S.; Jung, J.; Rosenstein, C.; Mathur,
A.; Altman, M. D.; Chan, G.; Xu, L.; Bachman, E.; Mo, J. R.;
Bouthillette, M.; Rush, T.; Dinsmore, C. J.; Marshall, C. G.; Young, J.
R. The discovery of tricyclic pyridone JAK2 inhibitors. Part 1: hit to
lead. Bioorg. Med. Chem. Lett. 2010, 20, 7421−7425. (d) William, A.
D.; Lee, A. C.; Blanchard, S.; Poulsen, A.; Teo, E. L.; Nagaraj, H.; Tan,
E.; Chen, D.; Williams, M.; Sun, E. T.; Goh, K. C.; Ong, W. C.; Goh,
S. K.; Hart, S.; Jayaraman, R.; Pasha, M. K.; Ethirajulu, K.; Wood, J.
M.; Dymock, B. W. Discovery of the macrocycle 11-(2-pyrrolidin-1-yl-
ethoxy)-14,19-dioxa-5,7,26-triaza-tetracyclo[19.3.1.1(2,6).1(8,12)]-
heptacosa-1(25),2(26),3,5,8,10,12(27),16,21,23-decaene (SB1518), a
potent Janus kinase 2/Fms-like tyrosine kinase-3 (JAK2/FLT3)
inhibitor for the treatment of myelofibrosis and lymphoma. J. Med.
Chem. 2011, 54, 4638−4658. (e) Ioannidis, S.; Lamb, M. L.; Wang, T.;
Almeida, L.; Block, M. H.; Davies, A. M.; Peng, B.; Su, M.; Zhang, H.
J.; Hoffmann, E.; Rivard, C.; Green, I.; Howard, T.; Pollard, H.; Read,
J.; Alimzhanov, M.; Bebernitz, G.; Bell, K.; Ye, M.; Huszar, D.; Zinda,
M. Discovery of 5-chloro-N2-[(1S)-1-(5-fluoropyrimidin-2-yl)ethyl]-
N4-(5-methyl-1H-pyrazol-3-yl)pyrimidine-2,4-diamine (AZD1480) as
a novel inhibitor of the Jak/Stat pathway. J. Med. Chem. 2011, 54,
262−276. (f) Wernig, G.; Kharas, M. G.; Okabe, R.; Moore, S. A.;
Leeman, D. S.; Cullen, D. E.; Gozo, M.; McDowell, E. P.; Levine, R.
L.; Doukas, J.; Mak, C. C.; Noronha, G.; Martin, M.; Ko, Y. D.; Lee, B.
H.; Soll, R. M.; Tefferi, A.; Hood, J. D.; Gilliland, D. G. Efficacy of
TG101348, a selective JAK2 inhibitor, in treatment of a murine model
of JAK2V617F-induced polycythemia vera. Cancer Cell 2008, 13, 311−
320. (g) Burns, C. J.; Bourke, D. G.; Andrau, L.; Bu, X.; Charman, S.
A.; Donohue, A. C.; Fantino, E.; Farrugia, M.; Feutrill, J. T.; Joffe, M.;
Kling, M. R.; Kurek, M.; Nero, T. L.; Nguyen, T.; Palmer, J. T.;
Phillips, I.; Shackleford, D. M.; Sikanyika, H.; Styles, M.; Su, S.;
Treutlein, H.; Zeng, J.; Wilks, A. F. Phenylaminopyrimidines as
inhibitors of Janus kinases (JAKs). Bioorg. Med. Chem. Lett. 2009, 19,
5887−5892. (h) Hexner, E. O.; Serdikoff, C.; Jan, M.; Swider, C. R.;
Robinson, C.; Yang, S.; Angeles, T.; Emerson, S. G.; Carroll, M.;
Ruggeri, B.; Dobrzanski, P. Lestaurtinib (CEP701) is a JAK2 inhibitor
that suppresses JAK2/STAT5 signaling and the proliferation of
primary erythroid cells from patients with myeloproliferative disorders.
Blood 2008, 111, 5663−5671.
(9) Kulagowski, J. J.; Blair, W.; Bull, R. J.; Chang, C.; Deshmukh, G.;
Dyke, H. J.; Eigenbrot, C.; Ghilardi, N.; Gibbons, P.; Harrison, T. K.;
Hewitt, P. R.; Liimatta, M.; Hurley, C. A.; Johnson, A.; Johnson, T.;
Kenny, J. R.; Bir Kohli, P.; Maxey, R. J.; Mendonca, R.; Mortara, K.;
Murray, J.; Narukulla, R.; Shia, S.; Steffek, M.; Ubhayakar, S.; Ultsch,
M.; van Abbema, A.; Ward, S. I.; Waszkowycz, B.; Zak, M.
Identification of Imidazo-pyrrolopyridines as Novel and Potent JAK1
Inhibitors. J. Med. Chem. 2012, 55, 5901−5921.
(10) We recently reported on a series of C-2 amine-containing
inhibitors capped with groups such as amides and sulfonamides.
Although these groups imparted consistent JAK1 selectivity, they also
produced large increases in TPSA. As such, the only P-loop
combinations with sufficient membrane permeability for acceptable
cell potency and oral bioavailability were highly lipophilic and
nonpolar (such as cycloalkanes and norbornanes). Labadie, S.;
Dragovich, P. S.; Barrett, K.; Blair, W. S.; Bergeron, P.; Chang, C.;
Deshmukh, G.; Eigenbrot, C.; Ghilardi, N.; Gibbons, P.; Hurley, C. A.;
Johnson, A.; Kenny, J. R.; Kohli, P. B.; Kulagowski, J. J.; Liimatta, M.;
Lupardus, P. J.; Mendonca, R.; Murray, J. M.; Pulk, R.; Shia, S.; Steffek,
M.; Ubhayakar, S.; Ultsch, M.; van Abbema, A.; Ward, S. I.; Zak, M.
Structure-based discovery of C-2 substituted imidazo-pyrrolopyridine
JAK1 inhibitors with improved selectivity over JAK2. Bioorg. Med.
Chem. Lett. 2012, 22, 7627−7633.
(11) (a) Kuntz, I. D.; Chen, K.; Sharp, K. A.; Kollman, P. A. The
maximal affinity of ligands. Proc. Natl. Acad. Sci U. S. A. 1999, 96,
9997−10002. (b) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand
efficiency: a useful metric for lead selection. Drug Discovery Today
2004, 9, 430−431.
(12) (a) Leeson, P. D.; Springthorpe, B. The influence of drug-like
concepts on decision-making in medicinal chemistry. Nature Rev. Drug
Discovery 2007, 6, 881−890. (b) Ryckmans, T.; Edwards, M. P.;
Horne, V. A.; Correia, A. M.; Owen, D. R.; Thompson, L. R.; Tran, I.;
Tutt, M. F.; Young, T. Rapid assessment of a novel series of selective
CB2 agonists using parallel synthesis protocols: a lipophilic efficiency
(LipE) analysis. Bioorg. Med. Chem. Lett. 2009, 19, 4406−4409.
(13) Descriptions of the biochemical and cell-based assays are
provided in the Supporting Information.
(14) Prior to synthesis, predicted LogD values were calculated using
the MoKa sofware (www.moldiscovery.com, (ref 14a)) and a custom
pKa model (ref 14b). (a) Milletti, F.; Storchi, L.; Sforna, G.; Cruciani,
G. New and original pKa prediction method using grid molecular
interaction fields. J. Chem. Inf. Model. 2007, 47, 2172−2181.
(b) Milletti, F.; Storchi, L.; Goracci, L.; Bendels, S.; Wagner, B.;
Kansy, M.; Cruciani, G. Extending pKa prediction accuracy: high-
throughput pKa measurements to understand pKa modulation of new
chemical series. Eur. J. Med. Chem. 2010, 45, 4270−4279.
T
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(15) Ertl, P.; Rohde, B.; Selzer, P. Fast calculation of molecular polar
surface area as a sum of fragment-based contributions and its
application to the prediction of drug transport properties. J. Med.
Chem. 2000, 43, 3714−3717.
Stewart, K. D.; Hajduk, P. J. Cheminformatic Tools for Medicinal
Chemists. J. Med. Chem. 2010, 53, 4830−4841. (o) Rogers, D.; Hahn,
M. Extended-Connectivity Fingerprints. J. Chem. Inf. Model. 2010, 50,
742−754. (p) Ortwine, D. F.; Aliagas, I. Physicochemical and DMPK
in silico models: facilitating their use by medicinal chemists. Mol.
Pharmaceutics 2013, 10, 1153−1161.
(18) Williams, N. K.; Bamert, R. S.; Patel, O.; Wang, C.; Walden, P.
M.; Wilks, A. F.; Fantino, E.; Rossjohn, J.; Lucet, I. S. Dissecting
specificity in the Janus kinases: the structures of JAK-specific inhibitors
complexed to the JAK1 and JAK2 protein tyrosine kinase domains. J.
Mol. Biol. 2009, 387, 219−232.
(19) Chrencik, J. E.; Patny, A.; Leung, I. K.; Korniski, B.; Emmons, T.
L.; Hall, T.; Weinberg, R. A.; Gormley, J. A.; Williams, J. M.; Day, J. E.;
Hirsch, J. L.; Kiefer, J. R.; Leone, J. W.; Fischer, H. D.; Sommers, C.
D.; Huang, H. C.; Jacobsen, E. J.; Tenbrink, R. E.; Tomasselli, A. G.;
Benson, T. E. Structural and thermodynamic characterization of the
TYK2 and JAK3 kinase domains in complex with CP-690550 and
CMP-6. J. Mol. Biol. 2010, 400, 413−433.
(16) In vitro metabolic stability in the presence of liver microsomes
and hepatocytes is expressed as hepatic clearance (CLH). CLH is
derived from the well stirred model according to the following
equation: CLH = (CLint·LBF)/(CLint + LBF), where CLint is intrinsic
clearance and LBF is liver blood flow. The fraction unbound in the
incubation and in blood are assumed to be 1. Human CLH values
(units = mL min−1 kg−1) are categorized are as follows: <6.2 = low,
6.2−15 = moderate, >15 = high. Further details on the in vitro
metabolic stability experiments used to measure CLint may be found in
the Supporting Information and in: Halladay, J. S.; Wong, S.; Jaffer, S.
M.; Sinhababu, A. K.; Khojasteh-Bakht, S. C. Metabolic stability screen
for drug discovery using cassette analysis and column switching. Drug
Metab. Lett. 2007, 1, 67−72.
(17) QSAR models used to prospectively predict metabolic stability
have previously been described (refs 17a−e). Similarly, an internal
HLM stability model was developed using experimental data from
18000 compounds spanning a variety of Genentech projects and
scaffolds. A regression model was built in Pipeline Pilot (ref 17f) using
a Support Vector Machine (refs 17g,17h) methodology implemented
in the R-Statistics (ref 17i) package. The following descriptors,
available in Pipeline Pilot, were used: ALogP, molecular weight,
number of H-bond donors, number of H-bond acceptors, number of
rotatable bonds, number of atoms, number of rings, number of
aromatic rings, molecular surface area, molecular polar surface area,
AlogP counts (ref 17j), E-state keys (ref 17k), MDL Public Keys (refs
17l,17m), and FCFP4 fingerprints (refs 17n,17o). The predicted
clearance was converted to the probability of a compound being stable
based on a total clearance cutoff of 13 mL min−1 kg−1 as described in
ref 17p. The structures of and predicted and experimental HLM
stability data associated with the compounds in Figure 2d are tabulated
in the Supporting Information. (a) Lee, P. H.; Cucurull-Sanchez, L.;
Lu, J.; Du, Y. H. J. Development of in silico models for human liver
microsomal stability. J. Comput.-Aided Mol. Des. 2007, 21, 665−673.
(b) Ekins, S.; Gupta, R. R.; Gifford, E. M.; Liston, T.; Waller, C. L.;
Hohman, M.; Bunin, B. A. Using Open Source Computational Tools
for Predicting Human Metabolic Stability and Additional Absorption,
Distribution, Metabolism, Excretion, and Toxicity Properties. Drug
Metab. Dispos. 2010, 38, 2083−2090. (c) Hu, Y. B.; Unwalla, R.;
Denny, R. A.; Bikker, J.; Di, L.; Humblet, C. Development of QSAR
models for microsomal stability: identification of good and bad
structural features for rat, human and mouse microsomal stability. J.
Comput.-Aided Mol. Des. 2010, 24, 23−35. (d) Sakiyama, Y.; Yuki, H.;
Moriya, T.; Hattori, K.; Suzuki, M.; Shimada, K.; Honma, T. Predicting
human liver microsomal stability with machine learning techniques. J.
Mol. Graphics Modell. 2008, 26, 907−915. (e) Langowski, J.; Long, A.
Computer systems for the prediction of xenobiotic metabolism. Adv.
Drug Delivery Rev. 2002, 54, 407−415. (f) Pipeline Pilot, v7.5; Accelrys
Inc.: San Diego, CA; http://accelrys.com/products/pipeline-pilot/.
(g) Warmuth, M. K.; Liao, J.; Ratsch, G.; Mathieson, M.; Putta, S.;
Lemmen, C. Active learning with support vector machines in the drug
discovery process. J. Chem. Inf. Comput. Sci. 2003, 43, 667−673.
(h) Ivanciuc, O. Applications of support vector machines in chemistry.
Rev. Comput. Chem 2007, 23, 291−400. (i) R Project for Statistical
Computing, (R-3.2.1); www.r-project.org/. (j) Ghose, A. K.;
Viswanadhan, V. N.; Wendoloski, J. J. Prediction of hydrophobic
(lipophilic) properties of small organic molecules using fragmental
methods: an analysis of ALOGP and CLOGP methods. J. Phys. Chem.
1998, 102, 3762−3772. (k) Hall, L. H.; Kier, L. B. The E-state as the
basis for molecular structure space definition and structure similarity. J.
Chem. Inf. Comput. Sci. 2000, 40, 784−791. (l) Durant, J. L.; Leland, B.
A.; Henry, D. R.; Nourse, J. G. Reoptimization of MDL keys for use in
drug discovery. J. Chem. Inf. Comput. Sci. 2002, 42, 1273−1280.
(m) Henry, D. R.; Durant, J. L. Optimization of MDL substructure
search keys for the prediction of activity and toxicity. Chemom.
Chemoinf. 2005, 894, 145−156. (n) Muchmore, S. W.; Edmunds, J. J.;
(20) The maximum achievable oral bioavailability (Fmax) is the
product of the fraction absorbed (Fa), the fraction escaping intestinal
metabolism (Fg), and the fraction escaping hepatic metabolism (Fh), as
expressed by the equation: Fmax = Fa·Fg·Fh. Assuming a best case
scenario of no intestinal wall metabolism, and complete absorption of
the oral dose (Fg and Fa = 1), the equation is simplified to Fmax = Fh.
Because Fh = 1 − CL/Qh, where CL is clearance and Qh is hepatic
blood flow, it is apparent that the maximum achievable bioavailability
should be controlled by clearance according to the equation: Fmax = 1
− CL/Qh. Hepatic blood flows of 70, 31, and 44 mL min−1 kg−1 were
used for rat, dog, and cyno Fmax calculations, respectively. See: Rautio,
J.; Kumpulainen, H.; Heimbach, T.; Oliyai, R.; Oh, D.; Jarvinen, T.;
̈
Savolainen, J. Prodrugs: design and clinical applications. Nature Rev.
Drug Discovery 2008, 7, 255−270.
(21) Detailed experimental procedures were previously published in
ref 7.
(22) Meerwein, H. Triethyloxonium fluoroborate. Org. Synth. 1973,
5, 1080.
(23) (a) In some instances when the imidate formation with
triethyloxonium tetrafluoroborate was performed in THF, polymer-
ization of the solvent occurred to form a viscous mass and the reaction
could not be continued. Similar observations were previously reported:
Vartak, A. P.; Crooks, P. A. Org. Process Res. Dev. 2009, 13, 415. (b)
Imidate formation can also be carried out in DCM: Kiessling, A. J.;
McClure, C. K. Synth. Commun. 1997, 27, 923.
(24) Clososki, G. C.; Rohbogner, C. J.; Knochel, P. Direct
Magnesiation of Polyfunctionalized Arenes and Heteroarenes Using
(tmp)2Mg·2LiCl. Angew. Chem., Int. Ed. 2007, 46, 7681−7684.
(25) All amines were either purchased from commercial sources,
synthesized as described in the references in the Experimental Section
or synthesized by the methods in the Supporting Information.
(26) Broo, A.; Holm, P.; Judkins, R.; Li, L.; Lindstedt-Alstermark, E.;
Sandberg, P.; Swanson, M.; Weidolf. L.; Mrickmann, K. Non-anilinic
derivatives of isothiazol-3(2H)-one 1,2-dioxides as Liver X receptor
modulators. WO2006073366, 2006.
(27) Prepared as described in the Supporting Information.
(28) Savage, S.; Babu, S.; Zak, M.; Mao, Z.; Cao, J.; Ge, Y.; Ma, D.;
Jiang, G. Kilogram synthesis of (S)-3-aminopyran from L-glutamic acid.
Synlett 2013, 24, 987−990.
(29) Brocklehurst, C. E.; Laumen, K.; La Vecchia, L.; Shaw, D.;
Vogtle, M. Diastereoisomeric salt formation and enzyme-catalyzed
̈
kinetic resolution as complementary methods for the chiral separation
of cis-/trans-enantiomers of 3-aminocyclohexanol. Org. Process Res.
Dev. 2011, 15, 294−300.
(30) Sekiguchi, Y.; Kanuma, K.; Omodera, K.; Busujima, T.; Tran, T.;
Han, S.; Casper, M.; Kramer, B. A.; Semple, G.; Zou, N. Quinoline,
tetrahydroquinoline and pyrimidine derivatives as MCH antagonist.
EP1464335, 2004.
(31) Sekiguchi, Y.i; Kanuma, K.; Omodera, K.; Busujima, T.; Tran,
T.; Han, S.; Casper, M.; Kramer, B. A. Preparation of novel
quinazolines as MCH receptor antagonists. WO2004087680, 2004.
U
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(32) Thompson, W.; Young, S. D.; Phillips, B. T.; Munson, P.;
Whitter, W.; Liverton, N.; Dieckhaus, C.; Butcher, J.; Mccauley, J. A.;
Mcintyre, C. J.; Layton, M. E.; Sanderson, P. E. 4-Cycloalkylaminopyr-
azolopyrimidine NMDA/NR2B antagonists. US20050054658. 2005.
(33) Aahlin, K.; Arvidsson, P. I.; Besidski, Y.; Nilsson, L. I.
Preparation of cyano group containing isoindoline derivatives for
treating pain disorders. WO2009145719, 2009.
V
dx.doi.org/10.1021/jm4004895 | J. Med. Chem. XXXX, XXX, XXX−XXX