Sulfoximine-substituted trifluoromethylpyrimidine analogs as inhibitors of proline-rich tyrosine kinase 2 (PYK2) show reduced hERG activity

Article abstract of DOI:10.1016/j.bmcl.2009.04.093

The synthesis, in vitro properties, and in vivo pharmacokinetics for a series of sulfoximine-substituted trifluoromethylpyrimidines as inhibitors of proline-rich tyrosine kinase, a target for the possible treatment of osteoporosis, are described. These co

Full text of DOI:10.1016/j.bmcl.2009.04.093

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3253–3258
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Sulfoximine-substituted trifluoromethylpyrimidine analogs as inhibitors
of proline-rich tyrosine kinase 2 (PYK2) show reduced hERG activity
*
Daniel P. Walker , Michael P. Zawistoski, Molly A. McGlynn, Jian-Cheng Li, Daniel W. Kung,
Peter C. Bonnette, Amy Baumann, Leonard Buckbinder, Janet A. Houser, Jason Boer, Anil Mistry,
Seungil Han, Li Xing and Angel Guzman-Perez
Pfizer Global Research and Development, Eastern Point Road, Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis, in vitro properties, and in vivo pharmacokinetics for a series of sulfoximine-substituted tri-
fluoromethylpyrimidines as inhibitors of proline-rich tyrosine kinase, a target for the possible treatment
of osteoporosis, are described. These compounds were prepared as surrogates of the corresponding sul-
fone compound 1. Sulfone 1 was an attractive PYK2 lead compound; however, subsequent studies deter-
mined this compound possessed high dofetilide binding, which is an early indicator of cardiovascular
safety. Surprisingly, the corresponding sulfoximine analogs displayed significantly lower dofetilide bind-
ing, which, for N-methylsulfoximine (S)-14a, translated to lower activity in a patch clamp hERG K+ ion
channel screen. In addition, compound (S)-14a shows good oral exposure in a rat pharmacokinetic model.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 17 March 2009
Revised 17 April 2009
Accepted 21 April 2009
Available online 24 April 2009
Dedicated to Professor Paul A. Grieco on the
occasion of his 65th birthday
Keywords:
PYK2
FAK
Kinase
Osteoporosis
hERG
Sulfoximine
Proline-rich tyrosine kinase 2 (PYK2) and focal adhesion kinase
(FAK) are non-receptor tyrosine kinases, and together they consti-
tute the FAK subfamily. Unlike FAK expression, which is ubiqui-
tous, PYK2 expression is relatively restricted, with highest levels
in the brain and hematopoietic system. Recent pharmacological
and PYK2 knockout data demonstrate a link between inhibition
of PYK2 and bone anabolism.1 Hence, PYK2 represents a novel ana-
bolic approach for treating osteoporosis. Currently FDA-approved
oral agents to treat osteoporosis include bisphosphonates, estro-
gen, and selective estrogen receptor modulators (SERMs), all of
which are antiresorptive in nature and are insufficient for restoring
bone in critically osteoporotic patients. Parathyroid hormone 1–34
(Teriparatide)2 is the only approved bone anabolic agent; however,
it is an injectable that increases bone turnover with a balance
favoring bone formation. Considering the usual challenges associ-
ated with protein-based therapeutics, such as inconvenient routes
of administration, an orally bioavailable small-molecule PYK2
inhibitor represents an attractive anabolic therapy for patients
afflicted by low bone mass and offers a new opportunity to impact
this unmet medical need in the aging population.
Our prior work on PYK2 inhibitors focused on a trifluorometh-
ylpyrimidine (TFMP) series.3 Based on this work, PF-431396, a
FAK selective inhibitor, was transformed into sulfone 1, a potent
PYK2 inhibitor with modest (17-fold) selectivity against FAK (Chart
1). Sulfone 1 was further attractive in that it was stable in human
liver microsomes (HLM) and negative in
a high-throughput
reactive metabolite assay (RMA). However, subsequent studies
on this compound found that it possessed significant activity in a
dofetilide binding assay.4 The dofetilide binding assay has been
shown to be an effective screening tool for hERG blockade and
O
S
O
S
O
O
R1
CF3
CF3
N
N
H3C
N
H
N
H
N
N
H
N
H
N
N
H
2a (R1 = Me)
N
N
1
2b (R1 = cyclopropyl)
H
CF3
SO2Me
N
N
N
O
N
H
N
N
H
PF-431396
Chart 1. Selected PYK2 inhibitors.
* Corresponding author. Tel.: +1 636 247 1192.
E-mail address: daniel.p.walker@pfizer.com (D.P. Walker).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.04.093

3254
D. P. Walker et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3253–3258
Table 1
Biological data for arylsulfoximine-substituted analogs
R2
R1
N O
R2
R1
N O
R2
H3C
N
O
S
CF3
S
CF3
S
CF3
N
N
N
S
R
N
H
N
N
H
N
H
N
N
H
N
H
3a-d
N
N
H
N
N
N
14a
14b
, (R)-
(R)-
(S)-14a, (S)-14b, (S)-15
R1
R2
PYK2 IC50 (nM)
PYK2 Cell IC50 (nM)
FAK IC50 (nM)
HLM Ehd
MDCK (A > B)e Papp ꢀ 10ꢁ6 cm/s
Dofetilide % bindingf
a
b
c
Compound
PF-431396
1
2a
2b
3a
3b
3c
3d
(S)-14a
(S)-14b
(S)-15
(R)-14a
(R)-14b
311 (n = 12)
95 (n = 4)
140 (n = 4)
140, 260
67 (n = 4)
32 (n = 3)
22 (n = 3)
65, 69
43 (n = 5)
16 (n = 3)
21, 33
67 (n = 3)
63, 69
63 (n = 9)
60 (n = 4)
88 (n = 3)
95, 270
76, 110
10, 13
1.0 (n = 12)
1300, 2000
3800 (n = 3)
2100, 5800
1800, 2900
1100
ND
ND
1100, 1700
790
ND
0.78
<0.29
0.66
0.74
0.46
0.54
0.78
<0.29
0.45
0.58
0.58
0.52
0.64
9.2
9.5
5.3
4.8
4.4
4.7
6.6
1
5.1
8.3
7.8
5.7
9.3
41
48
50
76
11
5.9
8.4
12
1.1
ND
2
Me
Et
i-Pr2
H
Me
Et
Me
Me
Et
34
320 (n = 3)
29 (n = 3)
22 (n = 3)
24, 46
98 (n = 3)
37, 67
Me
Me
Et
Me
Me
2300
2400
4
ND
a
PYK2 fluorescence polarization (FP) protocol as described in Ref. 1. IC50 values represent the concentration to inhibit 50% of the phosphorylation of a peptide substrate
relative to vehicle control. Numbers indicate IC50 values generated from individual 8-point concentration response relationships in triplicate.
b
PYK2 LI-COR cellular assay as described in Ref. 19. Numbers indicate IC50 values generated from 9-point concentration response relationships in quadruplicate.
FAK fluorescence polarization (FP) protocol as described in Ref. 1 using FAK kinase domain construct as described in Ref. 20. IC50 values represent the concentration to
c
inhibit 50% of the phosphorylation of a peptide substrate relative to vehicle control. Numbers indicate IC50 values generated from individual 8-point concentration response
relationships in triplicate.
d
Eh = Hepatic extraction ratio, which is obtained from dividing estimated hepatic blood clearance of test compounds by the human hepatic blood flow of 20 mL/min/kg.
Protocols for measuring half-lives in HLM and subsequent scaling to blood clearance have been published (see: Ref. 21).
e
Madin Darby canine kidney cell permeability assay, apical to basel (A > B) permeability expressed as Papp
.
f
[3H]-Dofetilide binding assay as described in Ref. 4. Numbers indicate the percentage inhibition of [3H]-dofetilide binding to the hERG protein stably expressed on HEK-
293 cells following a 10 M dose of test compound.
l
proarrythmia.5,6 With an increased awareness by the pharmaceuti-
cal industry and regulatory agencies on the liabilities associated
with drug-induced QT prolongation, we felt it was important to
minimize the hERG activity for compounds within this chemical
series. We detail below our efforts to identify a novel compound
with the same promising biological profile as sulfone 1 but with re-
duced hERG liability. This work led to the discovery of a novel ser-
ies of arylsulfoximine-substituted TFMPs.7
During our search of possible surrogates for the sulfone moiety
in 1, we briefly considered sulfonamides. We prepared a 25 com-
pound library of N-alkyl-, N-cycloalkyl- and N-heteroalkylsulfona-
mides. The best compounds to emerge from this library in terms
of PYK2 enzyme and cellular activities were N-methyl
sulfonamide 2a and N-cyclopropylsulfonamide 2b (Chart 1). These
sulfonamides showed essentially equipotent PYK2 enzyme and cel-
lular activities to that of sulfone 1 (Table 1). In fact, most of the
compounds in the sulfonamide library showed equal PYK2 enzyme
activity to that of sulfone 1, although many displayed diminished
PYK2 cellular activity (data not shown). The X-ray co-crystal struc-
ture of compound 2a bound to PYK2 shows that the sulfonamide
moiety of 2a is partially solvent exposed (Fig. 1).8 The binding
mode of sulfonamide 2a with PYK2 is quite similar to the binding
mode of other related TFMPs, such as PF-431396 (Chart 1).3,9 Given
the structural similarities between compounds 1 and 2a, it is likely
that the sulfone moiety in 1 is also partially solvent exposed when
bound to PYK2. Thus, it was not surprising that compounds 1 and
2a have similar PYK2 enzyme activities. Unfortunately, neither sul-
fonamide 2a nor 2b showed any improvement in dofetilide bind-
ing, and both analogs were less stable in HLM compared to
sulfone 1 (Table 1).
We next turned our attention to sulfoximines, which are monoi-
mino analogs of sulfones. Sulfoximines are constitutionally and
O
O
HN
O
O
S
N
S
Ph
NHAc
HN
H
H
F
Antimicrobial agent11a
H
Ph
OH
N
NH
O
H
N
N
S
HO
OH
N
CONHtBu
O
Low calcemic
vitamin D analog11c
HIV-1 Protease
inhibitor11b
Figure 1. Crystal structure of compound 2a (magenta) bound to PYK2. Dashed lines
show hydrogen bond contacts to the hinge region.
Chart 2. Biologically active substances possessing a sulfoximine.

D. P. Walker et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3253–3258
3255
O
S
aqueous acid–base workup. Subjection of the sulfoxide to in situ
generated hydrazoic acid afforded racemic sulfoximine rac-8.13
Mono-methylation of the sulfoximine nitrogen in rac-8 under
Eschweiler–Clarke conditions gave rise (98%) to N-methylsulfoxi-
mine rac-10a.14 N-Ethylsulfoximine rac-10b and N-iso-propylsul-
foximine rac-10c were prepared via alkylation with the
appropriate alkyl iodide.15 Initial attempts to prepare N-Boc
sulfoximine rac-10d using standard conditions [(Boc)2O, Et3N,
MeOH] or [(Boc)2O, aq NaOH, MeOH] failed to provide any of the
desired product. However, the conditions of Bolm16 employing
tBuOK/tBuOH together with di-tert-butyl dicarbonate proved quite
effective. Transfer hydrogenation of the nitro group in rac-10a–d
afforded the corresponding amines rac-12a–d. Regioselective
addition of anilines rac-12a–d to the 2-position of 2,4-dichloro-
5-trifluoromethylpyrimidine was accomplished via the use of
ZnCl2.17 Treatment of chloropyrimidines rac-13a–d with enantio-
merically pure (1R,2R)-N,N-dimethylcyclopentane-1,2-diamine
gave rise (89–95%) to analogs 3a–d, wherein each analog
S
a
b
R1
R1
54-77%
86-87%
NO2
4 (R1 = Me)
5 (R1 = Et)
NO2
6 (R1 = Me)
7 (R1 = Et)
c (R2 = Me)
d (R2 = Et)
R2
N
O
HN
R1
O
S
g
S
R1
e (R2 = i-Pr2)
90-98%
f (R2 = Boc)
NO2
NO2
rac-10a-d (R1 = Me)
rac-8 (R1 = Me)
rac-9 (R1 = Et)
rac-11 (R1 = Et, R2 = Me)
R2
H3C
N
R2
H3C
N
O
S
O
S
h
CF3
Cl
N
51-75%
NH2
N
H
N
rac-12a-d
rac-13a-d
represents
a 50/50 diastereomeric mixture of (R)- and (S)-
R2
N
O
S
i (R2 = Me, Et, i-Pr2)
i, j (R2 = Boc)
sulfoximines.18
CF3
H3C
N
Biological data forsulfoximineanalogs 3a–d is detailedin Table 1.
Sulfone 1 and PF-431396 are shown for comparative purposes. The
sulfoximine analogs all showed similar PYK2 enzyme activity com-
pared to the corresponding sulfone analog. Based on the similarities
in structure for the sulfoximine analogs versus the sulfonamide ana-
log and knowledge of the binding mode of sulfonamide 2a to PYK2
(Fig. 1), it was not surprising that these compounds have very similar
enzyme activities. The N-methyl and N-ethyl sulfoximine analogs
show 30–35-fold selectivity against FAK, which is similar to the
FAK selectivity shown for the sulfone and sulfonamide analogs. We
were pleased to see that all the sulfoximines 3a–d were active in a
cell-based PYK2 assay, although sulfoximine 3d showed a fivefold
right shift in potency on going from the enzyme to the cell assay.
The diminished cellular activity of 3d is likely due to poor cell pene-
tration and was consistent with the compound’s weak performance
in an MDCK cell permeability assay (Table 1). Sulfoximine analogs
3a, 3b and 3d showed good to moderate stability in HLM, whereas
N-isopropyl sulfoximine 3c was unstable. Fortunately, all the sulfox-
imine analogs showed significantly lower dofetilide binding than
either sulfone 1 or sulfonamides 2a,b. The lower dofetilide binding
for the sulfoximine analogs was unexpected given the subtle differ-
ences between these analogs and sulfone 1 or sulfonamides 2a,b.
Based on the overall biological activities shown by N-methylsulfox-
imine 3a and N-ethylsulfoximine 3b, these analogs were identified
for further profiling.
2
R
N
N
4N
H
H
R
89-95%
N
3a-d
Scheme 1. General synthetic route for sulfoximine analogs. Reagents and condi-
tions: (a) H5IO6 (1.05 equiv), FeCl3 (0.03 equiv), CH3CN, room temperature, 30 min;
(b) NaN3 (1.1 equiv), concd H2SO4, CHCl3, 0 °C ? 48 °C, 24 h; (c) 37% aq CH2O,
HCO2H, 100 °C, 24 h, 95–98%; (d) NaH, DMF, 0 °C ? room temperature, 30 min; EtI,
1 h, 78%; (e) NaH, DMF, 0 °C ? room temperature, 30 min; 2-iodopropane,
lW,
80 °C, 20 min, 24%; (f) (Boc)2O, tBuOK, tBuOH, reflux, 48 h, 97%; (g) HCO2NH4
(excess), 20% Pd(OH)2/C (cat.), MeOH, reflux, 2 h; (h) 2,4-dichloro-5-trifluorometh-
ylpyrimidine, ZnCl2 (1.0 M solution in Et2O), 1,2-dichloroethane–tBuOH (1:1),
ꢁ5 °C, 1 h; add aniline 12, ꢁ5 °C, 1 h; Et3N ꢁ5 °C ? 60 °C, 24 h; (i) (1R,2R)-N,N-
dimethylcyclopentane–1,2-diamine, Na2CO3, DMF, 85 °C, 3 h; (j) HCl (4.0 M solu-
tion in dioxane), MeOH, room temperature, 4 h; NaHCO3.
configurationally stable compounds that have gained considerable
attention in synthetic organic chemistry.10 Recently, a number of
reports detailing the biological activities of sulfoximine-substi-
tuted analogs for the treatment of various diseases have appeared
in the literature (Chart 2).11
The preparation of type 3 sulfoximines is described in Scheme 1.
Thus, oxidation of sulfide 4 with periodic acid12 gave the corre-
sponding sulfoxide 6. Under these conditions ca. 3–4% of the corre-
sponding sulfone was formed. This was of little consequence as the
sulfone impurity could be easily removed in the next step via an
R2
R1
N
O
R2
R1
N
O
S
S
CF3
48-49%
b, c, d
N
S
S
NO2
)-10a (R1 = Me, R2 = Me)
N
H
N
N
H
41-63%
(S
(S)-10b (R1 = Me, R2 = Et)
)-11 (R1 = Et, R2 = Me)
N
(S
(S
S
a (R1 = Me, R2 = Me)
b (R1 = Me, R2 = Et)
)-14
)-14
R2
R1
N O
(
S
)-15 (R1 = Et, R2 = Me)
a
(
S
NO2
rac-10a (R1 = Me, R2 = Me)
rac-10b (R1 = Me, R2 = Et)
rac-11 (R1 = Et, R2 = Me)
R2
R2
R1
N
O
N
O
S
CF3
S
b, c, d
R1
N
R
R
N
H
N
N
H
NO2
44-53%
48-49%
(R)-10a (R1 = Me, R2 = Me)
N
(R)-14a (R1 = Me, R2 = Me)
(R)-10b (R1 = Me, R2 = Et)
(R)-14b (R1 = Me, R2 = E
(
)-11 (R1 = Et, R2 = Me)
R
Scheme 2. Preparation of diastereomerically pure sulfoximines. Reagents and conditions: (a) Chromatographic resolution [Chiralpak AS column, heptane–EtOH (70:30)]; (b)
HCO2NH4 (excess), 20% Pd(OH)2/C (cat.), MeOH, reflux, 2 h; (c) 2,4-dichloro-5-trifluoromethylpyrimidine, ZnCl2 (1.0 M solution in Et2O), 1,2-dichloroethane–tBuOH (1:1),
ꢁ5 °C, 1 h; add aniline, ꢁ5 °C, 1 h; Et3N ꢁ5 °C ? 60 °C, 24 h; (d) (1R,2R)-N,N-dimethylcyclopentane–1,2-diamine, Na2CO3, DMF, 85 °C, 3 h.

3256
D. P. Walker et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3253–3258
+H2N
Me
O
O
-
HN
SO3
O
b
S
S
Me
46%
NO2
NO2
(S)-(+)-8
16
O
HN
a
S
c or d
Me
NO2
rac-8
O
HN
O
R
N
S
S
Me
Me
38%
NO2
NO2
(S)-10a (R = Me)
(R)-(-)-8
(S)-10b (R = Et)
Scheme 3. Resolution of sulfoximine enantiomers. Reagents and conditions: (a) Chromatographic resolution [Chiralpak AS column, heptane–EtOH (70:30)]; (b) (ꢁ)-R-10-
camphorsulfonic acid, MeOH, reflux, 10 min; (c) 37% aq CH2O, HCO2H, 100 °C, 24 h, 96%; (d) NaH, DMF, 0 °C ? room temperature, 30 min; EtI, 1 h, 71%.
The absolute configuration of N-methylsulfoximine (S)-10a was
unambiguously established as (S) by the sequence outlined in
Scheme 3. Thus, chromatographic resolution of sulfoximine rac-8
using a Chrialpak AS column gave rise to sulfoximine enantiomers
(S)-(+)-823 and (R)-(ꢁ)-8. Salt formation of (S)-(+)-8 with (ꢁ)-R-10-
camphorsulfonic acid afforded a crystalline solid 16. Single crystal
X-ray analysis of sulfoximine 16 showed that it possessed the (S)-
configuration (Fig. 2).24 N-Methylation of (S)-(+)-8 afforded sulfox-
imine (S)-10a, whose spectral properties were identical in all re-
spects to (S)-10a prepared according to Scheme 2. Similarily, N-
ethylation of sulfoximine (S)-(+)-8 afforded sulfoximine (S)-10b
(Scheme 3), which was identical in all respects to (S)-10b prepared
according to Scheme 2.
The biological activity for individual diastereomers prepared in
Scheme 2 is shown in Table 1. For the N-methylsulfoximine ana-
logs, (S)-14a and (R)-14a, there was a modest preference for the
S-sulfoximine diastereomer in the PYK2 cellular assay. Otherwise,
Figure 2. ORTEP diagram of compound 16.
no significant differences in activity were observed between the
two N-methyl diastereomers for any of the assays in Table 1. Like-
wise, the same trend was observed for the corresponding N-ethyl-
sulfoximines in that sulfoximine (S)-14b was slightly more potent
against PYK2 than (R)-14b.
Prior to submitting sulfoximines 3a and 3b to additional
biological assays, it was desired to prepare these analogs as dia-
stereomerically pure compounds. Toward this end, chromato-
The final analog to be prepared in this study was N-methyl-S-
ethylsulfoximine (S)-15.25 This compound was prepared from ethyl
4-nitrophenyl sulfide26 (5) by methods similar to those described
for the other sulfoximine analogs (Schemes 1 and 2). Compound
(S)-15 displayed similar potency in the PYK2 enzyme and cellular
assays to that of sulfoximines (S)-14a and (S)-14b. Sulfoximine
(S)-15 showed low dofetilide binding, which was consistent with
the other sulfoximine analogs in the table. However, sulfoximine
(S)-15 showed decreased stability in HLM compared to the
corresponding methyl sulfoximine analog (S)-14a. Based on the
graphic resolution of N-methyl sulfoximine rac-10a using
a
Chiralpak AS column afforded enantiomers (S)-10a and (R)-10a
(Scheme 2). Diastereomerically pure sulfoximine analogs (S)-
14a and (R)-14a were prepared in three steps from sulfoximines
(S)-10a and (R)-10a, respectively. Diastereomerically pure N-eth-
ylsulfoximines (S)-14b and (R)-14b were obtained in an identical
manner to that described for the corresponding N-methylsulfox-
imines except N-ethylsulfoximine rac-10b was used as a starting
material.
Table 2
In vitro and in vivo data for selected compounds
RLM Eha
RMAb (HML)
hERG IC50
(
lM)
rPPBd (fu)
Rat pharmacokinetic datae
c
Compound
MRTf(h)
Vdss (L/kg)
Cmax
(
lM)
AUCinf
(l
M h)
Cave
(l
M)
g
CL (mL/min/kg)
1
<0.30
<0.30
Negative
Negative
4.7
20.3
0.30
0.36
42
74
3.7
2.3
13
13
ND
0.72
ND
5.0
ND
0.21
(S)-14a
a
RLM, rat liver microsomes; in vitro compound stabilities (Eh) are reported as the fraction of compound remaining as compared to the concentration at t0.
RMA: Reactive metabolite assay as described in Ref. 27 using GSH–EE as a trapping agent in NADPH-supplemented HLM.
hERG Patch clamp screen as described in Ref. 22. IC50 values represent the concentration to inhibit 50% of hERG current (IKr). Numbers represent IC50 values generated
b
c
from 3-point concentration response relationships in duplicate.
d
In vivo rat plasma protein binding, expressed as fraction unbound (fu), determined at a drug concentration of 1 lg/mL.
Clearance (CL), half-life and volume of distribution (Vdss) were generated following a 1.0 mg/kg iv dose in Spague–Dawley rats. Cmax, AUCinf and Cave were determined
e
following a 10 mg/kg po dose in Sprague–Dawley rats. Numbers represent the means of two experiments.
f
MRT: mean residence time.
Cave = AUCinf/t, where t = dosing interval in h.
g

D. P. Walker et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3253–3258
3257
data in Table 1, sulfoximine (S)-14a was chosen for further
profiling.
rina Zhao and John Soglia for generating reactive metabolite data.
We thank Robert Depianta for the chromatographic resolution of
various sulfoximines. We thank the ADME CoE group for in vitro
HTS ADME data. The Pfizer Institutional Animal Care and Use Com-
mittee 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.
The bioactivation potential of sulfoximine (S)-14a was assessed
in a high-throughput reactive metabolite assay (RMA) (Table 2).
This screen examines the bioactivation of drug candidates in
HLM via the detection of glutathione- (GSH) and/or glutathione
ethyl ester- (GSH–EE) captured metabolites.27 Bioactivation of
drug candidates can lead to toxicological liabilities.28 We were
pleased to find that the sulfoximine analog (S)-14a was negative
in this assay.
References and notes
In vitro cardiovascular safety was assessed in a patch-clamp
hERG K+ channel screen (Table 2).22 The hERG IC50 for compound
1. Buckbinder, L.; Crawford, D. T.; Qi, H.; Ke, H.-Z.; Olson, L. M.; Long, K. R.;
Bonnette, P. C.; Baumann, A. P.; Hambor, J. E.; Grasser, W. A.; Pan, L. C.; Owen, T.
A.; Luzzio, M. J.; Hulford, C. A.; Gebhard, D. F.; Paralkar, V. M.; Simmons, H. A.;
Kath, J. C.; Roberts, W. G.; Smock, S. L.; Guzman-Perez, A.; Brown, T. A.; Li, M.
Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 10619.
2. Neer, R. M.; Arnaud, C. D.; Zanchetta, J. R.; Prince, R.; Gaich, G. A.; Reginster, J.
Y.; Hodsman, A. B.; Eriksen, E. F. N. Engl. J. Med. 2001, 344, 1434.
3. Walker, D. P.; Bi, F. C.; Kalgutkar, A. S.; Bauman, J. N.; Zhao, S. X.; Soglia, J. R.;
Aspnes, G. E.; Kung, D. W.; Klug-McLeod, J.; Zawistoski, M. P.; McGlynn, M. A.;
Oliver, R.; Dunn, M.; Li, J.-C.; Richter, D. T.; Cooper, B. A.; Kath, J. C.; Hulford, C.
A.; Autry, C. L.; Luzzio, M. J.; Ung, E. J.; Roberts, W. G.; Bonnette, P. C.;
Buckbinder, L.; Mistry, A.; Griffor, M. C.; Han, S.; Guzman-Perez, A. Bioorg. Med.
Chem. Lett. 2008, 6071.
(S)-14a was 20.3
lM; the corresponding sulfone analog (1) had
an IC50 of 4.7 M in the same assay. This represents a 4.3-fold
l
improvement in the hERG activity for sulfoximine (S)-14a, which
is significant considering the two compounds showed similar
PYK2 cell potencies and similar plasma protein binding (Table 2).
Given the similar log D values for (S)-14a and 1,29 the origin of
reduced hERG activity for (S)-14a is not likely related to the lipo-
philicities of the two molecules. The reduction in hERG activity
of (S)-14a could be attributed to the sulfoximine moeity disrupting
the interaction of the adjacent phenyl ring within the hERG chan-
nel binding site.30
4. Greengrass, P. M.; Stewart, M.; Wood, C. M. Patent Cooperation Treaty
WO2003021271, March 13, 2003.
5. (a) Finlayson, K.; Turnbaull, L.; January, C. T.; Sharkey, J.; Kelly, J. S. Eur. J. Pharmacol.
2001,430,147;(b)Diaz,G.J.;Daniell,K.;Leitza,S.T.;Martin,R.L.;Su,Z.;McDermott,
J. S.; Cox, B. F.; Gintant, G. A. J. Pharm. Toxicol. Methods 2004, 50, 187.
6. A good correlation between dofetilide binding and hERG activity has been
observed within the TFMP chemical series based on prior work in our lab (data
not shown).
7. (a) While this work was on-going, several applications describing a related set
of arylsulfoximine-substituted pyrimidines were published in the patent
literature, Cf: Luecking, U.; Nguyen, D.; Von Bonin, A.; Von Ahsen, O.;
Krueger, M.; Briem, H.; Kettschau, H.; Prien, O.; Mengel, A.; Krolikiewicz, K.;
Boemer, U.; Bothe, U.; Hartung, I. Patent Cooperation Treaty WO 2007/071455,
June 28, 2007.; (b) Hartung, I.; Bothe, U.; Kettschau, G.; Luecking, U.; Mengel,
A.; Krueger, M.; Thierauch, K.-H.; Lienau, P.; Boemer, U. Patent Cooperation
Treaty WO 2008/155140, December 24, 2008.
8. The atomic coordinates for the X-ray structure of PYK2 in complex with
compound 2a have been deposited into the RCSB protein data bank: RCSB ID
code rcsb052639, and under PDB ID code 3H3C.
9. Han, S.; Mistry, A.; Chang, J. S.; Cunningham, D.; Griffor, M.; Bonnette, P. C.;
Wang, H.; Chrunyk, B. A.; Aspner, G. E.; Walker, D. P.; Brosius, A. D.; Buckbinder,
L. J. Biol. Chem. 2009, 284, 13193.
Plasma concentration–time profiles were obtained from rats
receiving a parenteral dose (1 mg/kg, iv) of sulfoximine (S)-14a.
The pharmacokinetic (PK) parameters of (S)-14a in rat best fit a
non-compartmental model (Table 2). The clearance (CL) of
(S)-14a (74 mL/min/kg) exceeded the hepatic blood flow in rat,
which was significantly higher than that predicted by HLM or rat
liver miscrosomes (Table 2). It is possible compound (S)-14a is
undergoing non-cytochrome P450-mediated metabolism in vivo.
This hypothesis is supported by the fact that the rat oral bioavail-
ability of sulfoximine (S)-14a was 100%. Analog (S)-14a displayed a
volume of distribution at steady state (Vdss) exceeding the total
water volume in the rat (13 L/kg). The mean-residence time
(MRT) of sulfoximine (S)-14a was roughly half that of the corre-
sponding sulfone (1), indicating the sulfoximine compound was
metabolically less stable than the corresponding sulfone. This is
consistent with the HLM data, which also predicted that sulfoxim-
ine (S)-14a was metabolically less stable than sulfone 1. However,
despite the high in vivo CL of sulfoximine (S)-14a, the oral PK char-
acteristics of this compound were sufficient to produce an average
free plasma concentration (Cavg) of 76 nM following a single oral
10 mg/kg dose of the test compound. Thus, the average free plasma
concentration of (S)-14a over 24 h exceeded the compound’s PYK2
cell IC50 (29 nM) by 2.6-fold.
10. (a) Okamura, H.; Bolm, C. Chem. Lett. 2004, 33, 482; (b) Reggelin, M.; Zur, C.
Synthesis 2000, 1; (c) Johnson, C. Aldrichim. Acta 1985, 18, 3.
11. (a) Hester, J. B. Jr.; Alexander, D. L. Patent Cooperation Treaty WO 2001/46185,
June 28, 2001.; (b) Raza, A.; Sham, Y. Y.; Vince, R. Bioorg. Med. Chem. Lett. 2008,
5406; (c) Kahraman, M.; Sinishtaj, S.; Dolan, P. M.; Kensler, T. W.; Peleg, S.;
Saha, U.; Chuang, S. S.; Bernstein, G.; Korczak, B.; Posner, G. H. J. Med. Chem.
2004, 47, 6854.
12. Kim, S. S.; Nehru, K.; Kim, S. S.; Won, D.; Jung, H. C. Synthesis 2002, 2484.
13. Misati, F.; Fair, T. W.; Reiner, L. J. Am. Chem. Soc. 1951, 73, 459.
14. Jonsson, E. U.; Johnson, C. R. J. Am. Chem. Soc. 1971, 93, 5308.
15. Raguse, B.; Ridley, D. D. Aust. J. Chem. 1986, 39, 1655.
In summary, we have described the synthesis and SAR for a no-
vel series of sulfoximine-substituted TFMPs. These compounds
showed equal potency to sulfone 1 in the PYK2 enzyme assay
and showed modest FAK selectivity. Except for compound 3d, all
sulfoximine analogs showed good PYK2 cellular activity and mod-
erate stability in HLM. Unexpectedly, the sulfoximine analogs
showed significantly less dofetilide binding than the corresponding
sulfone or sulfonamide compounds. The improvement in dofetilide
binding for sulfoximine 14a translated to a 4.3-fold improvement
in a hERG patch clamp K+ channel assay. Furthermore, sulfoximine
14a was found to have good oral exposure in a rat pharmacokinetic
model. Future plans include profiling compound (S)-14a in addi-
tional safety-related assays as well as an animal efficacy model
for osteoporosis.
16. Bolm, C.; Kahmann, J. D.; Moll, G. Tetrahedron Lett. 1997, 1169.
17. Kath, J. C.; Richter, D. T.; Luzzio, M. J. US Patent 7,122,670, October 16, 2006.
18. Separation of the sulfoximine diastereomers for analogs 3a–d using silica gel
chromatography was unsuccessful.
19. Full experimental details around the PYK2 LI-COR cellular assay will be
published elsewhere in due course.
20. Slack-Davis, J. K.; Martin, K. H.; Tighman, R. W.; Iwanicki, M.; Ung, E. J.; Autry,
C.; Luzzio, M. J.; Cooper, B.; Kath, J. C.; Roberts, W. G.; Parsons, J. T. J. Biol. Chem.
2007, 282, 14845.
21. Obach, R. S. Drug Metab. Dispos. 1999, 27, 1350.
22. Dubin, A. E.; Nasser, N.; Rohrbacher, J.; Herman, A. N.; Marrannes, R.;
Grantham, C.; Van Rossem, K.; Cik, M.; Chaplan, S. R.; Gallacher, D.; Xu, J.;
Guia, A.; Byrne, N. G.; Mathes, C. J. Biomol. Screen 2005, 10, 168.
23. Analytical data for sulfoximine (S)-(+)-8: first eluting enantiomer, >98% ee; mp
153–154 °C; ½a 2D5
ꢂ
18 (c = 1, MeOH); 1H NMR (400 MHz, DMSO-d6) d 8.39 (d, 2H,
J = 8.0 Hz), 8.16 (d, 2H, J = 8.0 Hz), 4.58 (br s, 1H), 3.14 (s, 3H).
24. The data for compound 16 were collected on a Bruker APEX diffractometer at
Pfizer Groton Laboratories, and all crystallographic calculations were facilitated
þ
by the SHELXTL system: C7H9N2O ꢃ C10H15S1O4ꢁ ꢃ H2O; Fw = 450.52; monoclinic;
space group P2(1); unit cell 3dimensions: a = 8.703(3) Å, b = 6.959(2) Å, c =
Acknowledgments
17.243(6) Å with b = 93.099(10); volume = 1042.9(6) Å3; Z = 2; Dcalcd
=
1.435 Mg/m3; absorption coefficient = 2.735 mmꢁ1
F2 = 1.091; final R indices [I > 2r(I)]: R1 = 0.0428, wR2 = 0.1093.
; F(0 0 0) = 476; GOF on
We thank Jon Bordner and Ivan Samardjiev for solving the X-ray
crystal structure of compound 16. We thank Bernard Fermini and
Shuya Wang for generating hERG patch clamp data. We thank Sab-
25. The absolute configuration of (S)-11, which is an intermediate of S-
ethylsulfoximine (S)-15, was unambiguously established via single crystal X-

3258
D. P. Walker et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3253–3258
ray analysis utilizing methods similar to those described in Scheme 3 (data not
shown).
26. Pasto, D. J.; Cottard, F. J. Org. Chem. 1994, 59, 4642.
27. Soglia, J. R.; Harriman, S. P.; Zhao, S.; Barberia, J.; Cole, M. J.; Boyd, J. G.; Contillo,
L. G. J. Pharm. Biomed. Anal. 2004, 36, 105.
Harriman, S. P. Curr. Drug Metab. 2005, 6, 161; (b) Evans, D. C.; Watt, A. P.;
Nicoll-Griffith, D. A.; Baillie, T. A. Chem. Res. Toxicol. 2004, 17, 3; (c) Ju, C.;
Uetrecht, J. P. Curr. Drug Metab. 2002, 3, 367.
29. For sulfone 1: measured logD (pH 7.4) = 1.5. For sulfoximine (S)-14a: measured
log D (pH 7.4) = 1.7.
28. (a) Kalgutkar, A. S.; Gardner, I.; Obach, R. S.; Shaffer, C. L.; Callegari, E.; Henne,
K. R.; Mutlib, A. E.; Dalvie, D. K.; Lee, J. S.; Nakai, Y.; O’Donnell, J. P.; Boer, J.;
30. See: Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. J. Med. Chem. 2006, 49,
1. and references cited therein.