Discovery, Structure-Activity Relationship, and Binding Mode of an Imidazo[1,2-a]pyridine Series of Autotaxin Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.7b00647

Autotaxin (ATX) is a secreted enzyme playing a major role in the production of lysophosphatidic acid (LPA) in blood through hydrolysis of lysophosphatidyl choline (LPC). The ATX-LPA signaling axis arouses a high interest in the drug discovery industry as it has been implicated in several diseases including cancer, fibrotic diseases, and inflammation, among others. An imidazo[1,2-a]pyridine series of ATX inhibitors was identified out of a high-throughput screening (HTS). A cocrystal structure with one of these compounds and ATX revealed a novel binding mode with occupancy of the hydrophobic pocket and channel of ATX but no interaction with zinc ions of the catalytic site. Exploration of the structure-activity relationship led to compounds displaying high activity in biochemical and plasma assays, exemplified by compound 40. Compound 40 was also able to decrease the plasma LPA levels upon oral administration to rats.

Full text of DOI:10.1021/acs.jmedchem.7b00647

Subscriber access provided by CARLETON UNIVERSITY
Article
Discovery, Structure-Activity Relationship and Binding
Mode of Imidazo[1,2-a]pyridine Series of Autotaxin Inhibitors
Agnès Joncour, Nicolas Desroy, Christopher Housseman, Xavier Bock, Natacha Bienvenu, Laëtitia
Cherel, Virginie Labeguere, Christophe Peixoto, Denis Annoot, Luce Lepissier, Jörg Heiermann,
Willem Jan Hengeveld, Gregor Pilzak, Alain Monjardet, Emanuelle Wakselman, Veronique Roncoroni,
Sandrine Le Tallec, René Galien, Christelle David, Nele Vandervoort, Thierry Christophe, Katja
Conrath, Mia Jans, Alexandre Wohlkonig, Robert Touitou, Damien Fleury, Lionel Vercheval,
Patrick Mollat, Nicolas Triballeau, Ellen Van der Aar, Reginald Brys, and Bertrand Heckmann
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00647 • Publication Date (Web): 21 Jul 2017
Downloaded from http://pubs.acs.org on July 21, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Discovery, StructureꢀActivity Relationship and
Binding Mode of Imidazo[1,2ꢀa]pyridine Series of
Autotaxin Inhibitors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†,*
†
†
†
Agnès Joncour, Nicolas Desroy, Christopher Housseman, Xavier Bock, Natacha Bienvenu,
†
†
†
†
†
Laëtitia Cherel, Virginie Labeguere, Christophe Peixoto, Denis Annoot, Luce Lepissier, Jörg
○
○
○
†
Heiermann, Willem Jan Hengeveld, Gregor Pilzak, Alain Monjardet, Emanuelle
†
†
†
†
†
Wakselman, Veronique Roncoroni, Sandrine Le Tallec, René Galien, Christelle David, Nele
‡
‡
‡
‡
║
Vandervoort, Thierry Christophe, Katja Conrath, Mia Jans, Alexandre Wohlkonig, Robert
†
†
†
†
†
Touitou, Damien Fleury, Lionel Vercheval, Patrick Mollat, Nicolas Triballeau, Ellen van der
‡
‡
†,
Aar, Reginald Brys and Bertrand Heckmann *
†
Galapagos SASU, 102 avenue Gaston Roussel, 93230 Romainville, France
‡
Galapagos NV, Generaal De Wittelaan L11 A3, 2800 Mechelen, Belgium
○
Mercachem, Kerkenbos 1013, 6546 Nijmegen, The Netherlands
║
VIB Structural Biology Research Center, Vrije Universiteit Brussel, 1050 Brussel, Belgium
KEYWORDS Autotaxin, ENPPꢀ2, Lysophosphatidic acid, Lysophosphatidyl choline, FSꢀ3.
ABSTRACT
ACS Paragon Plus Environment
1

Page 3 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Autotaxin (ATX) is a secreted enzyme playing a major role in the production of lysophosphatidic
acid (LPA) in blood through hydrolysis of lysophosphatidyl choline (LPC). The ATXꢀLPA
signaling axis arouses a high interest in the drug discovery industry as it has been implicated in
several diseases including cancer, fibrotic diseases and inflammation among others. An
imidazo[1,2ꢀa]pyridine series of ATX inhibitors was identified out of a high throughput
screening (HTS). A coꢀcrystal structure with one of these compounds and ATX revealed a novel
binding mode with occupancy of the hydrophobic pocket and channel of ATX but no interaction
with zinc ions of the catalytic site. Exploration of the structureꢀactivity relationship led to
compounds displaying high activity in biochemical and plasma assays, exemplified by
compound 40. Compound 40 was also able to decrease the plasma LPA levels upon oral
administration to rats.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Autotaxin (ATX) also known as ENPPꢀ2 (ectonucleotide pyrophosphatase / phosphodiesterase 2)
or lysophospholipase D is a secreted protein that belongs to the ENPP family of enzymes which
is composed of seven members. ATX is the only enzyme of the ENPP family with
lysophospholipase D (lysoPLD) activity and is responsible for the hydrolysis of lysophosphatidyl
1
choline (LPC), an abundant plasma phospholipid, towards lysophosphatidic acid (LPA). LPA
represents a family of bioactive lipids of varying fatty acyl chain length that signal through at
2
least 6 LPA receptors (LPA ) to control a broad range of cellular events (Figure 1).
1
ꢀ6
ACS Paragon Plus Environment
2

Journal of Medicinal Chemistry
Page 4 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. (A) ATXꢀmediated conversion of LPC into LPA and summary of cellular events
triggered by LPA receptor activation. (B) Examples of abundant LPA species in plasma.
In recent years numerous ATX inhibitors were described in the literature and patented with
possible application in the treatment of diverse pathologies such as cancer, fibrotic diseases,
3
inflammation, pain and angiogenesis among others. At Galapagos, an ATX drug discovery
program was initiated using a classical screening approach, aiming at developing novel
therapeutics for fibrotic diseases. The monitoring of ATXꢀmediated cleavage of LPC requires
either LCꢀMS/MS detection methods for LPA determination or a cascade of biochemical
transformations to measure the amount of choline released (Figure 2). As both techniques were
not deemed suitable for highꢀthroughput screening (HTS) purposes the use of an alternative
fluorogenic substrate was envisaged. A screening campaign was performed on a diverse set of
approximately 100,000 compounds using a biochemical assay with the FSꢀ3 fluorogenic
4
substrate as LPC surrogate. Cleavage of FSꢀ3 by ATX leads to the release of a fluorophore,
ACS Paragon Plus Environment
3

Page 5 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibition of which can be directly measured by fluorescence detection. Several chemical series
of ATX inhibitors were identified out of the screening campaign. Optimization and structure
activity relationship (SAR) of an imidazo[1,2ꢀa]pyridine series to identify lead compound 40 and
discussion of the binding modes of ATX inhibitors will be described in this article. Structural
modifications of compound 40 leading to the discovery of clinical candidate GLPG1690 (55)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ꢀ7
have recently been disclosed.
Figure 2. Methods to assess ATX activity in vitro and in plasma samples.
RESULTS AND DISCUSSION
Structureꢀactivity relationship
Out of HTS efforts a Nꢀbenzylꢀ2ꢀethylꢀNꢀmethylꢀimidazo[1,2ꢀa]pyridinꢀ3ꢀamine series hit series
was identified with best compounds displaying an IC around 30 nM with the FSꢀ3 substrate
5
0
based assay (FSꢀ3 assay) (Figure 3). Since plasma naturally contains ATX and LPC but not
membraneꢀbound lipid phosphate phosphatases responsible for LPA degradation, incubation of
plasma at 37°C leads to the production of LPA that can be measured by LCꢀMS/MS. Monitoring
the inhibition of LPA production by compounds in this plasma assay allowed to evaluate their
ACS Paragon Plus Environment
4

Journal of Medicinal Chemistry
Page 6 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibitory activity and the impact of plasma proteins on this activity. Best Nꢀbenzylꢀ2ꢀethylꢀNꢀ
methylꢀimidazo[1,2ꢀa]pyridinꢀ3ꢀamine derivatives were able to show roughly 50% reduction of
LPA production at 10 µM. Those compounds suffered from poor metabolic stability which
precluded their use in vivo. Compound 1 with improved pharmacokinetic properties was
identified as an early lead compound: 1 had a biochemical IC of 122 nM (FSꢀ3 assay) and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
ꢀ1
ꢀ1
displayed 39% bioavailability in rat and low clearance (0.15 L h kg ). However this compound
only weakly inhibited the production of LPA in rat plasma, probably due to an insufficient
potency.
imidazo[1,2-a]pyridine series
1
Figure 3. Hit series and compound 1.
Exploration of the substitution in position 6 of the imidazo[1,2ꢀa]pyridine led to the
identification of very potent inhibitors 2ꢀ6 of ATX mediated FSꢀ3 cleavage (Table 1). Despite
nanomolar activity in the FSꢀ3 assay, those compounds displayed only micromolar activity in the
biochemical assay using natural substrate LPC. These results clearly indicated divergence in the
structureꢀactivity relationships with the natural substrate LPC or with the artificial substrate FSꢀ
3. In order to assess the biological relevance of the data returned by biochemical assays, the
inhibition of LPA production in plasma was evaluated. Compounds from Table 1 appeared
weakly active in this plasma assay, which illustrates the strong disconnect with the results
ACS Paragon Plus Environment
5

Page 7 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
obtained in the biochemical FSꢀ3 assay. In contrast a better correlation with little to no shift was
observed when the potency of compounds for inhibiting the conversion of LPC in vitro was
compared to the potency in the plasma assay. At this point we hypothesized that the probe
substrate used in the FSꢀ3 assay, although mimicking LPC as a substrate, might bind differently
to the enzyme owing to the steric bulk associated with the fluorophore and quencher groups. A
different binding mode of the substrates used in the different ATX assays might result in
differences in the ability of ATX inhibitors to compete with these, possibly explaining
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
discrepancies between the outcome of the FSꢀ3 assay and phenotypic assays. As a consequence
further explorations were carried out using LPC as substrate to assess the activity of compounds.
a
Table 1. Inhibitory activity of compounds 1ꢀ6 in biochemical and plasma assays
Compound
R
X
ATX FSꢀ3
ATX LPC
Plasma
activity
IC , nM)
b
b
(IC , nM)
(IC , nM)
5
0
50
(
5
0
c
1
Cl
122 ± 17
>10,000
6,555
> 50,000
12,555
2
Cl
5
ACS Paragon Plus Environment
6

Journal of Medicinal Chemistry
Page 8 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
c
c
3
4
F
F
2
8
>10,000
6,800
> 50,000
> 50,000
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c, d
5
6
Cl
F
9 ± 6
1,210 ± 212
2,359 ± 404
< 5,000
6,080
12 ± 2
a
ATX FSꢀ3 and ATX LPC activities were measured using FSꢀ3 and LPC 16:0 as substrate
respectively. Plasma activity was evaluated by incubating rat plasma at 37°C for 2 h and
comparing LPA 18:2 production in presence or absence of compound. Standard deviations
indicated when IC50 determination was replicated at least twice. Compounds evaluated at
concentrations of 0.5 µM, 5 µM and 50 µM (n = 3). Compounds with IC indicated > 50,000
nM showed less than 50% reduction of LPA production after incubation at all concentrations.
7% reduction of LPA production at 0.5 µM, 67% at 5 µM, 100% at 50 µM.
b
c
50
d
The better potencies in the LPC assay of compounds 5 and 6 versus 2 and 4 respectively
suggested that a basic amine and an extended substituent at position 6 bearing an Hꢀbond
accepting or donating group could impart ATX inhibitory activity. As depicted in Table 2,
imidazo[1,2ꢀa]pyridine derivatives 7ꢀ22 bearing a substituted piperidine or piperazine at position
6 were explored. Compound 8 was identified as a submicromolar inhibitor of ATX, the methyl
sulfonamide in compound 8 provided approximately a 2ꢀfold gain of activity versus acetamide
group in compound 7. Compound 8 was also active in rat plasma with an IC of 1,033 nM. Sideꢀ
50
chain extension with a basic group as in compound 9 further improved activity. Alternatively Nꢀ
substitution with an ethyl acetate group on the piperidine or the piperazine ring provided
ACS Paragon Plus Environment
7

Page 9 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 10 and 11 which displayed notable biochemical potencies. These data suggested that
a substituent bearing a basic nitrogen atom and a distal Hꢀbond accepting group was preferred at
position 6. However the latter compounds were weakly active in plasma due to plasma instability
arising probably from ester cleavage, parent carboxylic acid derivative 12 being poorly active.
Investigation of heteroaromatic rings as replacement of the ester function in compounds 13ꢀ16
led to a loss of activity in most cases. Only compound 16 retained interesting biochemical
potency but the compound had low activity in plasma (IC = 4,100 nM). Interestingly, whereas
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
0
unsubstituted and monosubstituted amides were detrimental for activity (compounds 17 and 18),
replacement of the ester function by a disubstituted amide as in compounds 19ꢀ22 afforded
potent compounds in the biochemical and plasma assays. Compounds 19 and 20 displayed IC50
of 476 nM and 115 nM respectively in plasma.
a
Table 2. SAR in position 6 of the imidazo[1,2ꢀa]pyridine ring
Compound
R
ATX LPC
b
(
IC , nM)
50
7
1,624
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8
9
710 ± 42
357 ± 53
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
1
2
3
4
5
6
7
480 ± 102
280 ± 70
1
1
1
1
1
1
1
9,163 ± 4,527
1,521 ± 280
1,708 ± 611
> 4,100
771 ± 108
2,372 ± 659
ACS Paragon Plus Environment
9

Page 11 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
8
9
1,614 ± 292
261 ± 32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
0
86 ± 20
21
231 ± 73
22
246 ± 23
a
b
ATX LPC activities were measured using LPC 16:0 as substrate. Standard deviations
indicated when IC determination was replicated at least twice.
5
0
In parallel, exploration was carried out on the substituent in position 3 of the imidazo[1,2ꢀ
a]pyridine scaffold. Early investigations showed that the thiazole ring was important for activity:
replacing the thiazole ring by either a phenyl ring or heteroaromatic rings such as 1,2,4ꢀ
oxadiazole or 1,3,4ꢀoxadiazole rings led to significant loss of activity (data not shown).
Therefore investigations focused on the substitution of the thiazole ring as in compounds 23ꢀ39
(
Table 3). Compound 23 illustrates the loss of activity when the phenyl and thiazole rings were
fused. Cyclohexyl replacing 4ꢀfluorophenyl ring in compound 24 maintained activity, albeit with
ACS Paragon Plus Environment
10

Journal of Medicinal Chemistry
Page 12 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a slight loss of potency. The 4ꢀfluorophenyl ring could also be replaced by a 4ꢀchloroꢀ3ꢀpyridine
moiety without any loss of activity in compound 25. Replacing the paraꢀfluorine substituent on
the phenyl ring by other halogens, a trifluoromethyl group (26) or a methoxy substituent (27)
was tolerated. However bulky substituents such as tertꢀbutyl group (28) had a detrimental impact
on activity. Diꢀhalogenated compounds 29 and 30 were less potent than the parent compound 8.
The importance of the para substitution was well established by the significant loss in potency
with the meta, metaꢀdifluoro compound 31. Interestingly compound 32 bearing an orthoꢀmethyl
substituent displayed similar potency compared to analogue 8. This good result inspired the
preparation of rigidified tricyclic analogs 33 and 34, however the latter compounds only
maintained similar level of activity and were not pursued any further. When the orthoꢀmethyl
substituent of 32 was replaced by a nitrile group in compound 35, a remarkable improvement of
activity was obtained. Although introduction of a methyl group in position 5 of the thiazole
decreased potency (36 versus 8), introduction of a nitrile group in 37 afforded a very potent
compound. Replacing the nitrile group by a carboxamide (38) or a carboxylic acid (39) resulted
in a drop of potency.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Table 3. SAR on the thiazole ring
Compound
R
ATX LPC
b
(
IC , nM)
50
ACS Paragon Plus Environment
11

Page 13 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
710 ± 42
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
3
> 10,000
2
2
4
1,904 ± 359
5
6
7
493 ± 11
1,001 ± 71
1,684 ± 161
CF3
2
2
N
S
O
N
S
ACS Paragon Plus Environment
12

Journal of Medicinal Chemistry
Page 14 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
8
7,609 ± 707
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
29
1,899 ± 650
2,021 ± 433
4,246
3
3
0
1
32
533 ± 157
804 ± 125
33
ACS Paragon Plus Environment
13

Page 15 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
4
5
1,212 ± 25
126 ± 37
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
36
2,266 ± 344
3
3
7
8
138 ± 26
1,410 ± 388
39
> 10,000
a
ATX LPC activities were measured using LPC 16:0 as substrate. b Standard deviations
indicated when IC50 determination was replicated at least twice.
ACS Paragon Plus Environment
14

Journal of Medicinal Chemistry
Page 16 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Taken together these data indicated that a paraꢀsubstituted 4ꢀphenylthiazole substituent was
preferred at position 3 and the introduction of a nitrile group on the ortho position of the phenyl
ring or on the thiazole ring afforded a significant improvement of activity; our hypothesis to
explain this gain of activity will be discussed later in this article. Structural modifications in
positions 3 and 6 of the imidazo[1,2ꢀa]pyridine scaffold were combined and compound 40,
displaying potencies of 27 nM and 22 nM in the biochemical and rat plasma assays, respectively,
was identified as lead compound in the series (Figure 4).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
40
Figure 4. Structure of compound 40.
Coꢀcrystal structure and binding mode discussion
The first experimental structures of ATX (rat ortholog) were concurrently published by two
different groups in 2011 revealing important structural features that explain the uncommon
biochemical properties of this enzyme within the ENPP family as well as the binding mode of
13ꢀ15
different LPA ligands and inhibitors.
Based on sequence and structural homology, ATX is
composed of 2 somatomedin Bꢀlike domains followed by the catalytic ENPP domain and a
nucleaseꢀlike domain maintained in close contact with the catalytic domain thanks to a peptide
ACS Paragon Plus Environment
15

Page 17 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
linker (or “lasso”) and a disulfide bridge. The catalytic domain reveals 3 pockets arranged in a Tꢀ
2
10
shaped groove: a catalytic pocket that exhibits a nucleophilic threonine residue (Thr in the
human ortholog) and 2 adjacent zinc ions, a hydrophobic pocket that accommodates the
hydrophobic tail of a variety of LPC substrates and a soꢀcalled hydrophobic channel (or
“tunnel”) which is proposed to play an important role in the transport and local delivery of LPA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
6ꢀ18
products and other hydrophobic substances.
In order to resolve the binding mode of
compound 9, a dedicated construct of human ATX was produced and purified in house providing
a batch of protein suitable for crystallography (see Supporting Information). Coꢀcrystals were
obtained and diffracted with a maximum resolution of 2.7 Å, enabling the construction of an
experimental model useful to assist and verify drug design hypothesis. Compound 9 binds to the
catalytic domain with its imidazo[1,2ꢀa]pyridine core occupying a pivotal spot in the center of
the Tꢀshaped groove, orienting each of its 3 substituents towards each of the 3 pockets (Figure
5). The ethyl group in position 2 points in the direction of the catalytic pocket, while substituents
in positions 3 and 6 occupy the hydrophobic pocket and hydrophobic channel respectively. In
addition to Van der Waals contacts essentially driven here by hydrophobic effects, 9 displays a
2
61
hydrogen bond between a sulfonamide oxygen and the indole NH of Trp and potentially a
second one, albeit not certain given the electron density, between the basic nitrogen of the C6
2
55
257
substituent and the backbone carbonyl groups of Trp and Gly .
ACS Paragon Plus Environment
16

Journal of Medicinal Chemistry
Page 18 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. (A) Overall structure of ATX with compound 9 shown in CPK with green carbon
atoms. Somatomedin Bꢀlike, catalytic and nucleaseꢀlike domains are shown in blue, green and
pink, respectively. Disulfide bridges are depicted as yellow bars and the two zinc ions of the
catalytic site are shown as magenta spheres. The arrow indicates the viewing angle of Figure 5B
and the dashed line the clipping plane for the site surface. (B) Focus on the binding pocket
showing details of the binding mode of 9. The surface delineating the site is colored by
ACS Paragon Plus Environment
17

Page 19 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hydrophobicity using in a blueꢀwhiteꢀbrown spectrum (blue for hydrophilic, brown for
hydrophobic). For clarity the surface was clipped vertically (see Figure 5A). Dashed lines
highlight interactions between compound 9 and ATX, hydrophobic contacts in pink, hydrogen
bonding in green.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Since the publication of the first structures of ATX, a variety of endogenous ligands and
synthetic inhibitors has been described, examples of which are shown in Figure 6. Interestingly
they exhibit a diversity of binding modes which, at this stage, can be classified into 4 types
illustrated in Figure 7.
Figure 6. Structures of ATX inhibitors. PDB codes of coꢀcrystal structures with ATX are
indicated in brackets (note: reported PDB code for compound 64 (5mom) doesn’t match the
intended structure).
ACS Paragon Plus Environment
18

Journal of Medicinal Chemistry
Page 20 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Type I inhibitors (such as LPA species, HAꢀ155 (56), PFꢀ8380 (57) or “compound 2” (58) in
1
3,14,19
Figure 6) mimic the binding mode of LPC substrate and occupy the catalytic site.
generally have a zinc binding group and a hydrophobic tail lodged in the hydrophobic pocket
Figure 7A). Such inhibitors leave the hydrophobic channel unoccupied which explains why
They
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
many structures categorized as Type I inhibitors have a secondary compound occupying the
channel, generally a substance from the solution with amphiphilic properties such as LPA or
cholesterol derivatives. In fact, all crystallographic structures with type I inhibitors published so
far exhibit an electronic density in the channel but not all have a modelled compound in this area,
probably because of partial occupancy. Type II inhibitors occupy the hydrophobic pocket by
largely exploiting its intrinsic plasticity (Figure 7B). Indeed, in most cases this induced fit
significantly restructures the binding pocket involving sideꢀchain conformation variations of
214
217
274
275
307
residues such as Phe , Leu , Phe , Phe and Tyr . Interestingly, type II inhibitors jam the
bottleneck between the hydrophobic channel and the catalytic site but do not fully occupy any of
them. PATꢀ078 (59), PATꢀ494 (60) and, to a certain extent, CRT0273750 (61), represent such
20,21
Type II inhibitors.
As for type I inhibitors, the remaining pockets left vacant allow binding of
an LPA molecule. By contrast, type III inhibitors bind to the hydrophobic channel leaving the
hydrophobic pocket and catalytic site unoccupied (Figure 7C). They comprise PATꢀ347 (63),
PATꢀ505 (65), PATꢀ048 (66), and the steroids 7αꢀhydroxycholesterol and tauroursodeoxycholic
1
8,20,22
acid (TUDCA, 62).
The binding modes of compounds 55 and 9 differ from that of other
inhibitors and can therefore be categorized as type IV inhibitors (Figure 7D). Indeed by
occupying both the hydrophobic pocket and the hydrophobic channel, the two sites enabling the
binding of LPC, the type IV inhibitors preclude the binding of LPA and bile acid derivatives to
ATX and avoid any interactions of substrates with the zinc atoms. Recently another type IV
ACS Paragon Plus Environment
19

Page 21 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitor (64) has been designed by grafting a “tail” fragment extracted from a type I inhibitor
2
3
(57) on a type III inhibitor (62).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Schematic representation of the autotaxin binding pocket and the different inhibitor
binding modes reported to date. Type I, II, III and IV inhibitors are represented in green, purple,
cyan and orange respectively on the schemas with matching colors for carbon atoms on the
exemplified structures. The grey dotted lines represent secondary ligands modeled next to the
inhibitor as shown with grey carbons on the illustrations. Some structures lack secondary ligand
models but exhibit a positive electron density in the same region (FoꢀFc map at 3 sigma in green,
2
FoꢀFc map at 1 sigma in blue). Zinc ions are depicted in magenta. (A) Binding mode of type I
inhibitors LPA 22:6 (a), 56 (b), 58 (c). (B) Binding mode of type II inhibitors 60 (d), 59 (e), 61
ACS Paragon Plus Environment
20

Journal of Medicinal Chemistry
Page 22 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
f). (C) Binding mode of type III inhibitors 63 (g), 62 (h). (D) Binding mode of type IV
inhibitors 55 (i), compound 9 (j).
Such diversity of binding modes might impact the properties of ATX inhibitors: recently, on
top of LPA formation, a role for LPA transport and delivery to its cognate receptors has been
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6ꢀ18
suggested for ATX.
Whereas all types of ATX inhibitors block the formation of LPA, type
IV and possibly some type II inhibitors such as compound 61 could also prevent LPA transport
and delivery to LPA receptors through ATX channel occupancy. Consequently, ATX inhibitors
could display various modes of action and subsequently different therapeutic effects depending
on their binding mode, with some inhibitors allowing while others blocking the binding of fatty
2
4
acids (LPA and derivatives).
The reported SAR of the series confirms the binding mode determined experimentally. In
particular whereas position 6 of the imidazo[1,2ꢀa]pyridine scaffold tolerates a broad range of
substituents without significant change of potency, slight structural modification of the
substituent at position 3 has a strong impact on activity. Indeed, the former appears to occupy the
hydrophobic channel with access to the bulk solvent in contrast to the latter, which is lodged in
an enclosed hydrophobic pocket within the structure of the enzyme. An interesting modification
which boosted the potency is worth mentioning: by adding a nitrile group on the substituent in
position 3, either on the thiazole or on the terminal fluorophenyl moiety, the potency could be
systematically increased by a factor 7 on average (matchedꢀmolecular pair analysis, data not
shown). This effect could be wellꢀexplained by analyzing the role of water before and after the
2
5
binding event using WaterMap. This tool simulates the dynamic behavior of water molecules
enabling, among others, the identification of hydration sites that are energetically unfavorable.
When the calculation is run without compound 9 in the ATX crystal structure (apo map, Figure
ACS Paragon Plus Environment
21

Page 23 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
A), WaterMap locates a high energy water molecule (+7.5 kcal/mol) close to the positions
occupied by the hydrogen atoms of the thiazole moiety and the phenyl group of the bound
ligand. This indicates that the overall energy of the system should benefit from the binding of a
ligand that displaces the otherwise constrained water molecule. Although compounds which lack
the cyano group (such as compound 9) don’t seem to allow for a water molecule to occupy the
same site (holo map, Figure 8B), a high energy “vacuum bubble” is created instead, penalizing
the global free energy of the system. By adding a nitrile group either on the phenyl moiety
(compound 35) or on the thiazole moiety (compounds 37, 40), not only the high energy water
molecule can be released into the bulk solvent but the pocket is also fully occupied by the ligand.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. WaterMap results in the hydrophobic pocket. (A) Apo map showing the high energy
water molecule highlighted by a dashed circle. Water molecules are colored based on their
calculated free energy using a blue (negative, stable)ꢀgreen (neutral)ꢀred (positive, unstable)
spectrum. Albeit absent during the calculation, the ligand is shown in sticks with white carbon
atoms. (B) Holo map highlighting the high energy vacuum bubble (crosses circled in black) left
by compounds (balls and sticks with orange carbon atoms). (C) Model of a compound with a
nitrile group on the thiazole.
ACS Paragon Plus Environment
22

Journal of Medicinal Chemistry
Page 24 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In vivo reduction of LPA production by compound 40
As shown in Table 4, compound 40 displays low to moderate clearance in rodents and low oral
bioavailability. As conversion of LPC into LPA by ATX is the major source of circulating LPA
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
in blood, in vivo inhibition of this process can be evaluated by quantification of LPA levels in
plasma. Compound 40 and other compounds within the chemical series generally displayed little
to no potency shift between the biochemical and plasma assays despite a plasma protein binding
greater than 99%. The high potency of compound 40 in plasma could therefore compensate for
its low oral bioavailability. In order to evaluate in vivo ATX inhibitory activity of compound 40,
reduction of plasma LPA levels upon compound 40 administration to rats was monitored. Rats
received a single dose of compound 40 (5 mg/kg) and blood was sampled at 0.5, 1 and 3 hours
postꢀdosing for the determination of plasma levels of both compound 40 and LPA 18:2.
Percentage of LPA reduction was calculated by comparing LPA 18:2 levels in animals treated
with compound 40 or with vehicle at each time point. As seen in Table 5 compound 40 was able
to cause a more than 70% reduction of LPA 18:2 levels over 3 hours, in line with the high
potency of the compound in the rat plasma assay (IC50 = 22 nM), which is approximately
covered by the observed plasma levels for compound 40. The higher reduction of LPA 18:2 level
at 3 h compared to vehicleꢀtreated animals could be due to the presence of an active metabolite
which was not investigated at this stage. This result clearly demonstrated in vivo target
engagement with compound 40.
a
Table 4. In vivo pharmacokinetic properties for compound 40
ꢀ1
ꢀ1
mouse iv Cl (L h kg )
1.2
2.0
mouse iv t (h)
1/2
ACS Paragon Plus Environment
23

Page 25 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ꢀ1
mouse iv V (L kg )
1.7
ss
mouse po F (%)
12
ꢀ1
ꢀ1
rat iv Cl (L h kg )
rat iv t (h)
3.1 (6)
2.1 (4)
8.3 (11)
15 (22)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1/2
ꢀ1
rat iv V (L kg )
ss
rat po F (%)
a
Data determined from 5 mg/kg po and 1 mg/kg iv doses (n = 3). Between brackets is the
coefficient of variation in percentage, CV%. No CV% is available for mice data because of
composite blood sampling on a group of six mice.
Table 5. In vivo reduction of plasma LPA 18:2 levels in rats after a single po dose of
a
compound 40 at 5 mg/kg
Time point
Compound 40 plasma
concentration (nM)
% reduction of plasma
LPA 18:2 level
0
.5 h
67 ± 15
34 ± 5
74 ± 7
82 ± 5
97 ± 1
1
h
h
3
58 ± 16
a
Rats were dosed orally at 5 mg/kg with compound 40 under fasted conditions. Values for
plasma concentration and LPA reduction are mean values of 3 animals. Percentage of reduction
of LPA 18:2 level was calculated by comparing levels of LPA 18:2 in animals treated with
compound 40 or vehicle at each time point.
Chemistry
The general synthesis of imidazo[1,2ꢀa]pyridineꢀ3ꢀylꢀ4ꢀarylꢀthiazolꢀ2ꢀamines is depicted in
Scheme 1. 3ꢀFormylaminoꢀimidazo[1,2ꢀa]pyridine derivatives 42aꢀc were prepared by the 3ꢀ
9
ꢀ11
component GroebkeꢀBlackburnꢀBienaymé reaction
of 5ꢀsubstituted 2ꢀaminopyridine
1
2
compounds 41aꢀc with propanal and Walborsky’s reagent (1,1,3,3ꢀtetramethylbutyl isocyanide)
ACS Paragon Plus Environment
24

Journal of Medicinal Chemistry
Page 26 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in the presence of MgCl . Alkylation and acidic deformylation with methanolic HCl of
2
compounds 42aꢀc afforded the 3ꢀalkylaminoꢀimidazo[1,2ꢀa]pyridine derivatives 43aꢀc.
Aminothiazole derivatives 44aꢀh were accessed from 43aꢀc via the formation of the thiourea
intermediate using FmocꢀNCS and piperidine, and subsequent reaction with corresponding
bromoꢀ or chloroꢀacetophenones.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. General synthesis of (2ꢀethylimidazo[1,2ꢀa]pyridineꢀ3ꢀyl)ꢀ4ꢀarylꢀNꢀalkylthiazolꢀ2ꢀ
a
amine derivatives 44aꢀh
4
4a, R1 = COOMe, Ar = p-C H Cl
6 4
44b, R1 = Br, Ar = p-C H F
6
4
a
b
c
4
4c, R1 = I, Ar = p-C H F
6 4
4
4d, R1 = I, Ar = p-C H t-Bu
6 4
4
4e, R1 = I, Ar = p-C H CF
6 4 3
41a-c
41a, R1 = COOMe
42a-c
43a-c
44a-h
44f, R H OMe
= I, Ar = p-C
1 6 4
44g, R1 = I, Ar = 3,4-C H F
6 3
42a, R1 = COOMe
42b, R1 = Br
42c, R1 = I
43a, R1 = COOMe
43b, R1 = Br
43c, R1 = I
2
2
4
4h, R1 = I, Ar = 3,5-C H F
41b, R1 = Br
1c, R1 = I
6 3
4
a
Reagents and conditions: (a) (1) 1,1,3,3ꢀtetramethylbutyl isocyanide, EtCHO, MgCl , nꢀ
2
BuOH, 130°C, (2) HCOOH 80°C, 56ꢀ78% over 2 steps; (b) (1) MeI, NaH, DMF, rt, (2) HCl,
MeOH, rtꢀreflux 37ꢀ80% over 2 steps; (c) (1) FmocꢀNCS, TEA, DCM, rt then piperidine, rt, (2)
bromoꢀ or chloroacetophenone, EtOH, reflux, 27ꢀ74% over 2 steps.
The synthesis of compounds 1ꢀ6 is represented in Scheme 2. Intermediate 44a was saponified
with NaOH or LiOH and subjected to an EDCIꢀmediated coupling with the corresponding
amines to give compounds 1 and 2. Compound 5 was obtained from 44a by reduction of the ester
to alcohol with LAH followed by a oneꢀpot mesylation/displacement of the mesylate with 4ꢀ
aminoꢀpiperidineꢀ1ꢀcarboxylic acid ethyl ester. Buchwald coupling of 44b and 44c with
morpholine and 2ꢀpiperazinꢀ1ꢀylꢀethanol afforded compounds 4 and 6 respectively. Compound
44b also reacted in a Suzuki coupling with 3ꢀpyridylboronic acid to give 3.
ACS Paragon Plus Environment
25

Page 27 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 2. Synthesis of compounds from Table 1
1, R =
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5, R =
1
a or b
2
, R =
1
44a, R1 = CO Me
2
3, R1=
c or d
4
4b, R = Br
1
4
, R1=
d
6
, R1=
4
4c, R = I
1
a
Reagents and conditions: (a) (1) NaOH, MeOH/H O or LiOH, THF/H O, rt, (2) Amine,
EDC.HCl, TEA, HOBt, DMF, rt, 15ꢀ74% over 2 steps (for 1 and 2); (b) (1) LiAlH , THF, rt, (2)
2
2
4
MsCl, TEA, DCM, rt, then amine, rt, 25% over 2 steps (for 5); (c) pyridineꢀ3ꢀboronic acid,
Na CO , Pd(PPh ) , dioxane/H O, 90°C, 53% (for 3); (d) Amine, tꢀBuONa, Pd dba , Xantphos
2
3
3 4
2
2
3
or DavePhos, toluene or dioxane, 90ꢀ120°C, 7ꢀ16% (for 4 and 6).
The syntheses of substituted piperidine derivatives 7ꢀ10, 17ꢀ20 and 22 from intermediate 44c
are depicted in Scheme 3. Suzuki coupling on 44c with the Nꢀtertꢀbutoxycarbonylꢀ1,2,3,6ꢀ
tetrahydropyridinꢀ4ꢀyl)boronic acid pinacol ester followed by boc deprotection afforded
intermediate 45. Acylation of 45 with acetyl chloride and subsequent reduction of the
tetrahydropyridine double bond allowed the synthesis of 7. Intermediate 46 was synthesized with
good yield from 44c via a Suzuki coupling with Nꢀtertꢀbutoxycarbonylꢀ1,2,3,6ꢀ
tetrahydropyridinꢀ4ꢀyl)boronic acid pinacol ester, reduction of the tetrahydropyridine double
bond using H /Pd(OH) system and final boc deprotection using HCl in MeOH. Intermediate 46
2
2
could then react with a wide range of alkylating agents in presence of strong or weak bases to
ACS Paragon Plus Environment
26

Journal of Medicinal Chemistry
Page 28 of 78
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
afford ester compound 10 and amides 17ꢀ20 and 22. Chloro acetamide alkylating agents were
typically prepared by reaction of chloroacetyl chloride with the corresponding amine as
described for the synthesis of intermediate 47. Reaction of methane sulfonyl chloride with 46
afforded sulfonamide 8. Compound 9 was obtained from 46 by sulfonylation with 3ꢀ
chloropropaneꢀ1ꢀsulfonyl chloride followed by chlorine displacement with dimethylamine in
presence of NaI.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 3. Synthesis of compounds 7ꢀ10, 17ꢀ20 and 22
a, b
c
4
5, R =
7, R1 =
1
44b, R1 = Br
8
, R2
=
4
4c, R1 = I
19, R2
=
=
a, d, b
10, R2
=
e or f
20, R2
4
6, R =
1
or g or h
1
7, R =
2
i
1
8, R =
22, R2 =
2
9, R1
=
j
47
a
Reagents and conditions: (a) NꢀtertꢀButoxycarbonylꢀ1,2,3,6ꢀtetrahydropyridinꢀ4ꢀyl)boronic
acid pinacol ester, Na CO , Pd(PPh ) , dioxane/H O, 85°C, quant.; (b) HCl, THF/H O or
2
3
3 4
2
2
dioxane/MeOH or MeOH, 0°Cꢀrt, 71ꢀ100%; (c) (1) acetyl chloride, TEA, DCM, rt, 91% (2) H2,
Pd/C, EtOH, 60°C, 34% (for 7); (d) H , Pd(OH) , DCM/EtOH, rt, 79%; (e) ethyl 2ꢀ
2
2
bromoacetate, NaH, DMF, rt, 56% (for 10); (f) MsCl, TEA, DCM, 0°Cꢀrt, 17% (for 8); (g) R X,
2
K CO , MeCN, 70ꢀ100°C, 40ꢀ100% (for 17, 18, 19, 22); (h) 2ꢀchloroꢀ1ꢀpyrrolidinꢀ1ꢀylꢀ
2
3
ACS Paragon Plus Environment
27

Page 29 of 78
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ethanone, TEA, MeCN, 190°C (microwave irradiation), 45% (for 20); (i) (1) 3ꢀchloropropaneꢀ1ꢀ
sulfonyl chloride, TEA, DCM, rt, 99%, (2) Dimethylamine, NaI, THF, 150°C (microwave
irradiation), 59% over 2 steps (for 9); (j) chloroacetyl chloride, K CO , H O, DCM, 0°Cꢀrt, 70%.
2
3
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Variation of the piperazine substitution (compounds 11ꢀ16 and 21) was achieved from
intermediate 48 which was obtained from 44b by a Buchwald coupling with Nꢀbocꢀpiperazine
using DavePhos as ligand followed by boc cleavage with HCl (Scheme 4). Intermediate 48 could
be alkylated with ethyl chloroacetate to give compound 11, and subsequent ester hydrolysis with
LiOH afforded carboxylic acid 12. Syntheses of compounds 13, 14 and 21 were performed by
alkylation of intermediate 48 with corresponding alkyl chloride reagents. Whereas 13 and 14
could be obtained at room temperature in presence of TEA, heating at 160°C under microwave
irradiation was required to get 21 in a moderate yield. Finally, intermediate 48 underwent
reductive amination with the corresponding aldehydes to provide compounds 15 and 16.
a
Scheme 4. Synthesis of piperazine derivatives 11ꢀ16 and 21
ACS Paragon Plus Environment
28