Synthesis, structure-property relationships and pharmacokinetic evaluation of ethyl 6-aminonicotinate sulfonylureas as antagonists of the P2Y12 receptor

Article abstract of DOI:10.1016/j.ejmech.2013.04.007

The present paper describes the development of a new series of P2Y ₁₂ receptor antagonists based on our previously reported piperazinyl urea series 1 (IC₅₀ binding affinity = 0.33 μM, aq solubility <0.1 μM, microsomal CLint (HLM) ≥300 μM/min/mg). By replacement of the urea functionality with a sulfonylurea group we observed increased affinity along with improved stability and solubility as exemplified by 47 (IC ₅₀ binding affinity = 0.042 μM, aq solubility = 90 μM, microsomal CLint (HLM) = 70 μM/min/mg). Further improvements in affinity and metabolic stability were achieved by replacing the central piperazine ring with a 3-aminoazetidine as exemplified by 3 (IC₅₀ binding affinity = 0.0062 μM, aq solubility = 83 μM, microsomal CLint (HLM) = 28 μM/min/mg). The improved affinity observed in the in vitro binding assay also translated to the potency observed in the WPA aggregation assay (47: 19 nM and 3: 9.5 nM) and the observed in vitro ADME properties translates to the in vivo PK properties observed in rat. In addition, we found that the chemical stability of the sulfonylureas during prolonged storage in solution was related to the sulfonyl urea linker and depended on the type of solvent and the substitution pattern of the sulfonyl urea functionality.

Full text of DOI:10.1016/j.ejmech.2013.04.007

Accepted Manuscript
Synthesis, Structure-Property Relationships and Pharmacokinetic Evaluation of Ethyl
6-Aminonicotinate Sulfonylureas as Antagonists of the P2Y12 Receptor
Peter Bach, Jonas Boström, Kay Brickmann, J.J.J. van Giezen, Robert D. Groneberg,
Darren M. Harvey, Michael O’Sullivan, Fredrik Zetterberg
PII:
S0223-5234(13)00239-0
10.1016/j.ejmech.2013.04.007
EJMECH 6117
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 13 December 2012
Revised Date: 1 April 2013
Accepted Date: 2 April 2013
Please cite this article as: P. Bach, J. Boström, K. Brickmann, J.J.J. van Giezen, R.D. Groneberg, D.M.
Harvey, M. O’Sullivan, F. Zetterberg, Synthesis, Structure-Property Relationships and Pharmacokinetic
Evaluation of Ethyl 6-Aminonicotinate Sulfonylureas as Antagonists of the P2Y12 Receptor, European
Journal of Medicinal Chemistry (2013), doi: 10.1016/j.ejmech.2013.04.007.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
1
Synthesis, Structure-Property Relationships and Pharmacokinetic
Evaluation of Ethyl 6-Aminonicotinate Sulfonylureas as Antagonists of
the P2Y12 Receptor
Peter Bach,a, Jonas Boström,a Kay Brickmann,a J. J. J. van Giezen,b Robert D.
Groneberg,c Darren M. Harvey,c Michael O’Sullivan,c and Fredrik Zetterberga,
aDepartment of Medicinal Chemistry, AstraZeneca R&D Mölndal, Pepparedsleden 1, S-43183 Mölndal,
Sweden
bDepartment of Bioscience, AstraZeneca R&D Mölndal, Pepparedsleden 1, S-43183 Mölndal, Sweden
cArray Biopharma, 3200 Walnut Street, Boulder, CO 80301, USA
1

ACCEPTED MANUSCRIPT
2
Graphical abstract
Abstract
The present paper describes the development of a new series of P2Y12 receptor antagonists based
on our previously reported piperazinyl urea series 1 (IC50 binding affinity = 0.33 µM, aq.
solubility <0.1µM, microsomal CLint (HLM) ≥300 µM/min/mg). By replacement of the urea
functionality with a sulfonylurea group we observed increased affinity along with improved
stability and solubility as exemplified by 47 (IC50 binding affinity = 0.042 µM, aq solubility = 90
µM, microsomal CLint (HLM) = 70 µM/min/mg). Further improvements in affinity and
metabolic stability was achieved by replacing the central piperazine ring with an 3-
aminoazetidine as examplified by 3 (IC50 binding affinity = 0.0062 µM, aq. solubility = 83 µM,
microsomal Clint (HLM) = 28 µM/min/mg). The improved affinity observerved in the in vitro
binding assay also translated to the potency observed in the WPA aggregation assay (47: 19 nM
and 3: 9.5 nM) and the observed in vitro ADME properties translates to the in vivo PK properties
observed in rat. In addition, we found that the chemical stability of the sulfonylureas during
prolonged storage in solution was related to the sulfonyl urea linker and depended on the type of
solvent and the substitution pattern of the sulfonyl urea functionality.

ACCEPTED MANUSCRIPT
3
Keywords
P2Y12 receptor, anti-platelet, sulfonylureas, solubility, stability.
1. Introduction
The platelet P2Y12 receptor, also known as the ADP receptor, plays an important role in the
amplification phase of platelet aggregation. P2Y12 is a 7-transmembrane, G-protein-coupled
receptor that is activated by adenosine diphosphate (ADP) [1]. When platelets are stimulated, they
release ADP from their dense-granules. ADP interacts with the P2Y12 receptor and causes a
down-regulation of intracellular adenylyl-cyclase activity [2]. This results in decreased cyclic-
AMP levels and prolonged calcium signalling which both stabilize the formed aggregates. The
important role of the P2Y12 receptor in platelet function makes it an attractive target for the
development of novel anti-platelet therapies [3].
The first class of compounds to show benefit in clinical studies were thienopyridine pro-drugs
like ticlopidine and clopidogrel, whose active metabolites bind irreversibly to the receptor [4].
However, preclinical data suggested that reversible binding would not only lead to a faster off-set
of effect, but also improved separation between the anti-thrombotic effect and bleeding risk [5, 6].
The clinical benefits of ticagrelor, the first reversibly binding, direct-acting P2Y12 antagonist,
were demonstrated in the phase III PLATelet inhibition and patient Outcomes (PLATO) trial; in
comparison with clopidogrel, ticagrelor significantly reduced the rate of the primary composite
endpoint of myocardial infarction, stroke, and death from vascular causes [7]. Ticagrelor was
developed via a medicinal chemistry program starting from adenosine triphosphate (ATP), the
natural antagonist of the P2Y12 receptor [8]. Other series of P2Y12 antagonists are for example

ACCEPTED MANUSCRIPT
4
piperazinyl-glutamate-pyridines [9], thienopyrimidines [10], anthroquinones [11], adenosine
analogs [12], and dinucleoside polyphosphates and nucleotides [13].
The present paper describes the discovery of a novel series of ethyl 6-aminonicotinate
sulfonylureas as potent antagonists of the P2Y12 receptor. The sulfonylureas were developed from
our recently reported series of piperazinyl-pyridine ureas [14], exemplified by compound 1
(Figure 1). Many compounds in the urea series had potencies in the submicromolar range, but the
urea compounds generally suffered from low solubility and low metabolic stability in liver
microsomes. Compounds to be tested in vivo by oral administration must be dissolved in order to
be absorbed from the intestinal fluid, thus low solubility often results in low systemic exposure
and poor in vivo activity [15]. Issues with low solubility can be adressed at different stages in the
drug discovery process. In the early stage, one strategy is molecular design, in the later stage
formulation can be an option, however neither of these have any guarantee of succes. Our strategy
to use molecular design to increase the solubility ultimately led to the discovery of the series of
sulfonyl ureas presented here. Likewise, our attempts to increase the microsomal stability of the
novel series by replacement of the ethyl ester functionality in structures like 2 and 3 will be
described.
Compounds in the sulfonylurea series are composed of a pyridine moiety (A), that is substituted
in the 6-position with a cyclic amine (B ring), exemplified by a piperazine 2 (Figure 1) and an
azetidine 3, but also other mono- and bicyclic B-rings have been incorporated into the structure.
A sulfonyl urea linker connects the A-B system to an aryl group (C). From a synthetic
perspective, compounds like 2 and 3 were attractive target molecules since building blocks A, B
and C could be cou pled by efficient parallel synthesis procedures.
[insert Figure 1]

ACCEPTED MANUSCRIPT
5
In the present study, SAR (structure-activity relationships) arising from variations of the different
parts of the molecules is presented. Firstly, the variations of the linker that led to the discovery of
the sulfonyl urea linker will be described. To study the sulfonyl ureas further, three different A-
rings (all ethyl nicotinates) with variations of the pyridine 2- and 5-substituents have been
employed. Variations of the B-ring, based on shape matching to the azetidinyl and piperazinyl
compounds, were made to test the effects of changing ring type, size, rigidity, and substitution
pattern, while substituted phenyls, benzyls and 5-chloro-2-thienyl were introduced as C-rings.
Then follows an investigation of the chemical stability in solution of the sulfonyl ureas. Finally,
in vivo PK data in rat on six selected compounds are presented.
2. Chemistry
Four different 6-chloronicotinic acid or ester building blocks (A-rings) were used as starting
materials (Figure 2). The 2-CF3/5-CN building block (4) was described previously [14], the 2-
H/5-Cl building blocks (5 and 6) were commercially available, and the 2-CH3/5-CN building
block (7) was made by a modified literature procedure [16, 17] via the corresponding pyridone
[18]. The 6-chloro nicotinates 4, 5, and 7 were reacted with an amine (B-ring) to give 6-amino
nicotinates (A-B). Subsequently, these were treated with (derivatives of) sulfonyl isocyanates to
form sulfonyl ureas. Alternatively, a more convergent route was used, in which a B-ring and a C-
ring were joined via a sulfonyl urea linker before being coupled with an A-ring.
[insert Figure 2]

ACCEPTED MANUSCRIPT
6
A synthetically efficient method to introduce the 5-chloro-2-thienyl C-ring was needed. The 5-
chlorothiophene-2-sulfonamide (8, Scheme 1) was commercially available, however its
corresponding sulfonyl isocyanate was not. Consequently, chemistry procedures were developed
to convert 8 into the corresponding sulfonyl isocyanate or synthetic equivalents thereof.
Activation of 8 with N,N’-disuccinimide carbonate (DSC) [19], N,N'-carbonyldiimidazole (CDI),
or n-butyl isocyanate/phosgene [20] generated intermediates that were reacted in situ.
Alternatively, reaction of 8 with Troc-Cl (2,2,2-trichloroethyl chloroformate) [21] gave a stable
intermediate 9 that could be isolated and stored. Compound 9 became the reagent-of-choice due
to its easy preparation, handling, and the reasonably high yields obtained in reactions of 9 with A-
B systems.
[Insert Scheme 1]
Ethyl 2-CF3/5-CN/6-piperazinyl nicotinates were synthesized from the 6-chloro-nicotinate 4 as
outlined in Scheme 2 [22]. Compound 4 was treated with two different piperazines to form
piperazinyl-pyridines 10 and (±)-11, respectively. Treatment of 10 with the CDI-derivative of 8
gave 12, while treatment with commercially available sulfonyl isocyanates produced sulfonyl
ureas 13-18 in 18-84% yield. Treatment of (±)-11 with the CDI-derivative of 8 gave (±)-19 after
tert-butyl deprotection.
[Insert Scheme 2]
Ethyl 2-H/5-Cl/6-piperazinyl nicotinates were synthesized as outlined in Scheme 3. Urea
compound (±)-22 was prepared in three steps, starting from commerically available (±)-20.
Compounds 2, 24-26, and 28-32 were synthesized by treating the piperazinyl-pyridine 23 with

ACCEPTED MANUSCRIPT
7
different iso(thio)cyanates. Compound 27 was made by reaction of 23 with the CDI-derivative of
8. The nicotinic acid 33 was made by hydrolysis under basic conditions of ethyl nicotinate 2.
[Insert Scheme 3]
Ethyl 6-azetidinyl and 6-piperidinyl nicotinates featured by two N-H functionalities in the
sulfonylurea linker were synthesized as outlined in Scheme 4. Reaction of pyridines 5 and 7 with
tert-butyl azetidin-3-ylcarbamate or tert-butyl piperidin-4-ylcarbamate provided intermediates 34-
37. Boc-deprotection of 34 and 36 followed by reaction with Troc-derivative 9 produced 3 and
42, respectively, while Boc-deprotection of 34-37 followed by reaction with the appropriate
sulfonylisocyanates produced 38-41 and 43-45.
[Insert Scheme 4]
Compounds 47-53 with an ethyl 2-CH3/5-CN nicotinate A-ring were prepared as shown in
Scheme 5. 6-Chloro nicotinic ester 7 was coupled with a B-ring, followed by treatment of the
formed A-B system with the activated sulfonamide derivative of a C-ring. Reaction with the CDI-
derivative of 8 produced 47. Treatment with commercially available sulfonyl isocyanates gave
48, 49, and 53, while reaction with the Troc-derivative 9 provided 50, (±)-51, and 52.
[Insert Scheme 5]
Nicotinic esters 59-63 (Scheme 6) were prepared from 5,6-dichloronicotinic acid 6. Nucleophilic
aromatic substitution with piperazine in the 6-position of compound 6 with subsequent acid
catalyzed esterification gave 6-piperazinyl nicotinic ester intermediates 55-58. Compounds 55-57
were reacted with the CDI-derivative of 8 to give 59, 61, and 63, while 58 was reacted with the

ACCEPTED MANUSCRIPT
8
isocyanate derivative of 8 to produce 60. Given the low yields (7-57%) in the syntheses of 55-58,
an alternative esterification procedure was used in the synthesis of the i-propyl ester homolog 62.
This involved EDCI/HOBT activation of N-Boc-protected 6-piperazinyl nicotinic acid 64
followed by reaction with i-PrOH which gave the i-propyl ester 65 in high yield (92%), followed
by Boc-deprotection and coupling with Troc-derivative 9 to provide 62.
[Insert Scheme 6]
Ketone analogs 70 and 71 (Scheme 7) were synthesized via the Weinreb amide 67 by reaction
with propyl and cyclopropyl Grignard reagents, respectively, followed by deprotection of the
Boc-protected piperazine and reaction with 9. 3-Propyl sulfonyl analog 74 was made from
aminopyridine 72 while benzene analog 77 was made from the commercially available benzoic
acid derivative 75. In the latter case, nonaflate [23] served as a useful leaving group in a
Buchwald-Hartwig reaction between 76 and tert-butyl piperazine-1-carboxylate.
[Insert Scheme 7]
3. Results and discussion
Affinity in a binding assay and potency in a GTPγS functional assay were determined in a first
screen cycle [24]. The GTPγS functional assay is based on the principle that activation of a GPCR
(P2Y12) by an agonist (ADP) leads to conformational changes that favors binding of the receptor
to a G-protein. Thereby, an agonist-receptor-G-protein ternary complex is formed which induces
GDP bound on the G-protein α-subunit to exchange with GTPγ35S. The effect of an added
antagonist can thus be determined by the fraction of unbound GTPγ35S. Most compounds with an

ACCEPTED MANUSCRIPT
9
IC50-value lower than 0.2 µM in the GTPγS assay were characterized further by additional in vitro
screens. These included determination of the potency of the compounds in inhibiting fibrinogen-
induced aggregation in a washed platelet assay (WPA) [5] permeability in Caco-2 monolayers in
the apical to the basolateral direction (A to B) [25], and in vitro intrinsic clearance (metabolic
stability, CLint) in rat liver microsomes (RLM) and human liver microsomes (HLM) [26]. A
balancing of in vitro potency, permeability, microsomal stability, and structural diversity was
applied when selecting compounds for determination of PK parameters in vivo in rat. The
ultimate goal was that the selected set of compounds could be used for analyzing in vitro-in vivo
correlations and thereby enable the prediction of in vivo PK parameters for other compounds in
this series.
3.1 Improvement of solubility and study of the SAR of the linker
The sulfonylurea series was discovered from our efforts to improve the solubility of the ethyl 6-
aminonicotinate urea compounds [14], exemplified by 1 (Figure 3, Table 1). Structural changes
leading to reduced lipophilicity (78, 79) or increased conformational flexibility (80, 81) improved
the solubility compared to 1. Of these, only 80 gave a moderate increase (to 0.11 µM) in potency
compared to 1. Replacing the urea linker with thiourea (24), acylurea (25) or acylthiourea (26) did
not lead to any improvement in potency compared to 78. Comparison of thiourea 24 with urea 78
indicated that one strong hydrogen bond acceptor was needed for high binding affinity [27]. The
introduction of an additional strong hydrogen bond acceptor as in 25 led to lower affinity
compared to 82.

ACCEPTED MANUSCRIPT
10
Inspired by the fact that ADP, the natural agonist of P2Y12, contains an acidic side chain, attempts
were made to introduce acidic functionalities into structure 1. Introduction of a carboxylate as
exemplified by (±)-22 increased the solubility and maintained the affinity of compound 1.
Since sulfonamides had been reported as P2Y12 inhibitors [28], it was attempted to lower the pKa
of the urea linker of 1 by way of a sulfone. Thus, replacement of the urea linker of 1 with the
acidic sulfonylurea functionality, a bioisostere of the carboxylic acid [29], produced sulfonylurea
2.
Compound 2 showed a potency on par with urea 1 and provided a breakthrough in solving the
issue with low solubility. The high solubility of sulfonylurea 2 was a consequence of the
relatively low pKa of the NH proton in the sulfonylurea linker (pKa = 3.8, i.e. an acidity similar
to formic acid) which made the compound appear as an anion at pH 6.8, where solubility is
determined. Additional examples of an increase in solubility by replacement of the urea linker
with a sulfonylurea linker include the matched-pairs 28/83 (increase from 10 to 210 µM) and
30/84 (increase from <0.09 to 490 µM).
Compound 33, in which the ethyl ester of compound 2 had been replaced with the corresponding
carboxylic acid, was found to be inactive [30]. This was analogous to what was observed in the
urea series.
[Insert Figure 3]

ACCEPTED MANUSCRIPT
11
Table 1. Binding affinity, potency, solubility, and lipophilicity of compounds leading to sulfonyl urea 2
[31]. See Figure 3 for structures.a
Compd
No
Binding
IC50
Solubility
at pH6.8
(µM)
clogP
logD
GTPγS
IC50
pH7.4
(µM)
(µM)
0.33
0.18
0.68
12
0.68
0.22
2.2
<0.10
100
96
4.0
3.3
3.3
4.1
3.0
4.2
0.7
0.9
2.8
3.3
2.0
3.7
3.5
3.5
3.2
3.7
5.0
ND
ND
ND
ND
ND
ND
ND
<0.5
ND
3.7
1
2
(±)-22
24
25
26
28
30
33
78
79
80
81
82
83
84
3.7
ND
ND
ND
210b
490b
ND
40
9.9
5.3
2.4
0.99
0.26
0.30
ND
1.7
0.15
0.30
>33
0.89
0.92
0.47
1.3
2.6
90
0.11
0.51
1.7
2.2
5.0
15
5.0
2.4
ND
10b
ND
ND
ND
1.4
2.6
0.60
1.1
<0.09b
a ND = not determined.bSolubility at pH7.4.
3.2 SAR by structural variations of the A-, B- and C-rings
To understand the scope of the sulfonyl urea compounds, a library with structural variations of
the A-, B- and C-rings of compound 2 was synthesized (Table 2). Variations of the pyridine A-

ACCEPTED MANUSCRIPT
12
ring were made in the 2- and 5-positions and included the substitution patterns 2-CF3/5-CN, 2-
H/5-Cl, and 2-CH3/5-CN.
Pairwise comparison of the piperazinyls show that 2-CF3/5-CN pyridines had similar or higher
potency than the corresponding 2-H/5-Cl pyridines. In general, the 2-CF3/5-CN pyridines had
lower permeability in Caco-2 monolayers than the 2-H/5-Cl pyridines, by comparison of the
pairs 16/30, 17/31, and 18/32. In both piperazinyl sub-series 12-18 (2-CF3/5-Cl pyridines) and 27-
32 (2-H/5-Cl pyridines), monosubstitution on the phenyl C-ring in the 2- or 4-position with CH3,
Cl or F (4-position only) produced compounds with similar or lower potency than compound 2 in
both binding and GTPγS assays.
In the 3-CH3/5-CN pyridine sub-series 47-49, a benzyl C-ring (49) provided binding affinity and
potency in both the GTPγS and WPA assays similar to 5-chloro-2-thienyl (47), and 49 was the
only piperazinyl that has low clearance in HLM. In the azetidinyl (3 and 38-41) and piperidinyl
(42-45) sub-series the 5-Cl-2-thienyl remained the optimum C-ring to provide potency, as
exemplified by compounds 3 and 42 having GTPγS and WPA potencies similar to 47.
In general, compounds with a 5-Cl-thienyl C-ring had consistently high potency in WPA, but also
had lower permeability in Caco-2 monolayers than compounds of similar logD (compare 27/30,
47/49, and 3/38). Among the 2-H/5-Cl pyridines, azetidinyl 38 was similar in potency to the
comparable piperazinyl 2 while piperidinyl 43 had a remarkably lower potency than 2.
Comparison of pairs and triplets 3/42/47, 40/44, and 41/45/49, showed the azetidinyls and
piperazinyls to have similar logD, while the piperidinyls had logD 0.8-1.0 higher than the
azetidinyls which is the likely reason for the piperidinyls having higher permeability in Caco-2
monolayers. Overall, the azetidinyls were the sub-series with highest stability in HLM.

ACCEPTED MANUSCRIPT
13
Table 2. Effects on potency, solubility, logD, permeability in Caco-2 monolayers, and intrinsic clearance,
CLint, in rat liver microsomes and human liver microsomes by variations of the pyridine substituents R1
and R4, the central ring (B), and the sulfonyl urea substituent R6. a
O
O
O
R4
N
R4
N
R4
N
O
O
O
R1
N
R1
N
O
R1
N
O
O
R6
O
R6
O
R6
H
B
B
B
S
N
N
S
N
N
S
O
N
N
2, 12-18,
H
3, 38-41
H
O
42-45
H
H
O
O
27-32, 47-49
Compd R1
No
R4 Central
ring (B)
R6
Binding
IC50
WPA
IC50
logD
Caco-2 CLint
CLint
HLM
GTPγS
IC50
pH
7.4
A to B
(10-6
RLM
L/M/H L/M/H
(µM)
(µM)
(µM)
cm/s)
CF3 CN
CF3 CN
paz
paz
5-Cl-2-thienyl
Ph
0.079
0.17
0.43
0.69
0.12
0.068
0.23
0.18
0.090
0.15
0.87
0.30
0.52
0.37
0.039
0.22
0.38
0.47
0.090
0.097
0.21
0.22
0.10
0.26
0.99
0.30
0.38
0.47
0.038
0.12
ND
1.1 L (0.076) ND
ND
H
12
ND M (0.60)
H
ND
ND
H
13
14
15
16
17
18
2
CF3 CN
CF3 CN
CF3 CN
CF3 CN
CF3 CN
paz
2-CH3-Ph
2-Cl-Ph
4-CH3-Ph
4-Cl-Ph
4-F-Ph
ND
ND
1.3
ND
ND
ND
ND
H
paz
ND
paz
0.16
0.80
ND
M (0.22)
L (0.13)
paz
ND
H
H
paz
0.4 L (0.053)
ND M (0.28)
H
M
H
H
H
H
H
H
H
H
Cl
Cl
Cl
Cl
Cl
Cl
Cl
paz
Ph
1.1
H
paz
5-Cl-2-thienyl
2-CH3-Ph
2-Cl-Ph
4-CH3-Ph
4-Cl-Ph
4-F-Ph
0.26
ND
0.9
0.7
0.6
0.9
1.3
0.6
0.8
0.8
0.4
L (0.11)
ND
H
H
27
28
29
30
31
32
3
paz
H
H
paz
1.1
ND
ND
H
ND
H
paz
0.19
ND
H (2.5)
H (1.2)
M (0.54)
L (0.11)
H (3.9)
M (0.41)
paz
H
H
paz
ND
H
H
CH3 CN
Cl
CH3 CN
3-aze
3-aze
3-aze
5-Cl-2-thienyl 0.0063
0.025 0.0095
H
M
ND
L
H
Ph
Ph
0.43
0.38
0.19
ND
ND
L
38
39
0.062
0.19

ACCEPTED MANUSCRIPT
14
CH3 CN
CH3 CN
CH3 CN
3-aze
3-aze
4-pip
4-pip
4-pip
4-pip
paz
4-Cl-Ph
0.086
0.079
0.027
4.8
0.078
0.082
0.057
3.2
0.094
0.44
0.026
ND
1.1
1.1
1.6
1.5
2.1
1.9
H (1.2)
H (1.2)
H (6.9)
ND
M
M
H
L
M
H
40
41
42
43
44
45
47
48
49
Bn
5-Cl-2-thienyl
H
Cl
Ph
4-Cl-Ph
Bn
ND
H
ND
H
CH3 CN
CH3 CN
CH3 CN
CH3 CN
CH3 CN
0.19
0.26
0.21
ND
H (19)
H (30)
0.088
0.042
0.14
0.15
H
M
H
5-Cl-2-thienyl
Ph
0.034
0.13
0.019
ND
1.0 L (0.058)
H
paz
ND M (0.30) ND
0.9 M (0.37)
ND
L
paz
Bn
0.033
0.054
0.020
H
a paz = piperazinyl, 3-aze = 3-amino-azetidinyl, 4-pip = 4-amino-piperidinyl, WPA = washed platelet assay, in which
the ability of the compounds to inhibit fibrinogen-induced aggregation is determined, Caco-2 = permeability in the
apical (A) to basolateral (B) direction in adenocarcinoma cells from human colon, CLint = intrinsic clearance,
measured as metabolism in rat liver microsomes (RLM) and , human liver microsomes (HLM), L/M/H = low/
moderate/high. ND = not determined.
3.3 SAR of the central ring (B) and of the pyridine 3-substituent (R2)
Next, the B-ring was varied exploratively (Table 3). We expected that changes in the
conformation, rigidity, or size of the B-ring should modulate the binding mode, and thus
potentially increase the binding affinity, of the molecules. Low-energy conformations of
molecules with structurally different B-rings were shape-matched to compound 47 (Figure 4). As
seen from the figure, the compounds with alternative B-rings, that were selected for synthesis,
have functionalities that overlay the similar functionalities in 47. Compound (±)-19, featured by a
propanoic acid side chain on the B-ring, had potencies on level with 47 and was stable in HLM.
Introduction of a 7-membered B-ring in the form of a 1,4-diazepane (50) gave a 7-fold (to 0.63
µM) reduction in the binding affinity and a 15-fold (to 0.24 µM) reduction in the GTPγS potency
relative to 47. Compound (±)-51, featured by a 2-methyl substituent on the B-ring, had a potency
similar to 47. The compounds 52 and 53 with bi-cyclic B-rings both had lower potency than 47.

ACCEPTED MANUSCRIPT
15
[Insert Figure 4]
Table 3. Effects on binding affinity, potency, Caco-2 permeability, and intrinsic clearance in rat liver
microsomes and human liver microsomes by variations of the central ring.a
O
O
O
N
N
N
O
O
F
OH
S
Cl
Cl
Cl
N
N
N
N
N
N
H
S
H
N
S
N
N
F
F
N
N
47
S
(±)-19
O
O
O
O
O
O
O
O
N
O
O
O
Cl
N
N
H
H
S
S
N
N
N
S
(±)-51
50
S
O
O
O
O
O
O
N
O
N
O
N
O
Cl
N
N
N
N
H
H
N
S
53
N
N
S
S
52
O
O
O
O
O
O
Compd No
Binding
IC50
WPA
IC50
logD
pH
Caco-2
A to B
CLint
RLM
CLint
HLM
GTPγS
IC50
7.4
1.0
(10-6 cm/s)
L (0.058)
L/M/H
H
L/M/H
H
(µM)
(µM)
(µM)
0.042
0.034
0.019
47
(±)-19
50
0.095
0.63
0.064
2.9
0.11
0.24
0.037
0.69
0.58
0.031
ND
ND
0.9
1.6
ND
0.4
M (0.19)
ND
H
ND
H
L
ND
H
(±)-51
52
0.031
ND
M (0.31)
ND
ND
L
ND
L
0.24
ND
M (0.39)
53
a , WPA = washed platelet assay, in which the ability of the compounds to inhibit fibrinogen-induced aggregation is
determined, Caco-2 = permeability in the apical (A) to basolateral (B) direction in adenocarcinoma cells from human
colon, CLint = intrinsic clearance, measured as CYP450 catalyzed metabolism in rat liver microsomes (RLM) and
human liver microsomes (HLM), ), L/M/H = low/moderate/high. ND = not determined.

ACCEPTED MANUSCRIPT
16
Since the increased structural complexity (e.g. chirality) introduced by replacement of the
piperazine with other B-rings gave no significant increase in potency compared to compound 47,
the unsubstituted piperazine B-ring was retained in the subsequent variation of the pyridine 3-
ethoxycarbonyl substituent (R2) and replacement of the pyridine ring with a benzene ring (Table
4).
Minor structural variations of the pyridine 3-substituent were made, in part to study the SAR in
the pyridine 3-position and in part to potentially identify substituents that could provide
compounds with higher microsomal stability than the ethyl esters. Replacement of the 3-
ethoxycarbonyl substituent with 3-carboxamides had given compounds of low potency in the urea
series,14 thus carboxamides were not included in the present study. Instead, we included a number
of alkyl esters and ethyl ester isosteres in the form of two ketones and a sulfone.
Shortening of the ethyl ester in 27 to methyl ester 59 reduced the binding affinity 7-fold (to 0.65
µM) and the potency 32-fold (to 3.2 µM) in the GTPγS assay. Straight-chain elongations with n-
propyl (60) and n-butyl (61) esters or with an alkyl chain having branching away from the ester
functionality like in the iso-pentyl ester 63 gave potencies similar to the ethyl ester. Branching
close to the ester carbonyl like in the iso-propyl ester 62 reduced the binding affinity and GTPγS
potency 3-5 fold compared to the ethyl ester and clearance in RLM and HLM remained high. n-
Propyl and cyclopropyl ketone analogs 70 and 71 had 9-13 fold lower GTPγS potencies than
ethyl ester 27, however the ketones were stable in both RLM and HLM. Sulfone analog 74 had
more than a 100-fold lower GTPγS potency than 27. Replacement of the pyridine ring of 27 with
a benzene ring gave compound 77 that had a 7-fold lower binding affinity and a 18-fold lower
potency in the GTPγS assay compared to 27, which underlines the importance of the A-ring being
a pyridine rather than a benzene.
In summary, replacement of the pyridine 3-ethoxycarbonyl substituent and of the pyridine with
benzene did not improve the potency compared to compound 27. Although ketones 70 and 71 had

ACCEPTED MANUSCRIPT
17
a notably lower clearance in HLM, this was out-weighted by lower potencies, hence further
development was made with the 3-ethoxycarbonyl pyridines.
Table 4. Variations of the pyridine 3-ethoxycarbonyl substituent (R2) and replacement of pyridine (X = N)
with benzene (X = C).a
R2
Cl
N
Cl
X
H
S
N
N
S
O
O
O
Compd
No
R2
X
Binding
IC50
WPA
IC50
logD
Caco-2
A to B
CLint
RLM
CLint
HLM
GTPγS
IC50
pH7.4
(10-6 cm/s)
L/M/H
L/M/H
(µM)
(µM)
(µM)
-CO2Et
N
N
N
N
N
N
N
N
N
C
0.090
0.65
0.21
0.095
0.25
0.14
0.56
0.42
9.3
0.10
3.2
0.26
ND
ND
ND
ND
ND
ND
3.8
ND
0.9
1.6
2.3
1.5
2.7
1.3
1.2
ND
1.5
L (0.11)
ND
H
ND
ND
ND
H
H
ND
ND
H
27
59
60
61
62
63
70
71
74
77
-CO2Me
-CO2-n-Pr
0.24
0.15
0.53
0.11
0.89
1.3
ND
-CO2-n-Bu
-CO2-i-Pr
-CO2-i-Pn
-C(O)-n-Pr
-C(O)-cy-Pr
S(O)2-n-Pr
-CO2Et
M (0.50)
ND
H
M (0.44)
M (0.16)
L (0.12)
ND
ND
H
H
M
L
L
14
ND
ND
ND
H
ND
ND
0.66
1.8
ND
a
WPA = washed platelet assay, in which the ability of the compounds to inhibit fibrinogen-induced aggregation is
determined, Caco-2 = permeability in the apical (A) to basolateral (B) direction in adenocarcinoma cells from human
colon, CLint = intrinsic clearance, measured as metabolism in rat liver microsomes (RLM) and human liver
microsomes (HLM), L/M/H = low/moderate/high. ND = not determined.
-

ACCEPTED MANUSCRIPT
18
3.4 Stability of the sulfonyl ureas during prolonged storage in solution
During the course of the project we noticed that some of the sulfonyl ureas having a piperazine B-
ring were chemically instable during prolonged storage in acetonitrile (MeCN) or DMSO. For
example, 10% remained of 12 after three months and 52% remained of 16 after four months at 5
°C in MeCN. Mass analysis of the fragments showed that this instability was solely due to
cleavage of the sulfonyl urea linker, i.e. the ethyl ester functionality was stable under these
conditions. No instability was observed during storage at room temperature of any of the sulfonyl
ureas as solid materials. To investigate the stability during prolonged storage in solution of the
piperazine sub-series compared to the azetidine and piperidine sub-series, the stability of
solutions of two piperazines, 30 and 31, one azetidine, 38, and one piperidine, 43, in MeCN and
ethanol (EtOH), respectively, was studied by UV absorption in LC/MS during 21 days at room
temperature [32]. The pair 30/31 was chosen in order to disclose any electronic influence of the
C-ring (4-CH3-phenyl vs 4-Cl-phenyl) on the stability of the sulfonyl urea linker. EtOH and
MeCN were chosen as solvents to reveal any difference between stability in a nucleophilic
(EtOH) and a less nucleophilic (MeCN) solvent.
As seen from the graphs (Figure 5), solutions of the piperazines 30 and 31 were the least stable,
particularily in EtOH. During 21 days, the purity of 30 in EtOH decreased from 89% to 9% and in
MeCN from 89% to 80%, while the purity of 31 in EtOH decreased from 95% to 32% and in
MeCN from 95% to 74%. In marked contrast, the azetidine 38 was stable in both EtOH and
MeCN while the piperidine 43 was stable in MeCN and slightly instable in EtOH (decrease from
97% to 89% purity). Compound 30 (4-CH3, slightly electron rich phenyl) was less stable than 31
(4-Cl, electron poor phenyl) in EtOH, but more stable than 31 in MeCN. The latter observation is
in concord with the above-mentioned stability in MeCN of 16 (4-CH3, slightly electron rich
phenyl) vs 12 (5-Cl-2-thienyl, electron poor C-ring). Detected degradation products by LC-

ACCEPTED MANUSCRIPT
19
HRMS were the azetidinyl-, piperazinyl- or piperidinyl-pyridines, respectively, and, for solutions
in EtOH, the ethyl sulfonyl carbamates. No degradation of the nicotinic ethyl esters was observed.
The observation that the stability of sulfonyl ureas in EtOH depended on their substitution pattern
was also seen with tolbutamide derivatives [33]. In aqueous solution at neutral pH, where
sulfonamides (usual pKa ≈ 4) exist as anions, the stability may be higher since the formed anion
protects the linker against hydrolytic cleavage [34].
To conclude, long-time storage of solutions of piperidinyl and (most critically) piperazinyl
pyridines in acetonitrile and ethanol (and probably in other solvents as well) should be avoided.
[Insert Figure 5]
3.5 Property profiling including in vivo PK
A general trend of the sulfonyl urea compounds was high intrinsic clearance in RLM. In HLM,
the compounds showed more variation. This difference in intrinsic clearance reflected
interspecies differences in expression of metabolizing enzymes [35]. The metabolism could be
due to hydrolytic enzymes (esterases), CYP450, or other enzymes like oxidases. NADPH is
needed for CYP450 to function, so to uncover if the metabolism in RLM and HLM was CYP450
dependent, the metabolism of an array of compounds 2, 12, 13, 27, 38, 47, and 48 was determined
in assays from which NADPH had been excluded. No change in the degree of metabolism was
observed for these compounds. It was thus concluded that the metabolism in microsomes was not
CYP-dependent and was most likely due to the activity of esterases present in the liver
microsomes. The major metabolites of the ethyl esters 19 and 30 in RLM and HLM was
identified as the corresponding carboxylates. The interspecies difference in the activity of
esterases was underlined when testing the metabolic stability of compounds (±)-19, 27, 38, 39,

ACCEPTED MANUSCRIPT
20
and 41 in rat plasma and human plasma, respectively. In human plasma, the compounds were
stable [36] and showed a strong binding to human plasma protein with free fractions lower than
1.5% [37]. In rat plasma, the compounds were unstable and the corresponding carboxylate
metabolites could be identified by HRMS.
In vivo PK parameters in rat (Table 5) was determined for the six compounds 16, (±)-19, 27, 38,
39, and 41. This was done to understand if the microsomal clearance and Caco-2 permeability in
vitro could be correlated to the in vivo PK parameters of the compounds. Given our limited
ressources for in vivo PK determination it was attempted to include compounds that were diverse
with regards to structure and in vitro properties. For example was the moderately potent
compound 38 included due to its comparatively high permeability in Caco-2 monolayers.
Compounds (±)-19 and 41 had very low bioavailability (0.8-3%) while 16, 27, and 38 had
moderate (15-20%), and 39 comparatively high (39%) bioavailability. Both 16 and (±)-19 had a
moderate Caco-2 permeability, but 16 had a remarkably higher bioavailability (20% vs 0.8%),
despite being metabolically more labile than (±)-19. The high bioavailability of 39, explained by a
moderate Caco-2 permeability in combination with the lowest clearance (13 mL/min/kg),
contributed to 39 having the longest half-lives (1.7 h (iv) and 9.0 h (po)) of the six compounds.
The triad of compounds 38/39/41 had the same B-ring and linker and were thus particularly
interesting to compare. Compounds 38 and 39, that were equipotent in GTPγS, differed only in
the substitution pattern of the A-ring, and shifting from the 2-H, 5-Cl-pyridine (38) to the less
lipophilic 2-CH3, 5-CN-pyridine (39) led to reduced clearance (from 29 to 13 mL/min/kg) and
increased bioavailability (from 20 to 39%) and oral half-life (from 2.1 to 9.0 h). Compounds 39
and 41, that showed similar affinity and potency, differed only in the C-ring (phenyl vs benzyl)
and replacing the phenyl of 39 with the more lipophilic benzyl of 41 significantly increased the
clearance (from 13 to 31 mL/min/kg) and reduced both the bioavailability (from 39% to 3%) and
the oral half-life (from 9.0 h to 1.3 h).

ACCEPTED MANUSCRIPT
21
Overall was the observed clearance in vivo well-predicted by the intrinsic clearance in rat liver
microsomes, as seen from Table 5. It was not possible to predict the bioavailability from an
estimate using Caco-2 permeability and in vivo clearance data.
Table 5. Pharmacokinetic data in vivo (in rat) for six selected compounds.a
O
O
O
N
N
O
F
O
F
OH
S
Cl
N
N
N
N
H
H
N
S
F
F
N
N
F
F
N
S
(±)-19
16
O
O
S
O
O
O
O
O
O
Cl
R4
N
O
Cl
O
38: R2 = H, R4 = Cl, R6 = Ph
N
N
39: R2 = CH3, R4 = CN, R6 = Ph
R2
N
H
O
R6
O
N
N
41: R2 = CH3, R4 = CN, R6 = Bn
S
S
27
N
N
H
O
O
O
38, 39, 41
O
H
Compd N,
CL
t1/2
iv
N,
F (%)b
t1/2
WPA
IC50
logD
Caco-2
A to B
CLint
RLM
No
iv (mL/min/kg)
± SEM
po
± SEM
po
pH 7.4
(h)
(h)
(µM)
(10-6 cm/s)
L/M/H
16
(±)-19
27
2
2
1
2
1
1
89 (±2)
46 (±4)
36
1.5
0.62
0.92
0.17
1.7
2
2
2
3
3
2
20 (±10)
0.83 (±0.53)
15 (±0)
0.65
4.0
2.4
2.1
9.0
1.3
0.16
0.031
0.26
ND
1.3
ND
ND
0.8
0.4
1.1
M (0.22)
M (0.19)
L (0.11)
H (3.9)
H
H
H
29 (±2)
13
20 (±2)
ND
L
38
39 (±13)
3.0 (±1.1)
0.19
0.44
M (0.41)
H (1.2)
39
31
0.55
M
41
aIn vitro data are included for reference. CL = clearance, F = bioavailability, WPA = washed platelet assay, in which
the ability of the compounds to inhibit fibrinogen-induced aggregation is determined, Caco-2 = permeability in the
apical (A) to basolateral (B) direction in adenocarcinoma cells from human colon, CLint = intrinsic clearance,
measured as metabolism in rat liver microsomes (RLM), L/M/H = low/moderate/high. ND = not determined. bThe
compounds were administered as solutions in TEG/DMA/H2O 1:1:1 (39 and 41) or TEG/EtOH/H2O 50:5:45.

ACCEPTED MANUSCRIPT
22
4. Conclusions
In conclusion, we have discovered a series of ethyl 6-aminonicotinate sulfonylureas that are
antagonists of the P2Y12 receptor. The series was developed from our previously reported urea
linker series [14]. A molecular design strategy in which a urea linker was replaced with a sulfonyl
urea linker led to compounds with improved solubility and microsomal stability. The structural
features of the compounds facilitated the use of parallel synthesis methodology. SAR studies
showed that the 2-H/5-Cl-pyridines had similar or lower potency in the GTPγS assay than the
corresponding 2-CF3/5-CN pyridines. Replacement of the central piperazine with rings having
different size or rigidity than the piperazine gave no improvement in potency. The substituent on
the sulfonyl of the sulfonylurea linker was varied and presented 5-Cl-2-thienyl, 4-CH3-phenyl, 4-
Cl-phenyl and benzyl as the most promising substituents to provide potency in the GTPγS and
WPA assays. Replacement of the ethyl ester functionality with n-propyl and n-butyl esters gave
unchanged GTPγS potency, while ketone and sulfone analogs gave lower potency. Replacement
of the pyridine ring with a benzene ring gave significantly lowered potency.
The chemical stability of the sulfonyl urea linker by prolonged storage in solutions of acetonitrile
or ethanol depended on the type of solvent and the substitution pattern of the nitrogens in the
linker; thus 3-azetidinyl and 4-piperidinyl-pyridines were more stable than 4-piperazinyl-
pyridines.
In vivo PK parameters in rat were determined for six selected compounds. Of these, (±)-19 had the
highest potency (0.031 µM) in the WPA assay, but had low bioavailability (0.8%). Compared to
(±)-19, compound 39 had a much higher bioavailability (39%), but a 6-fold lower potency in the
WPA assay. In conclusion, although one single molecule that integrated all the desired properties
was not identified, the present study provided a solid foundation for continued research to

ACCEPTED MANUSCRIPT
23
optimize this type of compounds. Our further progress along this line will be reported in due
course.
5. Experimental Section
5.1 General methods. Commercially available chemicals were used as provided by the
commercial supplier. In general, reactions were performed under a nitrogen atmosphere.
Reactions in microwave reactors were performed by single node heating in a Smith Creator,
Smith Synthesizer, or an Emrys Optimizer using microwave vials from Personal Chemistry (now
Biotage). Concentration of solutions was done in vacuo on a rotary evaporator at temperatures
below 50 °C.
Thin-layer chromatography was made using TLC silica gel 60 F254 from Merck KGaA,
Darmstadt. Flash chromatography was performed with either standard glass- or plastic-columns
using Merck silica-gel grade 9385, 60Å (0.063-0.200 mm) from Sigma-Aldrich or on a Biotage
Horizon High Performance Flash Chromatography (HPFC) system using Biotage silica gel.
Preperative HPLC was performed on a Waters YMC-ODS AQS-3 120 Angstrom 3 x 500 mm or
on a Waters Delta Prep System using Kromasil C8, 10 µm columns. Gradient: 95% 0.1M
ammonium acetate buffer: 5% CH3CN → 100% CH3CN.
1H NMR measurements were performed on a Varian Mercury VX 400 spectrometer, operating at
a 1H frequency of 400 MHz or on Varian UNITY plus 400, 500 and 600 spectrometers, operating
at 1H frequencies of 400, 500 and 600 MHz, respectively. Chemical shifts are given in ppm with
the solvent as internal standard. Coupling constants are given in Hz. Note that with CD3OD as
NMR solvent the signals from exchangable protons are not observed, and thus not reported.

ACCEPTED MANUSCRIPT
24
The purity of screening compounds was determined by analytic HPLC. Low resolution mass
spectra (electrospray ionization) were acquired on a Waters ZQ quadrupole spectrometer coupled
to an Agilent Technologies 1100-series HPLC. The HPLC retention time were recorded through
standard gradient 5% to 95% acetonitrile (10 mM ammonium acetate, pH4) over 4 minutes using
a Synergy MAX-RP C12 (4 ꢀm; 50 mm x 3 mm ID) column with a flow rate of 2 mL/min. All
screening compounds had purity ≥ 95%.
5.2. Chemistry
5.2.1. Ethyl 5-chloro-6-(4-{[(phenylsulfonyl)amino]carbonyl}piperazin-1-yl)nicotinate (2).
23 (0.054 g, 0.20 mmol) was dissolved in THF (1.0 mL). TEA (0.030 g, 0.30 mmol) was added at
0 °C, followed by benzenesulfonyl isocyanate (0.048 g, 0.26 mmol). The reaction mixture was
stirred at 0 °C for 1 h and then at rt for 17 h. The reaction mixture was then stirred gently with
PS-TRIS (0.050 g, loading 4.35 mmol/g) and PS-NCO (0.100 g, loading 1.53 mmol/g) at rt for 1
h. The resins were filtered off and the solvents evaporated. The crude product was purified by
flash chromatography (EtOAc/pentane 1:1, then gradient to 100% EtOAc). Yield: 0.022 g (24%).
1H NMR (400 MHz, CDCl3): δ 1.39 (3H, t, J = 7.2 Hz), 3.49 (1H, s), 3.54-3.57 (8H, m), 4.38
(2H, q, J = 7.2 Hz), 7.56-7.59 (2H, m), 7. 62-7.67 (2H, m), 8.06-8.12 (1H, m), 8.15 (1H, d, J =
2.0 Hz), 8.75 (1H, d, J = 2.0 Hz). MS m/Z: 453 (M+1).
5.2.2. Ethyl 6-(3-(3-(5-chlorothiophen-2-ylsulfonyl)ureido)azetidin-1-yl)-5-cyano-2-
methylnicotinate (3). 34 (1.00 g, 2.77 mmol) was dissolved in DCM (10 mL). HCl (4 M in 1,4-
dioxane, 13.9 mL, 55.5 mmol) was added slowly and the reaction mixture was stirred at rt for 16
h. The mixture was concentrated to afford a solid, which was used crude assuming 100%
conversion.

ACCEPTED MANUSCRIPT
25
Ethyl 6-(3-aminoazetidin-1-yl)-5-cyano-2-methylnicotinate dihydrochloride (0.200 g, 0.60 mmol)
and 9 (0.336 g, 0.90 mmol) were dissolved in DMA (2 mL). DIPEA (1.05 mL, 6.0 mmol) was
added and the reaction mixture was heated to 100 °C for 1 h. The reaction mixture was cooled to
rt and concentrated. The crude was dissolved in EtOAc (40 mL) and washed with NH4Cl (2 x 40
mL, saturated, aq solution). The organic phase were dried (MgSO4) and concentrated. The crude
was purified by flash chromatography (30-50% EtOAc in hexanes, then 50% EtOAc in hexanes
with 0.5% AcOH). Yield: 0.034 g (11%). 1H NMR (400 MHz, CDCl3): δ 1.29 (3H, t, J = 7.1 Hz),
2.61 (3H, s), 4.11-4.19 (2H, m), 4.23 (2H, q, J = 7.1 Hz), 4.46-4.58 (3H, m), 4.85 (1H, br s), 7.26
(1H, d, J = 4.1 Hz), 7.40-7.49 (1H, m), 7.63 (1H, d, J = 4.1 Hz), 8.27 (1H, br s). HRMS: Calc. for
C19H21N5O5S2Cl (M+1)+, 498.0673; found: 498.0671.
5.2.3. Ethyl 6-chloro-5-cyano-2-methylnicotinate (7). Ethyl 5-cyano-2-methyl-6-oxo-1,6-
dihydropyridine-3-carboxylate (2.00 g, 9.70 mmol) [18(b)] was suspended in DCM (40 mL).
Oxalyl chloride (6.16 g, 4.25 mL, 48.5 mmol) was added. DMF (0.071 g, 0.080 mL, 0.97 mmol)
was added slowly and the reaction mixture was heated to reflux for 18 h. The mixture was cooled
to rt, diluted with DCM (250 mL) and poured onto ice-water (250 mL). The organic phase was
separated, washed with brine (250 mL), NaHCO3 (2 x 250 mL, saturated, aq solution), and brine
(250 mL), dried (Na2SO4) and concentrated. Yield: 2.18 g (98%). 1H NMR (400 MHz, CDCl3):
δ 1.42 (3H, t, J = 7.1 Hz), 4.40 (2H, q, J = 7.1 Hz), 2.91 (3H, s), 8.49 (1H, s). MS m/Z: 223 (M-1).
5.2.4. 2,2,2-Trichloroethyl [(5-chloro-2-thienyl)sulfonyl]carbamate (9). 8 (15.00 g, 75.9
mmol) was suspended in a bi-phasic mixture of NaOH (9.11 g, 55.4 mmol) in water (100 mL)
and DCM (250 mL). The reaction mixture was stirred rapidly at 0 °C and 2,2,2-trichloroethyl
chloroformate (30.1 mL, 133 mmol) was added drop-wise. The reaction mixture was slowly
allowed to reach rt and stirred for 18 h. HCl (concentrated) was added drop-wise till pH < 1,

ACCEPTED MANUSCRIPT
26
followed by dilution with DCM (500 mL). The organic phase was separated, dried (MgSO4) and
concentrated. The material was purified by flash chromatography (EtOAc/hexanes 1:3 to 1:1).
Yield: 20.67 g (73%) as a solid. 1H NMR (400 MHz, CDCl3): δ 4.76 (2H, s), 6.99 (1H, d, J = 4.2
Hz), 7.71 (1H, d, J = 4.2 Hz), 7.74 (1H, br s). MS m/Z: 372 (M-1).
5.2.5. Ethyl 5-cyano-6-piperazin-1-yl-2-(trifluoromethyl)nicotinate (10). 4 (1.00 g, 3.41
mmol) and piperazine (0.928 g, 10.8 mmol) were mixed in EtOH (3 mL) and TEA (0.727 g, 7.18
mmol) was added. The reaction mixture was heated in a microwave oven at 170 0C for 20
minutes. The mixture was diluted with DCM (200 mL), and the organic phase was washed with
NaHCO3 (saturated, aq solution), and brine, dried (Na2SO4) and concentrated. The crude was
purified by flash chromatography (DCM/MeOH 100:1 to 30:1). Yield: 0.751 g (67%). 1H NMR
(400, CD3OD): δ 1.36 (3H, t, J = 7.14 Hz), 2.93-2-99 (4H, m), 3.92-3.98 (4H, m), 4.34 (2H, q, J
= 7.22 Hz), 8.42 (1H, s). MS m/z: 329 (M+1).
5.2.6. (±)-Ethyl 6-[3-(3-tert-butoxy-3-oxopropyl)piperazin-1-yl]-5-cyano-2-
(trifluoromethyl)nicotinate ((±)-11). 4 (0.250 g, 0.90 mmol) and tert-butyl 3-piperazin-2-
ylpropanoate (0.192 g, 0.90 mmol) were dissolved in EtOH (2 mL). TEA (0.15 mL, 1.1 mmol)
was added. The solution was heated in a microwave oven at 150 °C for 20 minutes. The mixture
was concentrated, diluted with DCM (50 mL), washed with water (50 mL), dried (MgSO4) and
concentrated. The crude was purified by flash chromatography (DCM/MeOH 50:1). Yield: 0.162
g (40%).1H NMR (400 MHz, CDCl3): δ 1.36 (3H, t, J = 7.2 Hz), 1.44 (9H, s), 1.58-1.84 (3H, m),
2.35 (2H, t, J = 7.7 Hz), 2.75-2.83 (1H, m), 2.85-2.93 (2H, m), 3.10-3.16 (1H, m), 3.18-3.28 (1H,
m), 4.35 (2H, q, J = 7.2 Hz), 4.59-4.67 (2H, m), 8.34 (1H, s). MS m/z: 457 (M+H).

ACCEPTED MANUSCRIPT
27
5.2.7. Ethyl 6-[4-({[(5-chloro-2-thienyl)sulfonyl]amino}carbonyl)piperazin-1-yl]-5-cyano-2-
(trifluoromethyl)nicotinate (12). 9 (0.181 g, 0.91 mmol) and CDI (0.148 g, 0.91 mmol) were
dissolved in DCM (5 mL) and DIPEA (1.59 mL, 9.14 mmol) was added. The reaction mixture
was stirred at rt for 4 h. 10 (0.300 g, 0.91 mmol) was added and the mixture was heated at reflux
for 16 h. The mixture was cooled to rt and concentrated. Flash chromatography (3:7
EtOAc/hexanes to 99:1 EtOAc/AcOH) gave the product as a solid. Yield: 0.060 g (12%). 13C
NMR (151 MHz, d6-DMSO) δ 162.83, 157.10, 151.62, 148.75, 145.91 (q, JC-C-F = 35.1 Hz),
139.22, 135.85, 133.08, 127.10, 120.20 (q, JC-F = 275.5 Hz), 116.48, 114.83, 93.67, 61.76, 46.10,
40.03, 13.72. 1H NMR (400 MHz, d6-DMSO): δ 1.29 (3H, t, J = 7.1 Hz), 3.53-3.60 (4H, m), 3.86-
3.93 (4H, m), 4.29 (2H, q, J = 7.1 Hz), 7.24 (1H, d, J = 4.1 Hz), 7.62 (1H, d, J = 4.1 Hz), 8.57
(1H, s), 11.43 (1H, br s). HRMS: Calc. for C19H17N5O5F3S2Cl (M+1)+, 552.0390; found:
552.0411.
5.2.8. Ethyl 5-cyano-6-(4-(phenylsulfonylcarbamoyl)piperazin-1-yl)-2-
(trifluoromethyl)nicotinate (13). 10 (0.210 g, 0.640 mmol) was dissolved in DCM (10 mL).
Benzenesulfonylisocyante (0.352
g, 1.92 mmol) and TEA (0.10 mL, 1.0 mmol) were added and the reaction mixture was stirred
over night at rt. The mixture was concentrated and purified by preparative HPLC. Yield: 0.093 g
(27%). 1H NMR (400 MHz,CDCl3): δ 1.37 (3H, t, J = 7.2 Hz), 3.59-3.64 (4H, m), 3.67 (1H, s),
3.93-4.01 (4H, m), 4.37 (2H, q, J = 7.2 Hz), 7.52-7.60 (2H, m), 7.61-7.68 (1H, m), 8.05-8.07 (1H,
m), 8.07-8.10 (1H, m), 8.36 (1H, s). MS m/z: 510 (M-1).
5.2.9. Ethyl 5-cyano-6-[4-({[(2-methylphenyl)sulfonyl]amino}carbonyl)piperazin-1-yl]-2-
(trifluoromethyl)nicotinate (14). 10 (0.066 g, 0.20 mmol) and 2-methylbenzenesulfonyl
isocyanate (0.047 g, 0.24 mmol) were dissolved in DCM (1.0 mL) and TEA (0.080 mL, 0.60

ACCEPTED MANUSCRIPT
28
mmol) was added. The reaction mixture was stirred at rt for 14 h and then evaporated. The crude
material was purified by preparative HPLC. Yield: 0.066 g (62%). 1H NMR (400 MHz, CDCl3): δ
1.39 (3H, t, J = 7.1 Hz), 2.66 (1H, s), 2.69 (3H, s), 3.61-3.66 (4H, m), 3.93-3.98 (4H, m), 4.37
(2H, q, J = 7.1 Hz), 7.28-7.38 (2H, m), 7.44-7.51 (1H, m), 8.08-8.12 (1H, m), 8.36 (1H, s). MS
m/Z: 526 (M+1).
5.2.10. (±)-3-{1-({[(5-Chloro-2-thienyl)sulfonyl]amino}carbonyl)-4-[3-cyano-5-
[ethoxy(hydroxy)methyl]-6-(trifluoromethyl)pyridin-2-yl]piperazin-2-yl}propanoic acid
((±)-19). 8 (0.027 g, 0.14 mmol), CDI (0.034 g, 0.21 mmol), and DIPEA (0.10 mL, 0.58 mmol)
were dissolved in DCM (1 mL) and the mixture was stirred at rt for 3.5 h. A solution of (±)-11
(0.063 g, 0.14 mmol) and DIPEA (0.14 mL, 0.81 mmol) in DCM (1 mL) was added. The reaction
mixture was stirred at rt for 24 h and then in a sealed vial at 40 °C (oil bath heating) for 16 h. The
mixture was concentrated and the residue was purified by preparative HPLC. Yield: 0.056 g
(60%). 1H NMR (400 MHz, CDCl3): δ 1.37 (3H, t, J = 7.2 Hz), 1.52 (9H, s), 1.69-1.81 (1H, m),
1.83-1.95 (1H, m), 2.34-2.42 (2H, m), 3.09 (dt, 1H, J = 3.2 and 12.5 Hz), 3.36-3.50 (2H, m),
4.02-4.10 (1H, m), 4.29 (1H, d, J = 13.9 Hz), 4.37 (2H, q, J = 7.2 Hz), 4.51-4.66 (2H, m), 6.91
(1H, d, J = 4.0 Hz), 7.65 (1H, d, J = 4.0 Hz), 8.37 (1H, s). MS m/z: 680 (M+H).
(±)-Ethyl 6-[3-(3-tert-butoxy-3-oxopropyl)-4-({[(5-chloro-2-
thienyl)sulfonyl]amino)carbonyl)piperazin-1-yl]-5-cyano-2-(trifluoromethyl)nicotinate (0.056 g,
0.082 mmol) was dissolved in DCM (4 mL) and TFA (1 mL) was added. The reaction mixture
was stirred at rt for 1 h and then concentrated. The crude was purified by preparative HPLC.
Yield: 0.046 g (90%). 1H NMR (400 MHz, d6-DMSO): δ 1.27 (3H, t, J = 7.1 Hz), 1.53-1.75 (2H,
m), 2.02-2.14 (1H, m), 2.16-2.28 (1H, m), 3.14-3.40 (3H, m), 4.06-4.16 (1H, m), 4.26 (2H, q, J =
7.1 Hz), 4.30-4.45 (3H, m), 6.90 (1H, d, J = 3.8 Hz), 7.14 (1H, d, J = 3.8 Hz), 7.64 (1H, br s),
8.49 (1H, s). MS m/z: 624 (M+H).