Design, synthesis and structure-activity relationships of a novel class of sulfonylpyridine inhibitors of Interleukin-2 inducible T-cell kinase (ITK)

Article abstract of DOI:10.1016/j.bmcl.2014.10.020

Starting from benzylpyrimidine 2, molecular modeling and X-ray crystallography were used to design highly potent inhibitors of Interleukin-2 inducible T-cell kinase (ITK). Sulfonylpyridine 4i showed sub-nanomolar affinity against ITK, was selective versus Lck and its activity in the Jurkat cell-based assay was greatly improved over 2.

Full text of DOI:10.1016/j.bmcl.2014.10.020

Bioorganic & Medicinal Chemistry Letters 24 (2014) 5818–5823
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and structure–activity relationships of a novel class
of sulfonylpyridine inhibitors of Interleukin-2 inducible T-cell kinase
(ITK)
a
a
a
b
Giancarlo Trani a, , John J. Barker , Steven M. Bromidge , Frederick A. Brookfield , Jason D. Burch ,
Yuan Chen b, Charles Eigenbrot b, Alexander Heifetz a, M. Hicham A. Ismaili b, Adam Johnson b,
Thomas M. Krülle a, Colin H. MacKinnon a, Rosemary Maghames a, Paul A. McEwan a,
Christian A. G. N. Montalbetti a, Daniel F. Ortwine b, Yolanda Pérez-Fuertes a, Darshan G. Vaidya a,
Xiaolu Wang c, Ali A. Zarrin b, Zhonghua Pei b
⇑
a Discovery Chemistry, Evotec UK Ltd, 114 Innovation Drive, Milton Park, Abingdon, OX14 4RZ, UK
b Discovery Chemistry, Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080, USA
c Discovery Biology, Evotec AG, Manfred Eigen Campus, Essener Bogen 7, D-22419 Hamburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Starting from benzylpyrimidine 2, molecular modeling and X-ray crystallography were used to design
highly potent inhibitors of Interleukin-2 inducible T-cell kinase (ITK). Sulfonylpyridine 4i showed
sub-nanomolar affinity against ITK, was selective versus Lck and its activity in the Jurkat cell-based assay
was greatly improved over 2.
Received 29 August 2014
Revised 30 September 2014
Accepted 1 October 2014
Available online 12 October 2014
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
ITK inhibitor
T-cell receptor
Sulfonylpyridine
Interleukin-2 inducible T cell kinase (ITK, also known as Emt or
Tsk), is a member of the Tec family of tyrosine kinases and is mainly
expressed in T cells, mast cells and natural killer cells. In T cells, ITK
is activated after T-cell receptor (TCR) stimulation leading to phos-
The initial optimization work carried out on an HTS hit from the
Genentech compound collection (unpublished data) led to the
identification of benzylpyrimidine 2 (Fig. 1), a compound which
combines the features of the initial Genentech HTS hit 1a and
another ITK inhibitor 1b previously published by researchers at
phorylation of phospholipase C-c1 (PLC-c1), calcium mobilization,
IL-2 production, and cell proliferation and differentiation.1,2 Dele-
tion of ITK in mice results in reduced production of Th2 cytokines
such as IL-4, IL-5 and IL-13.3 ITK has been shown to play an impor-
tant role in the development of T-cell dependent late phase
responses of allergic asthma. In studies with ITKÀ/À mice, the
immunological symptoms of allergic asthma are attenuated and
lung inflammation, eosinophil infiltration and mucus production
are drastically reduced in response to challenge with the allergen
ovalbumin.4 Similar results were reported with a selective ITK
inhibitor5 indicating a potential role of ITK kinase activity in inflam-
matory processes. Therefore, selective inhibition of ITK could repre-
sent an attractive approach for the treatment of T-cell mediated
diseases.
GSK.6 Compound
(70 M) compared to 1b (61
2
had improved kinetic aqueous solubility
M) and showed high inhibitory
l
l
activity against ITK (Ki = 53 nM);7 however, when it was tested
against lymphocyte-specific kinase (Lck), another T-cell tyrosine
kinase which operates upstream of ITK in the TCR cascade, it
showed a complete lack of selectivity (3.5-fold) and its activity in
a Jurkat cell-based assay,8 as measured by inhibition of phosphor-
ylation of PLC-c1, was sub-optimal (IC50 = 2.6 lM). Here we
describe the further optimization of compound 2, which led to
the identification of a novel series of sulfonylpyridines as selective
ITK inhibitors with improved activity in the Jurkat cell-based assay
compared to 2.
In silico modeling of 2 in the ITK kinase domain gave rise to the
binding hypothesis which is summarized in Figure 2. The amino-
pyrazole portion forms three hydrogen bonds with Met438 and
Glu436 of the hinge region and the benzene ring of the substituent
at the 6-position of the pyrimidine is involved in an edge–face
⇑
Corresponding author.
E-mail address: Giancarlo.Trani@evotec.com (G. Trani).
http://dx.doi.org/10.1016/j.bmcl.2014.10.020
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

G. Trani et al. / Bioorg. Med. Chem. Lett. 24 (2014) 5818–5823
5819
Figure 1. Hit to lead studies.
Table 1
Data for sulfonylpyrimidine 3 and sulfonylpyridine 4a
Compound ITK Ki
Fold selectivity
over Lck
p-PLC
(nM)
c
1 IC50
Kin. Sol.
(lM)
(nM)
2
3
4a
53
177
4
3.5
1.5
34
2600
—
540
70
39
32
4-methanesulfonyl-pyridine (Fig. 3B) which lacks the nitrogen lone
pair ortho to the sulfonyl linker revealed an energy minimum at 90°.
In this system, the putative bioactive conformation was much clo-
ser to the minimum and only 1 kcal/mol higher in energy, thus
decreasing the enthalpic penalty of binding. Based on this hypoth-
esis we prepared the sulfonylpyridine 4a and we were pleased to
see that the experimental results reflected what was predicted in
the torsional scan. Compound 4a was not only very potent (ITK
Ki = 4 nM), showing 44-fold increase in potency over 3 and 13-fold
over 2, but it also showed improved selectivity against Lck (34-fold)
compared to 2 (3.5-fold).
In order to explore whether the binding energy of 4a could be
further improved, we prepared a range of heterocycles (Table 2).
Compound 5 where the pyrazole was replaced by an imidazole
displayed a 3-fold drop in potency and loss of Lck selectivity.
Thiazole 6a showed sub-nanomolar affinity for ITK, (Ki = 0.27 nM);
however, its affinity for Lck was also in the low nanomolar range.
We believe that the different shape of the thiazole which allows
for more effective hydrogen bond interactions with the hinge
sequence, the restricted rotation of the thiazole ring across the
NHAC bond and a higher desolvation penalty associated to the
pyrazole compared to the thiazole, all contributed to the 15-fold
increase in potency observed for compound 6a over 4a.
Figure 2. In silico modeling of compound 2 in the ITK kinase domain.
interaction with Phe437, which is proposed to account for the
majority of the potency boost over the initial HTS hit 1a. The same
interactions were also observed in the X-ray crystal structures of
two other ITK inhibitors (indazole and benzothiazole) discovered
at Genentech.9
We initially sought to replace the benzylic carbon with alterna-
tive linkers in order to modulate the angles between the pyrimidine
core and the phenyl ring, in an attempt to maximize the interaction
with Phe437 and selectivity against Lck, which contains a tyrosine
(Tyr318) in place of Phe437 in ITK. Sulfonylpyrimidine 3 (Table 1)
showed the highest affinity amongst the linkers explored (ether,
thioether, sulfone), with a Ki against ITK of 177 nM. Although less
active than 2, we envisioned that the loss in affinity could be the
result of an electronic repulsion between the oxygens of the sulfo-
nyl linker and the lone pair of the proximal nitrogen of the pyrim-
idine core, disfavoring the phenyl ring from adopting the desired
bioactive conformation required for interaction with Phe437. This
Benzothiazole 6b had similar potency to pyrazole 4a but
showed >100-fold selectivity against Lck. Although compounds
6a and 6b were respectively more active and selective than 4a,
they were also more lipophilic, and in the case of 6b contained
an additional aromatic ring. This led to poor kinetic solubility
observation was supported by
a
conformational analysis10
performed on the model system 4-methanesulfonyl-pyrimidine,
as shown in Figure 3A. The minimization of steric interactions
and electronic repulsions suggested an energy minimum at a
dihedral angle of 120°, while the optimal bioactive conformation
as predicted by the docking (Fig. 2) was circa 4 kcal/mol higher in
energy, at an angle of about 60°. A similar analysis performed on
which was reflected in a higher solubility forecast index (SFI),11
a
simple metric which has been used as a tool to predict solubility
via the sum of cLogD7.4 and aromatic ring count.
Given its favorable potency and selectivity profile, we further
explored the SAR of the more soluble pyrazole 4a (Table 3). We

5820
G. Trani et al. / Bioorg. Med. Chem. Lett. 24 (2014) 5818–5823
Figure 3. Dihedral torsional scan performed on 4-methanesulfonyl-pyrimidine (A) and on 4-methanesulfonyl-pyridine (B).
Table 2
Hinge binder exploration
Compound Heterocycle ITK Ki
Fold selectivity
over Lck
Kin. Sol.
(lM)
SFI
7.7
(nM)
H
N
N
N
N
4a
4
34
32
Me
Me
Me
5
14
3
6
6.7
8.1
N
6a
0.27
11
61
S
S
N
6b
8
106
61
10
Figure 4. X-ray structure of 4b in the ITK kinase domain (resolution 2.6 Å, PDB code
4QD6). Only selected residues of ITK are shown for clarity. Protein–ligand hydrogen
bonds are indicated by the dashed lines in yellow.
Table 3
SAR around the pyrazole hinge binder
minimally affected by the introduction of small substituents on
the phenyl ring, as shown by compounds 4c–g, with para-substitu-
tion (4d) preferred over meta- (4c). However, larger, lipophilic sub-
stituents led to a drop in potency (4f and 4g), presumably due to a
clash with the gatekeeper residue.
Compound
R
ITK Ki
(nM)
Fold selectivity
over Lck
Kin. Sol.
(lM)
SFI
In order to validate the binding hypothesis, we determined the
crystal structure of compound 4b bound in the ITK active site. The
structure, solved at 2.6 Å resolution, confirmed many of the fea-
tures predicted by the in silico model (Fig. 4).12 The aminopyrazole
moiety forms three hydrogen bonds with the hinge sequence
(Met438 and Glu436). The pyridine core is positioned in the same
plane of the aminopyrazole, sandwiched in a narrow lipophilic cleft,
and the sulfone linker positions the phenyl ring at a dihedral angle
of 60° from the pyridine core, giving a close edge–face interaction
with Phe437. The crystal structure also revealed that the hydroxyl
of the cyclohexanol group forms a hydrogen bond to a water mole-
cule which is further stabilized by hydrogen bonds with Ser499 and
Asp500.
4a
4b
Me
4
34
9
32
7.7
2.3
61
10.3
4c
5.3
11
61
10.1
OMe
OMe
4d
4e
4f
1.2
1.8
20
10
61
61
61
10.4
10.5
11.3
F
16
CF3
OCF3
n.d.
4g
4h
4i
61
n.d.
39
61
8.8
61
11.6
8.4
1.6
0.17
182
8.8
The phenyl ring at the 5-position of the pyrazole is involved in
an edge–face interaction with the gatekeeper Phe435. Unfortu-
nately, this only slightly improved ITK activity over 4a (compounds
4b–e), and was also associated with an equal increase in affinity for
4j
0.38
45
61
9.3
Lck, presumably due to a p–OH interaction with the Thr316 gate-
reasoned that ITK activity and selectivity against Lck could be
further improved by exploiting the difference in the gatekeeper
residue between the two kinases (Phe435 in ITK, Thr316 in Lck).
keeper in Lck. Furthermore, the presence of the extra aromatic ring
led to a deterioration in aqueous solubility of 4b–g relative to 4a,
and therefore we investigated the replacement of the phenyl ring
with alkyl groups.
A
phenyl group directly attached to the pyrazole ring
(compound 4b) showed similar potency to 4a. The activity was

G. Trani et al. / Bioorg. Med. Chem. Lett. 24 (2014) 5818–5823
5821
Scheme 1. Reagents and conditions: (a) SEM-Cl, DIPEA, CH2Cl2, RT, 20 h; (b) H2, Pd/C, ethanol, RT, 20 h, 41% over two steps; (c) Thiophenol, NaOH, acetone/water (1:1), RT,
1 h, 80%; (d) 8, DIPEA, ethanol, 85 °C, 37 h, 52%; (e) trans-4-aminocyclohexanol, 2-propanol, 150 °C in microwave, 4 h, 90%; (f) mCPBA, CH2Cl2, RT, 45 min, 47%; (g) HCl (4 M in
1,4-dioxane), RT, 20 h, 37%.
in no improvement of aqueous solubility compared to the aryl
analogs, thus suggesting that in order to obtain compounds with
measurable solubility a hard cut-off of SFI = 7 should be observed
in this series.
Compound 4i (Table 3) had the best ITK activity and selectivity
versus Lck and it was selected for further profiling. It showed mod-
erate to high clearance in human liver microsomes (16 mL/min/kg),
no P450 inhibition below 8 lM (against 2C9, 2C19, 2D6 and 3A4)
Scheme 2. Reagents and conditions: (a) thiophenol, Pd2(dba)3, xantphos, DIPEA,
1,4-dioxane, 110 °C, 1.5 h, 77%; (b) mCPBA, CH2Cl2, RT, 4 h, 90%.
and moderate permeability as measured in the MDR1/MDCK assay
(Papp apical to basolateral (AB) = 2.5 Â 10À6 cm/s, Papp basolateral to
apical (BA) = 12 Â 10À6 cm/s, BA/AB = 4.8). 4i also inhibited PLC-
c1
Gratifyingly, we found that aliphatic substituents (4h–j) were
well tolerated and in general led to improved ITK affinity and
increased Lck selectivity. Cyclobutane 4h had ITK Ki = 1.6 nM
and was 39-fold selective versus Lck. Increasing the size to cyclo-
pentane 4i led to a 9.5-fold increase in potency (ITK Ki = 0.17 nM)
compared to 4h and selectivity versus Lck was also improved
(182-fold). This boost in potency and selectivity is likely due to
the ability of the cyclopentane ring to establish more effective
lipophilic interactions with Phe435 in ITK than with Thr316 in
Lck. When the size of the alkyl substituent was increased further
(cyclohexane 4j), ITK activity started to plateau and selectivity to
decrease. Although alkyl pyrazoles 4h–j have a lower SFI compared
to the aryl analogs 4b–g due to the former having one less aromatic
ring, the increasing size of the alkyl groups necessary for better
activity and selectivity, also contributed to a higher LogD, resulting
phosphorylation in Jurkat cells with an IC50 = 180 nM, a 15-fold
improvement over 2. Invitrogen kinase selectivity profiling of 4i
at 0.1 lM (40 kinases) resulted in only 3 kinases showing greater
than 70% inhibition (CDK2, Flt3 and MuSK).
Compounds 3, 4a–j, 5 and 6a–b were prepared as described in
Schemes 1–4. Sulfonylpyrimidine 3 was prepared in five steps
starting from 2,4,6-trichloropyrimidine 9 (Scheme 1). Reaction
with thiophenoxide ion at room temperature followed by a second
nucleophilic displacement with SEM-protected aminopyrazole 8
afforded the 2-chloropyrimidine 10. A third nucleophilic substitu-
tion with trans-4-aminocyclohexanol, followed by oxidation of the
thioether to the corresponding sulfone and final deprotection of
the pyrazole with HCl at room temperature afforded the desired
pyrimidine sulfone 3. Aminopyrazole 8 (Scheme 1) was obtained
Scheme 3. Reagents and conditions: (a) DIPEA, DMSO, 100–110 °C, up to 36 h; (b) DIPEA, DMSO, 150–165 °C in microwave, up to 6 h, 23–62%; (c) trans-4-aminocyclohexanol,
DIPEA, DMSO, 120 °C, up to 20 h; (d) trans-4-aminocyclohexanol, DIPEA, DMSO, 170 °C in microwave, up to 2 h, 10–54%; (e) 1-methyl-4-aminoimidazole, DIPEA, DMSO,
120 °C, 2 h, 24%; (f) trans-4-aminocyclohexanol, DIPEA, DMSO, 120 °C, 20 h, 31%.

5822
G. Trani et al. / Bioorg. Med. Chem. Lett. 24 (2014) 5818–5823
Scheme 4. Reagents and conditions: (a) 2-amino-5-methylthiazole, Pd2(dba)3, xantphos, Na2CO3, 1,4-dioxane, 100 °C, 2 h, 52%; (b) 2-aminobenzothiazole, Pd2(dba)3,
xantphos, Na2CO3, 1,4-dioxane, 150 °C in microwave, 45 min, 40%; (c) 17a, trans-4-aminocyclohexanol, Pd2(dba)3, BINAP, NaOtBu, DME, 85 °C, 20 h, 35%; (d) 17b, trans-4-
aminocyclohexanol, DMSO, 150 °C in microwave, 1.5 h, 19%.
D.; Somers, D. O.; Tame, C. J.; Wayne, G.; Wilson, C.; Woolven, J. M. ACS Med.
Chem. Lett. 2013, 4, 948.
7. GST-ITK and LCK-HIS full length enzymes were from Invitrogen (PV3875 and
by reaction of 3-methyl-5-nitropyrazole 7 with (2-chloromethoxy-
ethyl)-trimethyl-silane followed by catalytic hydrogenation over
palladium on charcoal and final separation of the regioisomers by
chromatography.13
The building block 4-benzenesulfonyl-2,6-dichloropyridine 13
(Scheme 2) was prepared in two steps from commercially available
2,6-dichloro-4-iodopyridine 11 by palladium catalyzed cross-cou-
pling with thiophenol, followed by oxidation of thioether 12 with
meta-chloroperbenzoic acid.
P3043, respectively) and substrate was BLK peptide (Ac-EFPIYDFLPAKKK-NH2).
Reactions were carried out in a final volume of 51 l with 50 mM HEPES buffer
(pH 7.2), 15 mM MgCl2, 2 mM DTT, 0.015% Brij-35, 1 nM ITK or 0.5 nM LCK,
M substrate, 20 M ATP, and test article in a final concentration of 2%
DMSO. After 35 min incubation at RT, reactions were stopped upon addition of
10 of 30% TCA. Samples were centrifuged (4350 rpm, 4 °C, 5 min) and
subjected to LC/MS analysis on a Waters Acquity UPLC/TQD system equipped
with a Waters Acquity UPLC BEH C18 (2.1 Â 50 mm) 1.7 m column (injection
volume: 5 l, column temperature: 60 °C, flow rate: 1 ml/min, solvent A: 0.1%
l
2
l
l
ll
l
l
Compounds 4a–j were synthesized from 13 in two successive
nucleophilic displacements, initially with the desired 3-aminopyr-
azole followed by trans-4-aminocyclohexanol (Scheme 3). Amino-
pyrazoles 14a–f and 14h–j were all commercially available,
while 14g was prepared by cyclization of the corresponding 3-
oxo-3-(4-trifluoromethoxy-phenyl)-propionitrile with hydrazine
according to the literature procedure.14 Compound 5 (Scheme 3)
was prepared in an analogous manner to 4a–j, using 1-methyl-4-
aminoimidazole in the first nucleophilic displacement.
formic acid in LC/MS grade water, solvent B: 0.1% formic acid in LC/MS grade
ACN). Analytes were separated by applying a gradient from 15% to 32% solvent
B within 0.7 min and detected in positive mode ESI-MS/MS by MRM (multiple
reaction monitoring) of transitions 819.8/84.8 (BLK substrate) and 859.0/84.8
as well as 859.0/120.7 (phosphorylated BLK product). Ki values were
determined using the Morrison tight-binding equation
( William, J. W.;
Morrison, J. F. Methods Enzymol. 1979, 63, 437. ), modified to account for an
ATP-competitive mechanism of inhibition and for the concentration of active
kinase used.
8. Jurkat T-cells (400 K cells/well) were suspended in assay buffer (D-PBS + 0.1%
glucose + 0.22 lM sodium pyruvate), seeded in 96-well plate, treated with
compounds for 30 min, and stimulated with anti-CD3 Dynabeads (Invitrogen,
Attempts at installing the aminothiazole moiety by nucleophilic
substitution on dichloropyridine 13 in the presence of mild organic
bases were unsuccessful due to the low reactivity of the aminothi-
azole. Increasing the nucleophilicity by formation of the anion with
sodium hydride led to the preferential displacement of the sulfone
group at the 4-position. This issue was overcome by performing a
palladium catalyzed amination to afford intermediates 17a–b
(Scheme 4). The trans-4-aminocyclohexanol was then installed
either by nucleophilic displacement in DMSO or by palladium cat-
alyzed cross-coupling to afford compounds 6a–b.
In conclusion, starting from the non selective benzylpyrimidine
2, by judicious choice of the linker and optimization of small mol-
ecule conformations we have identified a new class of highly
potent and selective ITK inhibitors, as exemplified by compound
4i. The intrinsic potency, selectivity versus Lck and cellular activity
were all improved over the lead compound; however, further
improvements in physicochemical properties and in particular
solubility are required.
400 K beads/well) for 2 min at 37 °C. The cells were pelleted by brief spinning,
and after aspirating off media, 30
Technology) was added to each well, and plate was incubated at 4 °C for
30 min while shaking. After beads were quickly spinned down, 25 l of lysate
from each well were transferred to capture antibody (mouse monoclonal
antibody to human PLC 1, from Abcom) coated MSD goat anti-mouse 96-well
plate (MesoScale Discovery), and incubated at 4 °C overnight with shaking.
After 4Â wash with 150 l MSD Tris buffer (50 mM Tris–HCl, pH 7.5, 150 mM
NaCl, 0.02% Tween-20), 25 L of detection antibodies (1:2000 dilution mixture
of rabbit anti-pPLC 1, from Cell Signaling Technology and goat anti-rabbit
SULFO-Tag from MesoScale Discovery) to each well, and incubated for 2 h at
room temperature. After washing of the plate with 4 Â 150 L of MSD Tris
buffer, 150
l of 1Â MSD read buffer T was added to each well, and the plate
was read out on SECTOR Imager 6000 (MesoScale Discovery).
ll of 1Â lysis buffer (Cell Signaling
l
c
l
l
c
l
l
9. (a) Pastor, R. M.; Burch, J. D.; Magnuson, S.; Ortwine, D. F.; Chen, Y.; De La Torre,
K.; Ding, X.; Eigenbrot, C.; Johnson, A.; Liimatta, M.; Liu, Y.; Shia, S.; Wang, X.;
Wu, L. C.; Pei, Z. Bioorg. Med. Chem. Lett. 2014, 24, 2448; (b) Burch, J. D.; Lau, K.;
Barker, J. J.; Brookfield, F.; Chen, Y.; Chen, Y.; Eigenbrot, C.; Ellebrandt, C.;
Ismaili, M. H. A.; Johnson, A.; Kordt, D.; MacKinnon, C. H.; McEwan, P. A.;
Ortwine, D. F.; Stein, D. B.; Wang, X.; Winkler, D.; Yuen, P.-W.; Zhang, Y.; Zarrin,
A. A.; Pei, Z. J. Med. Chem. 2014, 57, 5714; (c) MacKinnon, C. H.; Lau, K.; Burch, J.
D.; Chen, Y.; Dines, J.; Ding, X.; Eigenbrot, C.; Heifetz, A.; Jaochico, A.; Johnson,
A.; Kraemer, J.; Kruger, S.; Krülle, T. M.; Liimatta, M.; Ly, J.; Maghames, R.;
Montalbetti, C. A. G. N.; Ortwine, D. F.; Pérez-Fuertes, Y.; Shia, S.; Stein, D. B.;
Trani, G.; Vaidya, D. G.; Wang, X.; Bromidge, S. M.; Wu, L. C.; Pei, Z. Bioorg. Med.
Chem. Lett. 2013, 23, 6331.
References and notes
10. The conformational analysis was performed using torsional sampling every 15°
around the SAHeterocycle bond. The energy of each conformer was calculated
by quantum mechanical methods at MP2/3-21G⁄ level of theory with
polarisable continuum model (PCM) of water using GAMESS.
1. Schaeffer, E. M.; Debnath, J.; Yap, G.; McVicar, D.; Liao, X. C.; Littman, D. R.;
Sher, A.; Varmus, H. E.; Lenardo, M. J.; Schwartzberg, P. L. Science 1999, 284,
638.
For details about GAMESS see ‘Advances in electronic structure theory:
2. Fowell, D. J.; Shinkai, K.; Liao, X. C.; Beebe, A. M.; Coffman, R. L.; Littman, D. R.;
Locksley, R. M. Immunity 1999, 11, 399.
3. Miller, A. T.; Wilcox, H. M.; Lai, Z.; Berg, L. J. Immunity 2004, 21, 67.
4. Mueller, C.; August, A. J. Immunol. 2003, 170, 5056.
GAMESS
a decade later’ Gordon, M. S.; Schmidt, M. W. In Theory and
Applications of Computational Chemistry: The First Forty Years, Dykstra, C. E.;
Frenking, G.; Kim, K. S.; Scuseria, G. E., Eds.; Elsevier: Amsterdam, 2005; pp
1167–1189.
5. Lin, T. A.; McIntyre, K. W.; Das, J.; Liu, C.; O’Day, K. D.; Penhallow, B.; Hung, C.
Y.; Whitney, G. S.; Shuster, D. J.; Yang, X.; Townsend, R.; Postelnek, J.; Spergel, S.
H.; Lin, J.; Moquin, R. V.; Furch, J. A.; Kamath, A. V.; Zhang, H.; Marathe, P. H.;
Perez-Villar, J. J.; Doweyko, A.; Killar, L.; Dodd, J. H.; Barrish, J. C.; Wityak, J.;
Kanner, S. B. Biochemistry 2004, 43, 11056.
6. Alder, C. M.; Baldwin, I. R.; Barton, N. P.; Campbell, A. J.; Champigny, A. C.;
Harling, J. D.; Maxwell, A. C.; Simpson, J. K.; Smith, I. E. D.; Tame, C. J.; Wilson,
C.; Woolven, J. M. WO2010/106016 A1; Alder, C. M.; Ambler, M.; Campbell, A.
J.; Champigny, A. C.; Deakin, A. M.; Harling, J. D.; Harris, C. A.; Longstaff, T.;
Lynn, S.; Maxwell, A. C.; Mooney, C. J.; Scullion, C.; Singh, O. M. P.; Smith, I. E.
11. Hill, A. P.; Young, R. J. Drug Discovery Today 2010, 15, 648.
12. The crystal structure of 4b in complex with ITK was determined from crystals
of ITK soaked with 2 mM compound for 4 days. Crystals of ITK (residues 357–
620) were prepared from recombinant protein prepared from baculovirus
infected SF9 cells. The protein was incubated on ice in the presence of 4 mM
AMP-PNP and 2 mM magnesium chloride for 2 h prior to crystal trials.
Crystallisation drops were set at 1:1 ratio of protein and precipitant
consisting of 0.2 M sodium nitrate, 0.1 M bis tris propane pH 8.5 and 20%
PEG 3350. Plate shaped crystals typically appear within a week at 4 °C and
require multiple rounds of macro-seeding to achieve sufficiently large crystals.

G. Trani et al. / Bioorg. Med. Chem. Lett. 24 (2014) 5818–5823
5823
Diffraction data were collected at beamline IO4-1 of Diamond Light Source
(Didcot, Oxfordshire, UK) at a wavelength of 0.917 Å. The structure was solved
by molecular replacement and refined using the program REFMAC and
rebuilding in the program COOT.
For details about COOT see: ‘Features and Development of Coot’ Emsley, P.;
Lohkamp, B.; Scott, W.; Cowtan, K. Acta Crystallogr. 2010, D66, 486.
13. The more polar, slow running regioisomer, was tentatively assigned as
structure 8 (Scheme 1) and was carried forward to the next stage.
For details about REFMAC see ‘Refinement of Macromolecular Structures by
the Maximum-Likelihood Method’ Murshudov, G. N.; Vagin, A. A.; Dodson, E. J.
Acta Crystallogr. 1997, D53, 240.
14. Haydar, S. N.; Bothmann, H.; Ghiron, C.; Maccari, L.; Micco, I.; Nencini, A.;
Zanaletti, R. WO 2010/009290 A1; Chem. Abstr. 2010, 152, 192109.