Imidazo[1,5-a]quinoxalines as irreversible BTK inhibitors for the treatment of rheumatoid arthritis

Article abstract of DOI:10.1016/j.bmcl.2011.09.008

Imidazo[1,5-a]quinoxalines were synthesized that function as irreversible Bruton's tyrosine kinase (BTK) inhibitors. The syntheses and SAR of this series of compounds are presented as well as the X-ray crystal structure of the lead compound 36 in complex with a gate-keeper variant of ITK enzyme. The lead compound showed good in vivo efficacy in preclinical RA models.

Full text of DOI:10.1016/j.bmcl.2011.09.008

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6258–6263
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Imidazo[1,5-a]quinoxalines as irreversible BTK inhibitors for the treatment
of rheumatoid arthritis
Kyung-Hee Kim a, , Andreas Maderna a, Mark E. Schnute a, Martin Hegen b, Shashi Mohan b,
⇑
Joy Miyashiro b, Laura Lin b, Evelyn Li b, Sean Keegan b, Jennifer Lussier b, Christopher Wrocklage b,
Cheryl L. Nickerson-Nutter b, Arthur J. Wittwer b, Holly Soutter c, Nicole Caspers c, Seungil Han c,
Ravi Kurumbail c, Kyri Dunussi-Joannopoulos b, John Douhan III b, Allan Wissner a
a Medicinal Chemistry, Pfizer, 200 Cambridge Park Drive, Cambridge, MA 02140, United States
b Inflammation and Autoimmunity, Pfizer, 200 Cambridge Park Drive, Cambridge, MA 02140, United States
c Structural Biology and Biophysics, Pfizer, 100 Eastern Point Road, Groton, CT 06340, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 July 2011
Revised 26 August 2011
Accepted 2 September 2011
Available online 10 September 2011
Imidazo[1,5-a]quinoxalines were synthesized that function as irreversible Bruton’s tyrosine kinase (BTK)
inhibitors. The syntheses and SAR of this series of compounds are presented as well as the X-ray crystal
structure of the lead compound 36 in complex with a gate-keeper variant of ITK enzyme. The lead com-
pound showed good in vivo efficacy in preclinical RA models.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
BTK
Autoimmune
Rheumatoid arthritis
Covalent kinase inhibitor
Imidazo[1,5-a]quinoxaline
Bruton’s tyrosine kinase (BTK), a nonreceptor cytoplasmic tyro-
sine kinase belonging to the Tec kinase family, has been shown to
be critical for B-cell proliferation, differentiation, and signaling.1
In human, defects in the BTK gene lead to X-linked agammaglobu-
linemia (XLA), which is characterized by a profound decrease in B
cell numbers and in serum immunoglobulin levels. BTK mutation
in mice resulted in X-linked immunodeficiency (xid), a similar
though less severe phenotype than XLA. Since B cells play an essen-
tial role in regulating immune response, inhibitors of BTK may be
effective in treating autoimmune diseases such as rheumatoid
arthritis (RA) and systemic lupus erythematosus (SLE). The evi-
dence that B cells play a central role in immune dysregulation in
RA is substantiated by the observation of clinical improvements
in RA patients receiving B-cell depleting antibody therapies such
as Rituximab. Currently, there are no small molecule BTK inhibitors
in clinical use for the treatment of autoimmune diseases. In this
communication, we will describe the discovery of a series of small
molecule inhibitors of BTK having efficacy in in vivo preclinical
models of RA.
Since the ATP binding site in kinases is highly conserved, it is
often difficult to find an ATP competitive inhibitor having sufficient
selectivity. In the present case, the ATP binding site of BTK shows a
close homology to that of other Src-family kinases, such as
lymphocyte-specific protein tyrosine kinase (LCK) and LYN. A lack
of selectivity of an inhibitor for BTK over these kinases could
potentially have negative consequences. BTK is a member of a group
of eleven tyrosine kinases (the Tec family kinases, EGFR, Jak3, ErbB2,
ErbB4, and BLK) that contain a conserved cysteine residue adjacent
to the ATP-binding site.2 This cysteine (Cys 481 in BTK) is a potential
nucleophilic site which could form a covalent adduct with an
electrophilic inhibitor. As the inhibitory activity of such an inhibitor
is dependent on the covalent interaction, this cysteine residue could
provide a handle for achieving the desired degree of selectivity.2a
Previously PCI-32765 (Fig. 1) has been reported to be a potent and
selective irreversible BTK inhibitor that binds covalently to Cys
481 and is efficacious in animal cancer models. It is currently under
clinical development in patients with
B cell non-Hodgkin
lymphoma.3
Our series of irreversible BTK inhibitors is based on the imi-
dazo[1,5-a]quinoxaline scaffold (Fig. 1). Some compounds of this
type have been previously reported as LCK inhibitors.4 Using
molecular modeling, we predicted that a Michael acceptor at-
tached via an amide linkage to the C8 position of this core structure
⇑
Corresponding author. Tel.: + 1 617 665 5647; fax: +1 617 665 5682.
E-mail
addresses:
khkim87@hotmail.com,
kyung-hee.kim@pfizer.com
(K.-H. Kim).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.09.008

K.-H. Kim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6258–6263
6259
O
O
O
O
F
NH2
NO2
HN
R3
NH
NO2
HN
R3
NH2
NO2
a
b
N2
R2
N
NH2
N
R3
8
N
N
R1
17
N
15
R4
16
N
4 X
R3 = H, F,
Cl
N
R3
7
O
N
O
N
HN
R3
N
N
HN
c
N
e, f
O
d
PCI-32765
Imidazo[1,5-a]quinoxaline
N
H
O
R3
NO2
BTK IC50 = 0.5 nM
LYN IC50 = 200 nM
EGFR IC50 = 5.6 nM
18
19
N
N
HN
R3
N
g
F3C
N
N
R4
Figure 1. The structure of the BTK inhibitor PCI-32765 and the proposed imidazo-
quinoxaline BTK inhibitor series.
O
R3
N
Cl
N
21
X
20
X= N, O, S
N
h
is properly oriented to allow a covalent interaction with Cys 481 of
the BTK enzyme. With this in mind, we decided to pursue the syn-
thesis of members of this series. The synthesis resulted in irrevers-
ible BTK inhibitors having high potency, good selectivity relative to
the Src-family kinases, and good in vivo efficacy in models of RA.
The syntheses of this class of compounds are illustrated in
Schemes 1 and 2 where the core structure was prepared using a
method previously reported.5 As shown in Scheme 1, the reaction
of 4-methoxy-2-nitroaniline 1 with ethyl glyoxylate provided com-
pound 2 which was then reacted with Tosmic to give the imidazole
3. Imidazole 3 was reduced with sodium hydrosulfite and cyclized
to give imidazo quinoxalinone 4. Selective nitration at the C8-
position was achieved using ammonium nitrate in sulfuric acid;
in this reaction, it is important to use no more than 1 equiv of
nitrate in order to avoid over nitration. The carbonyl group was
converted to chloride to give chloro derivative 5. The reaction of
this imidazo quinoxaline 5 with ortho-toluidine provided the
nitrogen linked arylated compound 6. The nitro group on
compound 6 was reduced to amino group to obtain compound 7
which was acylated to install a Michael acceptor providing the final
compounds 8–11. In order to prepare compounds with an alkyl
group on the nitrogen at C8 position, compound 7 was reacted
N
N
N
N
R4
O
R3
X
22-48
Scheme 2. Reagents: (a) CH3NH2, 40% in water, THF, reflux; (b) 50% ethyl
glyoxylate in toluene, methanol, reflux; (c) Tosmic, K2CO3, abs ethanol, reflux; (d)
Na2S2O4, acetic acid/water, reflux; (e) Hunig’s base, TFA-anhydride, rt; (f) POCl3,
reflux; (g) Method A: anilines, pyridine–HCl (cat.), amyl alcohol reflux, and then,
KOH, water/methanol, Method B: phenols, Cs2CO3 (3 equiv), DMF, or Method C:
thiols, Cs2CO3 (3 equiv), DMA; (h) Method A: (E)-4-(dimethylamino)but-2-enoic
acid HCl (1.0 equiv), POCl3 (1.0 equiv), ꢀ15 °C, DMA or Method B: (E)-4-(dimeth-
ylamino)but-2-enoic acid HCl, oxalyl chloride, DMF(trace amount), acetonitrile/
DMF.
with corresponding alkyl bromides to give intermediates 12 and
then subjected to acylation with the Michael acceptor to produce
the compounds 13 and 14. An alternative synthetic route for addi-
tional compounds is outlined in Scheme 2. As shown in Scheme 2,
three different starting materials 15 were used which differed
depending on the substituent on C7 position at the final com-
pounds. The methyl amino group was installed from the beginning
using methylamine to provide compounds 16. The later steps were
similar to the corresponding steps shown on Scheme 1 except for
the introduction of a trifluoroacetyl protecting group at the step
e to provide intermediates 20 after chlorination. Various nitrogen,
oxygen, or sulfur linked aryl group were coupled to provide com-
pounds 21. The methods used to incorporate these substituents
at the C4 position also resulted in deprotection of the trifluoroace-
tamide group. Finally, the Michael acceptor was introduced to
produce the final compounds 22–48.
O
O
O
O
O
NH2
NO2
a
b
N
NH
NO2
N
O
O
O
NO2
1
2
3
N
N
c
d, e
N
N
f
O2N
O
N
O
O
N
Cl
H
4
5
Table 1
Activities of imidazoquinoxaline compounds 8–11
N
N
O2N
O
N
H2N
O
N
N
g
H
N
N
N
R1
N
6
N
N
N
H
H
7
O
N
H
i
h
R2
HN
R2
N
N
a
a
Compound
R1
BTK IC50 (nM)
LYN IC50 (nM)
N
h
N
N
N
R1
N
O
N
N
O
N
H
O
8
0.847
46.6
21.3
H
O
12
via h
8-11, R2 = H
N
9
0.847
via i and h
13,
14,
O
2
R = Cyclopropylmethyl
O
2
R = Et
N
10
11
2.52
297
O
Scheme 1. Reagents: (a) 50% ethyl glyoxylate in toluene, methanol, reflux; (b)
Tosmic, K2CO3, abs ethanol, reflux; (c) Na2S2O4, acetic acid/water, reflux; (d)
NH4NO2, H2SO4; (e) POCl3, reflux; (f) o-toluidine, pyridine–HCl (cat.), amyl alcohol,
reflux; (g) H2, Pd/C, methanol; (h) acid chloride or mixed anhydride; (i) R2Br, Et3N,
DMF.
N
0.884
10.9
O
For experimental details regarding assay conditions see Supplementary data.
a
Enzyme IC50 values are averages of two or more experiments.

6260
K.-H. Kim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6258–6263
Table 2
Our initial efforts were focused on finding a proper Michael
acceptor for covalent binding with the BTK enzyme. On the basis
of our previous experience with irreversible inhibitors of EGFR/
Her2, which have the same conserved Cys residue, various Michael
acceptors were examined for covalent binding to Cys 481.7 As
shown in Table 1, compounds 8–11 each showed excellent BTK
inhibitory activity with good selectivity versus LYN kinase. Com-
pound 11 showed the best combination of properties such as po-
tency, potential reactivity, and physical properties, therefore, the
(E)-4-(dimethylamino)-N-methylbut-2-enamide group was chosen
for the further investigation.7a,c
As expected, most of the compounds synthesized inhibited the
in vitro proliferation of human B cells following stimulation
through the B cell receptor with anti-IgM. However, some of the
earliest compounds we prepared in this series, for example com-
pound 11, while functioning as a potent irreversible inhibitor of
BTK, did not inhibit B cell proliferation in cell culture, but, on the
contrary, appeared to stimulated B cell proliferation. We found that
we could overcome this issue simply by placing a small alkyl group
on the C8 amide nitrogen (see Table 2). A simple methyl group was
effective in eliminating the hyper-proliferation and increased the
BTK inhibitory potency. Consequently, all subsequent compounds
that we prepared in the series retained the methylated C8 amide
nitrogen. We also observed that compounds lacking a substituent
at the C7 position retained inhibitory activity (Table 2, compound
22).
Activities of imidazoquinoxaline compounds 11, 13, 14, and 22
R2
N
8
N
N
N
N
4
O
R3
7
N
H
Compound R2
R3
BTK IC50
(nM)
LYN IC50
(nM)
BTK B cell
IC50 (nM)
a
a
a,b
11
13
H
OCH3
0.884
10.9
1980
NAc
43.15
Cyclopropyl- OCH3 45.3
methyl
14
22
ethyl
OCH3 19.1
H 3.99
533
623
16.15
3.165
methyl
For experimental details regarding assay conditions see Supplementary data.
a
IC50 values are averages of two or more experiments.
IC50 for inhibition of B cell proliferation.
Hyper-proliferation was observed.
b
c
Subsequent to completion of this work, an internally-developed
in silico model noted compounds 6–14 and 21–48 to have some
structural similarity to chloracnegenic agents.6 A representative
sample of these compounds was selected for further screening.
Compounds 25, 36, 42, and 47 were tested (Xenobiotic Detection
Systems, Inc), in an in vitro CALUX assay, which is a sensitive test
that uses a genetically engineered mouse cell line containing the
firefly luciferase gene under transactivational control of the Aryl-
hydrocarbon receptor to screen for potential chloracnegens, such
as dioxin and dioxin-like compounds. The compounds tested failed
to induce activation of the Ah receptor, and were therefore consid-
We now turn our attention to modifications of the aryl group on
C4 position since we observed that the aryl group was very impor-
tant for the BTK activity. Compounds lacking an aryl group at this
position lost BTK and LYN inhibitory activity nearly completely
(Table 3, compound 23). The goal in optimizing the C4 aryl
ered negative for agonist activity (EC50 >10 lM compared to TCDD
(2,3,7,8-tetrachlorodibenzo-p-dioxin) with EC50 = 0.03 nM).
Table 3
Activities of imidazoquinoxaline compounds 22–48
N
8
N
N
R4
4
N
O
R3
N
X
7
a
a
a,b
Compound
R3
X
R4
BTK IC50 (nM)
LYN IC50 (nM)
BTK B cell IC50 (nM)
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
F
F
F
Cl
Cl
Cl
Cl
F
N
Cl
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
O
O
O
O
S
2-Me
No aryl group
3.99
4290
16.8
3.33
2.96
1.89
10.3
13.2
7.56
60.6
10.4
1.55
12.2
76.5
1.93
1.82
6.76
5.35
4.83
5.03
3.98
9.53
5.51
13.0
9.53
6.58
283
623
>50,000
1.55
1420
716
327
44.4
1670
378
2320
167
10.9
3.165
NA
NA
4.88
2.00
3.17
NA
53.7
4.12
NA
2-Me, 5-NHCOC6H4-m-CF3
2-Me, 5-F
2-Me, 4-F
2-Me, 3-F
2-Me, 6-Cl
2-Me, 5-Cl
2-Me, 4-Cl
2-Me, 3-Cl
2-Me, 6-OH
2-Me, 5-OH
2-Me, 4-OH
2-Me, 3-OH
2-Me
44.9
2.56
1.02
NA
350
490
624
3.41
6.75
23.5
8.08
2.17
4.40
5.84
2.41
17.4
144
32.3
228
651
2-Cl
2130
3080
3450
1300
2000
3390
1590
6020
7400
10,400
3730
19,600
2-Me, 5-F
2-Me, 4-F
2-Me
2-Cl
2-Me, 5-F
2-Me, 4-F
2-Me
2-Me, 5-F
2-Me, 4-F
2-Me, 4-F
2-Me
F
F
H
H
For experimental details regarding assay conditions see Supplementary data.
a
IC50 values are averages of two or more determinations.
IC50 for inhibition of B cell proliferation.
b

K.-H. Kim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6258–6263
6261
substituent was to identify irreversible BTK inhibitors that would
retain high potency in inhibiting the enzyme and in inhibiting B
cell proliferation while showing good selectivity with respect to
LYN inhibition (Table 3). The initial investigation was on the C7
unsubstituted series. Molecular modeling of our inhibitors having
smaller groups attached to the C4 aryl ring suggested that they
bound to the enzyme in the DFG-in configuration. Since we were
interested in investigating if compounds in our series could also
bind to BTK in the DFG-out configuration, we prepared the com-
pound 24 containing a well-known DFG-out type substituent at
the C4 position.8 While this compound retained BTK inhibitory
activity, it was found to be more potent for LYN inhibition. As a
consequence of this observation, we concentrated our efforts in
designing compounds that would only bind in the DFG-in configu-
ration. Molecular modeling predicted that only smaller substitu-
ents on the C4 aryl ring would be compatible with the DFG-in
binding mode and it was important that at least one of these sub-
stituents be located ortho to the C4 hetero-atom. Consequently, we
retained a methyl or chloro substituent in the ortho-position and
introduced fluoro, chloro, and hydroxy groups elsewhere on the
C4 aryl ring. In general, fluoro substituted compounds (25–27)
showed better BTK activity and selectivity over LYN compared to
the chloro or hydroxyl substituted compounds (28–35). Within
the fluoro substituted compounds, 5- or 4-substituents (25 and
26) showed better selectivity compared to the 3-F substituent
(27). A similar trend was observed with the chloro and hydroxyl
substituted compounds. Only fluoro substituted compounds 25
and 26 displayed better activity and selectivity relative to mono-
substituted compound 22.
ATP-site of BTK shares ꢁ70% sequence identity with ITK which is
another TEC family member kinase with ꢁ58% sequence identify
across the kinase domain. The crystal structure (Fig. 2) shows that
compound 36 binds at the active site by forming two direct hydro-
gen bonds at the kinase hinge region.9 The imidazo nitrogen (N2)
forms a hydrogen bond with the main chain amide nitrogen of
Met 438 in the hinge while the C4-amino group makes another
hydrogen bond with side chain oxygen of the gatekeeper Thr
435. In addition, the carbonyl oxygen adjacent to the Michael
acceptor forms a water-mediated hydrogen bond with Asp 445.
The toluene substituent binds into a hydrophobic pocket formed
N2
8
N
F
N
N
N
O
4
7
N
H
36
Introduction of fluoro or chloro substituents at C7 position re-
sulted in compounds that retained good BTK inhibitory activity
and sometimes improved selectivity over LYN inhibition relative
to the C7 unsubstituted analogs (36–43). We also examined
changes to the hetero-atom that linked the C4 aryl ring to the core
structure. In general, the oxygen linked compounds showed better
selectivity but decreased BTK potency compared to the nitrogen
linked compounds (36 vs 44, 38 vs 45, 39 vs 46, and 26 vs 47).
The sulfur linked compound (48) was much less active than oxygen
or nitrogen linked compounds.
Figure 2. X-ray structure of 36 in complex with a gate-keeper variant of ITK
enzyme in the ATP binding site.
While our focus is on the identification of an irreversible inhib-
itor of BTK, we were interested in the potency of the original
reversible scaffold. As shown in Table 4 the reversible compounds
without a Michael acceptor showed some BTK potency, but they
were much less potent than irreversible compounds and, in addi-
tion, they were significantly less selective against LYN. This shows
the significant impact of the covalent interaction on the BTK inhib-
itory potency and selectivity.
Table 5
Selectivity data for compounds 36 and 42
Compound
BTK IC50
(nM)
LYN IC50
(nM)
LCK IC50
(nM)
SRC IC50
(nM)
36
No
PI
W/
PI
No
PI
W/
PI
1.93
0.7
624
353
772
755
NA
298
84
In order to understand the molecular basis for the observed
SAR, we pursued crystallographic studies of a few compounds in
the series with a gate-keeper variant (Phe435Thr) of ITK. The
42
3.98
2.6
3390
3020
1620
410
2790
Table 4
Activities of reversible imidazoquinoxaline compounds 21
PI = pre-incubation.
N
8
HN
R3
N
N
R4
4
X
7
Table 6
EGFR Mechanistic differentiation of selected BTK inhibitors
21
a
a
Compound
BTK
No PI
IC50
BTK
W/PI
IC50
BTK
EGFR
No PI
IC50
EGFR
W/PI
IC50
EGFR
Compound
R3
X
R4
BTK IC50 (nM)
LYN IC50 (nM)
kinact/KI
kinact/KI
21a
21b
21c
21d
H
H
F
N
O
N
O
2-Me, 5-F
2-Me, 4-F
2-Me
72
1220
40
40
560
19
(1/(M.s))
(1/(M.s))
(nM)
(nM)
(nM)
(nM)
36
42
1.93
3.98
0.70
2.6
53,200
11,900
17.2
68
1.4
3.35
101,000
22,500
F
2-Me
670
130
For experimental details regarding assay conditions see Supplementary data.
a
PI = pre-incubation.
IC50 values are averages of two or more determinations.

6262
K.-H. Kim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6258–6263
Table 7
time it is not clear what dose would be needed to show efficacy in
Pharmacokinetic parameters of 36 and 42
treating RA and whether this dose will be high enough for any of
the EGFR side effects to be evident in patients treated with such
inhibitors. One could speculate that the dose needed to treat RA
could be significantly lower than the MTD used to treat cancer
and under these conditions the undesirable side effects due to
EGFR inhibition might be minimized.
The inhibition of BTK autophosphorylation at pY 223 in Ramos
cell by compound 36 was observed (IC50 = 48 nM), which is the evi-
dence that the inhibition of B-cell growth is due to the inhibiting
BTK.12 The pharmacokinetic properties of 36 and 42 were evalu-
ated following intravenous and oral delivery in mice. These studies
demonstrated that 36 and 42 exhibit favorable pharmacokinetic
properties with T1/2 of 1.5 h and 2.2 h (iv dose), AUC of 876 and
3622 h ng/mL (po dose), and oral bioavailability (%F) of 34 and
58, respectively (Table 7). While the T1/2 of these compounds in cir-
culation is relatively short, since these compounds are functioning
as irreversible inhibitors, the T1/2 in circulation may not be a good
measure of the effective half-life, which will be largely determined
by the rate of receptor turn over.
Compound
(mouse)
36
42
iv dose
po dose
iv dose
po dose
Dose (mg/Kg)
AUC0–1 (h ng/mL)
T1/2 (h)
CL (mL/min/kg)
Vss (L/kg)
Tmax (h)
Cmax (ng/ml)
F (%)
2
10
876
2.0
2
10
3622
2.7
510
1.5
63
6.2
1125
2.2
26.6
4.51
3.0
220
34
3.0
454
58
by Lys 391, Thr 435 and Leu 433 in the DFG-in mode. Also, we ob-
serve a tightly-bound glycerol molecule (from the crystallographic
cryo-solution) adjacent to the tricyclic core which forms a water
mediated hydrogen bond to N5 of the inhibitor. Most significantly,
Cys 442 (corresponding to Cys 481 in BTK) makes a covalent inter-
action with the Michael acceptor as predicted. This was subse-
quently confirmed by LC-Mass spectrometric studies which
showed the formation of a 1:1 stoichiometric complex formed by
36 with BTK.
Unlike reversible inhibitors, irreversible inhibitors typically dis-
play a time-dependent inhibition profile due to different enzyme
kinetics caused by covalent bond formation or protein conforma-
tional changes.10 In order to shed further light on the kinetics of
inhibition, we performed enzymatic assays whereby the com-
pounds were pre-incubated with the enzyme. The compounds in
our imidazo[1,5-a]quinoxaline series generally show good LYN/
LCK/Src selectivities versus BTK in the presence or absence of
pre-incubation, suggesting that the observed selectivities are not
solely the consequence of the covalent interaction (see Table 5
for selected compounds).
In vivo efficacy of 36 was examined in the NP-Ficoll model.13
NP-Ficoll is a T-independent type 2 (TI-2) antigen which can acti-
vate B-cells in the absence of T cell help. BTK-deficient mice are un-
able to mount TI-2 responses. Female C57BL/6 mice at 8–10 weeks
of age were immunized ip with NP(109)-Ficoll.14 Mice were dosed
with compound 36 as indicated or vehicle daily po for 7 days,
beginning 1 day prior to immunization. Serum was collected
6 days after immunization and NP-specific IgM and IgG3 antibody
titers were measured by ELISA. It is clear that compound 36 dem-
onstrates oral efficacy in the NP-Ficoll model in a dose responsive
manner (Fig. 3).
The efficacy of compound 36 was evaluated in the mouse
collagen-induced arthritis (CIA) model, an autoimmune disease
model which shares pathogenic similarities to human rheumatoid
arthritis. The model was performed as previously described.15
Compound 36 was examined in a semi-therapeutic model, which
involved assignment to treatment groups (n = 14/group) when
>10% of all the mice exhibited disease. Mice were scored in a
blinded manner for disease and dosed orally once each day for
22 days.16
We also monitored cross-over activity of compounds to EGFR
which also possesses a Cys residue corresponding to Cys 481 of
BTK. Off-target inhibition of EGFR has been observed during clini-
cal use of inhibitors of the ErbB kinase family (which are generally
dosed at the maximum tolerated dose (MTD) in treating cancer)
and could raise a concern for potential toxicity. We have found that
members of imidazoquinoxaline series show less selectivity for
EGFR when pre-incubated with the enzyme compared to studies
Treatment with compound 36 significantly inhibited the pro-
gression of the disease compared to the vehicle control at 10 and
3 mg/kg from day 7 and 5, respectively, to the end of the study.
The 1 mg/kg regimen was only significant at the end of the study
from day 19–22 post-treatment (Fig. 4).
where there is no pre-incubation (see Table
6 for selected
compounds). Poor selectivity with respect to EGFR under pre-
incubation conditions and a high kinact/KI ratio11, which is similar
or greater in magnitude to that of BTK, indicates that this series
could potentially inhibit EGFR in cells and tissues. However, at this
In summary, we have described the synthesis and SAR of a ser-
ies of imidazo[1,5-a]quinoxalines, which function as potent and
Anti-NP IgM
Anti-NP IgG3
p=0.2167
104
104
p=0.6991
p=0.7637
p=0.3309
p=0.1595
p=0.0010
103
p=0.0084
p=0.0002
p=0.0012
103
p=0.0007
102
101
102
Figure 3. Inhibition of TI-2 antibody responses to NP-Ficoll of 36.

K.-H. Kim et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6258–6263
6263
Spoerke, J. M.; Honigberg, L. A.; Young, P. R.; Dalrymple, S. A.; Palmer, J. T.
ChemMedChem 2007, 2, 58; (b) Honigberg, L. A.; Smith, A. M.; Sirisawad, M.;
Verner, E.; Loury, D.; Chang, B.; Li, S.; Pan, Z.; Thamm, D. H.; Miller, R. A.; Buggy,
J. J. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 13075.
4. Chen, P.; Doweyko, A. M.; Norris, D.; Gu, H. H.; Spergel, S. H.; Das, J.; Moquin, R.
V.; Lin, J.; Wityak, J.; Iwanowicz, E. J.; McIntyre, K. W.; Shuster, D. J.; Behnia, K.;
Chong, S.; de Fex, H.; Pang, S.; Pitt, S.; Shen, D. R.; Thrall, S.; Stanley, P.; Kocy, O.
R.; Witmer, M. R.; Kanner, S. B.; Schieven, G. L.; Barrish, J. C. J. Med. Chem. 2004,
47, 4517.
5. Chen, B.-C.; Zhao, R.; Bednarz, M. S.; Wang, B.; Sundeen, J. E.; Barrish, J. C. J. Org.
Chem. 2004, 69, 977.
6. Gawkrodger, D. J.; Harris, G.; Bojar, R. A. Br. J. Dermatol. 2009, 161, 939.
7. (a) Wissner, A.; Mansour, T. S. Arch. Pharm. Chem. Life Sci. 2008, 341, 465; (b)
Tsou, H.-R.; Overbeek-Klumpers, E. G.; Hallett, W. A.; Reich, M. F.; Floyd, B.;
Johnson, B. D.; Michalak, R. S.; Nilakantan, R.; Discafani, C.; Golas, J.; Rabindran,
S. K.; Shen, R.; Shi, X.; Wang, Y.-F.; Upeslacis, J.; Wissner, A. J. Med. Chem. 2005,
48, 1107; (c) Wissner, A.; Overbeek, E.; Reich, M. F.; Floyd, B.; Johnson, B. D.;
Mamuya, N.; Rosfjord, E. C.; Discafani, C.; Davis, R.; Shi, X.; Rabindran, S. K.;
Gruber, B. C.; Ye, F.; Hallett, W. A.; Nilakantan, R.; Shen, R.; Wang, Y.-F.;
Greenberger, L. M.; Tsou, H.-R. J. Med. Chem. 2003, 46, 49.
8. Okram, B.; Nagle, A.; Adrian, F. J.; Lee, C.; Ren, P.; Wang, X.; Sim, T.; Xie, Y.;
Wang, X.; Xia, G.; Spraggon, G.; Warmuth, M.; Liu, Y.; Gray, N. S. Chem. Biol.
2006, 13, 779.
9. PDB code: 3T9T.
Figure 4. Inhibition of BTK prevents disease in the murine collagen-induced
10. (a) Gierse, J. K.; Koboldt, C. M.; Walker, M. C.; Seibert, K.; Isakson, P. C. Biochem.
J. 1999, 339, 607; (b) Singh, J.; Dobrusin, E. M.; Fry, D. W.; Haske, T.; Whitty, A.;
McNamara, D. J. J. Med. Chem. 1997, 40, 1130; (c) Wang, Z.; Watt, W.; Brooks, N.
A.; Harris, M. S.; Urban, J.; Boatman, D.; McMillan, M.; Kahn, M.; Heinrikson, R.
L.; Finzel, B. C.; Wittwer, A. J.; Blinn, J.; Kamtekar, S.; Tomasselli, A. G. Biochim.
Biophys. Acta 2010, 1804, 1817.
arthritis (CIA) model.16
selective irreversible BTK inhibitors. Members of this series show
good efficacy in preclinical in vivo models of RA after oral dosing.
11. The efficiency of covalent inhibitors could be expressed as a function of the
kinact/KI ratio, see Ref. 10 for more detail.
Acknowledgments
KI
E þ I ¢ ½EIꢂ K!inact EIꢃ
The authors thank Eddine Saiah and Suvit Thaisrivongs for sup-
port of the program.
[EI] initial non covalent complex
EI⁄ final covalent complex
Supplementary data
12. Park, H.; Wahl, M. I.; Afar, D. E. H.; Turck, C. W.; Rawlings, D. J.; Tam, C.;
Scharenberg, A. M.; Kinet, J.-P.; Witte, O. N. Immunity 1996, 4, 515.
13. Ellmeier, W.; Jung, S.; Sunshine, M. J.; Hatam, F.; Xu, Y.; Baltimore, D.; Mano,
H.; Littman, D. R. J. Exp. Med. 2000, 192, 1611.
Supplementary data (experimental details for the biological as-
says, BTK autophosphorylation data, mass spectroscopic data of
binding of compound 36 with BTK, and kinase screening data for
compound 36) associated with this article can be found, in the on-
line version, at doi:10.1016/j.bmcl.2011.09.008.
14. The Pfizer institutional animal care and use committee reviewed and approved
the animal use in these studies. The animal care and use program is fully
accredited by the association for assessment and accreditation of laboratory
animal care, international.
15. Lin, T. H.; Hegen, M.; Quadros, E.; Nickerson-Nutter, C. L.; Appell, K. C.; Cole, A.
G.; Shao, Y.; Tam, S.; Ohlmeyer, M.; Wang, B.; Goodwin, D. G.; Kimble, E. F.;
Quintero, J.; Gao, M.; Symanowicz, P.; Wrocklage, C.; Lussier, J.; Schelling, S. H.;
Hewet, A. G.; Xuan, D.; Krykbaev, R.; Togias, J.; Xu, X.; Harrison, R.; Mansour, T.;
Collins, M.; Clark, J. D.; Webb, M. L.; Seidl, K. J. Arthritis Rheum. 2010, 62, 2283.
16. Arthritis was induced by immunizing DBA/1JBomTac mice with bovine type II
collagen (CII) emulsified in complete Freund’s adjuvant and by a boost 21 days
later with CII emulsified in incomplete Freund’s adjuvant, as described
previously.15 Disease severity was evaluated by scoring all four paws for
each animal, with a maximum possible score being 16. 0: No arthritis, 1: One
or two swollen digits, 2: Three or more swollen digits or mild to moderate
swelling of the entire paw, 3: Extensive swelling of the entire paw, 4:
Resolution of swelling, ankylosis of the paw. Values are the mean SEM. Mice
were dosed orally with compound 36 at 0.3, 1, 3, or 10 mg/kg QD. Disease
progression was significantly inhibited by 36 compared to the vehicle control
(P <0.05, by Student’s t-test) at 10 mg/kg (day 7–22), 3 mg/kg (day 5–22) and,
at the end of the study, 1 mg/kg (day 19–22).
References and notes
1. (a) Niiro, H.; Clark, E. A. Nat. Rev. Immunol. 2002, 2, 945; (b) Uckun, F. M.;
Tibbles, H. E.; Vassilev, A. O. Anti-Cancer Agents Med. Chem. 2007, 7, 624; (c) Di
Paolo, J. A.; Huang, T.; Balazs, M.; Barbosa, J.; Barck, K. H.; Bravo, B. J.; Carano, R.
A. D.; Darrow, J.; Davies, D. R.; DeForge, L.; Diehl, L.; Ferrando, R.; Gallion, S. L.;
Giannetti, A. M.; Gribling, P.; Hurez, V.; Hymowitz, S. G.; Jones, R.; Kropf, J. E.;
Lee, W. P.; Maciejewski, P. M.; Mitchell, S. A.; Rong, H.; Staker, B. L.; Whitney, J.
A.; Yeh, S.; Young, W. B.; Yu, C.; Zhang, J.; Reif, K.; Currie, K. S. Nat. Chem. Biol.
2011, 7, 41; (d) Chang, B. Y.; Huang, M. M.; Francesco, M.; Chen, J.; Sokolove, J.;
Magadala, P.; Robinson, W. H.; Buggy, J. J. Arthritis Res. Ther. 2011, 13, R115.
doi:10.1186/ar3400.
2. (a) Cohen, M. S.; Zhang, C.; Shokat, K. M.; Taunton, K. Science 2005, 308, 1318;
(b) Leproult, E.; Barluenga, S.; Moras, D.; Wurtz, J. M.; Winssinger, N. J. Med.
Chem. 2011, 54, 1347.
3. (a) Pan, Z.; Scheerens, H.; Li, S.-J.; Schultz, B. E.; Sprengeler, P. A.; Burrill, L. C.;
Mendonca, R. V.; Sweeney, M. D.; Scott, K. C. K.; Grothaus, P. G.; Jeffery, D. A.;