Purine derivatives as potent Bruton's tyrosine kinase (BTK) inhibitors for autoimmune diseases

Article abstract of DOI:10.1016/j.bmcl.2014.02.075

Investigation of various heterocyclic core isosteres of imidazopyrazines 1 & 2 yielded purine derivatives 3 & 8 as potent and selective BTK inhibitors. Subsequent SAR studies of the purine series led to the discovery of 20 as a leading compound. Compound

Full text of DOI:10.1016/j.bmcl.2014.02.075

Bioorganic & Medicinal Chemistry Letters 24 (2014) 2206–2211
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Purine derivatives as potent Bruton’s tyrosine kinase (BTK) inhibitors
for autoimmune diseases
⇑
Qing Shi , Andrew Tebben, Alaric J. Dyckman, Hedy Li, Chunjian Liu, James Lin, Steve Spergel,
James R. Burke, Kim W. McIntyre, Gilbert C. Olini, Joann Strnad, Neha Surti, Jodi K. Muckelbauer,
Chiehying Chang, Yongmi An, Lin Cheng, Qian Ruan, Katerina Leftheris, Percy H. Carter, Joseph Tino,
⇑
George V. De Lucca
Bristol-Myers Squibb Research and Development, P.O. Box 4000, Princeton, NJ 08543-4000, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Investigation of various heterocyclic core isosteres of imidazopyrazines 1 & 2 yielded purine derivatives 3
& 8 as potent and selective BTK inhibitors. Subsequent SAR studies of the purine series led to the discov-
ery of 20 as a leading compound. Compound 20 is very selective when screened against a panel of 400
kinases and is a potent inhibitor in cellular assays of human B cell function including B-Cell proliferation
and CD86 cell surface expression and exhibited in vivo efficacy in a mouse PCA model. Its X-ray co-crystal
structure with BTK shows that the high selectivity is gained from filling a BTK specific lipophilic pocket.
However, physical and ADME properties leading to low oral exposure hindered further development.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 1 January 2014
Revised 26 February 2014
Accepted 28 February 2014
Available online 13 March 2014
Keywords:
BTK
Autoimmune diseases
Purine
Bruton’s tyrosine kinase (BTK) is a member of the Tec family of
non-receptor tyrosine kinases. BTK is expressed in most hemato-
poietic cells such as B cells, mast cells, and macrophages but not
in T cells and plays a critical role in the signaling pathways down-
starting point, we first examined a variety of heterocyclic isosteres
of the imidazo[1,2-a]pyrazine core. As shown in Figure 1, purines 3
and 8, imidazo[4,5-c]pyridine 5, imidazo[1,2-b]pyridazine 9, and
imidazo[1,2-a]pyridine 10 are all good replacements of the original
core. Among all the cores that were examined, purine derivatives
displayed the most potent BTK inhibition.
stream of the B cell receptor as well as the Fc
cRIIa, FccRIIIa and
Fce
RI receptors.1 Mutations in the BTK gene lead to defects in B cell
development at the pre-B cell stage causing X-linked agammaglob-
ulinemia in humans (XLA) and less severe X-linked immunodefi-
ciency (xid) in mice.1g,h BTK deficient mice are resistant to mouse
animal models of lupus1i as well as collagen-induced arthritis
(CIA).1j A small molecule inhibitor of BTK activity represents an
intriguing approach to treat autoimmune diseases such as rheuma-
toid arthritis, multiple sclerosis and systemic lupus erythemato-
sus.2 In addition to autoimmune diseases, many pharmaceutical
companies have investigated small molecules irreversible covalent
inhibitors of BTK as an approach for B-Cell malignancies3 which
has recently culminated with the approval of Ibrutinib for Mantle
Cell Lymphoma (MCL) and Chronic Lymphocytic Leukemia (CLL).3d
At the beginning of our program we were interested in revers-
ible BTK inhibitors to target autoimmune diseases and were at-
tracted to reports of the imidazo[1,2-a]pyrazines represented by
compounds 1 and 2 in Figure 1.4 Using compounds 1 and 2 as a
N
N
N
N
N
N
N
N
N
N
N
O
5
3
4
N
BTK IC50
0.067 µM
BTK IC50
12.7 µM
BTK IC50
0.150 µM
O
NH
N
N
N
H
N
N
N
O
N
N
7
N
6
1
BTK IC50
>50 µM
BTK IC50
>50 µM
O
N
O
N
N
N
N
N
NH
N
N
N
N
N
N
H
N
N
10
9
8
O
BTK IC50
0.574 M
µ
BTK IC50
0.038 µM
BTK IC50
0.213
2
M
µ
⇑
Corresponding authors. Tel.: +1 6092523226 (Q.S.), +1 6092523121 (G.V.D).
E-mail addresses: qing.shi@bms.com (Q. Shi), george.delucca@bms.com
Figure 1. Heterocyclic replacement of the imidazo[1,2-a]pyrazine core.
(G.V. De Lucca).
http://dx.doi.org/10.1016/j.bmcl.2014.02.075
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

Q. Shi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2206–2211
2207
potency. Compounds lacking N-7, such as molecules 4, 6 and 7,
are inactive in BTK enzyme assay. Replacing the NH of the C-6 ani-
line of 8 with an oxygen also eliminates BTK enzyme potency (BTK
IC50 > 25 lM).
The model also shows that the benzene ring of the C-6 aniline
falls into a narrow groove formed by L408 and G480 and makes
hydrophobic interactions with these residues. To fit within the ste-
ric constraints imposed by L408 and G480, the aniline ring must re-
main nearly coplanar with respect to the bicyclic core. We
hypothesized that the aniline ring of compounds 3, 5 and 8 could
adopt a coplanar conformation due to their N-1 on the bicyclic core
that reduces steric interaction.5 On the other hand, the aniline ring
of 9 and 10 will be forced to adopt an out-of-plane conformation
due to a clash with the C–H on the bicyclic core adjacent to the
aniline substituent, consistent with their weaker potency. Our
Figure 2. Compound 8 docked into a BTK homology model.
Table 1
Evaluation of the para-substituent of the C-6 aniline
R
NH
5
7
6
N
1
N
H
N
8
9
N
N
3
2
4
O
R
Compound
BTK IC50 (lM)
O
11
8
0.064
0.038
0.025
0.028
0.017
0.143
0.043
OH
N
O
O
O
O
O
O
O
NH
N
12
13
14
15
16
N
N
N
OH
Figure 3. The ‘gatekeeper pocket’ in the BTK homology model.
N
Table 2
Optimization of the purine N-9 substituents
N
O
N
O
N
O
NH
O
N
N
N
H
N
17
0.087
N
H
N
N
R
O
O
OH
18
19
0.126
0.019
N
H
Selectivity (X)
R
N
N
Compound
BTK IC50
(
l
M) JAK2/
BTK
FLT3/
BTK
ITK/
BTK
Procedure for the BTK enzyme assay has been described in Ref. 6.
20
8
H
Me
0.008
0.038
383
4.9
51
3.3
6.4
0.13
21
0.17
281
0.34
0.16
In order to understand the dramatic potency differences among
the heterocyclic cores, we employed molecular modeling to study
the interactions of these compounds with BTK protein. Compound
8, as an example, was docked into a BTK homology model con-
structed from the crystal structure of the homologous kinase ITK
(Fig. 2). The model predicts that N-7 (see structure in Table 1 for
numbering) of the purine core and the NH of the C-6 aniline sub-
stituent form hydrogen bonds with M477 in the hinge region.
These two hydrogen bond interactions are critical to BTK enzyme
22
0.14
324
20
345
23
24
>50
29
1.0
1.7
0.10
1.7
1.0
2.4
O
O
O
25
0.97
42
36
OH

2208
Q. Shi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2206–2211
Table 3
Methyl substituent on the C-2 phenyl ring
O
N
O
NH
N
N
H
N
2
5
1
3
N
N
H
O
4
6
Selectivity (X)
SRC/BTK
R
Compound
BTK IC50
0.008
(l
M)
JAK2/BTK
383
FLT3/BTK
ITK/BTK
6.4
20
26
27
28
304
14
51
13
37
28
0.033
37
0.7
0.065
229
187
5
1.0
0.026
24
1.7
Table 4
Evaluation of the side chain
O
N
O
NH
N
N
R
N
H
N
Selectivity (X)
R
Compound
BTK IC50
0.008
(
lM)
JAK2/BTK
383
SRC/BTK
FLT3/BTK
60
BMX/BTK
ITK/BTK
6.4
H
N
20
29
304
222
4.5
2.3
O
H
N
0.032
993
155
6.9
O
F3C
H
N
30
31
32
33
0.048
0.083
0.169
0.109
36
35
16
5.8
20
14
9.0
2.0
1.3
1.2
6.5
10
O
H
N
10
11
O
H
N
5.4
4.5
3.1
5.4
3.9
9.2
O
H
N
O
H2N
34
35
0.077
0.035
5.4
5.0
7.1
11
3.8
6.0
0.7
0.8
2.1
2.1
H3C
H H
N N
O
36
37
38
39
0.018
0.049
0.027
0.041
13
5.6
74
18
26
9.2
92
11
9.1
1.3
16
1.0
0.8
1.8
2.4
5.2
2.4
45
H H
N N
O
H
N
S
O O
H
N
3.8
11
S
O O
hypothesis was confirmed with ab initio minimizations of model
systems consisting of the bicyclic core and unsubstituted aniline
ring at the B3LYP/6-31G⁄⁄ level of theory. As anticipated, the angle
between the plane of the bicyclic core and aniline ring was zero de-
grees for cores 3, 5 and 8 and 33 degrees for cores 9 and 10 at their
minimum energy conformations.
(Table 1). Indeed, very small effect on the BTK potency was ob-
served. The BTK IC50 fluctuated only within a small range from
17 nM to 143 nM as the substituent changed from acid to a variety
of amides to amine. This part of the molecule could therefore be
used as a handle to balance the physical properties of the overall
molecule.
The para-substituent of the C-6 aniline extends towards a sol-
vent exposed region in the model. This implies that the para-sub-
stituent would not affect the BTK enzyme potency. To verify, we
selected a wide range of hydrophilic groups as para-substituents
The ‘gatekeeper’ region then attracted our attention. Due to the
residue differences at this region among BTK (T474), Janus kinase 2
(JAK2) (M929), Fms-like tyrosine kinase 3 (Flt3) (F691), and Inter-
leukin-2-inducible T-cell kinase (ITK) (F434),7 we hypothesized

Q. Shi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2206–2211
2209
that growing the molecule into this ‘gatekeeper’ region (Fig. 3)
from the purine N-9 substituents would gain better selectivity
against these kinases. Different sized N-9 substituents from hydro-
gen to a large 3-phenylpropyl group were selected to probe this
area (Table 2). We observed that the off target kinase activities
did decrease with the size of the N-9 substituent. However, this
also decreased BTK potency. Of all the N-9 substituents examined,
2-phenylethyl 23 was optimal in balancing the kinase selectivity
over JAK2 (324 fold), Flt3 (20 fold), and ITK (345 fold) while main-
taining potency against BTK (140 nM). Compound 20, though bear-
ing the smallest N-9 substituent, gave the best BTK potency (8 nM)
along with excellent JAK2 selectivity (383 fold) (vide infra).
We then examined the substitution pattern of the central ben-
zene ring (Table 3). Lacking the 2-methyl group, compound 26
loses BTK potency and especially kinase selectivity. Walking the
methyl group to the 4- or 6-position, compounds 27 and 28 lose
BTK potency and kinase selectivity as well. The 2-methyl group
helps position the 1-benzamide and 3-purine groups, due to the
steric interactions, to get the best potency and selectivity. A methyl
group is also the ideal size to interact with L528 at the base of the
ATP binding site (Fig. 2).
Next, we studied the SAR of the side chain off the central ben-
zene ring (Table 4). We evaluated a variety of R groups including
alkyl, amines, amides, ureas and sulfonamides. Amides turned
out to be superior to ureas or sulfonamides in both BTK potency
and kinase selectivity. Among amides, the BTK potency and selec-
tivity increased with the lipophilicity of the side chain. The optimal
result was obtained with a tert-butylphenyl amide moiety in com-
pound 20.
Figure 4. X-ray co-crystal structure of compound 20 in human BTK (PDB code
4NWM). Hydrogen bond interactions between BTK and the bicyclic purine core of
compound 20 from the BTK/compound 20 co-crystal structure are highlighted.
Hydrogen bonds to compound 20 and within the water network are shown as
yellow and red dashed lines, respectively.
To better understand the SAR in a structural context, a co-crys-
tal structure of compound 20 and BTK was obtained (PDB code
4NWM).8 In agreement with the homology model, the co-crystal
structure shows that the N-7 of the purine core and the NH of
the C-6 aniline form hydrogen bonds with the backbone amide
NH and carbonyl of M477 (Fig. 4). The crystal structure reveals
additional interactions between purine N-3, N-9 and a water net-
work that bridges to a conserved lysine K430 and the gatekeeper
residue T474. This network would be disrupted by any N-9 substi-
tuent other than hydrogen, which explains the SAR in Table 2. In
contrast, Jak2, Flt3 and ITK have hydrophobic gatekeeper residues
that pack well against the methyl group at N-9 of compound 8,
thereby enhancing their potency and reducing BTK selectivity.
The structure also made clear the specific interactions with the
tert-butyl benzamide moiety (Fig. 5). The tert-butyl group is in a
well-defined hydrophobic pocket, that is, the specificity pocket.3b,9
This pocket is comprised of residues from the G-loop, Q412 and
F413, and the activation loop, L542, S543, V546, and Y551. The
benzene ring forms hydrophobic interactions with Q412 and
Y413. The amide carbonyl forms hydrogen bonds to the backbone
NH of F413 and the backbone carbonyl of G414 via a bridging
water molecule. Collectively, these interactions contribute to the
higher potency of compound 20 versus the other analogs lacking
these contacts.
The side chains of residues S543, V546, and Y551 are presum-
ably responsible for the high selectivity gained from filling the
BTK specificity pocket. In the literature, sequence analysis of 491
human kinases shows that the V546/Y551 pair occurs in only three
other kinases, BMX, ITK, and TXK while the triple S543/V546/Y551
is completely unique to BTK.9 The kinase selectivity profile of com-
pound 20 shown in Tables 4 and 5 is in accordance to this analysis.
Compound 20 was screened against the Ambit kinase panel of
over 400 kinases. Only 3.1% of the kinases showed less than 33%
Figure 5. The tert-butyl pocket from BTK/compound 20 co-crystal structure.
Compound 20 was also found to be very potent in cellular as-
says and to have a good correlation between enzyme potency
and cell based activities. In B cells stimulated through the B cell
receptor (BCR), compound 20 selectively inhibited signaling and
functional endpoints, including calcium flux and phosphorylation
of BTK in human Ramos B cells,6,10 production of IL-6 and prolifer-
ation of human tonsilar B cells,11,12 and surface expression of CD86
on human or mouse peripheral blood B cells13,14 (Table 6).
control binding at 1 lm. Consistent with this measure of selectiv-
ity, biochemical kinase assays against 40 kinases showed com-
pound 20 to be more potent against BTK than others, including
the other Tec family kinases (BMX, ITK, TEC and TXK) (Table 5).
Further profiling of compound 20 is listed in Table 7. Unfortu-
nately, the metabolic stability of 20 was low in human and rat liver

2210
Q. Shi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2206–2211
Table 5
0.20
Kinases selectivity of compound 20
BTK IC50: 0.008
BMX IC50: 0.037
l
l
M
M
M
SRC IC50: 2.434
JAK2 IC50: 3.068
JAK3 IC50: 33.740
l
lM
M
IKK-1 IC50: >50
IKK-2 IC50: >50
IRAK4 IC50: >50
l
l
l
M
M
M
0.15
0.10
0.05
0.00
ITK IC50: 0.052
TEC IC50: 0.014
l
l
M
lM
HER1 IC50
:
GSK-3B IC50: >50 lM
>16.670
InsR IC50: 9.142
l
M
TXK IC50: 0.055
l
M
l
M
IGF-1R IC50
17.910
:
l
M
FLT3 IC50: 0.479
l
M
MK2 IC50: 13.330 lM
MAPKAPK3 IC50
0.2130
:
lM
*
*
Table 6
Cell potency of compound 20
Vehicle
3
10
30
Compound 20, mg/kg, IP, -1hr
Cell assays
IC50 (lM)
Calcium flux in Ramos B cell
pBTK in Ramos B cells (western blot)
IL-6 in Human Tonsilar B cells
0.009
ꢀ0.050
0.030
O.D. values are the MEAN +/- SEM of 14 tissue samples per treatment
group
Human B cell proliferation
Human CD86-BCR surface expression
Mouse CD86-BCR surface expression
0.008
0.142
0.112
Figure 6. Compound 20 in mouse Passive Cutaneous Anaphylaxis (PCA) model.
Procedures of the above cell-based assays are described in Refs. 6,10–14.
O
N
O
H2N
Cl
N
Cl
N
N
a
O
N
N
O
N
N
b
N
HN
N
N
Table 7
N
N
N
Cl
Cl
97%
H
92%
THP
Partial in vivo and in vitro profile of compound 20
Cl
THP
40
Liver microsomes stability (%
remaining):
Human 14%; Rat 36%; Mouse 68%
41
42
CYP inhibition (IC50):
3A4 BFC 9.8
l
M, 3A4 BZR>40
lM;
c
O
B
H
2C8 1.4 M; 2C9 0.4
l
lM; 2C19
95%
N
O
3.4
7.7
l
M; 2D6>40
lM
O
O
O
Mouse Cytotox IC50
:
lM
N
N
Herg Flux IC50
PAMPA:
:
>80
lM
O
d
O
HN
N
N
HN
403 nm/s
<15 nm/s (both directions)
<0.001 mg/mL at pH 1 and 6
N
N
N
H
N
N
H
N
90%
Caco2:
aq. soluability:
Mouse PK (30 mg/kg dose, Methocel
suspension through IP
N
N
H
THP
O
O
Cmax: 3.802
lM; AUC (tot):
M⁄h; t1/2: 4.3 h
43
20
32.468
l
administration):
Scheme 1. Reagents and conditions: (a) DHP, TsOH, EtOAc, rt; (b) KOtBu, THF, 0 °C-
rt; (c) Pd(Ph3P)4, 2M aq K3PO4, toluene/EtOH, 100 °C; (d) 1N aq HCl, MeOH, rt.
microsomes while moderate in mouse. Biotransformation studies
showed the major metabolic pathway was the oxidation of the
tert-butylphenyl moiety. Due to its poor Caco2 permeability, poor
aqueous solubility and moderate mouse liver microsomal stability,
acceptable exposure in mouse was only obtained through IP
administration.
References and notes
1. (a) Mohamed, A.; Yu, L.; Bäckesjö, C.; Vargas, L.; Faryal, R.; Aints, A.;
Christensson, B.; Berglöf, A.; Vihinen, M.; Nore, B. F.; Smith, C. I. E. Immunol.
Rev. 2009, 228, 58; (b) Mao, C.; Zhou, M.; Uckun, F. M. J. Biol. Chem. 2001, 276,
41435; (c) Dinh, M.; Grunberger, D.; Ho, H.; Tsing, S. Y.; Shaw, D.; Lee, S.;
Barnett, J.; Hill, R. J.; Swinney, D. C.; Bradshaw, J. M. J. Biol. Chem. 2007, 282,
8768; (d) de Weers, M.; Verschuren, M. C.; Kraakman, M. E.; Schuurman, R. K.;
van Dongen, J. J.; Hendriks, R. W. Eur. J. Immunol. 1993, 23, 3109; (e) Jongstra-
Bilen, J.; Puig Cano, C.; Hasija, M.; Xiao, H.; Smith, C. I.; Cybulsky, M. I. J.
Immunol. 2008, 181, 288; (f) Kuehn, H. S.; Radinger, M.; Brown, J. M.; Ali, K.;
Vanhaesebroeck, B.; Beaven, M. A.; Metcalfe, D. D.; Gilfillan, A. M. J. Cell Sci.
2010, 123, 2576; (g) Tsukada, S.; Rawlings, D. J.; Witte, O. N. Curr. Opin.
Immunol. 1994, 6, 623; (h) Satterthwaite, A. B.; Witte, O. N. Immunol. Rev. 2000,
175, 120; (i) Steinberg, B. J.; Smathers, P. A.; Frederiksen, K.; Steinberg, A. D. J.
Clin. Invest. 1982, 70, 587; (j) Jansson, L.; Holmdahl, R. Clin. Exp. Immunol. 1993,
94, 459.
In vivo efficacy of compound 20 was then evaluated in an FceRI-
driven mouse Passive Cutaneous Anaphylaxis (PCA) model. The
procedure15 was modified from the technique described by Wer-
shil et al.16 At 10 mg/kg dose, compound 20 reduced the dye con-
tent by 74% compared to vehicle (Fig. 6) demonstrating good
in vivo efficacy.
A representative synthesis of purine analogs is exemplified by
the synthesis of 20 as shown in Scheme 1. Commercially available
2,6-dichloro-9H-purine 40 was selectively protected with a THP
group at the N-9 position to give 41. Reaction of 41 and the corre-
sponding aniline with potassium tert-butoxide afforded 42 regio-
selectively. Suzuki coupling between 42 and the corresponding
boronic ester yielded 43, which was then deprotected under acidic
condition to give 20.
2. (a) Pan, Z. Drug News Perspect. 2008, 21, 357; (b) Ruderman, E.; Pope, R. M.
Arthritis Res. Ther. 2011, 13, 125; (c) Hendriks, R. W. Nat. Chem. Biol. 2011, 7, 4;
(d) Xu, D.; Kim, Y. K.; Postelnek, J.; Vu, M. D.; Hu, D.; Liao, C.; Bradshaw, M.;
Hsu, J.; Zhang, J.; Pashine, A.; Srinivassan, D.; Woods, J.; Levin, A.; O’Mahony, A.;
Owens, T. D.; Lou, Y.; Hill, R. J.; Narula, S.; DeMartino, J.; Fine, J. S. J. Pharmacol.
Exp. Ther. 2012, 341, 90.
3. (a) Sheridan, C. Nat. Biotechnol. 2012, 30, 199; (b) Lou, Y.; Owens, T. D.;
Kuglstatter, A.; Kondru, R. K.; Goldstein, D. M. J. Med. Chem. 2012, 55, 4539; (c)
Herman, S. E.; Gordon, A. L.; Herlein, E.; Ramanunni, A.; Zhang, X.; Jaglowski, S.;
Flynn, J.; Jones, J.; Blum, K. A.; Buggy, J. J.; Hamdy, A.; Johnson, A. J.; Byrd, J. C.
Blood 2011, 117, 6287; (d) Cameron, F.; Sanford, M. Drugs 2014, 74, 263.
4. (a) Currie, K.S.; Kropf, J.E.; Darrow, J.W.; Desimone, R.W. WO/2006/053121.; (b)
Currie, K.S.; Desimone, R.W.; Mitchell, S.A.; Pippin, D.A.I.; Darrow, J.W.; Qian,
X.; Velleca, M.; Qian, D. US2006/0183746 A1.
Herein we reported purine derivatives as a class of potent BTK
inhibitors. Structure-activity relationship studies led to the discov-
ery of 20 as one of the most potent and selective compounds. Com-
pound 20 was also efficacious in a mouse PCA model. However, the
low oral exposure hindered its further development. This issue has
to be addressed to advance this class of BTK inhibitors.

Q. Shi et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2206–2211
2211
5. Shewchuk, L.; Hassel, A.; Wisely, B.; Rocque, W.; Holmes, W.; Veal, J.; Kuyper, L.
F. J. Med. Chem. 2000, 43, 133.
6. Liu, Q.; Batt, G.D.; De Lucca, V.G.; Shi, Q.; Tebben, A. US8084620.
7. van Linden, O. P. J.; Kooistra, A. J.; Leurs, R.; de Esch, I. J. P.; de Graaf, C. J. Med.
Chem. 2014, 57, 249.
ImmunoReasearch, cat# 109–006-129) and 10 ng/mL IL-4 (Peprotech, cat#
200–04). After an additional 4.5 h incubation, the supernatants were collected
and the levels of IL-6 measured by EIA.
12. Measure of Human B Cell Proliferation: Human B cells from tonsils, isolated as
described above, in media containing 10% FCS and various concentrations of
the test compound were stimulated with AffinPure F(ab0)2 goat anti-human
IgG+IgM (Jackson, Cat#109-006-127) and incubated at 37 °C for 72 h. The cells
were then labeled with [3H]thymidine and incubated overnight at 37 °C. Cells
were then harvested onto filter plates and the amount of [3H]thymidine
incorporation, as a measure of B cell proliferation, was determined by liquid
scintillation counting.
8. A baculovirus construct of His-TEV-hBTK(E396–S659) was expressed in Sf9
cells for 66 h post infection at 27 °C. The frozen cell paste was lysed under
nitrogen cavitation at 300–400 psi for 30 min at 4 °C and centrifuged at
100,000g for 30 min. The supernatant was purified on the Ni-NTA (Qiagen)
resin. His-TEV-hBTK of ꢁ80% purity was eluted at 0.3 M imidazole in a gradient
of 0.03–0.6 M imidazole. The fractions that contained His-TEV-hBTK were
pooled together, concentrated and loaded on a Superdex-200 SEC column (26/
60) to remove soluble aggregates. This material was then quantitatively
cleaved at an engineered site for TEV recognition, followed by chromatography
on nickel-chelate resin to remove the AcTEV protease (InVitrogen) and
uncleaved material with a 60% recovery. Final pure cleaved hBTK (E396–
S659) was buffer exchanged into 200 mM NaCl, 20 mM Tris/HCl, pH 8.0, and
1 mM TCEP. The final sample was aliquoted and stored at ꢂ80 °C at 0.34 mg/
13. Measure of Human CD86-BCR surface expression: The E-negative fraction of
human peripheral blood mononuclear cells, isolated after removal of
T
lymphocytes by rosetting with sheep red blood cells, in media containing
10% BCS and various concentrations of the testing compound were stimulated
for 18 h at 3 7°C with AffinPure F(ab0)2 Fragment Goat anti Human IgG + IgM
(Jackson Cat#109-006-127). The cells were then fixed and stained with FITC-
conjugated mouse anti-human CD20 antibody (BD Pharmingen 555622) and
APC-conjugated mouse anti-human CD86 monoclonal antibody (BD
Pharmingen 555660) and the amount of CD86 expression quantitated by the
mean fluorescence intensity (MFI) after gating on the viable (propidium iodide
exclusion) CD20-positive B cell population as measured by FACS analysis. For
analogous experiments measuring the effect of CD40 or TLR4 stimulation on
these endpoints, either human IZ-CD40L or LPS was used to stimulate the cells.
To determine the effect of the testing compound CD86 expression in memory B
cells, an additional marker was employed (PE-conjugated mouse anti-human
CD27, BD Pharmingen 555441) to identify CD27+CD20+memory B cells.
14. Measure of Mouse CD86-BCR surface expression: Mouse splenic B cells were
ml. For protein/compound complex formation, 4.4 ll of Compound 20 (25 mM
DMSO stock) was added to 2.0 ml of hBTK at 0.34 mg/ml and incubated at 4 °C,
concentrated to 8.5 mg/ml and run over a 26/60 Superdex-200 SEC column.
Crystals were grown at room temperature using hanging drop vapor diffusion
method. The drop consisted of 2 ll protein solution and 2 ll reservoir solution
containing 26% (w/v) methyl ether PEG 5000, 0.2 M Tris–HCl, pH 7.25. Macro
seeding was performed to initiate crystal growth. The crystals appeared within
a few days and continued to grow for 2–3 lweeks. Crystals were flash-cooled
in liquid nitrogen for data collection with 20% glycerol and 80% reservoir
solution as cryoprotectant. Diffraction data were collected by Shamrock
Structures, Inc. at IMCA-CAT, beamline 17ID at the Advanced Photon Source.
isolated using
a nylon wool column from homogenized spleens and
hBTK/Compound
a = 44.9 Å, b = 86.5 Å, c = 135.1 Å,
structure was determined by molecular replacement using
6
co-crystals belonged to the space group p212121:
= b = = 90.0°. The 2.0 Å resolution
previously
resuspended in media containing 10% FCS with various concentrations of the
a
c
testing compound at 37 °C before stimulating for 18 h with 10 lg/mL affinPure
a
F(ab0)2 fragment goat anti-mouse IgG + IgM (Jackson, Cat#115-006-068). Cells
were then fixed and stained with an APC-conjugated rat anti-mouse CD45R/
B220 and FITC-conjugated rat anti-mouse CD86 (B7–2) and the amount of
CD86 expression quantitated by the mean fluorescence intensity (MFI) after
determined in house BTK structure (unpublished results). The structure of
hBTK + Compound 20 has been deposited to RCSB with PDB code 4NWM.
9. Di Paolo, A. J.; Huang, T.; Balazs, M.; Barbosa, J.; Barck, H. K.; Bravo, J. B.; Carano,
A. D. R.; Darrow, J.; Davies, R. D.; DeForge, E. L.; Diehl, L.; Ferrando, R.; Gallion, L.
S.; Giannetti, M. A.; Gribling, P.; Hurez, V.; Hymowitz, G. S.; Jones, R.; Kropf, E.
J.; Lee, P. W.; Maciejewski, M. P.; Mitchell, A. S.; Rong, H.; Staker, L. B.;
Whiteney, J. A.; Yeh, S.; Young, B. W.; Yu, C.; Zhang, J.; Reif, K.; Currie, S. K. Nat.
Chem. Biol. 2011, 7, 41.
gating on the viable (propidium iodide exclusion) B220-positive
population as measured by FACS analysis.
B cell
15. Experimental procedure of mouse PCA study: Female CD-1 mice, 8–10 week of
age (Harlan), were anesthetized (Isoflurane, 3–5% in O2) and their backs were
shaved. Mice were injected intradermally with 40ng anti-DNP IgE monoclonal
10. Measure of pBTK in Ramos B cells: After a 1 h pre-incubation of Ramos B cells
in media containing 10% FCS with varying concentrations of the testing
compound at 37 °C, the cells were stimulated with anti-human IgM (F(ab0)2
antibody in 20
were challenged by intravenous injection of 100
serum albumin in 200 L saline containing 1% Evans blue dye. Vehicle (0.5%
Methocel with 0.1% Tween 80) or compound 20 was administered
intraperitoneally 1 h prior to antigen challenge. At 1 h after antigen
challenge, mice were euthanized and the skin of the back was removed. Full-
thickness skin biopsy punches (8 mm diam.) encompassing each of the two IgE
l
L saline into each of two sites on the back. 24 h later, the mice
lg
DNP-coupled human
l
fragment, Jackson ImmunoResearch, catalog #109-006-129) at 50
lg/ml for
exactly 2 min at 37 °C. Cells were then fixed and stained with an Alexa647-
conjugated anti-phospho-BTK antibody which specifically recognizes pY551
(BD Biosciences, 558134) and the amount of phospho-Y551 quantitated by the
mean fluorescence intensity (MFI) as measured by FACS analysis.
injections sites were collected along with
a ‘background’ biopsy from a
11. Measure of IL-6 in Human Tonsilar B Cells: Human tonsilar B cells were
isolated according to the general procedure of Gillooly et al. [Gillooly, K.M.;
Pattoli, M.A.; Taylor, T.L.; Chen, L.; Cheng, L.; Gregor, K.R.; Whitney, G.S.;
Susulic, V.; Watterson. S.H.; Kempson, J.; Pitts, W.J.; Booth-Lute, H.; Yang, G.;
Davies, P.; Kukral, D.W.; Strnad, J.; McIntyre, K.W.; Darienzo, C.J.; Salter-Cid, L.;
Yang, Z.; Wang-Iverson, D.B.; Burke, J.R. J. Pharmacol. Exp. Ther. 2009, 331, 349].
The isolated B cells were suspended in media containing 10% FBS and various
concentrations of the testing compound. After a 1 h incubation at 37 °C, the
separate uninvolved region of the skin. The biopsy specimens were placed into
separate tubes with 1 ml formamide and the Evans blue dye was extracted at
100°C for 3 h. An aliquot of the formamide supernatant was transferred to a
96-well flat-bottomed plate and the OD was read at 620 nm. The OD of the
‘background’ biopsy supernatant was subtracted from the OD of the ‘IgE’
biopsy supernatants to determine a net extravasation of dye.
16. Wershil, B. K.; Mekori, Y. A.; Murakami, T.; Galli, S. J. J. Immunol. 1987, 139,
2605.
cells were stimulated with 40 l
g/mL F(ab0)2 goat anti-human IgM (Jackson