Molecular design, synthesis, and structure-activity relationships leading to the potent and selective p56lck inhibitor BMS-243117

Article abstract of DOI:10.1016/S0960-894X(03)00380-9

A series of structurally novel benzothiazole based small molecule inhibitors of p56^lck were prepared to elucidate their structure-activity relationships (SARs), selectivity and cell activity in the T-cell proliferation assay. BMS-243117 (compound 2) is identified as a potent, and selective Lck inhibitor with good cellular activity (IC₅₀=1.1 μM) against T-cell proliferation.

Full text of DOI:10.1016/S0960-894X(03)00380-9

Bioorganic & Medicinal ChemistryLetters 13 (2003) 2145–2149
Molecular Design, Synthesis, and Structure–Activity Relationships
Leading to the Potent and Selective P56lck Inhibitor BMS-243117
Jagabandhu Das,* James Lin, Robert V. Moquin, Zhongqi Shen, Steven H. Spergel,
John Wityak, Arthur M. Doweyko, Henry F. DeFex, Qiong Fang, Suhong Pang,
SidneyPitt, Ding Ren Shen, GaryL. Schieven and Joel C. Barrish
Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543-4000, USA
Received 14 February2003; accepted 7 April 2003
lck
Abstract—A series of structurallynovel benzothiazole based small molecule inhibitors of p56
were prepared to elucidate their
structure–activityrelationships (SARs), selectivityand cell activityin the T-cell proliferation assay. BMS-243117 (compound
identified as a potent, and selective Lck inhibitor with good cellular activity(IC 50=1.1 mM) against T-cell proliferation.
# 2003 Elsevier Science Ltd. All rights reserved.
2) is
p56lck(Lck), a member of the Src familyof non-receptor
protein tyrosine kinases is expressed primarily in
T-lymphocytes and natural killer cells. In T cells, Lck is
absolutelyrequired for T-cell antigen receptor (TCR)
signaling, culminating in IL-2 gene expression and
effector functions.1 T-cells lacking Lck are shown to be
severely impaired in TCR tyrosine phosphorylation and
subsequent activation via the TCR.2 Selective inhibitors
of Lck mayhave potential therapeutic utilityin the
treatment of T cell mediated disorders such as auto-
immune and inflammatorydiseases and in the preven-
tion of solid organ transplant rejection.3
Accordingly, 2-amino-6-carbethoxybenzothiazole5 was
treated with excess of di-tert-butyl dicarbonate and a
catalytic amount of 4-dimethylaminopyridine in di-
chloromethane to form the corresponding tert-butyl
carbamate. Saponification with aqueous sodium
hydroxide formed acid 3 which was coupled to a wide
varietyof anilines 4 under standard coupling conditions
to form anilides 6 outlined in Table 3. Alternatively
carboxylic acid 3 was converted to its acid chloride
(CH2Cl2, oxalyl chloride, catalytic DMF, rt) and then
condensed with the anilines 4 (THF, i-Pr2NEt, 0 ꢀC) to
form anilides 5. Hydrolysis of the tert-BOC group with
trifluoroacetic acid afforded the 2-amino-benzothiazole
analogues 6. This synthetic route is quite general and
was used for the synthesis of related analogues (Table
2).
Over the past decade, development of small molecule Lck
inhibitors4 has been an area of major pharmaceutical
research. High-throughput screening of our in-house
compound collection identified thiazole 1 as a modestly
potent Lck inhibitor (Fig. 1). In this communication, we
describe structure activitystudies leading to a novel ben-
zothiazole template, that have culminated in the discovery
of BMS-243117 (2) as a potent, and selective Lck inhibitor
having excellent activityin vitro and good activityin a
T-cell proliferation assay. The synthesis, and SAR studies
leading to BMS-243117 and analogues are described.
A solution-phase parallel synthesis approach was used
to prepare a large number of carboxamides 7 (Table 4).
2-Aminobenzothiazole 6 was either treated with the
corresponding carboxylic acid anhydride in dichloro-
methane and pyridine or coupled with a carboxylic acid
under standard activating conditions to form carbox-
amides 7 (Scheme 2).
The synthesis of this class of compounds follows a gen-
eral synthetic route that is illustrated in Scheme 1.
A similar approach was utilized to prepare the urea
derivatives 8 (Scheme 3). 2-Aminobenzothiazole 6 was
treated with an isocyanate in dichloromethane and
pyridine to form urea 8. Alternatively 6 was treated with
phenyl chloroformate in aqueous THF in the presence
*Corresponding author. Tel.:+1-609-252-5068; fax:+1-609-252-6804;
e-mail: jagabandhu.das@bms.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00380-9

2146
J. Das et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2145–2149
of potassium bicarbonate to form a phenyl carbamate
which was treated with an amine in THF to form the
urea 8.
Compound 2 (BMS-243117) and its analogues were
tested for their abilityto inhibit the phosphorylation of
an exogenous substrate6 byhuman Lck enzyme (IC
Tables 1–5).
,
50
Figure 1. Lck inhibitoryactivityof thiazole 1 and benzothiazole 2.
With compound 1 as the initial lead, the effect of the
modification of the core thiazole ring was studied and
these results are summarized in Table 1. Thiazole 1 and its
tert-butylcarbamate derivative 9 were found to be equi-
potent in inhibiting human Lck. Replacement of the thia-
zole nucleus in 1 with a benzothiazole ring system (10)
resulted in 5-fold loss in potencybut the corresponding
tert-butylcarbamate analogue 5a was an order of magni-
tude more potent than 1. Similar replacement of the
Scheme 1. (a) CH2Cl2, DMAP, (BOC)2O, 72%; (b) MeOH, 1 N aq
NaOH, 91%; (c) DMF, HATU, i-Pr2NEt, 55–85%; (d) CF3CO2H,
72–95%.
Table 1. SAR for core nucleus modification
Compd
R
Lck inhibition
IC50, mM6
Compd
R
Lck inhibition
IC50, mM6
1
3.2
3.0
12
13
14
15
16
>12.5
7.25
9
10
5a
11
15
23
0.29
3.23
>12.5
>12.5
Table 2. SAR for benzothiazole core modification
Compd
R
Lck inhibition
IC50, mM6
Compd
R
Lck inhibition
IC50, mM6
5a
17
18
19
0.29
>12.5
>12.5
>12.5
20
21
22
23
4.4
>12.5
5.6
>12.5

J. Das et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2145–2149
2147
Table 3. SAR for aniline modification
Compd
R
Lck inhibition
IC50, mM6
Compd
R
Lck inhibition
IC50, mM6
5a
5b
5c
5d
5e
5f
2,4,6- Me3
2-Me,4-Br
2-Cl,6-Me
0.29
>12.5
0.07
5i
5j
5k
5l
5m
5n
5o
5p
2-Br,6-Me
5-Br, 2,4,6- Me3
2,6- Me2,3-iPr
2,6- (OMe)2
2,6-Me2,4-CO2Me
2-Br,4,6- Me2
4.27
>12.5
>12.5
>12.5
3.12
2,6- Me2
2-Me,6-(CH2)2OH
2-OH,6-Ac
2-Br,3,4,6- Me3
2,6-Me2,4-Br
0.09
>12.5
>12.5
>12.5
0.11
0.20
0.31
9.65
5g
5h
2,6-Me2,4-CH2OH
2,6-Me2,4-Ph
Table 5. SAR for urea modification
Table 4. SAR for amide modification
Compd
R
R1
Lck inhibition
IC50, mM6
Compd
R
R1
Lck inhibition
IC50, mM6
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
7p
7q
7r
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,6-Me2
Me
Ph
i-Propyl
Cyclopropyl
Cyclopropyl
i-Propyl
2-Me-cyclopropyl
Cyclopropyl
Cyclobutyl
2-Tetrahydrofuryl
o-MePh–CH2
2-Me-cyclopropyl
trans-2-Ph-cyclopropyl
1-Me-cyclopropyl
Cyclopropyl-CH2
0.12
>12.5
0.70
8a
8b
8c
8d
8e
8f
8g
8h
8i
8j
8k
8l
8m
8n
8o
8p
8q
2
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,4,6-Me3
2,6-Me2
Me
Ph
Bn
0.13
0.38
1.73
0.30
0.33
0.07
0.09
0.10
0.08
0.09
0.045
0.039
0.019
2.01
>3.125
0.123
0.087
0.004
0.055
0.03
0.02
0.015
1.80
Cyclopropyl
n-Bu
t-Bu
Cyclopropyl
t-Bu
1,1-Me2propyl
Cyclopropyl
Cyclopentyl
2-Furyl-CH2
a-Et-Bn
2-Pyridinyl
2-Pyrimidinyl
p-F-Ph
2,6-Me2
2,6-Me2
0.011
0.009
0.666
0.272
0.565
0.007
0.027
0.337
0.085
0.007
0.003
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2,6-Me2
2,6-Me2
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me
2-Cl,6-Me trans-2-Me2NCH2-cyclopropyl
2-Cl,6-Me trans-2-CH2OH-cyclopropyl
2-Cl,6-Me 2-Pyrrolidinyl-CH2-cyclopropyl
p-F-Bn
t-Bu
1,1-Me2propyl
Me2(N(CH2)2
0.011
8s
8t
8u
8v
2-Cl,6-Me N-Me-2-pyrrolidinyl-(CH2)2
2-Cl,6-Me 2,2,6,6-Me4-4-piperidinyl
0.007
0.004
Scheme 2. (a) (R1CO)2O, DMAP, THF, Py; (b) R1CO2H, HOAt,
EDAC, THF, i-Pr2NEt, 45 ꢀC.
In order to further optimize the potency, we investi-
gated the effect of substitution on the aniline ring of 5a
(Table 3). The methyl substituent at C4 can be replaced
with hydrogen (5d), bromine (5h) or hydroxymethyl
group (5o) without anyloss of activity. In contrast, a
carbomethoxygroup ( 5m) or a bulkyphenyl ring at this
position is not tolerated. Addition of a substituent at
either C3 or C5 is quite detrimental for activity( 5g, 5j,
5k). A methyl group at C2 of 5d can be substituted with
chlorine (5c) but replacement with bromine (5i) or
hydroxyethyl (5e) resulted in significant loss of potency.
The importance of 2,6-substitution in the aniline ring is
demonstrated bythe fact that compound 5b is more
than 150-fold less potent than 5h. This finding is con-
sistent with findings reported earlier with other Lck
inhibitors.4a,c It appears that onlya small alkly or
halogen substituent is tolerated at C2 and C6 since
replacement of a chlorine bybromine at C2 ( 5i) or
Scheme 3. (a) R1NCO, CH2Cl2, Py; (b) PhOCOCl, THF, aq KHCO3;
(c) R1NH2, THF.
thiazole nucleus with a benzimidazole, benzoxazole, or
7-azabenzothiazole ring system led to significant
attenuation in activityin vitro. Compound 5a served as
the starting point for further optimization.
Table 2 outlines the SAR observed with the carbox-
amide regio-isomers of compound 5a or effect of sub-
stitution in the phenyl ring of compound 5a. None of
these modifications led to anyimprovement in potency.

2148
J. Das et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2145–2149
introduction of polar functions (5f, 5l) has led to sig-
nificant losses of activity. Because of the optimal potency
of the 2,4,6-trimethylaniline (5a), 2,6-dimethylaniline
(5d), and 2-chloro-6-methylaniline (5c) analogues, fur-
ther optimization was carried out on these series.
Table 6. Enzyme selectivity of selected lck inhibitors
Compd
Lck
IC50, nM6
Srca Fyna Hcka Blka Lyna Fgra T-Cell prolif.
IC50, mM6
2
7l
7p
7q
8u
8v
4
7
7
3
7
4
158 32
57 1.5
57 23
960 84 330 60
92 46
650 47 100 10
NT NT 22
>600 300 150 17
325 16 185
1.1
5
3
1.58
3.53
1.45
0.96
2.80
Table 4 outlines the SAR observed with the carbox-
amide modification at the C2 amine of the benzothi-
azole nucleus. In the 2,4,6-trimethylaniline series,
acetamide 7a was roughlyequipotent as the carbamate
5a. Corresponding iso-propyl amide 7c was about 6-fold
less potent while the benzamide 7b was significantlyless
potent than 7a. Cyclopropyl amide 7d was the most
potent analogue in this series. A similar trend was
observed with the 2,6-dimethylanilines. Cyclopropyl
amides (7h, 7l, 7p–q) in the 2-chloro-6-methylaniline
series turned out to be the most potent lck inhibitors.
Replacement of the cyclopropyl group by cyclobutyl
(7I), tetrahydrofuryl (7j) or o-MeBn (7k) was detri-
mental for activityas was the introduction of a methyl
substituent at C1 of the cyclopropyl ring (7n) or inser-
tion of a methylene linker (7o). Substitution at C2 of the
cyclopropyl ring was well tolerated as was the intro-
duction of a polar functionalityat this position. Com-
pounds 7l, and 7p–q were identified to be the most
potent analogues in the carboxamide series.
12
200
5
3
7
84 0.5
6
aSelectivityratio: IC 50/IC50 (Lck); NT, not tested.
selective against several additional receptor and non-
receptor tyrosine kinases, and serine/threonine kinases
(>6000-fold selectivityover cdk2, IGF1R, Kdr, and
p38 kinases). Compound 2 inhibited anti-CD3/anti-
CD28 induced PBL proliferation with an IC50 of
1.1 mM.
A binding model for these benzothiazole Lck inhibitors
was developed based on the published coordinates of
the activated Lck kinase domain bound to ANP (phos-
phoaminophosphonic acid-adenylate ester,
a non-
hydrolyzable ATP mimic).7 As illustrated in Figure 2,
BMS-243117 (2) binds in an extended conformation to
the ATP-binding site of Lck. The 2,6-disubstituted ani-
line of 2 occupies a clearlydefined hydrophobic pocket
(common to all Src familykinases) which is not occu-
pied byATP. The 2,6-disubstitution skews the aniline
ring out of plane with respect to the benzothiazole core
and this in turn allows the phenyl ring to fit into the
narrow and angular hydrophobic pocket. The SAR
observed regarding the aniline substitution pattern in
this series is consistent with this proposed model. In
addition, compound 2 makes several productive hydro-
gen bond interactions within the active site: the NH of
the aniline is hydrogen bonded to Thr316 side-chain
hydroxyl, the benzothiazole nitrogen is in H-bond con-
tact with Tyr318, and the two NHs of the urea moiety
are in contact with Met319, and Glu320, respectively.
The high degree of selectivityexhibited bythe urea
analogue 2 is not clearlyunderstood but it maybe due
to differences in the amino acid side chains found at the
edge of the ATP binding pocket in other kinases. For
example, Lck residues Glu320 and Asn321 are replaced
Similarly, functionalization of the C2 amino group as a
urea also led to some verypotent Lck inhibitors in this
series (Table 5). Some of the interesting SAR findings
are: (1) in the 2,4,6-trimethylaniline series methyl urea
8a was more potent than the corresponding phenyl (8b),
n-butyl (8e), cyclopropyl (8d) and the benzyl (8c) ana-
logues. More importantly, replacement of the methyl by
tert-butyl group (8f) was well tolerated. The corre-
sponding analogue (8h) in the 2,6-dimethylaniline series
was equipotent to 8f as were the homologated analogue
8i and the cyclopropyl urea 8g. As in the carboxamide
series the ureas in the 2-chloro-6-methylaniline series
were the most potent Lck inhibitors. The cyclobutyl
(8k) and tetrahydrofurylmethyl (8l) ureas were roughly
equipotent, while the a-ethylbenzyl analogue (8m) was
5-fold more potent than the cyclopropyl analogue (8j).
Heteroaryl ureas (8n, and 8o) were significantlyless
potent. In contrast, the phenyl (8p) and benzyl (8q) urea
modifications were tolerated. The tert-butyl urea 2 was
one of the most potent inhibitors in this series. SAR
pattern observed with the urea analogues was in some
cases different from the one observed with the carbox-
amides. In general the cyclopropyl amide analogues (7d,
7e, and 7h) were significantlymore potent than the cor-
responding urea analogues (8d, 8g, and 8j). However, as
in the carboxamide series, introduction of polar func-
tionalityalso resulted in some of the most potent urea
analogues (8u, 8v).
The most potent carboxamide and urea analogues (2, 7l,
7p, 7q, 8u, and 8v) were tested for their selectivityagainst
several kinases and for cell activityin a T-cell prolifera-
tion assay( Table 6). All these compounds showed some
degree of selectivityagainst other members of the Src
familykinases. Urea 2 (BMS-243117) is the most selec-
tive Lck inhibitor in this series. In addition, 2 was highly
Figure 2. Proposed conformation of 2 bound to the ATP binding site
of Lck.

J. Das et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2145–2149
2149
in Src by Ser 320 and Lys321, respectively, and in Fyn
by Asn320 and Lys321, respectively.
Chem. Lett. 2002, 12, 3153. (c) Snow, R. J.; Cardozo, M. G.;
Morwick, T. M.; Busacca, C. A.; Dong, Y.; Eckner, R. J.;
Jacober, S.; Jakes, S.; Kapadia, S.; Lukas, S.; Panzenbeck, M.;
Peet, G. W.; Peterson, J. D.; Prokopowicz, A. S., III; Sellati,
R.; Tolbert, R. M.; Tschantz, M. A.; Moss, N. J. Med. Chem.
2002, 45, 3394. (d) Arnold, L. D.; Calderwood, D. J.; Dixon,
R. W.; Johnston, D. N.; Kamens, J. S.; Munschauer, R.;
Rafferty, P.; Ratnofsky, S. E. Bioorg. Med. Chem. Lett. 2000,
10, 2167. (e) Burchat, A. F.; Calderwood, D. J.; Hirst, G. C.;
Holman, N. J.; Johnston, D. N.; Munschauer, R.; Rafferty, P.;
Tometzki, G. B. Bioorg. Med. Chem. Lett. 2000, 10, 2171.
5. Dave, A. M.; Bhatt, K. N.; Undavia, N. K.; Trivedi, P. B.
J. Ind. Chem. Soc. 1988, 65, 365.
In summary, we have developed a novel series of
benzothiazole Lck inhibitors based on our initial thia-
zole lead 1. SAR studies identified several carboxamide
and urea analogues as potent and selective inhibitors.
BMS-243117 (2) was shown to be a potent, and selective
lck inhibitor with good potencyin a T-cell proliferation
assay.
6. (a) Lck enzyme assay: Recombinant Lck expressed as a His-
tagged protein in insect cells using a baculovirus expression
system and purified by nickel affinity chromatography was
incubated in kinase buffer (20 mM MOPS, pH 7, 10 mM
MgCl2) in presence of the test compound. The reaction was
initiated byaddition of substrates to the final concentration of
1 mM ATP, 3.3 mCi/mL [33P}g-APT, and 0.1 mg/mL acid
denatured enolase, and stopped after 10 min byaddition of
10% trichloroacetic acid, 100 mM sodium pyrophosphate fol-
lowed by2 mg/mL bovine serum albumin. The labeled enolase
protein substrate was precipitated at 4 ꢀC, harvested onto
Packard Unifilter plates and counted in a Topcount scintilla-
tion counter. (b) Enzyme assays for Src, Fyn, Hck, Blk, Lyn,
and Fgr were run under the same conditions as for Lck. (c)
PBL proliferation assay: A 96-well plate was coated with
monoclonal antibodyto CD3 (G19-4), the antibodywas
allowed to bind, and the plate was washed. Normal human
peripheral blood T-cells were added to the wells along with the
test compound and anti-CD28 (E.3) antibody. After 3 days,
[3H]-thymidine was added to the cells, after further incubation,
the cells were harvested and counted in a scintillation counter.
7. (a) Two crystal structures of the catalytic domain of Lck
(Tyr394 phosphorylated) were published. See: Yamaguchi, H.;
Hendrickson, W. A. Nature 1996, 384, 484. (b) Zhu, X.; Kim,
J. L.; Newcomb, J. R.; Rose, P. E.; Stover, D. R.; Toledo,
L. M.; Zhao, H.; Morgenstern, K. A. Structure 1999, 7, 651.
Protein Data Bank (1QPC pdb, Research Collaboratoryfor
Structural Bioinformatics.
References and Notes
1. (a) Molina, T. J.; Kishihara, K.; Siderovski, D. P.; van
Ewijk, W.; Narendran, A.; Timms, E.; Wakeham, A.; Paige,
C. J.; Hartman, K.-U.; Veillette, A.; Davison, D.; Mak, T. W.
Nature 1992, 357, 161. (b) Levin, D. S.; Anderson, S. J.; For-
bush, K. A.; Perlmutter, R. M. EMBO J. 1993, 12, 1671.
2. (a) Straus, D. B.; Weiss, A. Cell 1992, 70, 585. (b) Chan,
A. C.; Desai, D. M.; Weiss, A. C. Ann. Rev. Immunol. 1994,
12, 555. (c) Weiss, A.; Littman, D. Cell 1994, 76, 263. (d)
Hanke, J. H.; Gardner, J. P.; Dow, R. L.; Changelian, P. S.;
Brissette, W. H.; Weringer, E. J.; Pollok, B. A.; Connelly, P. A.
J. Biol. Chem. 1996, 271, 695. (e) Van Oers, N. S. C.; Lowin-
Kropt, B.; Finlay, D.; Connolly, K.; Weiss, A. Immunity 1996,
5, 429.
3. Hanke, J. H.; Pollok, B. A.; Changelian, P. S. Inflammation
Res. 1995, 44, 357.
4. (a) Chen, P.; Norris, D.; Iwanowicz, E. J.; Spergel, S. H.;
Lin, J.; Gu, H. H.; Shen, Z.; Wityak, J.; Pang, S.; De Fex,
H. F.; Pitt, S.; Doweyko, A. M.; Bassolino, D. A.; Roberge,
J. Y.; Poss, M. A.; Chen, B.-C.; Schieven, G. L.; Barrish, J. C.
Bioorg. Med. Chem. Lett. 2002, 12, 1361. (b) Chen, P.; Iwa-
nowicz, E. J.; Norris, D.; Gu, H. H.; Lin, J.; Moquin, R. V.;
Das, J.; Wityak, J.; Spergel, S. H.; De Fex, H.; Pang, S.; Pitt,
S.; Shen, D. R.; Schieven, G. L.; Barrish, J. C. Bioorg. Med.