The lab oddity prevails: Discovery of Pan-CDK inhibitor (R)-S-Cyclopropyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-methylpropyl]oxy}-5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)sulfoximide (BAY1000394) for the treatment of cancer

Article abstract of DOI:10.1002/cmdc.201300096

Lead optimization of a high-throughput screening hit led to the rapid identification of aminopyrimidine ZK304709, a multitargeted CDK and VEGF-R inhibitor that displayed a promising preclinical profile. Nevertheless, ZK304709 failed in phaseI studies due

Full text of DOI:10.1002/cmdc.201300096

CHEMMEDCHEM
FULL PAPERS
FULL PAPERS
The unconventional proposal to intro-
duce a sulfoximine group into a lead
series of aminopyrimidine pan-CDK in-
hibitors was initially met with much
skepticism. However, it was the intro-
duction of the sulfoximine group that fi-
nally allowed the project problems to
be overcome, culminating in the iden-
tification of BAY 1000394, a nanomolar
pan-CDK inhibitor that is currently
being investigated in phase I clinical
trials.
U. Lꢀcking,* R. Jautelat, M. Krꢀger,
T. Brumby, P. Lienau, M. Schꢁfer, H. Briem,
J. Schulze, A. Hillisch, A. Reichel,
A. M. Wengner, G. Siemeister
&& – &&
The Lab Oddity Prevails: Discovery of
Pan-CDK Inhibitor (R)-S-Cyclopropyl-S-
(4-{[4-{[(1R,2R)-2-hydroxy-1-
methylpropyl]oxy}-5-
(trifluoromethyl)pyrimidin-2-
yl]amino}phenyl)sulfoximide (BAY
1000394) for the Treatment of Cancer
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
1
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
DOI: 10.1002/cmdc.201300096
The Lab Oddity Prevails: Discovery of Pan-CDK Inhibitor
(R)-S-Cyclopropyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-
methylpropyl]oxy}-5-(trifluoromethyl)pyrimidin-2-
yl]amino}phenyl)sulfoximide (BAY 1000394) for the
Treatment of Cancer
Ulrich Lꢁcking,*[a] Rolf Jautelat,[b] Martin Krꢁger,[a] Thomas Brumby,[a] Philip Lienau,[c]
Martina Schꢂfer,[d] Hans Briem,[a] Julia Schulze,[e] Alexander Hillisch,[b] Andreas Reichel,[c]
Antje Margret Wengner,[f] and Gerhard Siemeister[f]
Lead optimization of a high-throughput screening hit led to
the rapid identification of aminopyrimidine ZK 304709, a multi-
targeted CDK and VEGF-R inhibitor that displayed a promising
preclinical profile. Nevertheless, ZK 304709 failed in phase I
studies due to dose-limited absorption and high inter-patient
variability, which was attributed to limited aqueous solubility
and off-target activity against carbonic anhydrases. Further
lead optimization efforts to address the off-target activity pro-
file finally resulted in the introduction of a sulfoximine group,
which is still a rather unusual approach in medicinal chemistry.
However, the sulfoximine series of compounds quickly re-
vealed very interesting properties, culminating in the identifi-
cation of the nanomolar pan-CDK inhibitor BAY 1000394,
which is currently being investigated in phase I clinical trials.
Introduction
Loss of cell-cycle control and increased resistance to apoptosis
are major hallmarks of cancer. Cyclin-dependent kinases (CDKs)
belong to a family of serine/threonine kinases which associate
with an activating cyclin regulatory subunit. Cell-cycle CDKs 1,
2, 4, and 6 are required for the correct timing and order of
events of the cell-division cycle, whereas non-cell-cycle CDKs 7
and 9 are involved in gene transcription via regulation of RNA
polymerase II activity. Deregulated CDK activity results in loss
of function of cell-cycle checkpoints and increased expression
of anti-apoptotic proteins, which has been directly linked to
the molecular pathology of cancer. Herein we report the iden-
tification of pan-CDK inhibitors ZK 304709 and BAY 1000394.
Starting from a high-throughput screening (HTS) hit with
moderate biological activity, the process of lead optimization
by a collaborative effort involving medicinal chemistry, phar-
macology, DMPK, structural biology, and computational
chemistry led to the rapid identification of aminopyrimidine
ZK 304709, a multitargeted CDK and vascular endothelial
growth factor receptor (VEGF-R) inhibitor that blocks tumor
cell proliferation and induces apoptosis through inhibition of
cell-cycle progression and tumor-induced angiogenesis. How-
ever, in phase I studies, ZK 304709 revealed dose-limited and
variable exposure in patients, and the study was terminated
before the maximum tolerated dose (MTD) could be deter-
mined.
[a] Dr. U. Lꢀcking, Dr. M. Krꢀger, Dr. T. Brumby, Dr. H. Briem
Bayer Pharma AG, Global Drug Discovery
Medicinal Chemistry, 13353 Berlin (Germany)
E-mail: ulrich.luecking@bayer.com
Thorough analysis of the parameters that could have con-
tributed to this outcome in the clinic led us to a revised target
profile that was strictly addressed, and finally resulted in the
identification of BAY 1000394. Aminopyrimidine BAY 1000394
is a balanced pan-CDK inhibitor with potent, broad-spectrum
antiproliferative activity against human cancer cell lines, and is
currently being investigated in a phase I clinical trial in patients
with advanced solid tumors.
[b] Dr. R. Jautelat, Prof. Dr. A. Hillisch
Bayer Pharma AG, Global Drug Discovery
Medicinal Chemistry, 42096 Wuppertal (Germany)
[c] Dr. P. Lienau, Dr. A. Reichel
Bayer Pharma AG, Global Drug Discovery
DMPK, 13353 Berlin (Germany)
[d] Dr. M. Schꢁfer
Bayer Pharma AG, Global Drug Discovery
Structural Biology, 13353 Berlin (Germany)
Results and Discussion
[e] Dr. J. Schulze
Search for low-molecular-weight CDK inhibitors
Bayer Pharma AG, Global Drug Discovery
Pharmaceutical Development, 13353 Berlin (Germany)
Since their discovery, CDKs have been considered strong pro-
spective targets for a new generation of anticancer drugs.[1]
Despite initial skepticism arising from the conserved active site
[f] Dr. A. M. Wengner, Dr. G. Siemeister
Bayer Pharma AG, Global Drug Discovery
TRG Oncology, 13353 Berlin (Germany)
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
2
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
of ATP binding proteins, numerous pharmaceutical companies
have initiated drug discovery efforts to identify low-molecular-
weight CDK inhibitors for cancer therapy. Several CDK inhibi-
tors with various inhibitory profiles against the panel of
CDKs,[2–4] of which some have additional non-CDK targets,[5,6]
have entered clinical trials;[7] however, the therapeutic effica-
cies of these agents are modest, and none have been ap-
proved for clinical use to date. This may be, at least in part,
due to insufficient potency, unfavorable pharmacokinetic and
physicochemical properties, off-target toxicities, and subopti-
mal dosing schedules.[8]
Scheme 1. Synthesis of 2,4-diaminopyrimidines. Reagents and conditions:
a) R4NH2, NEt3, MeCN, RT; b) R2NH2, 4n HCl in dioxane, MeCN, H2O, reflux.
tional group into the phenyl group at the 2-position of the
pyrimidine, as it was directed toward the exit of the ATP bind-
ing pocket. It was found that substitution of the phenyl group
at the para position with a sulfonamide group leads to a signifi-
cant improvement of CDK inhibitory properties and antiproli-
ferative activity. For compound 2 (Figure 1), IC50 values in the
nanomolar range at CDK2 (IC50 =40 nm) and in an MCF7 cell
proliferation assay (IC50 =450 nm) were recorded (Table 1). Co-
crystallization experiments revealed that the sulfonamide
group appears to form three additional hydrogen bonds with
the enzyme in vitro.[10] Similar results have been reported by
others.[11,12]
HTS of the compound collection of the former Schering
AG[9] (~520000 compounds) led to the identification of 2,4-dia-
minopyrimidine derivative 1 (Figure 1) with potent CDK2
By extensive variation at the 4-position of the pyrimidine, in-
itially relying on commercial building blocks, chiral 1,2-amino
alcohols were soon identified as suitable substituents.[13]
ZK 304709 (Figure 1), featuring a d-alaninol residue, revealed
potent inhibition of CDK2, CDK1, CDK4, CDK7, and CDK9 with
respective IC50 values of 4, 50 (Table 1), 61, 85, and 5 nm. The
compound was also shown to inhibit kinases associated with
Figure 1. Structures of HTS hit 1, para-sulfonamide analogue 2, and clinical
pan-CDK inhibitors ZK 304709 and BAY 1000394.
(IC50 =100 nm) and CDK1 (IC50 =100 nm) inhibitory
properties, and moderate antiproliferative activity in
vitro against the MCF7 cancer cell line (IC50 =
Table 1. In vitro activities of aminopyrimidine derivatives.
4000 nm, Table 1). This compound was selected as
Compound
IC50 [nm][a]
VEGF-R2
a lead structure for further optimization efforts.
CDK2/CycE
CDK1/CycB
MCF7
CA-II
1
5
100
40
4
10
13
9
11
10
8
12
10
10
6
2
16
2
9
3
4
5
9
8
100
19
50
10
14
4
10
15
19
56
115
31
8
3
110
3
11
2
5
ND
70
34
4000
450
266
30
182
48
89
130
51
ND
ND
514
403
529
ND
ND
ND
>10000
<300
Structural variation of the HTS hit leading to
ZK 304709
ZK 304709
6
7
8
9
10
13
14
15
16
17
18
21
29
31
32
59
HTS hit 1 and related derivatives were easily accessi-
ble by straightforward chemistry (Scheme 1). Starting
from the commercial building block 5-bromo-2,4-di-
chloropyrimidine (3), the 4-position was first selec-
tively functionalized under basic conditions to give
the corresponding 2-chloropyrimidine 4, then the
substituent at the 2-position was introduced under
acidic conditions to give the 2,4-diaminopyrimidine
5.
The 3D structure of lead compound 1 in complex
with CDK2 was determined in a co-crystallization ex-
periment[10] and X-ray crystallographic analysis. The
corresponding X-ray structure was used as a model
for further structural variations of aminopyrimidine
lead 1, based on the assumption that the binding
mode observed in the crystal would be similar to
that in cells.
158
122
110
570
76
500
73
280
240
17
70
32
55
79
79
71
163
82
59
129
60
1220
2700
113
130
100
1500
44
110
60
34
26
15
34
100
18
>10000
>10000
>10000
>10000
ND
ND
ND
33
44
5
7
9
8
4
9
>10000
>10000
>10000
>10000
ND
BAY 1000394
55
56
57
58
7
7
8
91
477
35
>10000
The X-ray structure of 1 in complex with CDK2 sug-
gested, for instance, the introduction of a polar func-
[a] ND: not determined.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
3
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
angiogenesis such as VEGF-R1, VEGF-R2, VEGF-R3, and PDGF-R,
with IC50 values of 10, 34, 1, and 27 nm, respectively; however,
kinases such as the EGF-R and insulin receptor were not inhib-
ited at pharmacologically relevant concentrations.[6]
the parent compound after incubation for 60 min of 98, 94,
and 89%, respectively. Inhibition of cytochrome P450 isoen-
zyme (CYP) activity by ZK 304709 was evaluated using re-
combinant human enzymes. ZK 304709 did not show signifi-
cant inhibition, with IC50 >10 mm for all tested CYPs. In the
Caco2 model, a permeability coefficient (Papp, ap-bas) of
56 nmsꢀ1 was determined, suggesting moderate intestinal per-
meability for the compound.
In MCF7 cell proliferation assays with ZK 304709, an IC50
value of 266 nm was recorded (Table 1). The compound was
further tested for its antiproliferative potential against a panel
of 25 human tumor cell lines. A mean concentration of 317 nm
was found to be sufficient to block the growth of human
tumor cell lines by 50% under standard assay conditions in vi-
tro. Fluorescence-activated cell sorting (FACS) analysis of MCF7
breast carcinoma cells showed that ZK 304709 can induce G1/S
arrest followed by apoptosis in a dose-dependent manner.[6]
Co-crystallization experiments of ZK 304709 in complex with
CDK2 (Figure 2) revealed a binding mode similar to that of
In an in vivo pharmacokinetic study in rats, ZK 304709
showed low blood clearance (0.24 Lhꢀ1 kgꢀ1), high volume of
distribution (4.7 Lkgꢀ1), and long half-life (3.3 h). After oral ad-
ministration, a moderate to high bioavailability of 68% was ob-
served; however, ZK 304709 showed a high blood/plasma ratio
of 6 in human whole blood, and up to 14 in rats, indicating ac-
cumulation in the cellular blood fraction. It was speculated
that binding to carbonic anhydrases (CAs) is the reason for this
behavior. This hypothesis was further substantiated by reports
on the erythrocyte affinity of N-unsubstituted sulfonamides.[15]
Indeed, we found that the IC50 value for inhibition of CA-II by
ZK 304709 is 514 nm (Table 1).
The in vivo efficacy of ZK 304709 was initially evaluated with
the HeLa–MaTu human cervical tumor xenograft model in
athymic mice. ZK 304709 was shown to induce regression of
even large tumor masses at oral doses, which did not induce
any significant body weight loss. Macroscopically, large, well-
vascularized tumor masses lost their vascularization within
a time frame of 4 days after initiation of treatment and started
to regress. Further in vivo antitumor activity was demonstrated
in a broad panel of human tumor xenograft models in athymic
mice. In these models, ZK 304709 activity was frequently supe-
rior to that of standard cytotoxic chemotherapies.[16,17] Thus,
ZK 304709 was selected for phase I clinical studies.
Phase I results for ZK 304709 led to revised project aims
A phase I dose-escalation study with ZK 304709 was conduct-
ed, starting at oral doses of 15 mgdayꢀ1 up to 360 mgdayꢀ1. It
was found that Cmax and AUC, measured in whole blood, in-
creased with dose proportionally up to a dose of 90 mgdayꢀ1.
At doses exceeding 90 mgdayꢀ1, a less-than-dose-proportional
increase in exposure was observed, which was accompanied
by high inter-patient variability. The reason behind this finding
was identified to be dose-limited absorption at oral doses
>90 mgdayꢀ1, and was attributed to the limited solubility of
the compound. Therefore, because the obtained exposure was
unlikely to result in meaningful pharmacological or clinical ac-
tivity, and exposure could not be further increased, the trial
was stopped before reaching the MTD, despite disease stabili-
zation having been observed in 10 patients for up to 12
cycles.[16,17] Another source of the observed variability was as-
cribed to the binding of the compound to CAs.
Figure 2. X-ray structure of ZK 304709 in complex with CDK2: the sulfona-
mide moiety at the para position forms two hydrogen bonds to the main
and side chain of Asp86, one water-mediated hydrogen bond to the carbon-
yl of Ile10, and one hydrogen bond to Lys89.
compound 2. With respect to physicochemical properties,
ZK 304709 revealed a logD (pH 7.5) value of 0.3 and a high
melting point of 2628C. The thermodynamic solubility in water
of 8 mgLꢀ1 at pH 7.4 was determined by an equilibrium shake-
flask method.[14] Consistent with a pKa value of 3.8 for the pro-
tonation of a pyrimidine nitrogen atom, the solubility was
shown to increase with decreasing pH.
The revised project aims therefore focused on the following
key aspects in the follow-up program: First, the limited absorp-
tion at high dose was to be addressed by decreasing dose size
by a significant improvement in antitumor potency, as well as
increased aqueous solubility; second, the follow-up compound
was to be devoid of CA-inhibitory properties.
In vitro pharmacokinetic studies with ZK 304709 revealed
high metabolic stability in liver microsome preparations of
human, mouse, and rat origin, resulting in recovery rates of
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
4
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Further lead optimization efforts
rat origin were recorded with recovery rates of the parent
compound after incubation for 60 min of 100, 100, and 82%,
respectively; however, in vivo, compound 10 failed to show su-
perior results in the HeLa–MaTu human cervical tumor xeno-
graft model (treatment versus control ratio (T/C)=18% at
25 mgkgꢀ1).
Extensive variation of the d-alaninol side chain at the 4-posi-
tion of the pyrimidine led to the finding that the correspond-
ing sec- and tert-alcohol derivatives, such as aminopyrimidines
6 and 7 (Figure 3), revealed significantly improved in vitro ac-
Thiophene derivative 9 revealed even further improved ac-
tivity in vitro (Table 1). With respect to metabolic stability, the
compound was largely stable in human and rat liver micro-
somes, with recoveries of 82 and 80%, respectively. In mouse
liver microsomes, however, compound 9 was rather unstable
showing recovery of only 12% after incubation for 60 min.
Therefore, no xenograft model in nude mice was performed
with compound 9.
Elimination of off-target activity against CAs
Figure 3. Aminopyrimidines with a sec- or tert-alcohol at the 4-position.
The goal of eliminating undesired inhibition of CAs was ad-
dressed by rational design. Compounds that contain an N-un-
substituted arenesulfonamide group are known inhibitors of
human CAs (especially CA-II) and have been used, for instance,
as diuretics for the treatment of glaucoma. The deprotonated
nitrogen atom and one of the oxygen atoms of the sulfona-
mide group coordinate to the zinc cation. In addition, the sul-
fonamide nitrogen forms a hydrogen bond to the hydroxy
group of Thr199 in the active center of CA-II. N-Unsubstituted
arenesulfonamides thus block the enzymatic function of CA-
II.[21] The clinical use of CDK inhibitors that contain an N-unsub-
stituted arenesulfonamide group may therefore be restricted
by the possibility of off-target inhibition of CAs and a resultant
spectrum of side effects.
tivities. The corresponding 4-oxa derivatives, such as com-
pound 8, demonstrated similar in vitro profiles.
Compound 6 has an IC50 value of 10 nm at CDK2 and CDK1
(Table 1) and was shown to be ninefold more active than
ZK 304709 against a panel of human and murine cancer cell
lines (mean IC50 =42 nm). Indeed, compound
6 revealed
potent antitumor efficacy at a therapeutic dose of 8 mgkgꢀ1
p.o. once daily (12.5-fold lower than ZK 304709) in various
tumor models of human and murine origin. Antitumor efficacy
of compound 6 at the therapeutic dose was almost indistin-
guishable in the tumor models evaluated from the antitumor
efficacy of ZK 304709 at 100 mgkgꢀ1; however, due to the N-
unsubstituted aromatic sulfonamide moiety, compound 6 in-
hibits CA-II (IC50 =403 nm) and has a blood/plasma ratio of 13
in humans. Additionally, compound 6 shows lower water solu-
bility than ZK 304709.
The binding mode of ZK 304709 to human CA-II was investi-
gated by docking experiments (Figure 5). The modeling studies
supported the hypothesis that binding to CA-II is mediated by
the sulfonamide group, and that ZK 304709 shows a good fit
with the CA-II binding pocket. Based on these studies, three
different strategies to eliminate the undesired off-target activi-
ty of the sulfonamide-substituted CDK inhibitors were pro-
posed.
When it became apparent that aromatic sulfonamide deriva-
tives such as ZK 304709 and compound 6 revealed interesting
antitumor activities, analogous heteroaromatic sulfonamides
became of interest. Relying on the established synthetic route
(Scheme 1) and published syntheses of the required ani-
lines,[18,19] a series of thiophene and pyridine derivatives, such
as compounds 9 and 10 (Figure 4), were prepared.[20]
Compound 10 revealed strong CDK inhibitory properties
and antiproliferative activity in vitro (Table 1). High metabolic
stability in liver microsome preparations of human, mouse, and
The first proposal was to reevaluate N-substituted sulfona-
mide derivatives, because the introduction of substituents at
the sulfonamide nitrogen atom should lead to a steric clash
with CA-II. A set of suitable compounds was already available,
as numerous substituted sulfonamide derivatives had already
been prepared over the course of the project. The reaction of
sulfonyl fluorides 11 with the desired amines had been relied
on for the final step of the synthesis, which was also suitable
for parallel synthesis approaches (Scheme 2).[13] However, up to
that point, no added benefit of the N-substituted derivatives
relative to the N-unsubstituted inhibitors had been identified.
Indeed, it was shown that the introduction of a simple
methyl group at the sulfonamide nitrogen was sufficient to
block the off-target activity at human CAs while conserving
the potent CDK and VEGF-R inhibitory properties. For instance,
compound 13 (Figure 6) revealed CDK2, CDK1, and VEGF-R2 in-
hibitory properties similar to its N-unsubstituted analogue 6
(Table 1). Moreover, the blood/plasma ratio of compound 13
Figure 4. Aminopyrimidines with a heteroaromatic sulfonamide at the 2-po-
sition.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
5
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Figure 6. N-Methyl-substituted sulfonamide 13 and ortho-substituted sulfo-
namides 14–18.
ties. Subsequent experiments in mice showed that even 2 h
after p.o. dosing of compound 13, the ratio of compound 13
to its metabolite 6 was 1:13. Similar effects were recorded in
rats, although the process of methyl cleavage was slower in
this species. Due to this pronounced metabolic cleavage of the
methyl-substituted sulfonamide in vivo, which was not reflect-
ed to this degree by in vitro metabolic assays using hepato-
cytes or liver microsomes, it was decided to discontinue work
on the N-substituted sulfonamide series.
A second strategy to eliminate the undesired off-target ac-
tivity at CAs was to introduce substituents at the ortho posi-
tion of the arenesulfonamide group. These substituents would
lead to repulsive interactions with the lower part of the tight,
funnel-shaped active site of CA-II (Figure 5). Relying on the es-
tablished synthetic route (Scheme 1), a series of ortho-substi-
tuted inhibitors 14–18 was prepared (Figure 6).[13] With respect
to in vitro activities at CDK2 and CDK1, the additional substitu-
ents at the ortho position were well tolerated. Compound 17,
for instance, revealed very potent inhibition of CDK2 and
CDK1, with IC50 values of 6 and 8 nm, respectively. Antiprolifer-
ative activities against MCF7 cells in vitro were in a range simi-
lar to the unsubstituted analogue 7 (Table 1). The in vitro CA-II
assay revealed a clear SAR with respect to the nature of the
substituent at the ortho position. Hydroxy and methyl groups
are not sufficient to completely block off-target activity at CA-
II. The derivatives 16 and 17 with chloride and methoxy
groups, however, showed respective IC50 values of 2700 and
>10000 nm. It was also shown that the introduction of
a second methyl group at the neighboring ortho position is an
additional means to eliminate CA-II inhibition (see compound
18, Table 1).[22] In the HeLa–MaTu human cervical tumor xeno-
graft model, however, compound 17 revealed a T/C value of
only 13% at a high daily oral dose of 75 mgkgꢀ1, and thus this
second strategy was also abandoned.
Figure 5. Modeling study of ZK 304709 in complex with human CA-II.
Scheme 2. Synthesis of N-substituted sulfonamides. Reagents and conditions:
a) 4-aminobenzenesulfonyl fluoride, 4n HCl in dioxane, MeCN, H2O, reflux;
b) R1R2NH, DMAP, butan-2-ol, 808C.
was determined to be approximately 0.5 in mice and rats,
which was in line with our hypothesis that the off-target activi-
ty at CAs is responsible for the enrichment of CA-inhibiting sul-
fonamide derivatives such as ZK 304709 in erythrocytes. With
respect to metabolic stability, compound 13 was shown to be
stable in human and rat liver microsomes with recovery rates
after incubation for 60 min of 88 and 87%, respectively. In
mouse liver microsomes, a moderate recovery of 51% was ob-
served. The compound revealed potent antitumor efficacy at
a daily oral dose of 5 mgkgꢀ1 in the HeLa–MaTu human cervi-
cal tumor xenograft model. Antitumor efficacy of compound
13 at the therapeutic dose was almost indistinguishable from
the antitumor efficacy of compound 6 at 5 mgkgꢀ1 p.o. once
daily; however, analysis of blood samples from the xenograft
experiments revealed that methyl-substituted derivative 13
had been rapidly metabolized in vivo to give the correspond-
ing unsubstituted sulfonamide 6 with its CA inhibitory proper-
A third concept to eliminate undesired CA affinity was to ex-
change the sulfonamide group for a different functional group
without CA inhibitory properties, while conserving inhibitory
activity against the CDKs. One rather unorthodox proposal was
to use a sulfoximine instead of a sulfonamide group.
The sulfoximine group was not described until the end of
the 1940s, when methionine sulfoximine was identified as the
toxic agent in agenized wheat flour.[23] Since then, tetrahedrally
hybridized sulfoximines have been described as constitutional-
ly and configurationally stable compounds that can be manip-
ulated without special care. Due to the amphoteric character
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
6
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
of the sulfoximine nitrogen atom, the acidic protons at the a-
position, and the potentially stereogenic sulfur atom, com-
pounds bearing this functional group are characterized by
a rich and versatile chemistry.[24] Nevertheless, with very few
exceptions, the use of sulfoximines in medicinal chemistry con-
texts has been quite limited; to the best of our knowledge, no
drug containing this functional group is on the market.[25–28]
Accordingly, the proposal to introduce a sulfoximine group to
the CDK inhibitors was initially met with much skepticism.
However, the proposed synthetic route for the very first model
compound, relying on the traditional synthesis of sulfoximines
using sodium azide and sulfuric acid,[29] led to model com-
pound 21. Thus, oxidation of sulfide 19, which was prepared
by the established synthetic route (Scheme 1), with 3-chloro-
peroxybenzoic acid (mCPBA) gave racemic sulfoxide 20
(Scheme 3). Finally, imination of sulfoxide 20 with sodium
azide in sulfuric acid and dichloromethane at 458C gave the
desired sulfoximine 21.
Scheme 4. Modular synthesis approach to sulfoximine derivatives. Reagents
and conditions: a) H5IO6, FeCl3, MeCN, RT, 50 min; b) NaN3, H2SO4, CHCl3,
458C, overnight; c) ClC(O)OEt, pyridine, RT, 4 h; d) TiCl3, THF, RT, 4 h, crude;
e) (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)amino]butan-2-ol, 4n HCl in di-
oxane, MeCN, H2O, 608C, 24 h; f) NaOEt, EtOH, reflux, 5 h; g) preparative
chiral HPLC.
mine nitrogen results in a sluggish reaction with low yields.
Cleavage of the protecting group using sodium ethanolate at
reflux gave sulfoximine 28 as a mixture of two stereoisomers.
Finally, separation of the diastereomers was achieved by prepa-
rative chiral HPLC to give the single stereoisomers 29 and
30.[30]
Both stereoisomers revealed similar in vitro activities, with
diastereomer 29 being slightly more active. Compound 29 was
shown to be a very potent inhibitor of CDK2 and CDK1 in vitro,
with IC50 values of 2 and 3 nm, respectively (Table 1). Addition-
ally, very potent antiproliferative activity against MCF7 cells
(IC50 =44 nm) was recorded. The thermodynamic solubility in
water, as determined by an equilibrium shake-flask method,
was 261 mgLꢀ1. This result was considered to be very promis-
ing, taking into account that many kinase inhibitors suffer
from low solubility, especially in water, which often leads to
various problems in the clinic.[31,32] Additionally, compound 29
was shown to be hydrolytically stable (pH 2 to pH 7.4). In vitro
pharmacokinetic studies with sulfoximine 29 demonstrated
high metabolic stability in human, mouse, and rat liver micro-
some preparations with recovery rates after incubation for
60 min of 90, 84, and 100%, respectively. In vivo rat pharmaco-
kinetic studies revealed low blood clearance (0.65 Lhꢀ1 kgꢀ1),
a high volume of distribution (2.3 Lkgꢀ1), and a long half-life
(4.0 h). The blood/plasma ratio was determined to be approxi-
mately 1 across species, which was in line with our hypothesis
that the off-target activity at CAs is responsible for the enrich-
ment of CA-inhibiting sulfonamide derivatives such as
ZK 304709 in erythrocytes. Additionally, sulfoximine 29 re-
vealed potent antitumor efficacy (T/C=6%) at a therapeutic
dose of 10 mgkgꢀ1 p.o. once daily (10-fold less than
ZK 304709) in the HeLa–MaTu human cervical tumor xenograft
model in athymic mice.
Scheme 3. Synthesis of model sulfoximine compound 21. Reagents and con-
ditions: a) mCPBA, CH2Cl2, RT, 90 min; b) NaN3, H2SO4, CH2Cl2, 458C, 16 h.
Racemate 21, indeed, revealed significant CDK2, CDK1, and
VEGF-R2 inhibitory properties, with IC50 values of 16, 110, and
70 nm, respectively (Table 1). Additionally, moderate antiproli-
ferative activity against MCF7 cells in vitro (IC50 =1500 nm) was
recorded. High water solubility (>120 mgLꢀ1 at pH 7.4) was
determined by turbidometric analysis. In addition, the com-
pound is hydrolytically stable (pH 2 to pH 7.4) and demonstrat-
ed high metabolic stability in vitro. Moreover, sulfoximine 21
revealed no CA-II inhibitory properties in vitro.
With these promising results in hand, it was decided to
modify the linear synthesis to gain access to more complex
test compounds. In the revised synthesis (Scheme 4), oxidation
of sulfide 22 to sulfoxide 23 was achieved by using periodic
acid and catalytic iron trichloride, which, in our hands, provid-
ed less over-oxidation to the corresponding sulfone than the
use of 3-chloroperoxybenzoic acid. Sulfoxide 23 was then al-
lowed to react with sodium azide in sulfuric acid and CHCl3 to
give sulfoximine 24. The sulfoximine nitrogen atom was pro-
tected (intermediate 25), and the nitro group was subsequent-
ly reduced with titanium trichloride to give aniline building
block 26. Coupling of aniline 26 with (2R,3R)-3-[(5-bromo-2-
chloropyrimidin-4-yl)amino]butan-2-ol, which was prepared ac-
cording to Scheme 1, was carried out under acidic conditions.
Earlier studies had revealed that the corresponding coupling
reaction without a protecting group at the amphoteric sulfoxi-
Accordingly, the sulfoximine series had developed from
a “lab oddity” to a very interesting class of pan-CDK inhibitors.
To gain further insight into this promising series and to investi-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
7
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
gate the effects on potency and permeability, it was decided
to initiate a limited fluorine scan,[33] and compounds 31, 32,
and 33 were designed (Figure 7).
tions of our established synthetic route led to surprises
(Scheme 6). In contrast to 5-bromo-2,4-dichloropyrimidine (3)
(Scheme 1), the reaction of 2,4-dichloro-5-(trifluoromethyl)pyri-
midine (36) with the required amino alcohol under basic con-
ditions led to a mixture of regioisomers 37 and 38 in a ratio of
approximately 1:1. The desired regioisomer 37 could be sepa-
rated by chromatography. The subsequent coupling reaction
with aniline 26 under acidic conditions gave compound 39 in
good yield. Surprisingly, although the reaction of the protected
sulfoximine 39 with sodium ethanolate in ethanol at 608C re-
sulted in the desired deprotection of the sulfoximine group,
this was accompanied by the transformation of the 5-(trifluoro-
methyl) group to the corresponding ortho-ethoxy ester to give
compound 40. Synthetic alternatives were therefore needed.
In a modified approach to the synthesis of 5-(trifluorome-
thyl)pyrimidine derivatives, a new, rhodium-catalyzed method
for the synthesis of sulfoximines, which had just been reported
by Okamura and Bolm,[36] was used. With this method, sulfox-
ides can be converted into the corresponding sulfoximines by
using rhodium(II) acetate dimer as a catalyst and 2,2,2-trifluor-
oacetamide in combination with iodobenzene diacetate and
MgO. In contrast to older, metal-catalyzed methods for the imi-
nation of sulfoxides, the desired (“free”) NH-sulfoximines can
easily be obtained by cleavage of the nitrogen–acyl bond of
the resulting N-(trifluoroacetyl)-protected derivatives. In addi-
tion, the imination reaction is stereospecific and proceeds with
retention of configuration at the stereogenic sulfur center.
Consequently, enantiomerically pure sulfoximines are accessi-
ble by starting from enantiopure sulfoxides.
Figure 7. Fluorine scan compounds.
The synthesis of S-(trifluoromethyl)sulfoximine 31 initially
proved difficult. The reaction of the trifluoromethyl sulfoxide
building block 34 with sodium azide in sulfuric acid under
standard conditions did not lead to the desired product 35. In-
terestingly, as described by Kondratenko et al.,[34] replacement
of sulfuric acid by oleum gave the required S-(trifluoromethyl)-
sulfoximine 35 in a clean and fast reaction[35] (Scheme 5).
Reaction of racemic sulfoxide 23 under the described condi-
tions gave the desired N-(trifluoroacetyl)-protected derivative
41 (Scheme 7). Because a protecting group for the amphoteric
sulfoximine nitrogen was needed for the coupling reaction
with (2R,3R)-3-{[2-chloro-5-(trifluoromethyl)pyrimidin-4-yl]ami-
no}butan-2-ol under acidic conditions, the N-(trifluoroacetyl)
group was not immediately deprotected. Reduction of 41 with
titanium trichloride gave the corresponding aniline in good
yield, and the coupling reaction with (2R,3R)-3-{[2-chloro-5-(tri-
fluoromethyl)pyrimidin-4-yl]amino}butan-2-ol under acidic con-
Scheme 5. Synthesis of S-(trifluoromethyl)sulfoximine 35. Reagents and con-
ditions: a) NaN3, oleum, 708C, 3 h.
Compound 31 was shown to be a potent inhibitor of CDK2
and CDK1 in vitro, with IC50 values of 9 and 11 nm, respectively;
however, the compound did not reveal any improvement over
analogue 29 with respect to antiproliferative activity against
MCF7 cells (IC50 =110 nm). Ortho-substituted fluorine derivative
32 was shown to be a very
potent inhibitor of CDK2 and
CDK1 in vitro, with IC50 values of
3 and 2 nm, respectively. Addi-
tionally, very potent antiprolifer-
ative activity against MCF7 cells
(IC50 =60 nm) was recorded;
however, in the HeLa–MaTu
human cervical tumor xenograft
model, compound 32 revealed
tolerability issues and a rather
weak efficacy at a tolerated dose
of 25 mgkgꢀ1 b.i.d. 2 days on/5
days off (T/C=31%).
The first attempt to synthesize
the desired 5-(trifluoromethyl)-
Scheme 6. Attempted synthesis of a 5-(trifluoromethyl) derivative. Reagents and conditions: a) (2R,3R)-3-aminobu-
tan-2-ol, NEt3, MeCN, RT, overnight; b) 26, 4n HCl in dioxane, MeCN, H2O, 608C, 6 h; c) NaOEt, EtOH, 608C, 24 h.
pyrimidine derivative 33 by
simply transferring the condi-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
8
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
quently, the 5-(trifluoromethyl)
substituent
was
introduced
using (trifluoromethyl)trimethyl-
silane (Ruppert’s reagent).[38] The
coupling reaction of 2-chloropyr-
imidine 50 with aniline 47 fur-
nished the fully protected prod-
uct 51. Cleavage of the benzyl
group was achieved by hydroge-
nation with palladium on char-
coal to provide 52. Finally, de-
protection of the sulfoximine
using potassium carbonate in
methanol at room temperature
gave the unprotected sulfoxi-
mine 53, which was separated
Scheme 7. Modified synthesis of the 5-(trifluoromethyl) derivative. Reagents and conditions: a) CF3C(O)NH2, PhI-
(OAc)2, MgO, [Rh2(OAc)4], CH2Cl2, RT, overnight; b) TiCl3, THF, RT, 3.5 h, crude; c) 37, 4n HCl in dioxane, MeCN,
608C, 4.5 h; d) K2CO3, MeOH, RT, 1 h; e) preparative chiral HPLC.
ditions did furnish the desired product 42 in good yield. Final-
ly, deprotection using potassium carbonate in methanol at
room temperature left the trifluoromethyl group at the 5-posi-
tion untouched and gave the free sulfoximine 33 as a mixture
of two diastereomers that was separated by preparative chiral
HPLC into the single stereoisomers 43 and 44.
The synthesis of the related 4-oxa derivatives had to be
slightly modified yet again. For example, the synthesis of
BAY 1000394 is shown in Scheme 8. Thus, reaction of 1-fluoro-
4-nitrobenzene (45) with cyclopropanethiol[37] generated thio-
by preparative chiral HPLC into the single stereoisomers 54
and BAY 1000394.
The Bolm research group’s new rhodium-catalyzed proce-
dure for the synthesis of sulfoximines not only provided access
to the new class of 5-(trifluoromethyl)pyrimidine inhibitors, but
also represented a real improvement, as the previously used
and potentially explosive combination of sodium azide and
sulfuric acid or even oleum could be replaced by a safe, relia-
ble, and easy-to-use method employing commercially available
reagents. This new approach also improved the potential for
large-scale production of a sul-
foximine candidate.
The new 4-aza- and 4-oxa-5-
(trifluoromethyl)pyrimidine deriv-
atives 44 and BAY 1000394 re-
vealed very potent inhibition of
CDK2 and CDK1 in vitro, and
very potent antiproliferative ac-
tivities against MCF7 cells were
recorded (Table 1). The com-
pounds were stable with respect
to hydrolysis and metabolism in
vitro. Moreover, the new com-
pounds displayed high thermo-
dynamic solubilities in water, as
determined by an equilibrium
shake-flask
method
(e.g.,
BAY 1000394: 182 mgLꢀ1). Simi-
lar results were recorded for
other compounds from the new
trifluoromethyl series (e.g., com-
pound 55, Figure 8 and Table 1).
Thus, the 5-(trifluoromethyl)-
Scheme 8. Synthesis of BAY 1000394. Reagents and conditions: a) cyclopropanethiol, NaH, THF, Et2O, 408C, 2 h;
b) H5IO6, FeCl3, MeCN, RT, 30 min; c) CF3C(O)NH2, PhI(OAc)2, MgO, [Rh2(OAc)4], CH2Cl2, RT, overnight; d) H2, Pd/C,
EtOH, THF, RT, 6.25 h; e) (2R,3R)-3-(benzyloxy)butan-2-ol, NaH, Et2O, 0!408C, 4 h; f) CF3Si(CH3)3, KF, CuI, NMP, THF,
808C, 5.5 h; g) 47, 4n HCl in dioxane, MeCN, 808C, 5 h; h) H2, Pd/C, EtOH, RT; i) K2CO3, MeOH, RT, 1.5 h; j) prepara-
tive chiral HPLC.
ether 46. Oxidation of 46 with periodic acid followed by the
rhodium-catalyzed imination gave the corresponding protect-
ed sulfoximine. Finally, hydrogenation of the nitro group em-
ploying palladium on charcoal resulted in the desired aniline
47. In parallel, commercially available 2,4-dichloro-5-iodopyri-
midine (48) was allowed to react with (2R,3R)-3-(benzyloxy)bu-
tan-2-ol under basic conditions to give the 2-chloropyrimidine
49 with very high regioselectivity for the 4-position. Subse-
pyrimidine series offered very similar in vitro activities to those
of compounds from the 5-bromopyrimidine series. Surprisingly,
marked differences between the 5-bromo series and the 5-(tri-
fluoromethyl) series were found in the subsequent in vivo effi-
cacy studies in the HeLa–MaTu xenograft model. Using the
more in vitro active single stereoisomers, the 5-(trifluorometh-
yl) derivatives 44, 55, and BAY 1000394 demonstrated higher
potencies and better antitumor efficacies than their 5-bromo
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
9
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
analogues 29, 56, and 57 in head-to-head studies at
the corresponding MTDs (Figure 9).[39]
Comparison of sulfoximines and sulfones
Very interesting results were also derived from a com-
parison of the 5-(trifluoromethyl)sulfoximine series
and the corresponding sulfone derivatives.[40] The ex-
change of the sulfoximine group for the correspond-
ing sulfone group resulted in very potent pan-CDK
inhibitors that were devoid of CA inhibitory proper-
ties. One of the most potent compounds of this
Figure 8. 5-(Trifluoromethyl)pyrimidines 44, 55 and BAY 1000394, and 5-bromo ana-
logues 29, 56 and 57.
Figure 9. Antitumor activities of 5-(trifluoromethyl)pyrimidines and 5-bromo analogues. Athymic mice with established HeLa–MaTu human cervical tumors
were treated p.o. with vehicle, with 5-bromopyrimidine 29 [a),b)], 56 [e),f)] or 57 [i),j)], or with 5-(trifluoromethyl)pyrimidine 44 [c),d)], 55 [g),h)] or
BAY 1000394 [k),l)], at the doses and schedules as indicated. Treatment was started at day 3 [a)–h)] or day 5 [i)–l)] after tumor inoculation. Tumor area (TA)
was plotted against time [a), c), e), g), i), k)]. Terminal tumor weights (TW) are depicted for the individual dosing groups [b), d), f), h), j), l)], and treatment
versus control ratios (T/C) were calculated. Error bars indicate standard deviation.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&10
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
series, sulfone 58 (Figure 10), for
human, mouse, and rat origin, resulting in recovery rates of
the parent compound after incubation for 60 min of 94, 100,
and 100%, respectively. BAY 1000394 did not show significant
inhibition of cytochrome P450 activity, with IC50 values
>20 mm for all tested CYPs. In the Caco2 model, a permeability
coefficient (Papp, ap-bas) of 79 nmsꢀ1 was determined, suggest-
ing good intestinal permeability of the compound. The plasma
protein binding of BAY 1000394 was investigated in vitro and
found to be between 93 and 95% in plasma from mouse, rat,
dog, and human.
example, revealed very potent in-
hibition of CDK2 and CDK1 in vitro
and very potent antiproliferative
activity
against
MCF7
cells
(Table 1). Against a panel of various
human and murine tumor cell
lines, sulfone 58 was shown to po-
tently inhibit proliferation with
a very balanced profile (mean IC50
on human tumor cells: 44 nm);
however, the thermodynamic solu-
bility of compound 58 in water is
Figure 10. Sulfone derivative
58.
In vivo pharmacokinetic investigations revealed that
BAY 1000394 has low blood clearance rates in mouse, rat, and
dog (0.51, 0.78, and 0.50 Lhꢀ1 kgꢀ1, respectively). The volume
of distribution is high across species (1.8–2.5 Lkgꢀ1), and its
half-life is intermediate (2.1 h, rat) to long (3.7 h, mouse; 4.3 h,
dog) across species. After oral administration, bioavailability is
moderate across species (48–56%).[41] BAY 1000394 shows
potent antitumor efficacy in human tumor models xenografted
onto nude mice at a therapeutic dose of 2 mgkgꢀ1 p.o. once
daily (50-fold lower than ZK 304709). In contrast to ZK 304709,
BAY 1000394 is efficacious upon intermittent dosing in human
tumor models xenografted on mice (2 days on/5 days off) at
2.5 mgkgꢀ1 b.i.d. or 5 mgkgꢀ1 once daily.
significantly lower than that of BAY 1000394 (58: 25 mgLꢀ1
versus BAY 1000394: 182 mgLꢀ1).
High metabolic stability in liver microsome preparations of
human, mouse, and rat origin were recorded for sulfone 58
with recovery rates of the parent compound after incubation
for 60 min of 93, 91, and 96%, respectively. In the daily dosing
scheme, the in vivo efficacy of sulfone 58 using the HeLa–
MaTu human cervical tumor xenograft model in athymic mice
was almost identical to that of BAY 1000394, albeit at a higher
dose (58: T/C=2% at 5 mgkgꢀ1 versus BAY 1000394: 3% at
2 mgkgꢀ1); however, in the intermittent dosing scheme, sul-
fone 58 was less efficacious than BAY 1000394 (58: T/C=25%
at 5 mgkgꢀ1 b.i.d. 2 days on/5 days off versus BAY 1000394:
2% at 2.5 mgkgꢀ1 b.i.d. 2 days on/5 days off).[41] Additionally,
sulfone 58 showed inferior results with respect to toxicity and
safety pharmacology, and therefore the sulfoximine series was
prioritized.
The favorable CYP inhibition profile and promising results
from antitumor combination studies with paclitaxel, docetaxel,
cetuximab, etoposide, and platinum derivatives in nude
mice[41] have demonstrated the potential of BAY 1000394 for
combination therapies in the clinic. BAY 1000394 was therefore
selected as a development candidate and is currently being in-
vestigated in a phase I clinical trial in patients with advanced
solid tumors (ClinicalTrials.gov identifier: NCT01188252).
Selection of the new clinical candidate BAY 1000394
Conclusions
Additional studies with BAY 1000394 revealed a very interest-
ing overall profile for this 4-oxa-5-(trifluoromethyl) sulfoximine
derivative.[41] The compound is a nanomolar pan-CDK inhibitor
(IC50 values: CDK1/CycB, 7 nm; CDK2/CycE, 9 nm; CDK4/CycD1,
11 nm; CDK9/CycT1, 5 nm), but shows no off-target inhibition
of CAs (CA-II, IC50 >10000 nm). BAY 1000394 potently inhibits
the proliferation of various human and murine tumor cell lines
with a very balanced profile (mean IC50 on human tumor cells:
16 nm, 19-fold more active than ZK 304709). Consistent with
its pan-CDK inhibition activity, BAY 1000394 is able to arrest
the cell cycle at multiple checkpoints (G1, S, G2, M) and to de-
crease the viability of primary B-cell chronic lymphocytic leuke-
mia (B-CLL) cells.
Starting from HTS hit 1, lead optimization efforts led to the
identification of aminopyrimidine ZK 304709, an oral multitar-
geted CDK inhibitor containing a sulfonamide group. The com-
pound showed a promising activity profile in various human
tumor xenografts, however, only at relatively high daily doses
of up to 100 mgkgꢀ1. In phase I clinical trials, blood concentra-
tions of ZK 304709 increased in a sub-proportional manner
with doses exceeding 90 mgdayꢀ1, with high inter-patient vari-
ability. Concentrations expected to deliver meaningful pharma-
cological activity were not reached, and the trials were termi-
nated before the MTD was determined. The most likely reasons
for the insufficient human pharmacokinetics of ZK 304709 are
its low solubility and the high doses required in administra-
tion.
This compound has a melting point of 1348C and high ther-
modynamic solubility in water (182 mgLꢀ1; 22-fold more than
ZK 304709), as determined by an equilibrium shake-flask
method. Solubility even improves significantly at very low pH
(0.1m HCl), as well as in organic solvents such as ethanol and
acetonitrile, and is best in solvents like acetone and PEG 400.
BAY 1000394 is generally stable with regard to hydrolysis in
water and in pH 7 buffer. The absolute configuration of the
compound was determined by X-ray crystallography.
The lessons from the clinic were addressed successfully in
a follow-up project strictly focusing on improved physicochem-
ical properties, pharmacological potencies, and pharmacokinet-
ic properties, while eliminating off-target activity against car-
bonic anhydrases. The corresponding lead optimization efforts
culminated in the identification of nanomolar pan-CDK inhibi-
tor BAY 1000394, featuring a sulfoximine group. BAY 1000394
overcomes the limitations of ZK 304709 and is characterized
by high solubility in water, even at neutral pH, and low effica-
In vitro pharmacokinetic studies with BAY 1000394 revealed
high metabolic stability in liver microsome preparations of
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&
11
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
cious oral doses in the range of 1.5–2.5 mgkgꢀ1 in mice.
BAY 1000394 is efficacious in human tumor xenograft models,
including patient-derived models, which are considered to be
highly predictive of human tumor response. The compound is
active in mice and rats upon once-daily dosing, as well as
upon intermittent dosing, providing valuable options to opti-
mize human dosing schedules with respect to efficacy and tol-
erability. Furthermore, BAY 1000394 has proven antitumor ac-
tivity in xenograft models resistant to standard-of-care drugs
such as doxorubicin, cisplatin, or paclitaxel,[41] and has demon-
strated its potential for combination treatment with other anti-
tumor drugs on the market. BAY 1000394 is currently being in-
vestigated in phase I clinical trials.
33P]ATP (0.2 mCi per well), poly(Glu4Tyr)n (24 ngmLꢀ1; Sigma–Aldrich,
Germany), DMSO (1.25%, v/v)]. The reaction was stopped by the
addition of a solution of EDTA in H2O (0.25m; 15 mL, pH 7.5). A
quantity (15 mL) of each reaction mixture was transferred to P30
filter sheets (Wallac, Germany), which were washed with phospho-
rous acid (0.5%; 3ꢃ10 min) and subjected to scintillation counting
with MeltiLex A scintillator sheets (Wallac, Germany) and a Micobeta
counter (Wallac, Germany). Half-maximal inhibition (IC50) of VEGF-
R2 kinase activity was determined as the inhibitor concentration
resulting in 50% substrate phosphorylation between maximal sub-
strate phosphorylation in the presence of solvent and background
phosphorylation in the presence of inactivated VEGF-R2 kinase.
MCF7 cell proliferation assay
MCF7 cells (ATCC, USA) were seeded in RPMI 1640 medium (Bio-
chrome, Germany) supplemented with fetal calf serum (10%; PAA,
Austria), l-glutamine (2 mm), estradiol (0.1 nm), and insulin
(1 UmLꢀ1) at 5000 cells per well in 96-well plates. Cells were al-
lowed to adhere for 24 h, and then fresh growth medium plus
compound was added. The final concentration of DMSO was 0.5%.
After four days of continuous incubation, the cells were fixed with
glutaraldehyde and stained with crystal violet, and the absorbance
was recorded at l 595 nm using Tecan Sunrise equipment. All
measurements were performed in quadruplicate. The values were
normalized to the absorbance of solvent-treated cells (100%), and
the absorbance of a reference plate which was fixed at the time
point of compound application (0%). Half-maximal growth inhibi-
tion (IC50) was determined as the compound concentration re-
quired to achieve 50% inhibition of cell growth.
Experimental Section
Kinase assays
CDK1: GST-tagged recombinant human CDK1 kinase (5 ngmLꢀ1
)
complexed with GST-tagged human cyclin B expressed in SF-9 cells
(ProQinase, Germany) was incubated for 10 min at 228C in the
presence of various concentrations of test compounds in a Tris·HCl
assay buffer [40 mL, 50 mm, pH 8.0, MgCl2 (10 mm), dithiothreitol
(1.0 mm), Na3VO4 (0.1 mm), PEG 20000 (0.025%, v/v), ATP (0.5 mm),
[g-33P]ATP (0.2 mCi per well), calf thymus histone type IIIS (1 mm;
Sigma–Aldrich, Germany), dimethyl sulfoxide (DMSO; 1.25%, v/v),
Nonidet P40 (0.05%)]. The reaction was stopped by the addition of
a solution of EDTA in H2O (0.25m; 10 mL, pH 7.5). A quantity (15 mL)
of each reaction mixture was transferred to P30 filter sheets
(Wallac, Germany), which were washed with phosphorous acid
(0.5%; 3ꢃ10 min) and subjected to scintillation counting with Mel-
tiLex A scintillator sheets (Wallac, Germany) and a Micobeta coun-
ter (Wallac, Germany). Half-maximal inhibition (IC50) of CDK1/cy-
clin B activity was determined as the inhibitor concentration result-
ing in 50% substrate phosphorylation between maximal substrate
phosphorylation in the presence of solvent and background phos-
phorylation in the presence of inactivated CDK1.
CA-II inhibition assay
Inhibition of human CA-II (Sigma–Aldrich) was determined with
the published 4-nitrophenyl acetate hydrolysis technique[42] using
a Tecan Rainbow 96-well spectrophotometer.
CDK2: GST-tagged recombinant human CDK2 kinase (1.25 ngmLꢀ1
)
complexed with GST-tagged human cyclin E expressed in SF-9 cells
(ProQinase, Germany) was incubated for 10 min at 228C in the
presence of various concentrations of test compounds in a Tris·HCl
assay buffer [40 mL, 50 mm, pH 8.0, MgCl2 (10 mm), dithiothreitol
(1.0 mm), Na3VO4 (0.1 mm), PEG 20000 (0.2%, v/v), ATP (0.5 mm), [g-
33P]ATP (0.2 mCi per well), calf thymus histone type IIIS (1 mm;
Sigma–Aldrich, Germany), DMSO (1.25%, v/v), Nonidet P40
(0.05%)]. The reaction was stopped by the addition of a solution
of EDTA in H2O (0.25m; 10 mL, pH 7.5). A quantity (15 mL) of each
reaction mixture was transferred to P30 filter sheets (Wallac, Ger-
many), which were washed with phosphorous acid (0.5%; 3ꢃ
10 min) and subjected to scintillation counting with MeltiLex A
scintillator sheets (Wallac, Germany) and a Micobeta counter
(Wallac, Germany). Half-maximal inhibition (IC50) of CDK2/cyclin E
activity was determined as the inhibitor concentration resulting in
50% substrate phosphorylation between maximal substrate phos-
phorylation in the presence of solvent and background phosphory-
lation in the presence of inactivated CDK2.
HeLa–MaTu mammary tumor model
Housing and handling of animals was in strict compliance with Eu-
ropean and German Guidelines for Laboratory Animal Welfare. For
tumor xenograft studies, female athymic nu/nu mice (Taconic),
50 days old, average body weight 20–22 g, were used after an ac-
climatization period of 14 days. Feeding and drinking was ad libi-
tum 24 h per day. HeLa–MaTu human cervical carcinoma cells ob-
tained from Epo GmbH (Berlin, Germany) were grown in RPMI
1640 medium (Biochrome, Germany) supplemented with fetal calf
serum (10%; PAA, Austria). Female NMRI nude mice with a body
weight of ~22 g each were given a s.c. implant of 1.5ꢃ106 HeLa–
MaTu carcinoma cells, suspended in 50% Matrigel/50% culture
medium (without fetal calf serum), in the inguinal region. Tumors
were allowed to grow to the predetermined size of ~22 mm2.
When the tumors were at approximately the correct size, the ani-
mals were evenly distributed between treatment and control
groups; treatment was started and continued until the untreated
control group had to be euthanized due to large tumor size. Treat-
ment of each animal was based on individual body weight. Tumor
response, by measurement of tumor area (product of the longest
diameter and its perpendicular) with a caliper, and body weight
were determined twice a week.
VEGF-R2: GST-tagged recombinant human VEGF-R2 kinase domain
purified from SF-9 cells was incubated for 10 min at 228C in the
presence of various concentrations of test compounds in a Tris·HCl
assay buffer [40 mL, 40 mm, pH 7.5, MgCl2 (10 mm), MnCl2 (1 mm),
dithiothreitol (1.0 mm), PEG 20000 (0.025%, v/v), ATP (8 mm), [g-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&12
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
desired product (21.5 g, 81 mmol, 81%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=8.21 (s, 1H), 7.05 (d, 1H), 4.86 (t, 1H), 4.16
(m, 1H), 3.41 (m, 2H), 1.17 ppm (d, 3H).
Docking
The FlexX program (version 1.8)[43] implemented with the SYBYL 6.7
software package,[44] was applied to perform the docking experi-
ments. The X-ray structure of hCA-II complexed with a phenylsulfo-
namide (PDB ID: 1CNX) was used.[45] For the docking procedure, all
water molecules and ligands were removed. All FlexX default pa-
rameters, as implemented in the SYBYL 6.7 software, were used.
The final manually selected docking pose was energy-minimized
using the MMFF94 force field[46] in SYBYL.
4-[(5-Bromo-4-{[(R)-1-hydroxypropan-2-yl]amino}pyrimidin-2-
yl)amino]benzenesulfonamide hydrochloride (ZK 304709): A so-
lution of (R)-2-[(5-bromo-2-chloropyrimidin-4-yl)amino]propan-1-ol
(531 mg, 2.00 mmol) in MeCN (4.0 mL) was added under stirring to
a suspension of 4-aminobenzenesulfonamide (347 mg, 2.02 mmol)
in MeCN (2.0 mL), H2O (0.5 mL), and HCl in dioxane (4N; 0.5 mL).
The mixture was stirred at reflux for 16 h. After cooling, the precipi-
tate was collected by filtration, washed with H2O and dried to give
1
X-ray structure of ZK 304709 in complex with CDK2
the desired product (682 mg, 1.56 mmol, 78%): H NMR (400 MHz,
[D6]DMSO, 300 K): d=10.61 (s, 1H), 8.28 (s, 1H), 7.78 (m, 4H), 7.45
(d, 1H), 7.20 (s, 2H), 4.30 (br, 2H), 3.53 (m, 2H), 1.21 ppm (d, 3H);
13C NMR (100 MHz, [D6]DMSO, 300 K): d=158.6, 152.7, 146.9, 141.3,
138.5, 126.9, 119.6, 93.2, 63.7, 50.2, 16.5 ppm; ESI-HRMS: m/z [M+
H]+ calcd for C13H17N5O3SBr: 402.0235, found: 402.0234; mp:
2628C; specific optical rotation: 16.0ꢁ0.168 (DMSO, 589 nm,
208C).
Crystals of CDK2 in complex with ZK 304709 were obtained as de-
scribed previously.[10] Data were collected at the synchrotron facility
at the ESRF in Grenoble, France. The structure was solved using
the program AMoRe[47] and refined using REFMAC.[48] The crystallo-
graphic data for the structure have been deposited with the RCSB
Protein Data Bank (PDB) with access code 4BGH.
Methyl N-(5-bromo-2-chloropyrimidin-4-yl)-d-alaninate: NEt3
(33.5 mL, 240 mmol) was added to a stirred solution of 5-bromo-
2,4-dichloropyrimidine (3; 22.8 g, 100 mmol) and methyl d-alanina-
te·HCl (14.0 g, 100 mmol) in THF (300 mL) and DMF (75 mL) at 08C.
The mixture was slowly warmed to room temperature under stir-
ring. After 48 h, the solvents were removed in vacuo and the resi-
due was purified by chromatography (hexane/EtOAc, 4:1!2:1) to
give the desired product (25.5 g, 86.1 mmol, 86%): 1H NMR
(400 MHz, CDCl3, 300 K): d=8.2 (s, 1H), 6.1 (d, 1H), 4.8 (m, 1H), 3.8
(s, 3H), 1.6 ppm (d, 3H).
Synthetic procedures
General methods and materials: All air- and moisture-sensitive re-
actions were carried out in oven-dried (at 1208C) glassware under
an inert atmosphere of argon. All reactive liquid reagents were
transferred by syringe or cannula and were added into the flask
through a rubber septum. 1H NMR and 13C NMR spectra were re-
corded at room temperature on Bruker Avance spectrometers op-
1
erating at 300, 400, 500, and 600 MHz for H NMR, and at 75, 100,
125, and 150 MHz for 13C NMR. HRMS were recorded on a Waters
QTOF Premier instrument with electrospray ionization. Optical rota-
tions were recorded on a JASCO P2000 polarimeter. Solvents for
extraction and chromatography were reagent grade and used as
received. Commercial reagents were used without purification. The
purity of all target compounds was >97% as determined by
1H NMR spectroscopy.
(R)-2-[(5-Bromo-2-chloropyrimidin-4-yl)amino]propanal:
A solu-
tion of diisobutylaluminum hydride in toluene (1.2m; 310 mL) was
added to a stirred solution of methyl N-(5-bromo-2-chloropyrimi-
din-4-yl)-d-alaninate (40.0 g, 135.8 mmol) in toluene (800 mL) at
ꢀ788C. After 30 min, the reaction was carefully quenched with
MeOH. The mixture was warmed to room temperature and diluted
with tert-butyl methyl ether (1000 mL). The mixture was washed
successively with aqueous HCl (1n; 3ꢃ100 mL), saturated aqueous
NaHCO3 solution (3ꢃ) and saturated aqueous NaCl solution (3ꢃ).
The organic phase was dried (MgSO4), filtered and concentrated by
evaporation. The residue was purified by chromatography (hexane/
EtOAc, 4:1!1:1) to give the desired product (22.5 g, 83.9 mmol,
5-Bromo-2-chloro-N-(prop-2-yn-1-yl)pyrimidin-4-amine: Prop-2-
yn-1-amine (290 mg, 5.3 mmol) was added to a stirred solution of
5-bromo-2,4-dichloropyrimidine (3; 1000 mg, 4.4 mmol) and NEt3
(0.73 mL, 5.3 mmol) in MeCN (4.5 mL) at 08C. The mixture was
stirred at room temperature for 17 h. The precipitate was collected
by filtration, washed with H2O and dried (Na2SO4) to give the de-
sired product (769 mg, 3.1 mmol, 70%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=8.27 (s, 1H), 8.11 (t, 1H), 4.07 (dd, 2H),
3.04 ppm (t, 1H).
1
62%): H NMR (400 MHz, CDCl3, 300 K): d=9.6 (s, 1H), 8.2 (s, 1H),
6.3 (d, 1H), 4.8 (m, 1H), 1.5 ppm (d, 3H).
(2R,3R)-3-[(5-Bromo-2-chloropyrimidin-4-yl)amino]butan-2-ol:
Copper(I) bromide–dimethyl sulfide complex (32.7 g, 159 mmol)
was introduced under nitrogen atmosphere into Et2O (1000 mL),
and the mixture was cooled to ꢀ788C. Over a period of ~25 min,
a solution of methyllithium in Et2O (1.6m; 200 mL) was added
dropwise, and then the cooling bath was removed. The mixture
was stirred for 40 min, and the temperature increased to ꢀ358C.
The mixture was cooled to ꢀ558C and (R)-2-[(5-bromo-2-chloropyr-
imidin-4-yl)amino]propanal (18.9 g, 71.5 mmol) was added over
a period of 20 min. The mixture was stirred for 6 h at ꢀ558C, then
the cooling bath was filled with dry ice again, and the mixture
stirred overnight. Saturated aqueous NH4Cl solution (200 mL) was
added dropwise, and the mixture was warmed to room tempera-
ture. The reaction mixture was diluted with Et2O (500 mL), the or-
ganic phase was separated, and the aqueous phase was extracted
with Et2O. The combined organic phases were washed with satu-
rated aqueous NH4Cl solution and saturated aqueous NaCl solu-
tion, dried (Na2SO4), filtered and concentrated by evaporation. The
4-{[5-Bromo-4-(prop-2-yn-1-ylamino)pyrimidin-2-yl]amino}benze-
nesulfonamide (2): A solution of 5-bromo-2-chloro-N-(prop-2-yn-1-
yl)pyrimidin-4-amine (247 mg, 1.00 mmol) in MeCN (2.0 mL) was
added under stirring to a suspension of 4-aminobenzenesulfona-
mide (172 mg, 1.00 mmol) in MeCN (1.0 mL), H2O (0.25 mL) and
HCl in dioxane (4n; 0.25 mL). The mixture was stirred at reflux for
17 h. After cooling, the precipitate was collected by filtration,
washed with H2O and dried (Na2SO4) to give the desired product
(333 mg, 0.84 mmol, 84%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
d=10.06 (s, 1H), 8.16 (s, 1H), 8.05 (t, 1H), 7.87 (m, 2H), 7.68 (m,
2H), 7.18 (br, 2H), 4.11 (m, 2H), 3.13 ppm (t, 1H).
(R)-2-[(5-Bromo-2-chloropyrimidin-4-yl)amino]propan-1-ol: (R)-2-
Aminopropan-1-ol (9.4 g, 125 mmol) was added to a stirred solu-
tion of 5-bromo-2,4-dichloropyrimidine (3; 22.8 g, 100 mmol) and
NEt3 (17 mL, 125 mmol) in MeCN (100 mL) at 08C. The mixture was
stirred overnight at room temperature. The precipitate was collect-
ed by filtration, washed with H2O and dried (Na2SO4) to give the
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&13
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
residue was purified by chromatography (hexane/EtOAc, 4:1!1:1)
to give the desired product (8.4 g, 30.0 mmol, 42%): 1H NMR
(400 MHz, CDCl3, 300 K): d=8.1 (s, 1H), 5.8 (d, 1H), 4.2 (m, 1H), 3.9
(m, 1H), 2.0 (d, 1H), 1.3 (d, 3H), 1.2 ppm (d, 3H).
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]thiophene-2-sulfonamide (9): A solution of (R)-3-
[(5-bromo-2-chloropyrimidin-4-yl)amino]-2-methylbutan-2-ol
(294 mg, 1.00 mmol) in MeCN (2.0 mL) was added under stirring to
a
solution
of
4-aminothiophene-2-sulfonamide
(267 mg,
4-[(5-Bromo-4-{[(2R,3R)-3-hydroxybutan-2-yl]amino}pyrimidin-2-
yl)amino]benzenesulfonamide (6): The compound was prepared
in a manner similar to that described in the preparations of 2 and
ZK 304709, using (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)ami-
1.50 mmol) in MeCN (6.0 mL), H2O (0.2 mL) and HCl in dioxane
(4N; 0.4 mL). The mixture was stirred at reflux for 18 h. After cool-
ing, the mixture was concentrated and the residue was purified by
chromatography (CH2Cl2/EtOH, 9:1) to give the desired product
(183 mg, 0.42 mmol, 42%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
d=9.77 (s, 1H), 8.07 (s, 1H), 7.63 (m, 4H), 6.02 (d, 1H), 4.82 (s, 1H),
4.05 (m, 1H), 1.06 ppm (m, 9H); ESI-HRMS: m/z [M+H]+ calcd for
C13H19N5O3S2Br: 436.0113, found: 436.0118.
1
no]butan-2-ol and 4-aminobenzenesulfonamide: H NMR (400 MHz,
[D6]DMSO, 300 K): d=9.64 (s, 1H), 8.07 (s, 1H), 7.84 (m, 2H), 7.65
(m, 2H), 7.11 (s, 2H), 6.11 (d, 1H), 4.98 (d, 1H), 4.05 (m, 1H), 3.75
(m, 1H), 1.17 (d, 3H), 1.05 ppm (d, 3H); ESI-HRMS: m/z [M+H]+
calcd for C14H19N5O3SBr: 416.0392, found: 416.0391.
5-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]pyridine-2-sulfonamide (10): A solution of 5-[(5-
bromo-4-chloropyrimidin-2-yl)amino]pyridine-2-sulfonamide
(R)-3-[(5-Bromo-2-chloropyrimidin-4-yl)amino]-2-methylbutan-2-
ol: A solution of MeMgBr in Et2O (3m; 30 mL, 90 mmol) was added
dropwise to a stirred solution of methyl N-(5-bromo-2-chloropyri-
midin-4-yl)-d-alaninate (2.95 g, 10.0 mmol) in THF (150 mL) at 08C.
The mixture was stirred at room temperature for 2.5 h, then dilut-
ed with saturated aqueous NH4Cl solution (30 mL). H2O was added,
and the mixture was extracted with EtOAc (3ꢃ). The combined or-
ganic phases were dried (Na2SO4), filtered and concentrated. The
residue was purified by chromatography (hexane/EtOAc, 4:1!1:1)
to give the desired product (2.81 g, 9.5 mmol, 95%): 1H NMR
(400 MHz, CDCl3, 300 K): d=8.1 (s, 1H), 5.9 (d, 1H), 4.2 (m, 1H), 1.8
(br, 1H), 1.2 ppm (m, 9H).
(55 mg, 0.15 mmol) and (R)-3-amino-2-methylbutan-2-ol (62 mg,
0.60 mmol) in DMF (2.5 mL) was stirred at 508C for 5 h. The mix-
ture was concentrated, and the residue was purified by chroma-
tography (CH2Cl2/MeOH, 9:1) to give the desired product (58 mg,
1
0.13 mmol, 87%): H NMR (400 MHz, [D6]DMSO, 300 K): d=9.88 (s,
1H), 8.92 (m, 1H), 8.37 (m, 1H), 8.12 (s, 1H), 7.84 (m, 1H), 7.25 (br,
2H), 6.13 (d, 1H), 4.87 (s, 1H), 4.06 (m, 1H), 1.16 ppm (m, 9H); ESI-
HRMS: m/z [M+H]+ calcd for C14H20N6O3SBr: 431.0501, found:
431.0503.
4-[(5-Bromo-4-{[(2R,3R)-3-hydroxybutan-2-yl]amino}pyrimidin-2-
yl)amino]-N-methylbenzenesulfonamide (13): The compound was
prepared in a manner similar to that described in the preparations
of 2 and ZK 304709, using (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-
yl)amino]butan-2-ol and 4-amino-N-methylbenzenesulfonamide:
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.73 (s, 1H), 8.12 (s, 1H),
7.91 (m, 2H), 7.64 (m, 2H), 7.20 (q, 1H), 6.18 (d, 1H), 5.01 (br, 1H),
4.09 (m, 1H), 3.80 (m, 1H), 2.39 (d, 3H), 1.22 (d, 3H), 1.08 ppm (d,
3H); ESI-HRMS: m/z [M+H]+ calcd for C15H21N5O3SBr: 430.0548,
found: 430.0556.
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]benzenesulfonamide (7): The compound was pre-
pared in a manner similar to that described in the preparations of
2 and ZK 304709, using (R)-3-[(5-bromo-2-chloropyrimidin-4-yl)ami-
no]-2-methylbutan-2-ol and 4-aminobenzenesulfonamide: 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=9.68 (s, 1H), 8.11 (s, 1H), 7.86 (m,
2H), 7.69 (m, 2H), 7.14 (s, 2H), 6.08 (d, 1H), 4.83 (s, 1H), 4.09 (m,
1H), 1.17 ppm (m, 9H); ESI-HRMS: m/z [M+H]+ calcd for
C15H21N5O3SBr: 430.0548, found: 430.0549.
(2R,3R)-3-[(5-Bromo-2-chloropyrimidin-4-yl)oxy]butan-2-ol: NaH
(55% dispersion; 480 mg, 11.0 mmol) was added portionwise to
a stirred solution of (2R,3R)-butane-2,3-diol (1.35 g, 15.0 mmol) in
THF (50 mL) at 08C. The mixture was stirred for 10 min at room
temperature. The resulting solution was added to a stirred solution
of 5-bromo-2,4-dichloropyrimidine (3; 2.27 g, 10.0 mmol) in THF
(25 mL) at 08C. The mixture was slowly warmed to room tempera-
ture and stirred for 12 h. The solvent was removed in vacuo, and
the residue was purified by chromatography (hexane/EtOAc, 1:1)
to give the desired product (2.29 g, 8.1 mmol, 81%): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=8.44 (s, 1H), 5.18 (q, 1H), 3.96 (q,
1H), 2.02 (d, 1H), 1.40 (d, 3H), 1.28 ppm (d, 3H).
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]-2-hydroxybenzenesulfonamide (14): The com-
pound was prepared in a manner similar to that described in the
preparations of 2 and ZK 304709, using (R)-3-[(5-bromo-2-chloro-
pyrimidin-4-yl)amino]-2-methylbutan-2-ol and 4-amino-2-hydroxy-
benzenesulfonamide: 1H NMR (400 MHz, [D6]DMSO, 300 K): d=
10.46 (s, 1H), 9.70 (s, 1H), 8.12 (s, 1H), 7.53 (m, 1H), 7.36 (m, 1H),
7.28 (m, 1H), 6.80 (s, 2H), 6.28 (d, 1H), 4.09 (m, 1H), 1.17 ppm (m,
9H); ESI-HRMS: m/z [M+H]+ calcd for C15H21N5O4SBr: 446.0498,
found: 446.0500.
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]-2-methylbenzenesulfonamide (15): The com-
pound was prepared in a manner similar to that described in the
preparations of 2 and ZK 304709, using (R)-3-[(5-bromo-2-chloro-
pyrimidin-4-yl)amino]-2-methylbutan-2-ol and 4-amino-2-methyl-
benzenesulfonamide: 1H NMR (400 MHz, [D6]DMSO, 300 K): d=
10.35 (s, 1H), 8.25 (s, 1H), 7.78 (m, 1H), 7.72 (m, 1H), 7.56 (m, 1H),
7.27 (s, 2H), 6.83 (d, 1H), 4.10 (m, 1H), 2.57 (s, 3H), 1.18 ppm (m,
9H); ESI-HRMS: m/z [M+H]+ calcd for C16H23N5O3SBr: 444.0705,
found: 444.0708.
4-[(5-Bromo-4-{[(2R,3R)-3-hydroxybutan-2-yl]oxy}pyrimidin-2-
yl)amino]benzenesulfonamide (8): A solution of HCl in dioxane
(4n; 0.8 mL) was added to a mixture of (2R,3R)-3-[(5-bromo-2-
chloropyrimidin-4-yl)oxy]butan-2-ol (4.50 g, 16.0 mmol) and 4-ami-
nobenzenesulfonamide (2.75 g, 16.0 mmol) in butan-1-ol (120 mL)
and MeOH (12 mL) at room temperature. The mixture was stirred
for 43 h at 608C. After cooling, the precipitate was collected by fil-
tration, washed with H2O and dried (Na2SO4) to give the desired
product (3.91 g, 9.4 mmol, 59%): 1H NMR (400 MHz, [D6]DMSO,
300 K): d=10.08 (s, 1H), 8.44 (s, 1H), 7.86 (m, 2H), 7.74 (m, 2H),
7.20 (s, 2H), 5.22 (m, 1H), 3.85 (m, 1H), 1.29 (d, 3H), 1.14 ppm (d,
3H); ESI-HRMS: m/z [M+H]+ calcd for C14H18N4O4SBr: 417.0232,
found: 417.0239.
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]-2-chlorobenzenesulfonamide (16): The com-
pound was prepared in a manner similar to that described in the
preparations of 2 and ZK 304709, using (R)-3-[(5-bromo-2-chloro-
pyrimidin-4-yl)amino]-2-methylbutan-2-ol and 4-amino-2-chloro-
1
benzenesulfonamide: H NMR (400 MHz, [D6]DMSO, 300 K): d=9.86
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&14
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(s, 1H), 8.23 (m, 1H), 8.14 (s, 1H), 7.83 (m, 1H), 7.63 (m, 1H), 7.38
(s, 2H), 6.13 (d, 1H), 4.87 (s, 1H), 4.09 (m, 1H), 1.20 ppm (m, 9H);
ESI-HRMS: m/z [M+H]+ calcd for C15H20N5O3SClBr: 464.0159,
found: 464.0162.
(d, 6H); ESI-HRMS: m/z [M+H]+ calcd for C14H19N5OSBr: 384.0494,
found: 384.0504.
(R,S)-1-(Ethylsulfinyl)-4-nitrobenzene (23): Iron(III) chloride (0.43 g,
2.64 mmol) was added to a solution of 1-(ethylsulfanyl)-4-nitroben-
zene (22; 16.86 g, 92.02 mmol) in MeCN (75 mL), and the mixture
was stirred at room temperature for 10 min. H5IO6 (22.44 g,
98.44 mmol) was added portionwise to the stirred reaction mixture
so that the temperature remained <308C. The mixture was stirred
for 50 min, then was added to a stirred mixture of CH2Cl2 (170 mL),
ice water (500 mL) and Na2S2O3·5H2O (100 g). The mixture was ex-
tracted with CH2Cl2 (2ꢃ). The combined organic phases were dried
(Na2SO4), filtered and concentrated. Recrystallization from EtOAc/
hexane gave the desired product (12.49 g, 62.69 mmol, 68%):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.35 (m, 2H), 7.88 (m, 2H),
3.12 (m, 1H), 2.84 (m, 1H), 0.99 ppm (t, 3H).
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]-2-methoxybenzenesulfonamide (17): The com-
pound was prepared in a manner similar to that described in the
preparations of 2 and ZK 304709, using (R)-3-[(5-bromo-2-chloro-
pyrimidin-4-yl)amino]-2-methylbutan-2-ol and 4-amino-2-methoxy-
benzenesulfonamide: 1H NMR (400 MHz, [D6]DMSO, 300 K): d=
10.55 (s, 1H), 8.30 (s, 1H), 7.65 (m, 2H), 7.22 (m, 1H), 6.93 (m, 3H),
4.12 (m, 1H), 3.89 (s, 3H), 1.16 ppm (m, 9H); ESI-HRMS: m/z [M+
H]+ calcd for C16H23N5O4SBr: 460.0654, found: 460.0651.
4-[(5-Bromo-4-{[(R)-3-hydroxy-3-methylbutan-2-yl]amino}pyrimi-
din-2-yl)amino]-2,6-dimethylbenzenesulfonamide (18): The com-
pound was prepared in a manner similar to that described in the
preparations of 2 and ZK 304709, using (R)-3-[(5-bromo-2-chloro-
pyrimidin-4-yl)amino]-2-methylbutan-2-ol and 4-amino-2,6-dime-
(R,S)-S-Ethyl-S-(4-nitrophenyl)sulfoximide (24): Concentrated
H2SO4 (14.7 mL) was added dropwise to a stirred suspension of
(R,S)-1-(ethylsulfinyl)-4-nitrobenzene (23; 11.50 g, 57.8 mmol) and
NaN3 (4.22 g, 64.9 mmol) in CHCl3 (59 mL) under a nitrogen atmos-
phere at 08C. The mixture was slowly heated to 458C and stirred
overnight. After cooling, the mixture was diluted with ice water
and CHCl3. The organic phase was separated and discarded. The
aqueous phase was extracted with CHCl3 and the organic phase
was discarded. Under ice cooling, the aqueous phase was cautious-
ly basified with aqueous NaOH solution (2n) and extracted with
CH2Cl2 (2ꢃ). The combined organic phases were washed with
dilute aqueous NaCl solution, dried (Na2SO4), filtered and concen-
trated to give the desired product (5.52 g, 25.8 mmol, 45%):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.37 (m, 2H), 8.09 (m, 2H),
4.56 (s, 1H), 3.18 (q, 2H), 1.04 ppm (t, 3H).
1
thylbenzenesulfonamide: H NMR (400 MHz, [D6]DMSO, 300 K): d=
10.44 (s, 1H), 8.19 (s, 1H), 7.47 (s, 2H), 7.20 (s, 2H), 6.95 (d, 1H),
4.10 (m, 1H), 2.58 (s, 6H), 1.20 ppm (m, 9H); ESI-HRMS: m/z [M+
H]+ calcd for C17H25N5O3SBr: 458.0861, found: 458.0848.
5-Bromo-N4-isopropyl-N2-[4-(methylsulfanyl)phenyl]pyrimidine-
2,4-diamine hydrochloride (19): A solution of 4-(methylsulfanyl)a-
niline (2.0 mL, 16.3 mmol) in MeCN (10 mL) was added to a solution
of
5-bromo-2-chloro-N-isopropylpyrimidin-4-amine
(4.08 g,
16.3 mmol) in MeCN (20 mL) at room temperature. A solution of
HCl in dioxane (4N; 4.1 mL) and H2O (4.1 mL) were added, and the
mixture was refluxed for 16 h. After cooling, the precipitate was
collected by filtration, washed with H2O and dried (Na2SO4) to give
the desired product (4.94 g, 12.7 mmol, 78%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=10.39 (s, 1H), 8.18 (s, 1H), 7.88 (br, 1H), 7.49
(d, 2H), 7.29 (d, 2H), 4.30 (m, 1H), 2.50 (s, 3H), 1.21 ppm (d, 6H).
(R,S)-N-(Ethoxycarbonyl)-S-ethyl-S-(4-nitrophenyl)sulfoximide
(25): Under a nitrogen atmosphere, ethyl chloroformate (5.67 mL,
59.6 mmol) was added dropwise to a stirred solution of (R,S)-S-
ethyl-S-(4-nitrophenyl)sulfoximide (24; 2.75 g, 12.8 mmol) in pyri-
dine (120 mL) at room temperature. The mixture was stirred for
4 h, then poured onto dilute aqueous NaCl solution. The mixture
was extracted with EtOAc (3ꢃ). The combined organic phases were
dried (Na2SO4), filtered and concentrated. The residue was purified
by chromatography (hexane/EtOAc, 1:1) to give the desired prod-
(R,S)-5-Bromo-N4-isopropyl-N2-[4-(methylsulfinyl)phenyl]pyrimi-
dine-2,4-diamine (20): 5-Bromo-N4-isopropyl-N2-[4-(methylsulfa-
nyl)phenyl]pyrimidine-2,4-diamine·HCl (19; 1.77 g, 4.6 mmol) was
taken up in CH2Cl2 (40 mL) and 3-chloroperoxybenzoic acid (55%;
1.73 g, 5.5 mmol) was added under stirring. The mixture was stirred
at room temperature for 90 min, then diluted with CH2Cl2. The mix-
ture was washed with saturated aqueous NaHCO3 solution and sa-
turated aqueous NaCl solution. The organic phase was dried
(Na2SO4), filtered and concentrated. The residue was purified by
chromatography (CH2Cl2/EtOH, 9:1) to give the desired product
(553 mg, 1.50 mmol, 33%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
d=9.55 (s, 1H), 8.08 (s, 1H), 7.90 (d, 2H), 7.53 (d, 2H), 6.53 (d, 1H),
4.35 (m, 1H), 2.70 (s, 3H), 1.25 ppm (d, 6H).
1
uct (3.12 g, 10.9 mmol, 85%): H NMR (400 MHz, [D6]DMSO, 300 K):
d=8.48 (m, 2H), 8.15 (m, 2H), 3.92 (m, 2H), 3.69 (m, 2H), 1.12 ppm
(m, 6H).
(R,S)-S-(4-Aminophenyl)-N-(ethoxycarbonyl)-S-ethylsulfoximide
(26): Titanium(III) chloride solution (~10% in 20–30% HCl; 147 mL)
was added dropwise to a stirred solution of (R,S)-N-(ethoxycarbon-
yl)-S-ethyl-S-(4-nitrophenyl)sulfoximide (25; 3.10 g, 10.8 mmol) in
THF (220 mL) at room temperature. The mixture was stirred for 4 h,
then cooled to 08C. The mixture was basified (pH 9–10) by drop-
wise addition of aqueous NaOH solution (32%) under stirring. The
resulting pulp was diluted with H2O and EtOAc. After complete ad-
dition of the aqueous NaOH solution, the mixture was extracted
with EtOAc (3ꢃ). The combined organic phases were washed with
dilute aqueous NaCl solution, dried (Na2SO4), filtered and concen-
trated to give the crude product (3.31 g) that was used without
further purification: 1H NMR (400 MHz, [D6]DMSO, 300 K): d=7.47
(m, 2H), 6.67 (m, 2H), 6.20 (s, 2H), 3.90 (m, 2H), 3.42 (q, 2H),
1.10 ppm (m, 6H).
(R,S)-S-(4-{[5-Bromo-4-(isopropylamino)pyrimidin-2-yl]amino}-
phenyl)-S-methylsulfoximide (21): NaN3 (40 mg, 0.63 mmol) was
added to (R,S)-5-bromo-N4-isopropyl-N2-[4-(methylsulfinyl)phenyl]-
pyrimidine-2,4-diamine (20; 192 mg, 0.50 mmol) in CH2Cl2
(1.00 mL) at room temperature, and the mixture was cooled to
08C. Then, concentrated H2SO4 (0.13 mL) was cautiously added
under stirring. The mixture was stirred at 458C for 16 h. After cool-
ing, aqueous NaOH solution (2n; 2.00 mL) was cautiously added,
and the mixture was extracted with EtOAc (2ꢃ). The combined or-
ganic phases were dried (Na2SO4), filtered and concentrated. The
residue was purified by chromatography (CH2Cl2/EtOH, 9:1) to give
the desired product (38 mg, 0.10 mmol, 20%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=9.70 (s, 1H), 8.08 (s, 1H), 7.90 (d, 2H), 7.77 (d,
2H), 6.62 (d, 1H), 4.35 (m, 1H), 3.99 (s, 1H), 3.03 (s, 3H), 1.29 ppm
(R,S)-S-[4-({5-Bromo-4-[(1R,2R)-(2-hydroxy-1-methylpropyl)ami-
no]pyrimidin-2-yl}amino)phenyl]-N-(ethoxycarbonyl)-S-ethylsul-
foximide (27): A solution of HCl in dioxane (4N; 0.95 mL) was
added to a mixture of (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)a-
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&15
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
mino]butan-2-ol (1.07 g, 3.81 mmol) and crude (R,S)-S-(4-amino-
phenyl)-N-(ethoxycarbonyl)-S-ethylsulfoximide (26; 1.17 g) in MeCN
(15 mL) and H2O (0.95 mL) at room temperature. The mixture was
stirred for 24 h at 608C. After cooling, the mixture was diluted with
saturated aqueous NaCl solution and extracted with EtOAc (3ꢃ).
The combined organic phases were dried (Na2SO4), filtered and
concentrated. The residue was purified by chromatography
(CH2Cl2/EtOH, 95:5) to give the desired product (1.00 g, 2.00 mmol,
(32): The compound was prepared in a manner similar to that de-
scribed in the preparations of 27 and 28, using (2R,3R)-3-[(5-
bromo-2-chloropyrimidin-4-yl)amino]butan-2-ol
and
(R,S)-S-(4-
amino-2-fluorophenyl)-N-(ethoxycarbonyl)-S-ethylsulfoximide:
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.92 (s, 1H), 8.16 (s, 1H),
7.90 (m, 1H), 7.71 (m, 1H), 7.61 (m, 1H), 6.23 (d, 1H), 5.03 (d, 1H),
4.45 (s, 1H), 4.06 (m, 1H), 3.78 (m, 1H), 3.13 (s, 3H), 1.22 (d, 3H),
1.10 ppm (d, 3H); ESI-HRMS: m/z [M+H]+ calcd for
C15H20N5O2SBrF: 432.0505, found: 432.0505.
1
52%): H NMR (400 MHz, [D6]DMSO, 300 K): d=9.92 (s, 1H), 8.13 (s,
1H), 7.97 (m, 2H), 7.72 (m, 2H), 6.27 (d, 1H), 4.10 (m, 1H), 3.92 (m,
2H), 3.80 (m, 1H), 3.55 (q, 2H), 1.23 (d, 3H), 1.10 ppm (m, 9H).
(2R,3R)-3-{[2-Chloro-5-(trifluoromethyl)pyrimidin-4-yl]amino}bu-
tan-2-ol (37): NEt3 (72.2 mL, 520.7 mmol) was added dropwise to
2,4-dichloro-5-(trifluoromethyl)pyrimidine (36; 56.5 g, 260.4 mmol)
and (2R,3R)-3-aminobutan-2-ol·HCl (32.7 g, 260.4 mmol) in MeCN
(1035 mL) at 08C. The mixture was heated slowly overnight to
room temperature. The mixture was added to a semi-concentrated
aqueous NaCl solution and extracted with EtOAc (2ꢃ). The com-
bined organic phases were dried (Na2SO4), filtered and concentrat-
ed by evaporation. The residue was purified by chromatography
(hexane/EtOAc, 0–100%) to give the desired product (18.6 g,
(R,S)-S-[4-({5-Bromo-4-[(1R,2R)-(2-hydroxy-1-methylpropyl)ami-
no]pyrimidin-2-yl}amino)phenyl]-S-ethylsulfoximide (28): A solu-
tion of sodium ethanolate in EtOH (0.35m; 11.6 mL) was added to
a stirred solution of (R,S)-S-[4-({5-bromo-4-[(1R,2R)-(2-hydroxy-1-
methylpropyl)amino]pyrimidin-2-yl}amino)phenyl]-N-(ethoxycar-
bonyl)-S-ethylsulfoximide (27; 1.00 g, 2.0 mmol) in EtOH (3.9 mL) at
room temperature. The mixture was stirred at reflux for 5 h and
then at room temperature overnight. Saturated aqueous NaCl solu-
tion was added, and the mixture was extracted with EtOAc (2ꢃ).
The combined organic phases were dried (Na2SO4), filtered and
concentrated. The residue was purified by chromatography
(CH2Cl2/EtOH, 9:1) to give the desired product (0.80 g, 1.87 mmol,
1
68.97 mmol, 26%): H NMR (400 MHz, [D6]DMSO, 300 K): d=8.38 (s,
1H), 6.71 (d, 1H), 5.00 (d, 1H), 4.08 (m, 1H), 3.71 (m, 1H), 1.12 (d,
3H), 1.01 ppm (d, 3H).
(R,S)-N-(Ethoxycarbonyl)-S-ethyl-S-[4-({4-[(1R,2R)-(2-hydroxy-1-
methylpropyl)amino]-5-(trifluoromethyl)pyrimidin-2-yl}amino)-
phenyl]sulfoximide (39): A solution of HCl in dioxane (4n;
0.012 mL) was added to a mixture of (R,S)-S-(4-aminophenyl)-N-
(ethoxycarbonyl)-S-ethylsulfoximide (26; 92 mg, 0.36 mmol) and
(2R,3R)-3-{[2-chloro-5-(trifluoromethyl)pyrimidin-4-yl]amino}butan-2-
ol (37; 100 mg, 0.37 mmol) in MeCN (2.2 mL) and H2O (0.2 mL). The
mixture was stirred for 6 h at 608C. After cooling, the mixture was
diluted with EtOAc and washed with saturated aqueous NaHCO3
solution. The organic phase was dried (Na2SO4), filtered and con-
centrated by evaporation. The residue was purified by chromatog-
raphy (CH2Cl2/EtOH, 95:5) to give the desired product (129 mg,
1
94%): H NMR (400 MHz, [D6]DMSO, 300 K): d=9.71 (s, 1H), 8.11 (s,
1H), 7.90 (m, 2H), 7.71 (m, 2H), 6.13 (d, 1H), 5.01 (d, 1H), 4.08 (m,
1H), 3.93 (s, 1H), 3.78 (m, 1H), 3.03 (q, 2H), 1.22 (d, 3H), 1.10 ppm
(m, 6H); ESI-HRMS: m/z [M+H]+ calcd for C16H23N5O2SBr: 428.0756,
found: 428.0749.
The mixture of diastereomers was separated into single stereoiso-
mers by preparative HPLC [column: Chiracel OJ 20 mm, 250ꢃ
50.8 mm; eluents: A: hexane +0.1% DEA, B: EtOH; gradient: iso-
cratic 15% B; flow: 80 mLminꢀ1; detector: UV 300 nm; tempera-
ture: RT]: tR (min): 34.0 (diastereomer 1: compound 29), 43.7 (dia-
stereomer 2: compound 30).
1
0.26 mmol, 70%): H NMR (400 MHz, [D6]DMSO, 300 K): d=10.18 (s,
(R,S)-S-(4-Nitrophenyl)-S-(trifluoromethyl)sulfoximide
(35):[35]
1H), 8.27 (s, 1H), 7.99 (m, 2H), 7.75 (m, 2H), 6.09 (d, 1H), 5.07 (d,
1H), 4.14 (m, 1H), 3.87 (m, 2H), 3.76 (m, 1H), 3.50 (q, 2H), 1.19 (d,
3H), 1.05 ppm (m, 9H).
Oleum (20% SO3; 5.8 mL) was carefully added to an ice-cooled
batch of 1-nitro-4-[(R,S)-(trifluoromethyl)sulfinyl]benzene (34;
1.36 g, 5.7 mmol). Then, NaN3 (0.36 g, 5.5 mmol) was carefully
added portionwise with stirring and the mixture was heated at
708C for 90 min. After cooling, additional NaN3 (0.36 g, 5.5 mmol)
was added, and the mixture was stirred for 90 min at 708C. After
cooling, the mixture was carefully poured into ice water. The pre-
cipitate was collected by filtration, washed with H2O and dried to
give the desired product (0.63 g, 2.5 mmol, 44%): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=8.49 (m, 2H), 8.31 (m, 2H),
7.32 ppm (s, 1H).
(R,S)-S-Ethyl-S-[4-({4-[(1R,2R)-(2-hydroxy-1-methylpropyl)amino]-
5-(triethoxymethyl)pyrimidin-2-yl}amino)phenyl]sulfoximide
(40): A solution of sodium ethanolate in EtOH (1.53m; 0.73 mL)
was added to a stirred solution of (R,S)-N-(ethoxycarbonyl)-S-ethyl-
S-[4-({4-[(1R,2R)-(2-hydroxy-1-methylpropyl)amino]-5-(trifluorome-
thyl)pyrimidin-2-yl}amino)phenyl]sulfoximide
(39;
110 mg,
0.23 mmol) in EtOH (1.6 mL) at room temperature. The mixture
was stirred at 608C for 24 h. Saturated aqueous NaCl solution was
added, and the mixture was extracted with EtOAc (2ꢃ). The com-
bined organic phases were dried (Na2SO4), filtered and concentrat-
ed. The residue was purified by chromatography (CH2Cl2/EtOH, 9:1)
to give the product (39 mg, 0.08 mmol, 35%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=9.65 (s, 1H), 8.01 (s, 1H), 7.92 (m, 2H), 7.68
(m, 2H), 6.53 (d, 1H), 4.92 (d, 1H), 4.08 (m, 1H), 3.89 (s, 1H), 3.75
(m, 1H), 3.27 (q, 6H), 3.02 (q, 2H), 1.07 ppm (m, 18H).
(R,S)-S-[4-({5-Bromo-4-[(1R,2R)-(2-hydroxy-1-methylpropyl)ami-
no]pyrimidin-2-yl}amino)phenyl]-S-(trifluoromethyl)sulfoximide
(31): (R,S)-S-(4-Aminophenyl)-S-(trifluoromethyl)sulfoximide (60 mg,
0.27 mmol) and (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)amino]-
butan-2-ol (75 mg, 0.27 mmol) in butan-1-ol (3.0 mL) and MeOH
(0.3 mL) were stirred at 908C for 48 h. The solvents were removed
in vacuo and the residue was purified by chromatography (CH2Cl2/
EtOH, 8:2) to give the desired product (40 mg, 0.09 mmol, 33%):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=9.99 (s, 1H), 8.12 (s, 1H),
8.02 (m, 2H), 7.68 (m, 2H), 6.46 (s, 1H), 6.18 (d, 1H), 4.98 (d, 1H),
4.07 (m, 1H), 3.75 (m, 1H), 1.17 (d, 3H), 1.05 ppm (d, 3H); ESI-
HRMS: m/z [M+H]+ calcd for C15H18N5O2SBrF3: 468.0317, found:
468.0313.
(R,S)-S-Ethyl-S-(4-nitrophenyl)-N-(trifluoroacetyl)sulfoximide (41):
Rhodium(II) acetate dimer (211 mg, 0.48 mmol) was added to
a stirred suspension of (R,S)-1-(ethylsulfinyl)-4-nitrobenzene (23;
1270 mg,
6.4 mmol),
2,2,2-trifluoroacetamide
(1585 mg,
14.0 mmol), iodobenzene diacetate (3388 mg, 10.5 mmol) and
MgO (1130 mg, 28.0 mmol) in CH2Cl2 (38 mL) under argon. The
mixture was stirred overnight at room temperature, then was fil-
tered through Celite with suction. The filtrate was concentrated by
(R,S)-S-[4-({5-Bromo-4-[(1R,2R)-(2-hydroxy-1-methylpropyl)ami-
no]pyrimidin-2-yl}amino)-2-fluorophenyl]-S-methylsulfoximide
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&16
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
evaporation. The residue was purified by chromatography (CH2Cl2/
EtOH, 98:2) to give the desired product (1520 mg, 4.9 mmol, 77%):
1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.52 (m, 2H), 8.22 (m, 2H),
3.99 (m, 2H), 1.16 ppm (t, 3H).
d=8.41 (m, 2H), 7.98 (m, 2H), 2.60 (m, 1H), 1.01 (m, 3H), 0.86 ppm
(m, 1H).
(R,S)-S-Cyclopropyl-S-(4-nitrophenyl)-N-(trifluoroacetyl)sulfoxi-
mide: Rhodium(II) acetate dimer (1.58 g, 3.58 mmol) was added to
a stirred suspension of (R,S)-1-(cyclopropylsulfinyl)-4-nitrobenzene
(R,S)-S-(4-Aminophenyl)-S-ethyl-N-(trifluoroacetyl)sulfoximide:
Titanium(III) chloride solution (~15% in ~10% HCl; 88.5 mL) was
added slowly, with ice cooling, to a solution of (R,S)-S-ethyl-S-(4-ni-
trophenyl)-N-(trifluoroacetyl)sulfoximide (4.05 g, 13.1 mmol) in THF
(191 mL). The mixture was stirred for 3.5 h at room temperature,
diluted with EtOAc and then washed with a semi-concentrated
aqueous NaCl solution (3ꢃ). The organic phase was dried (Na2SO4),
filtered and concentrated by evaporation to give the desired prod-
(15.1 g,
71.53 mmol),
2,2,2-trifluoroacetamide
(17.8 g,
157.37 mmol), iodobenzene diacetate (38.0 g, 118.02 mmol) and
MgO (12.7 g, 314.73 mmol) in CH2Cl2 (801 mL) under argon. The
mixture was stirred overnight at room temperature, then filtered
through Celite with suction. The filtrate was concentrated under
reduced pressure. The residue was purified by chromatography
(hexane/EtOAc, 2:1) to give the desired product (18.0 g,
55.97 mmol, 78%): 1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.49
(m, 2H), 8.25 (m, 2H), 3.56 (m, 1H), 1.51 (m, 1H), 1.41 (m, 1H),
1.18 ppm (m, 2H).
1
uct (3.17 g, 11.3 mmol, 86%): H NMR (400 MHz, [D6]DMSO, 300 K):
d=7.48 (m, 2H), 6.68 (m, 2H), 3.64 (m, 2H), 1.06 ppm (t, 3H).
(R,S)-S-Ethyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-methylpropyl]amino}-
5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)sulfoximide (33):
A solution of HCl in dioxane (4n; 0.36 mL) was added to (R,S)-S-(4-
(R,S)-S-(4-Aminophenyl)-S-cyclopropyl-N-(trifluoroacetyl)sulfoxi-
mide (47): 10% Pd-C (50% moisture; 1.40 g) was added to a solu-
tion of (R,S)-S-cyclopropyl-S-(4-nitrophenyl)-N-(trifluoroacetyl)sulfox-
imide (6.9 g, 21.44 mmol) in EtOH (214 mL) and THF (39 mL), and
the mixture was hydrogenated for 1 h under atmospheric pressure
at 258C. Additional 10% Pd-C (50% moisture; 1.40 g) was added,
and the mixture was hydrogenated for a further 4.5 h under at-
mospheric pressure. The mixture was filtered, 10% Pd-C (50%
moisture; 1.40 g) was added to the filtrate, and the mixture was
hydrogenated for 45 min. The mixture was filtered and the filtrate
was concentrated under reduced pressure to give the desired
product (5.8 g, 19.91 mmol, 93%): 1H NMR (400 MHz, [D6]DMSO,
300 K): d=7.53 (m, 2H), 6.71 (m, 2H), 6.40 (br, 2H), 3.21 (m, 1H),
1.28 (m, 2H), 1.08 ppm (m, 2H).
aminophenyl)-S-ethyl-N-(trifluoroacetyl)sulfoximide
(400 mg,
1.43 mmol) and (2R,3R)-3-{[2-chloro-5-(trifluoromethyl)pyrimidin-4-
yl]amino}butan-2-ol (37; 385 mg, 1.43 mmol) in MeCN (7.0 mL), and
the mixture was stirred for 4.5 h at 608C, then evaporated to dry-
ness. MeOH (19.0 mL) and K2CO3 (608 mg, 4.40 mmol) were added,
and the mixture was stirred for 1 h at room temperature. The re-
sulting mixture was diluted with saturated aqueous NaCl solution
and extracted with EtOAc (2ꢃ). The combined organic phases were
dried (Na2SO4), filtered and concentrated by evaporation to give
1
the desired product (590 mg, 1.41 mmol, 98%): H NMR (400 MHz,
[D6]DMSO, 300 K): d=10.10 (s, 1H), 8.30 (s, 1H), 7.96 (m, 2H), 7.78
(m, 2H), 6.10 (d, 1H), 5.11 (d, 1H), 4.19 (m, 1H), 4.00 (s, 1H), 3.82
(m, 1H), 3.07 (q, 2H), 1.24 (d, 3H), 1.06 ppm (m, 6H); ESI-HRMS: m/
z [M+H]+ calcd for C17H23N5O2SF3: 418.1525, found: 418.1519.
(2R,3R)-3-(Benzyloxy)butan-2-ol: Potassium tert-butylate (5.00 g,
44.6 mmol) was added to a solution of (2R,3R)-butane-2,3-diol
(4.00 g, 44.4 mmol) in THF (300 mL) at room temperature, and the
mixture was held at reflux for 15 min. The mixture was cooled to
~508C and benzyl bromide (5.3 mL, 44.6 mmol) was added. The re-
sulting mixture was held at reflux for 3 h, and was then stirred
overnight at room temperature. The mixture was diluted with
EtOAc and concentrated aqueous NaCl solution, and then washed
with aqueous HCl solution (1n; 1ꢃ) and concentrated aqueous
NaCl solution (2ꢃ). The organic phase was dried (Na2SO4), filtered
and concentrated under reduced pressure. The residue was puri-
fied by chromatography (hexane/EtOAc, 1:1) to give the desired
product (3.41 g, 18.9 mmol, 43%): 1H NMR (400 MHz, [D6]DMSO,
300 K): d=7.35 (m, 4H), 7.28 (m, 1H), 4.52 (m, 3H), 3.67 (m, 1H),
3.37 (m, 1H), 1.05 (d, 3H), 1.01 ppm (d, 3H).
The mixture of diastereomers was separated into the single stereo-
isomers by preparative HPLC [column: Chiralpak AD-H 5 mm, 250ꢃ
20 mm; eluent: hexane/2-propanol, 6:4; buffer: hexane/0.1% DEA;
flow: 20.0 mLminꢀ1; detector: UV 280 nm; temperature: RT]: tR
(min): 6.2–6.8 (stereoisomer 1: compound 43), 7.2–8.9 (stereoiso-
mer 2: compound 44).
1-(Cyclopropylsulfanyl)-4-nitrobenzene (46): NaH (60%; 1.78 g,
44.6 mmol) was added in portions to a solution of cyclopropane-
thiol[37] (3.00 g, 40.5 mmol) in THF (100 mL)/Et2O (100 mL), and the
mixture was stirred for 30 min at room temperature. Then, 1-
fluoro-4-nitrobenzene (45; 6.00 g, 38.7 mmol) was added in por-
tions. The mixture was stirred for 2 h at 408C, then cooled and
poured into H2O. The resulting mixture was extracted with ben-
zene (3ꢃ). The combined organic phases were concentrated under
reduced pressure and the residue was purified by chromatography
(hexane/EtOAc, 95:5) to give the desired product (4.6 g, 23.6 mmol,
61%): 1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.12 (m, 2H), 7.54
(m, 2H), 2.35 (m, 1H), 1.16 (m, 2H), 0.61 ppm (m, 2H).
4-{[(1R,2R)-2-(Benzyloxy)-1-methylpropyl]oxy}-2-chloro-5-iodo-
pyrimidine (49): NaH (55%; 2.07 g) was added in portions to
a
stirred solution of (2R,3R)-3-(benzyloxy)butan-2-ol (8.55 g,
47.4 mmol) in Et2O (56 mL) at 08C. After 10 min, the ice bath was
removed, and the mixture was stirred for an additional 3 min at
room temperature. The suspension formed was added at 08C to
a solution of 2,4-dichloro-5-iodopyrimidine (48; 6.52 g, 23.7 mmol)
in MeCN (65 mL). The mixture was stirred for 4 h at 408C, then
dilute aqueous NaCl solution was added. The mixture was extract-
ed with EtOAc (2ꢃ). The combined organic phases were dried
(Na2SO4), filtered and concentrated. The residue was purified by
chromatography (hexane/EtOAc, 4:1) to give the desired product
(R,S)-1-(Cyclopropylsulfinyl)-4-nitrobenzene: Iron(III) chloride
(179 mg, 1.11 mmol) was added to a mixture of 1-(cyclopropylsul-
fanyl)-4-nitrobenzene (46; 7.20 g, 36.88 mmol) in MeCN (140 mL),
and the mixture was stirred for 15 min at room temperature. H5IO6
(9.25 g, 40.57 mmol) was added in portions at 258C. The mixture
was stirred for 30 min, then poured into a cooled, saturated aque-
ous Na2S2O3 solution under stirring. The mixture was extracted
with EtOAc (2ꢃ). The combined organic phases were dried
(Na2SO4), filtered and concentrated. The residue was purified by
chromatography (hexane/EtOAc, 1:1) to give the desired product
(5.93 g, 28.07 mmol, 76%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
1
(4.12 g, 9.8 mmol, 41%): H NMR (400 MHz, [D6]DMSO, 300 K): d=
8.78 (s, 1H), 7.29 (m, 5H), 5.27 (m, 1H), 4.64 (d, 1H), 4.53 (d, 1H),
3.73 (m, 1H), 1.30 (d, 3H), 1.19 ppm (d, 3H).
4-{[(1R,2R)-2-(Benzyloxy)-1-methylpropyl]oxy}-2-chloro-5-(tri-
fluoromethyl)pyrimidine
(50):
Copper(I)
iodide
(3.82 g,
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&17
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
20.0 mmol), KF (0.97 g, 16.7 mmol) and (trifluoromethyl)trimethylsi-
lane (2.47 mL, 16.7 mmol) were added to a stirred solution of 4-
{[(1R,2R)-2-(benzyloxy)-1-methylpropyl]oxy}-2-chloro-5-iodopyrimi-
dine (49; 4.66 g, 11.1 mmol) in NMP (15.8 mL) and THF (15.8 mL) at
room temperature. The mixture was stirred for 5.5 h at 808C. After
cooling, the mixture was added to dilute aqueous NaCl solution,
and the resulting mixture was extracted with EtOAc (2ꢃ). The com-
bined organic phases were dried (Na2SO4), filtered and concentrat-
ed under reduced pressure. The residue was purified by chroma-
tography (hexane/EtOAc, 4:1) to give the desired product (2.17 g,
6.0 mmol, 54%): 1H NMR (400 MHz, [D6]DMSO, 300 K): d=8.81 (s,
1H), 7.21 (m, 5H), 5.40 (m, 1H), 4.57 (d, 1H), 4.42 (d, 1H), 3.70 (m,
1H), 1.28 (d, 3H), 1.13 ppm (d, 3H).
The mixture of diastereomers was separated into the single stereo-
isomers by preparative HPLC [column: Chiralpak IA 5 mm, 250ꢃ
30 mm; eluent: hexane/EtOH, 8:2; flow: 40.0 mLminꢀ1; detector:
UV 254 nm; temperature: RT]: tR (min): 10.8–13.4 (stereoisomer 1:
compound 54), 13.6–18.5 (stereoisomer 2: BAY 1000394; mp:
1348C; specific optical rotation: ꢀ5.8ꢁ0.208 (DMSO, 589 nm,
208C).
(R)-3-{[2-Chloro-5-(trifluoromethyl)pyrimidin-4-yl]amino}-2-meth-
ylbutan-2-ol: (R)-3-Amino-2-methylbutan-2-ol (3.6 g, 35.0 mmol)
was added dropwise to a solution of 2,4-dichloro-5-(trifluorome-
thyl)pyrimidine (36; 7.6 g, 35.0 mmol) in MeCN (139 mL). NEt3
(9.7 mL, 70.1 mmol) was then added dropwise at 08C, and the mix-
ture was heated slowly overnight to room temperature, then was
stirred for a further 2 days at room temperature. The mixture was
added to a semi-concentrated aqueous NaCl solution and extract-
ed with EtOAc (2ꢃ). The combined organic phases were dried
(Na2SO4), filtered and concentrated by evaporation. The residue
was purified by preparative HPLC [column: XBridge C18 5 mm, 150ꢃ
20 mm; eluents: A: H2O/0.2% NH3, B: MeCN; gradient: 70% A+
30% B (2 min) !40% A+60% B (2–12 min) !1% A+99% B (12–
12.1 min); flow: 50.0 mLminꢀ1; detector: DAD (200–400 nm) TAC,
ESI+-MS (160–1000 m/z) TIC; temperature: RT; tR (min): 5.6–6.4] to
give the desired product (3.0 g, 10.7 mmol, 30%): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=8.42 (s, 1H), 6.52 (d, 1H), 5.01 (s,
1H), 4.10 (m, 1H), 1.11 ppm (m, 9H).
(R,S)-S-(4-{[4-{[(1R,2R)-2-(Benzyloxy)-1-methylpropyl]oxy}-5-(tri-
fluoromethyl)pyrimidin-2-yl]amino}phenyl)-S-cyclopropyl-N-(tri-
fluoroacetyl)sulfoximide (51): A solution of HCl in dioxane (4n;
0.96 mL) was added to 4-{[(1R,2R)-2-(benzyloxy)-1-methylpropy-
l]oxy}-2-chloro-5-(trifluoromethyl)pyrimidine (50; 1.39 g, 3.85 mmol)
and (R,S)-S-(4-aminophenyl)-S-cyclopropyl-N-(trifluoroacetyl)sulfoxi-
mide (47; 1.35 g, 4.62 mmol) in MeCN (18.8 mL), and the mixture
was stirred for 5 h at 808C. After cooling, the mixture was diluted
with EtOAc and washed with saturated aqueous NaHCO3 solution
and saturated aqueous NaCl solution, dried (Na2SO4), filtered and
concentrated under reduced pressure. The residue was purified by
chromatography (hexane/EtOAc, 4:1) to give the desired product
1
(1.32 g, 2.14 mmol, 56%): H NMR (400 MHz, [D6]DMSO, 300 K): d=
(R,S)-S-Cyclopropyl-S-(4-{[4-{[(R)-2-hydroxy-1,2-dimethylpropyl]a-
mino}-5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)sulfoxi-
mide: A solution of HCl in dioxane (4N; 0.34 mL) was added to
a mixture of (R,S)-S-(4-aminophenyl)-S-cyclopropyl-N-(trifluoroace-
tyl)sulfoximide (47; 400 mg, 1.37 mmol) and (R)-3-{[2-chloro-5-(tri-
10.71 (s, 1H), 8.84 (s, 1H), 8.08 (m, 2H), 7.93 (m, 2H), 7.26 (m, 5H),
5.52 (m, 1H), 4.62 (d, 1H), 4.51 (d, 1H), 3.78 (m, 1H), 3.35 (m, 1H),
1.37 (m, 5H), 1.16 ppm (m, 5H).
(R,S)-S-Cyclopropyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-methylpropy-
l]oxy}-5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)-N-(tri-
fluoroacetyl)sulfoximide (52): 10% Pd-C (1.64 g) was added to
fluoromethyl)pyrimidin-4-yl]amino}-2-methylbutan-2-ol
(388 mg,
1.37 mmol) in MeCN (6.0 mL), and the mixture was stirred for 3.5 h
at 608C and then was evaporated to dryness. MeOH (29.4 mL) and
K2CO3 (950 mg, 6.84 mmol) were added, and the mixture was
stirred for 1 h at room temperature. The mixture was diluted with
saturated aqueous NaCl solution and extracted with EtOAc (2ꢃ).
The combined organic phases were dried (Na2SO4), filtered and
concentrated by evaporation to give the desired product (600 mg,
a
solution of (R,S)-S-(4-{[4-{[(1R,2R)-2-(benzyloxy)-1-methylpropy-
l]oxy}-5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)-S-cycloprop-
yl-N-(trifluoroacetyl)sulfoximide (51; 1.31 g, 2.12 mmol) in EtOH
(66 mL), and the mixture was hydrogenated under atmospheric
pressure at room temperature. The mixture was filtered and con-
centrated under reduced pressure to give the desired product
1
1
(0.88 g, 1.67 mmol, 79%): H NMR (400 MHz, [D6]DMSO, 300 K): d=
1.35 mmol, 98%): H NMR (400 MHz, [D6]DMSO, 300 K): d=10.08 (s,
10.65 (s, 1H), 8.58 (s, 1H), 8.04 (m, 2H), 7.89 (m, 2H), 5.28 (m, 1H),
4.86 (d, 1H), 3.82 (m, 1H), 3.35 (m, 1H), 1.45 (m, 5H), 1.15 ppm (m,
5H).
1H), 8.30 (s, 1H), 7.94 (m, 2H), 7.80 (m, 2H), 6.07 (d, 1H), 4.95 (s,
1H), 4.16 (m, 1H), 4.02 (s, 1H), 2.62 (m, 1H), 1.20 (m, 6H), 1.10 (m,
4H), 0.89 ppm (m, 3H).
(R,S)-S-Cyclopropyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-methylpropy-
l]oxy}-5-(trifluoromethyl)pyrimidin-2-yl]amino}phenyl)sulfoxi-
mide (53): K2CO3 (1130 mg, 8.20 mmol) was added to (R,S)-S-cyclo-
propyl-S-(4-{[4-{[(1R,2R)-2-hydroxy-1-methylpropyl]oxy}-5-(trifluoro-
methyl)pyrimidin-2-yl]amino}phenyl)-N-(trifluoroacetyl)sulfoximide
(52; 863 mg, 1.64 mmol) in MeOH (35 mL), and the mixture was
stirred for 1.5 h at room temperature. The resulting mixture was di-
luted with saturated aqueous NaCl solution and extracted with
EtOAc (3ꢃ). The combined organic phases were dried (Na2SO4), fil-
tered and concentrated under reduced pressure to give the de-
sired product (709 mg, 1.64 mmol, 99%): 1H NMR (400 MHz,
[D6]DMSO, 300 K): d=10.50 (s, 1H), 8.59 (s, 1H), 7.94 (m, 2H), 7.84
(m, 2H), 5.32 (m, 1H), 4.91 (d, 1H), 4.07 (s, 1H), 3.86 (m, 1H), 2.63
(m, 1H), 1.30 (d, 3H), 1.11 (m, 4H), 0.91 ppm (m, 3H); 13C NMR
(100 MHz, [D6]DMSO, 300 K): d=166.1, 161.1, 157.0, 143.2, 136.7,
128.6, 125.1, 122.4, 119.2, 76.7, 67.1, 34.3, 18.1, 14.3, 5.9, 4.9 ppm;
ESI-HRMS: m/z [M+H]+ calcd for C18H22N4O3SF3: 431.1365, found:
431.1366.
The mixture of diastereomers was separated into the single stereo-
isomers by preparative HPLC [column: Chiralpak AD-H 5 mm, 250ꢃ
20 mm; eluent: hexane/2-propanol, 6:4; buffer: hexane/0.1% DEA;
flow: 20.0 mLminꢀ1; detector: UV 280 nm; temperature: RT]: tR
(min): 5.1–6.3 (stereoisomer 1), 8.0–10.8 (stereoisomer 2: com-
pound 55).
(R)-S-{4-[(5-Bromo-4-{[(R)-2-hydroxy-1,2-dimethylpropyl]amino}-
pyrimidin-2-yl)amino]phenyl}-S-cyclopropyl-N-(ethoxycarbonyl)-
sulfoximide: A solution of HCl in dioxane (4n; 0.07 mL) was
added to (R)-3-[(5-bromo-2-chloropyrimidin-4-yl)amino]-2-methyl-
butan-2-ol (988 mg, 3.35 mmol) and (R)-S-(4-aminophenyl)-S-cyclo-
propyl-N-(ethoxycarbonyl)sulfoximide[39] (750 mg, 2.79 mmol) in
butan-1-ol (16.50 mL) and MeOH (1.65 mL), and the mixture was
stirred for 3 days at 608C. After cooling, the mixture was evaporat-
ed to dryness and purified by chromatography (CH2Cl2/EtOH, 9:1)
to give the desired product (319 mg, 0.61 mmol, 22%): 1H NMR
(400 MHz, [D6]DMSO, 300 K): d=9.96 (s, 1H), 8.13 (s, 1H), 7.92 (m,
2H), 7.73 (m, 2H), 6.28 (d, 1H), 4.05 (m, 1H), 3.84 (m, 2H), 2.96 (m,
1H), 1.05 ppm (m, 16H).
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&18
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
(R)-S-{4-[(5-Bromo-4-{[(R)-2-hydroxy-1,2-dimethylpropyl]amino}-
pyrimidin-2-yl)amino]phenyl}-S-cyclopropylsulfoximide (56):
line·HCl (559 mg, 2.69 mmol) in MeCN (11 mL) was stirred for 16 h
at 808C. After cooling, the mixture was diluted with EtOAc and
washed with saturated aqueous NaHCO3 solution and saturated
aqueous NaCl solution, dried (Na2SO4), filtered and concentrated
under reduced pressure. The residue was purified by chromatogra-
phy (hexane/EtOAc, 1:1) to give the desired product (770 mg,
A
freshly prepared solution of sodium ethanolate in EtOH (1.5m;
2.0 mL) was added to a solution of (R)-S-{4-[(5-bromo-4-{[(R)-2-hy-
droxy-1,2-dimethylpropyl]amino}pyrimidin-2-yl)amino]phenyl}-S-cy-
clopropyl-N-(ethoxycarbonyl)sulfoximide (319 mg, 0.61 mmol) in
EtOH (4.3 mL), and the mixture was stirred for 18 h at 608C. After
cooling, the mixture was added to saturated aqueous NaCl solu-
tion and extracted with EtOAc (3ꢃ). The combined organic phases
were dried (Na2SO4), filtered and concentrated by evaporation.
Final recrystallization (CH2Cl2/EtOAc) gave the desired product
(215 mg, 0.47 mmol, 77%): 1H NMR (400 MHz, [D6]DMSO, 300 K):
d=9.68 (s, 1H), 8.08 (s, 1H), 7.86 (m, 2H), 7.70 (m, 2H), 6.07 (d,
1H), 4.82 (s, 1H), 4.05 (m, 1H), 3.93 (s, 1H), 2.55 (m, 1H), 1.15 (m,
6H), 1.10 (m, 3H), 1.03 (m, 1H), 0.83 ppm (m, 3H); analytical HPLC
[column: Chiralpak AD-H 5 mm, 150ꢃ4.6 mm; eluent: hexane/EtOH,
8:2; flow: 1.0 mLminꢀ1; detector: PDA 280 nm; temperature: 258C]:
tR (min): 11.64.
1
1.55 mmol, 69%): H NMR (400 MHz, [D6]DMSO, 300 K): d=10.55 (s,
1H), 8.58 (s, 1H), 7.95 (m, 2H), 7.83 (m, 2H), 7.22 (m, 5H), 5.48 (m,
1H), 4.57 (d, 1H), 4.46 (d, 1H), 3.72 (m, 1H), 3.12 (s, 3H), 1.31 (d,
3H), 1.15 ppm (d, 3H).
(2R,3R)-3-{[2-{[4-(Methylsulfonyl)phenyl]amino}-5-(trifluorome-
thyl)pyrimidin-4-yl]oxy}butan-2-ol (58): 10% Pd-C (152 mg) was
added to a solution of 4-{[(2R,3R)-3-(benzyloxy)butan-2-yl]oxy}-N-[4-
(methylsulfonyl)phenyl]-5-(trifluoromethyl)pyrimidin-2-amine
(750 mg, 1.51 mmol) in EtOH (20 mL), and the mixture was hydro-
genated under atmospheric pressure at room temperature for
30 min. Additional 10% Pd-C (152 mg) was added, and the mixture
was hydrogenated for 90 min. Additional 10% Pd-C (152 mg) was
again added, and the mixture was hydrogenated for 60 min. Final-
ly, additional 10% Pd-C (152 mg) was added, and the mixture was
hydrogenated for 15 min. The mixture was filtered and the filtrate
was concentrated under reduced pressure to give the desired
product (512 mg, 1.26 mmol, 83%): 1H NMR (400 MHz, [D6]DMSO,
300 K): d=10.55 (s, 1H), 8.57 (s, 1H), 7.96 (m, 2H), 7.83 (m, 2H),
5.27 (m, 1H), 4.86 (d, 1H), 3.82 (m, 1H), 3.14 (s, 3H), 1.25 (d, 3H),
1.07 ppm (d, 3H).
(R,S)-S-{4-[(5-Bromo-4-{[(2R,3R)-2-hydroxy-1-methylpropyl]oxy}-
pyrimidin-2-yl)amino]phenyl}-S-cyclopropyl-N-(trifluoroacetyl)-
sulfoximide: A solution of HCl in dioxane (4n; 1.5 mL) was added
to a mixture of (2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)oxy]bu-
tan-2-ol (845 mg, 3.00 mmol) and (R,S)-S-(4-aminophenyl)-S-cyclo-
propyl-N-(trifluoroacetyl)sulfoximide (47; 877 mg, 3.00 mmol) in
MeCN (13.1 mL), and the mixture was stirred for 5 h at 808C. Fur-
ther
(2R,3R)-3-[(5-bromo-2-chloropyrimidin-4-yl)oxy]butan-2-ol
(422 mg, 1.50 mmol) was added, and the mixture was stirred at
808C. After 3 h, additional HCl in dioxane (4n; 0.75 mL) was
added, and the mixture was stirred further at 808C. After 33 h, ad-
ditional (R,S)-S-(4-aminophenyl)-S-cyclopropyl-N-(trifluoroacetyl)sul-
foximide (47; 175 mg, 0.60 mmol) was added, and the mixture was
stirred finally for 18 h at 808C. After cooling, the mixture was con-
centrated and the residue was purified by chromatography
(CH2Cl2/EtOH, 9:1) to give the desired product (740 mg, 1.38 mmol,
Acknowledgements
The authors thank K. Sauvageot-Witzku, R. Golde, M. Mçwes, P.
Friese, M. Prior, M. Skalmowski, K.-D. Joachim, A. Bieseke, K. Bur-
meister, K. Kauffeldt and C. Wegner for technical support, O.
Schenk for HPLC separations, J. Lorenz and G. Depke for analyti-
cal support, C. Gambietz for help with the sulfoximine nomencla-
ture, and N. Heinrich for design of the table of contents graphic.
1
38%): H NMR (400 MHz, [D6]DMSO, 300 K): d=10.31 (s, 1H), 8.44
(s, 1H), 8.01 (m, 2H), 7.84 (m, 2H), 5.19 (m, 1H), 4.88 (d, 1H), 3.81
(m, 1H), 3.31 (m, 1H), 1.40 (m, 1H), 1.29 (m, 4H), 1.08 ppm (m,
5H).
(R,S)-S-{4-[(5-Bromo-4-{[(2R,3R)-2-hydroxy-1-methylpropyl]oxy}-
Keywords: antitumor agents · biological activity · CDK ·
medicinal chemistry · sulfoximines
pyrimidin-2-yl)amino]phenyl}-S-cyclopropylsulfoximide:
K2CO3
(950 mg, 6.84 mmol) was added to a solution of (R,S)-S-{4-[(5-
bromo-4-{[(2R,3R)-2-hydroxy-1-methylpropyl]oxy}pyrimidin-2-yl)ami-
no]phenyl}-S-cyclopropyl-N-(trifluoroacetyl)sulfoximide
(735 mg,
[1] J. Wesierska-Gadek, I. Chamrad, V. Krystof, Future Med. Chem. 2009, 1,
1561–1581.
[2] D. Parry, T. Guzi, F. Shanahan, N. Davis, D. Prabhavalkar, D. Wiswell, W.
Seghezzi, K. Paruch, M. P. Dwyer, R. Doll, A. Nomeir, W. Windsor, T. Fisch-
mann, Y. Wang, M. Oft, T. Chen, P. Kirschmeier, E. M. Lees, Mol. Cancer
Ther. 2010, 9, 2344–2352.
[3] D. W. Fry, P. J. Harvey, P. R. Keller, W. L. Elliott, M. Meade, E. Trachet, M.
Albassam, X. Zheng, W. R. Leopold, N. K. Pryer, P. L. Toogood, Mol.
Cancer Ther. 2004, 3, 1427–1438.
[4] S.-H. Chao, K. Fujinaga, J. E. Marion, R. Taube, E. A. Sausville, A. M. Send-
erowicz, B. M. Peterlin, D. H. Price, J. Biol. Chem. 2000, 275, 28345–
28348.
[5] C. Albanese, R. Alzani, N. Amboldi, N. Avanzi, D. Ballinari, M. G. Brasca,
C. Festuccia, F. Fiorentini, G. Locatelli, W. Pastori, V. Patton, F. Roletto, F.
Colotta, A. Galvani, A. Isacchi, J. Moll, E. Pesenti, C. Mercurio, M. Ciomei,
Mol. Cancer Ther. 2010, 9, 2243–2254.
[6] G. Siemeister, U. Luecking, C. Wagner, K. Detjen, C. McCoy, K. Bosslet,
Biomed. Pharmacother. 2006, 60, 269–272.
1.37 mmol) in MeOH (29 mL), and the mixture was stirred for 1.5 h
at room temperature. The mixture was diluted with saturated
aqueous NaCl solution and extracted with EtOAc (3ꢃ). The com-
bined organic phases were dried (Na2SO4), filtered and concentrat-
1
ed to give the desired product (607 mg, 1.37 mmol, 99%): H NMR
(400 MHz, [D6]DMSO, 300 K): d=10.08 (s, 1H), 8.40 (s, 1H), 7.86 (m,
2H), 7.75 (m, 2H), 5.19 (m, 1H), 4.87 (d, 1H), 3.97 (s, 1H), 3.82 (m,
1H), 2.56 (m, 1H), 1.25 (d, 3H), 1.09 (d, 3H), 1.05 (m, 1H), 0.86 ppm
(m, 3H).
The mixture of diastereomers was separated into the single pure
stereoisomers by preparative HPLC [column: Chiralpak IC 5 mm,
250ꢃ20 mm; eluent: hexane/EtOH, 7:3; flow: 20.0 mLminꢀ1; detec-
tor: UV 280 nm; temperature: RT]: tR (min): 9.6–11.2 (stereoisomer
1), 11.3–14.9 (stereoisomer 2: compound 57).
4-{[(2R,3R)-3-(Benzyloxy)butan-2-yl]oxy}-N-[4-(methylsulfonyl)-
phenyl]-5-(trifluoromethyl)pyrimidin-2-amine: A mixture of 4-
{[(1R,2R)-2-(benzyloxy)-1-methylpropyl]oxy}-2-chloro-5-(trifluorome-
thyl)pyrimidine (50; 810 mg, 2.25 mmol) and 4-(methylsulfonyl)ani-
[7] S. Lapenna, A. Giordano, Nat. Rev. Drug Discovery 2009, 8, 547–566.
[8] E. A. Musgrove, C. E. Caldon, J. Barraclough, A. Stone, R. L. Sutherland,
Nat. Rev. Cancer 2011, 11, 558–572.
[9] J. Schamberger, M. Grimm, A. Steinmeyer, A. Hillisch, Drug Discovery
Today 2011, 16, 636–641.
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&19
&
ÞÞ
These are not the final page numbers!

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[10] U. Lꢁcking, G. Siemeister, M. Schꢂfer, H. Briem, M. Krꢁger, P. Lienau, R.
Jautelat, ChemMedChem 2007, 2, 63–77.
[11] M. Anderson, J. F. Beattie, G. A. Breault, J. Breed, K. F. Byth, J. D. Culshaw,
R. P. A. Ellston, S. Green, C. A. Minshull, R. A. Norman, R. A. Pauptit, J.
Stanway, A. P. Thomas, P. J. Jewsbury, Bioorg. Med. Chem. Lett. 2003, 13,
3021–3026.
[12] S. Verma, D. Nagarathnam, J. Shao, L. Zhang, J. Zhao, Y. Wang, T. Li, E.
Mull, I. Enyedy, C. Wang, Q. Zhu, M. Altieri, J. Jordan, T.-T.-A. Dang, S.
Reddy, Bioorg. Med. Chem. Lett. 2005, 15, 1973–1977.
[31] J. D. Oslob, S. A. Heumann, C. H. Yu, D. A. Allen, S. Baskaran, M. Bui, E.
Delarosa, A. D. Fung, A. Hashash, J. Hau, S. Ivy, J. W. Jacobs, W. Lew, J.
Maung, R. S. McDowell, S. Ritchie, M. J. Romanowski, J. A. Silverman, W.
Yang, M. Zhong, T. Fuchs-Knotts, Bioorg. Med. Chem. Lett. 2009, 19,
1409–1412.
[32] M. Ishikawa, Y. Hashimoto, J. Med. Chem. 2011, 54, 1539–1554.
[33] J. Olsen, P. Seiler, B. Wagner, H. Fischer, T. Tschopp, U. Obst-Sander,
D. W. Banner, M. Kansy, K. Mꢁller, F. Diederich, Org. Biomol. Chem. 2004,
2, 1339–1352.
[13] T. Brumby, R. Jautelat, O. Prien, M. Schaefer, G. Siemeister, U. Luecking,
(Schering AG), WO 2002/096888, 2002.
[34] N. V. Kondratenko, O. A. Radchenko, L. M. Yagupol’ski, J. Org. Chem.
USSR 1984, 20, 2051–2052.
[14] E. H. Kerns, L. Di, Drug-like Properties: Concepts, Structure Design and
Methods, Academic Press, Burlington, 2008, pp. 276–286.
[15] W. F. Bayne, F. T. Tao, G. Rogers, L. C. Chu, F. Theeuwes, J. Pharm. Sci.
1981, 70, 75–81.
[16] J. S. Graham, R. Plummer, C. McCoy, K. Kowal, H. Wiesinger, K. Detjen, H.
Calvert, B. Wiedenmann, J. Cassidy, Eur. J. Cancer 2008, 44, 2162–2168.
[17] E. N. Scott, A. L. Thomas, L. R. Molife, S. Ahmed, S. Blagden, P. C. Fong,
K. Kowal, C. McCoy, H. Wiesinger, W. Steward, J. De Bono, Cancer Che-
mother. Pharmacol. 2009, 64, 425–429.
[18] W. T. Caldwell, E. C. Kornfeld, J. Am. Chem. Soc. 1942, 64, 1695–1698.
[19] H. Burton, W. A. Davy, J. Chem. Soc. 1948, 525–527.
[20] U. Luecking, M. Krueger, R. Jautelat, O. Prien, G. Siemeister, A. Ernst,
(Schering AG), WO 2003/076437, 2003.
[21] A. Casini, F. Abbate, A. Scozzafava, C. T. Supuran, Bioorg. Med. Chem.
Lett. 2003, 13, 2759–2763.
[35] This method also gave very good results in other imination reactions,
but was later abandoned due to safety concerns.
[36] H. Okamura, C. Bolm, Org. Lett. 2004, 6, 1305–1307.
[37] E. Block, A. Schwan, D. A. Dixon, J. Am. Chem. Soc. 1992, 114, 3492–
3499.
[38] G. K. S. Prakash, A. K. Yudin, Chem. Rev. 1997, 97, 757–786.
[39] U. Luecking, G. Siemeister, P. Lienau, R. Jautelat, J. Schulze, (Bayer Scher-
ing Pharma AG), EP 2179991, 2010.
[40] U. Luecking, G. Siemeister, P. Lienau, R. Jautelat, J. Schulze, (Bayer Scher-
ing Pharma AG), EP 2179992, 2010.
[41] a) G. Siemeister, U. Luecking, A. M. Wengner, P. Lienau, W. Steinke, C.
Schatz, D. Mumberg, K. Ziegelbauer, Mol. Cancer Ther. 2012, 11, 2265–
2273; b) U. Luecking, G. Siemeister, A. Wengner, (Bayer Schering
Pharma AG), DE 102010014427, 2011.
[42] Y. Pocker, J. T. Stone, Biochemistry 1967, 6, 668–678.
[43] M. Rarey, B. Kramer, T. Lengauer, G. Klebe, J. Mol. Biol. 1996, 261, 470–
489.
[22] A. Cleve, U. Luecking, C. West, H. Briem, G. Siemeister, M. Krueger, R.
Jautelat, P. Lienau, (Schering AG), WO 2006/034872, 2006.
[23] H. R. Bentley, E. E. McCermott, J. Pace, J. K. Whitehead, T. Moran, Nature
1949, 163, 675–676.
[44] SYBYL 6.7, Tripos Inc., 1699 S. Hanley Road, Suite 303, St. Louis, MO
63144-2913 (USA), 2000.
[24] M. Reggelin, C. Zur, Synthesis 2000, 1–64.
[45] P. A. Boriack, D. W. Christianson, J. Kingery-Wood, G. M. Whitesides, J.
Med. Chem. 1995, 38, 2286–2291.
[46] T. A. Halgren, J. Comput. Chem. 1996, 17, 490–519.
[47] J. Navaza, Acta Crystallogr. Sect. A 1994, 50, 157–163.
[48] M. D. Winn, G. N. Murshudov, M. Z. Papiz, Methods Enzymol. 2003, 374,
300–321.
[25] H. H. Bailey, Chem.-Biol. Interact. 1998, 111 – 112, 239–254.
[26] P. J. O’Dwyer, T. C. Hamilton, F. P. LaCreta, J. M. Gallo, D. Kilpatrick, T.
Halbherr, J. Brennan, M. A. Bookman, J. Hoffman, R. C. Young, R. L.
Comis, R. F. Ozols, J. Clin. Oncol. 1996, 14, 249–256.
[27] I. R. Ager, A. C. Barnes, G. W. Danswan, P. W. Hairsine, D. P. Kay, P. D. Ken-
newell, S. S. Matharu, P. Miller, P. Robson, D. A. Rowlands, R. Tully, R.
Westwood, J. Med. Chem. 1988, 31, 1098–1115.
[28] R. Pothmann, Drugs Future 1982, 7, 478–480.
[29] C. R. Johnson, M. Haake, C. W. Schroeck, J. Am. Chem. Soc. 1970, 92,
6594–6598.
[30] U. Luecking, M. Krueger, R. Jautelat, G. Siemeister, (Schering AG), WO
2005/037800, 2005.
Received: February 27, 2013
Revised: April 19, 2013
Published online on && &&, 0000
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 20
&20
&
ÞÞ
These are not the final page numbers!