Discovery of 2-[1-(4,4-Difluorocyclohexyl)piperidin-4-yl]-6-fluoro-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (NMS-P118): A Potent, Orally Available, and Highly Selective PARP-1 Inhibitor for Cancer Therapy

Article abstract of DOI:10.1021/acs.jmedchem.5b00680

The nuclear protein poly(ADP-ribose) polymerase-1 (PARP-1) has a well-established role in the signaling and repair of DNA and is a prominent target in oncology, as testified by the number of candidates in clinical testing that unselectively target both PARP-1 and its closest isoform PARP-2. The goal of our program was to find a PARP-1 selective inhibitor that would potentially mitigate toxicities arising from cross-inhibition of PARP-2. Thus, an HTS campaign on the proprietary Nerviano Medical Sciences (NMS) chemical collection, followed by SAR optimization, allowed us to discover 2-[1-(4,4-difluorocyclohexyl)piperidin-4-yl]-6-fluoro-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (NMS-P118, 20by). NMS-P118 proved to be a potent, orally available, and highly selective PARP-1 inhibitor endowed with excellent ADME and pharmacokinetic profiles and high efficacy in vivo both as a single agent and in combination with Temozolomide in MDA-MB-436 and Capan-1 xenograft models, respectively. Cocrystal structures of 20by with both PARP-1 and PARP-2 catalytic domain proteins allowed rationalization of the observed selectivity.

Full text of DOI:10.1021/acs.jmedchem.5b00680

Page 1 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Felder, Eduard; Oncology, Nerviano Medical Sciences, Chemical Core
Technologies
Donati, Daniele; Oncology, Nerviano Medical Sciences, Medicinal Chemistry
Montagnoli, Alessia; Oncology, Nerviano Medical Sciences, Cell Biology
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 88
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 of 2ꢀ[1ꢀ(4,4ꢀ
difluorocyclohexyl)piperidinꢀ4ꢀyl]ꢀ6ꢀfluoroꢀ3ꢀoxoꢀ
,3ꢀdihydroꢀ1Hꢀisoindoleꢀ4ꢀcarboxamide (NMSꢀ
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
2
P118): a Potent, Orally Available and Highly
Selective PARPꢀ1 Inhibitor for Cancer Therapy
a
a
a
b
a
Gianluca Papeo, * Helena Posteri, Daniela Borghi, Alina A. Busel, Francesco Caprera,
a
a
a
a
a
Elena Casale, Marina Ciomei, Alessandra Cirla, Emiliana Corti, Matteo D’Anello, Marina
a
a±
a
a
b
Fasolini, Barbara Forte, Arturo Galvani, Antonella Isacchi, Alexander Khvat, Mikhail Y.
b
⊥
a
a
a
a
c
Krasavin, Rosita Lupi, Paolo Orsini, Rita Perego, Enrico Pesenti, Daniele Pezzetta, Sonia
a
a
a
a
a±
Rainoldi, Federico RiccardiꢀSirtori, Alessandra Scolaro, Francesco Sola, Fabio Zuccotto,
a
a
a
Eduard R. Felder, Daniele Donati and Alessia Montagnoli
a
Oncology, Nerviano Medical Sciences S.r.l., Viale Pasteur 10, 20014 Nerviano, Milan, Italy
b
Chemical Diversity Research Institute, Rabochaya St. 2 Khimki, Moscow Region, Russia
114401
c
Accelera S.r.l., Viale Pasteur 10, 20014 Nerviano, Milan, Italy
1
ACS Paragon Plus Environment

Page 3 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT
The nuclear protein poly(ADPꢀribose) polymeraseꢀ1 (PARPꢀ1) has a wellꢀestablished role in the
signaling and repair of DNA and is a prominent target in oncology, as testified by the number of
candidates in clinical testing that unselectively target both PARPꢀ1 and its closest isoform
PARPꢀ2. The goal of our program was to find a PARPꢀ1 selective inhibitor that would
potentially mitigate toxicities arising from crossꢀinhibition of PARPꢀ2. Thus, an HTS campaign
on the proprietary Nerviano Medical Sciences (NMS) chemical collection, followed by SAR
optimization allowed us to discover 2ꢀ[1ꢀ(4,4ꢀdifluorocyclohexyl)piperidinꢀ4ꢀyl]ꢀ6ꢀfluoroꢀ3ꢀoxoꢀ
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
2,3ꢀdihydroꢀ1Hꢀisoindoleꢀ4ꢀcarboxamide (NMSꢀP118, 20by). NMSꢀP118 proved to be a potent,
orally available and highly selective PARPꢀ1 inhibitor endowed with excellent ADME and
pharmacokinetic profiles and high efficacy in vivo both as a single agent and in combination with
temozolomide, in MDAꢀMBꢀ436 and CAPANꢀ1 xenograft models, respectively. Coꢀcrystal
structures of 20by with both PARPꢀ1 and PARPꢀ2 catalytic domain proteins allowed
rationalization of the observed selectivity.
Introduction
The 113 KDa nuclear protein poly(ADPꢀribose) polymeraseꢀ1 (PARPꢀ1) is the most abundant
and wellꢀcharacterized member of the diphtheria toxinꢀlike ADPꢀribosyltransferase (ARTD)
1
2,3
family of enzymes. PARPꢀ1 (aka ARTDꢀ1) exerts its multifaceted biological roles through the
construction of shortꢀlived negatively charged ADPꢀribose homopolymers [poly(ADPꢀribose),
PAR] either on PARPꢀ1 itself (autoꢀmodification) or on different acceptor proteins (heteroꢀ
modification). Transferral of monomer units to the growing poly(ADPꢀribose) chain exploits the
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 88
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
intermediacy of a reactive oxonium ion species which, in turn, arises from the PARPꢀ1ꢀcatalyzed
+
4
detachment of nicotinamide from nicotinamide adenine dinucleotide (NAD ). The wellꢀ
documented involvement of PARPꢀ1 in the signaling and repair of DNA damage as well as its
overꢀactivation in several pathological contexts, triggered a number of aggressive medicinal
chemistry programs aimed at the discovery of PARPꢀ1 inhibitors, potentially useful in
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
therapeutic areas as diverse as stroke, cardiac ischemia, diabetes, inflammation and cancer.
Presently, the primary focus of these efforts still concerns oncology, as testified by the
6
tremendous amount of preꢀclinical and clinical data produced in the field. From an historical
perspective, PARPꢀ1 inhibitors entered the arena as promising coꢀadjuvant components of
standard chemoꢀ and radiotherapy regimens. Later, the discovery that tumor cell lines bearing
deficiencies or mutations in DNAꢀrepair genes (e.g. BRCA1 or BRCA2) do not tolerate PARPꢀ1
7
inhibition, fuelled the application of PARP inhibitors as single agent therapies in breast and
ovarian BRCAꢀmutated cancer settings. More recently, the discovery of new potential
8
9
10
combinative synergisms (e.g. PI3K, NAMPT and EGFR inhibitors) as well as the broadening
1
1
12
of “synthetic lethality” contexts (e.g. PTEN and ATM mutations, MSI colorectal cancer
1
3
14
phenotypes and Ewing’s sarcomas ) in which the inhibition of PARPꢀ1 can be therapeutically
valuable has further raised interest in this target. Currently, eight PARP inhibitors are at different
stages of clinical investigation, targeting several tumors types either in combination or as single
5
agents. These clinical candidates, as well as all known preꢀclinical PARPꢀ1 inhibitors, were
+
designed to imitate the nicotinamide portion of NAD , with which they compete for the
corresponding PARPꢀ1 binding site. Accordingly, PARPꢀ1 inhibitors are systematically endowed
with two peculiar nicotinamideꢀmimic motifs: (1) a rotationally constrained primary amide as,
1
5
for instance, in clinical candidates such as Veliparib (Abbott, Figure 1) and Niraparib
16
(
Merck/Tesaro, Figure 1) or (2) an amide embedded in a ring as in Rucaparib (Pfizer/Clovis,
3
ACS Paragon Plus Environment

Page 5 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
17
18
Figure 1), Olaparib (AstraZeneca, now marketed as Lynparza, Figure 1) and Talazoparib
1
9
(
BioMarin, Figure 1). Within the PARP family of enzymes, none of these inhibitors selectively
inhibits PARPꢀ1. For instance, all the clinical candidates and the vast majority of reported PARP
inhibitors also interact with PARPꢀ2 (84% identity and 90% similarity within the PARP
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
,3a
20
signature motif ) with similar potencies, as recently independently described by Wahlberg
2
1
1
and ourselves. PARPꢀ2 (aka ARTDꢀ2) is a 62 KDa nuclear protein and, like PARPꢀ1, is
involved in DNA single strand break repair. However, its contribution to the total DNA damage
2
2
induced PARP activity is minimal (5 to 10%). Specifically, PARPꢀ2 has been postulated to
participate in later steps of the DNA repair process, by recognizing gaps and flaps and by the
3
delayed and persistent accumulation at UV laser induced damaged sites. Moreover, PARPꢀ2
depleted cells show increased sensitivity to ionizing radiation, indicating a role in the IR induced
DNA damage and suggesting a potential application for PARPꢀ2 inhibitors in combination with
23
irradiation. Aside from PARPꢀ1 and ꢀ2 overlapping functions, as clearly demonstrated by the
ꢀ/ꢀ
embryonic lethality in double knockout mice, PARPꢀ2 single knockout mice show impaired
ꢀ/ꢀ
spermatogenesis, adipogenesis and thymopoiesis. These PARPꢀ2 mice furthermore suffer from
increased neuronal loss after ischemic damage and are more prone to develop pancreatitis
2
2
following chemical insult. In addition, loss of PARPꢀ2 has recently been shown to shorten
erythrocytes lifespan and to impair differentiation of erythroid progenitors, leading to chronic
2
4
anemia. As all these physiological functions of PARPꢀ2 can potentially result in undesirable
sideꢀeffects following its inhibition, we reasoned that a potent and highly selective PARPꢀ1
inhibitor, if equally efficacious in cancer therapy as are dual PARPꢀ1/ꢀ2 one, might represent a
significant advancement in the field. Crucial data that prompted us to embark upon this strategy
7
a
25
were the findings of Bryant and Sharp, proving that PARPꢀ1 genetic depletion is sufficient to
induce death of BRCA2ꢀ and BRCA1ꢀdeficient tumor cell lines respectively, together with the
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 88
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
26,27
feasibility of PARPꢀ1 selective inhibitors (1 and 2 (BYK204165), Figure 2),
which were
however designed for different therapeutic applications and did not progress to clinical testing.
Herein we describe the synthesis and SAR investigation of isoindolinoneꢀ4ꢀcarboxamide
derivatives, leading to the discovery of 2ꢀ[1ꢀ(4,4ꢀdifluorocyclohexyl)piperidinꢀ4ꢀyl]ꢀ6ꢀfluoroꢀ3ꢀ
oxoꢀ2,3ꢀdihydroꢀ1Hꢀisoindoleꢀ4ꢀcarboxamide (NMSꢀP118, 20by, Figure 2). NMSꢀP118 is a
potent (K = 0.009 ꢀM) PARPꢀ1 inhibitor, showing 150ꢀfold selectivity over PARPꢀ2 (K = 1.39
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
D
D
ꢀ
M). NMSꢀP118 possesses excellent pharmacokinetic profile and nearly complete oral
bioavailability both in mice and rats. It proved to be highly efficacious in vivo both as single
agent in MDAꢀMBꢀ436 human breast cancer tumors and in combination with temozolomide in
CAPANꢀ1 human pancreatic tumors growing as xenografts in the mouse. The compound is well
tolerated at highly efficacious doses and is endowed with an excellent ADME profile.
Chemistry
Two synthetic pathways were exploited to access the planned isoindolinoneꢀ4ꢀcarboxamide
derivatives. The first one capitalizes on an intramolecular DielsꢀAlder reaction as the key step
2
8
(
Scheme 1A). Thus furfural was initially subjected to a reductive amination protocol in the
presence of suitable amines (6aꢀu) by preꢀforming the Schiff bases which were then reduced with
sodium borohydride. The corresponding properly functionalized furanylmethylamines (7aꢀu)
reacted with maleic anhydride in a twoꢀstep fashion, by first delivering the transient maleic acid
2
9
monoamide which then underwent an intramolecular DielsꢀAlder reaction. As anticipated, the
process is highly stereoselective, with only the exoꢀisomer being detected in the reaction mixture.
The exoꢀ tricyclic adducts (8aꢀu) were then dehydrated by exposure to concentrated aqueous
hydrochloric acid at reflux, followed by conversion of the resulting carboxylic acids (9aꢀu) into
the desired primary amides using the ammonium salt of hydroxybenzotriazole in the presence of
5
ACS Paragon Plus Environment

Page 7 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
EDCI as the coupling agent (10aꢀu). Despite the lack of a common intermediate, the synthetic
sequence depicted in Scheme 1A is nonetheless straightforward, since isoindolinones precursors
(7aꢀu)ꢀ(9aꢀu) did not require any purification step. Further manipulation of the eastern portion of
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
the isoindolinone core was achieved (Scheme 1B) by elaboration of the piperidinꢀ4ꢀylꢀ moiety of
3
0
11 after debenzylation of 10u. Compound 11 was then subjected to either acylation with acyl
chlorides in pyridine or reductive alkylation in different conditions according to the carbonylꢀ
counterpart affording compound 12a-c or 13a-ag respectively. To further investigate SAR within
this chemical class, a second synthetic route to access benzoꢀsubstituted isoindolinoneꢀ4ꢀ
carboxamide analogues was pursued (Scheme 2). The key step of this alternative approach is the
ortho iodination of properly functionalized 2ꢀmethylbenzoic acids (14aꢀd) through CꢀH
3
1
activation, as described by Yu et al. Intermediates 15aꢀd were then transformed into the
corresponding methyl esters 16aꢀd before being subjected to a radical bromination at the benzylic
position. Crude benzyl bromides underwent a one pot nucleophilic displacement – ring closure
2
8c, 30, 32
33
sequence
in the presence of easily accessible amines (17xꢀy), thus yielding iodoꢀ
isoindolinone intermediates 18. A Rosenmund – von Braun protocol was subsequently applied to
efficiently replace the iodine atom with a cyano group. Finally, the resulting isoindolinoneꢀ4ꢀ
carbonitriles (19) were smoothly converted into the corresponding primary amides through a
34
recently disclosed indium (III) chlorideꢀcatalyzed transfer of water from acetaldoxime in
refluxing toluene, producing compounds 20bx and 20(a-d)y.
Results and Discussion
By using our recently reported PARPꢀ1 full length (FL) fluorescence polarization displacement
2
1
assay, a high throughput screening (HTS) campaign was conducted on the NMS proprietary
chemical collection. Two isoindolinoneꢀ4ꢀcarboxamide derivatives (10a and 10b, Table 1)
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 88
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
emerged from this screening, possessing good biochemical activity on PARPꢀ1 (each showing
4
K = 0.09 ꢀM). The isoindolinone core is a known privileged scaffold for PARP inhibition,
D
35
either as a feature embedded within a tetracyclic structure (3, Figure 2) or also, after appropriate
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
+
32
linkage to adenosine, in NAD ꢀresembling compounds such as EBꢀ47 (4, Figure 2). Such
inhibitors rely on the isoindolinone lactam moiety as a nicotinamide mimic. Subsequently,
Abbott disclosed a series of 2ꢀsubstituted isoindolinoneꢀ4ꢀcarboxamide derivatives (e.g. 5, Figure
3
0
2
), in which the carbonyl oxygen atom of the scaffold freezes the primary amide into its
biologically active conformation through an intramolecular hydrogen bond, according to the
alternative PARPꢀ1 pharmacophore discussed above. All these compounds proved to be potent
2
1,30
dual PARPꢀ1/ꢀ2 inhibitors.
The two low molecular weight compounds which emerged from our HTS campaign are highly
36
efficient ligands of PARPꢀ1 (10a: MW = 218 Da, Binding Efficiency Index (BEI) = 32.3,
Ligand Efficiency (LE) = 0.6; 10b: MW = 248 Da, BEI = 28.4, LE = 0.54). Both compounds
exhibited the desired mechanism of action in HeLa cells, i.e. both were able to inhibit PARPꢀ1
dependent PAR synthesis in cells at micromolar concentrations following H O ꢀinduced DNA
2
2
2
1
damage (PAR assay, Table 1). However, a fluorescence polarization displacement assay using
PARPꢀ2 showed that both 10a and 10b also significantly bind this isoform (Table 1). Despite the
moderate cellular activity and the lack of selectivity, the high intrinsic potency of these two hits
deserved further investigation. The Xꢀray coꢀcrystal structures of 10b in complex with both
human PARPꢀ1 (hPARPꢀ1, Figure 3A) and human PARPꢀ2 (hPARPꢀ2, Figure 3B) catalytic
domains were solved in order to gain insight into the corresponding binding modes. As
3
0
anticipated, the pseudo 7ꢀmembered ring, arising from a γꢀturnꢀlike intramolecular hydrogen
bond, locked the isoindolinoneꢀ4ꢀcarboxamide core into the anti conformation. This
7
ACS Paragon Plus Environment

Page 9 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intramolecular interaction minimizes negative entropic contribution to the binding energy
balance, and allows 10b to establish the usual network of hydrogen bonds within the proteins’
catalytic domains (i.e. with Gly863 and Ser904 in hPARPꢀ1 and with Gly429 and Ser470 in
hPARPꢀ2) and πꢀstacking (with Tyr907 in hPARPꢀ1 and Tyr473 in hPARPꢀ2) of a prototypical
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
4
nicotinamideꢀmimic inhibitor. In order to corroborate the presence of an intramolecular
1
hydrogen bond not only in the crystal form, but also in solution, HNMR spectra of 10b in
solvents with different polarity were recorded (Figure 4). The chemical shift difference (ꢁδ =
3
.08 ppm) between the downfield intramolecularly bonded amide proton and the upfield nonꢀ
bonded one observed in a polar solvent such as DMSOꢀd (Figure 4A) is fully consistent with
6
3
7
this kind of interaction. As expected, this difference was further exacerbated (ꢁδ = 5.44 ppm)
in a non polar solvent such as CDCl (Figure 4B). These data were further consolidated by a
3
3
8
quantitative assessment of the intramolecular hydrogen bond strength (See Supporting
Information for details). A closer inspection of the Xꢀray coꢀcrystal structures of compound 10b
revealed that the methoxypropylꢀ side chain did not actively participate in binding of the inhibitor
either to PARPꢀ1 or to PARPꢀ2, thus affording the possible opportunity of increasing PARPꢀ1
potency by tailoring the substituent onto the lactam nitrogen atom. To this purpose, an overlay of
coꢀcrystal structures of 10b and the selective quinazolinone inhibitor 1 (PDB Code: 1UK0)
within hPARPꢀ1 catalytic domain was analyzed (Figure 5). This exercise clearly ruled out the
possibility to achieve selectivity against PARPꢀ2 by engaging contacts within the adenineꢀribose
2
6a
(
AD) binding site, as the substituents departing from 10b and 1 point at different portions of
the protein (see gray arrows in Figure 5). Nevertheless, compounds dramatically different in
27
shape such as BYK204165 (2) indicated that selectivity against PARPꢀ2 might be achieved by
targeting regions of the PARPꢀ1 catalytic domain other than the AD site. Thus, a systematic
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 88
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
exploration of this chemical class was undertaken alongside determination of compound
2
1
selection criteria, based on biochemical potency (PARPꢀ1 K ≤ 0.1 ꢀM ) and selectivity (PARPꢀ
D
3
9
2
K /PARPꢀ1 K ratio of about 100 ) thresholds. Inhibitors endowed with these features would
D D
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
then be tested to quantify their ability to suppress H O ꢀinduced PAR synthesis in HeLa cells,
2
2
which is exclusively PARPꢀ1 dependent (PAR assay, see above). Derivatives possessing IC ≤
5
0
0.1 ꢀM in this cellular assay would be considered eligible for preclinical profiling. Good oral
bioavailability of the selected compounds was considered mandatory for further development.
After having established these cutoff criteria, HTS hits 10a and 10b were employed as a footprint
to design a first array of isoindolinones. This set of compounds was assembled by varying: (a) the
length of the aliphatic spacer connecting the nitrogen atom of the bicyclic core to the terminal
eastern substituent and (b) the nature of the eastern substituent itself. Results are reported in
Table 1. Thus, within the alkoxyꢀsubstituted series, shortening the spacer proved to be
detrimental in terms of biochemical potency, regardless of whether the alkoxy group was linear
or whether the oxygen atom was embedded in a cycle (compare 10c and (±)ꢀ10d with 10b). On
the contrary, a hydroxyethyl substituent (10e) restored potency but did not possess PARPꢀ1 vs.
PARPꢀ2 selectivity. The biochemical activity of phenylꢀsubstituted derivatives increased one
order of magnitude upon going from benzyl (10f) to 2ꢀphenylethyl (10g), with the latter being
unacceptably potent also against PARPꢀ2. A further elongation of the spacer (10h) impacted
neither potency nor selectivity. By replacing the phenyl ring present in 10g with a pyridinꢀ2ꢀyl
moiety (10i) an increase in potency, mainly against PARPꢀ2, was observed, with a K on both
D
2
1
isoforms below the assay sensitivity limit. By capitalizing on this result, we then investigated a
series of isoindolinones encompassing differently substituted tertiary aliphatic amines located at
a variable distance from the scaffold. Both the morpholinꢀ4ꢀylꢀethylꢀ (10j) and the 1ꢀpiperidinꢀ1ꢀ
9
ACS Paragon Plus Environment

Page 11 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ylꢀethylꢀ (10l) derivatives showed a lower potency against PARPꢀ1 than the corresponding higher
homologues (10k and 10m). Furthermore, compounds 10k and 10m showed, respectively, 160ꢀ
fold and > 110ꢀfold selectivity towards PARPꢀ1 over PARPꢀ2 and were thus also profiled in the
cellular PAR assay. While isoindolinone 10k showed no significant inhibition of PAR synthesis
in cells, the IC value of 10m, although modest, prompted us to further decorate the piperidinyl
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
moiety in an effort to obtain more potent derivatives. The introduction of a 2,6ꢀdimethyl
substitution (cisꢀ10n), as well as an additional piperidinyl ring (10o) resulted in a dramatic drop
of activity. On the contrary, a lipophilic substituent such as a benzyl group in the 4 position of the
piperidinyl ring (10p) restored the biochemical potency on both enzymes. The
tetrahydroisoquinolinꢀ2ꢀylꢀethyl derivative (10r) proved to be more potent against PARPꢀ1 and
more selective towards PARPꢀ2 then the corresponding isomeric tetrahydroquinolinꢀ2ꢀylꢀethyl
substituted isoindolinone (10q). Unfortunately, 10r showed negligible activity in the cellular
PAR assay. A further investigation was then conducted by moving the linker connecting the
isoindolinone core to the piperidinyl ring from the 1ꢀ to the 4 position of the latter. The resulting
derivatives were garnished with a lipophilic substituent onto the piperidinyl nitrogen atom,
affording compounds 10s and 10t. While derivative 10s strongly inhibited PARPꢀ1 but its
selectivity towards PARPꢀ2 was below our arbitrary threshold, the corresponding regioisomer
10t showed high biochemical potency on the target, excellent selectivity and cellular activity.
Compound 10t was then subjected to in vitro ADME profiling and pharmacokinetic analysis in
mice (Table 2). Disappointingly, despite the excellent solubility and good permeability, 10t
showed high clearance in vivo (230 mL/min/kg) and negligible (< 1%) oral bioavailability thus
preventing any further development. The preliminary SAR information emerged from this first
cohort of isoindolinones underlines that a threeꢀ to fourꢀcarbon unit spacer between the bicyclic
core and a basic center generally delivers the most potent and selective PARPꢀ1 inhibitors. This
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 88
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
observation prompted us to investigate the modulation of the potency and selectivity by
embedding the spacer into a piperidine ring, whose nitrogen was initially decorated with a benzyl
group (10u, Table 1). This lipophilic moiety was carefully chosen either because of its impact on
the inhibitory activity within the series (see 10p, 10s and 10t, Table 1) and because it can be
easily replaced by standard chemistry (Scheme 1B). Compound 10u proved to be quite potent
against PARPꢀ1 and exceedingly selective. However, its ability in inhibiting the PARylation in
cells was trifling (IC = 2 µM). A second array of piperidinꢀ4ꢀyl substituted isoindolinones was
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
then synthesized in order both to implement the biochemical and cellular activity and the
pharmacokinetic profile of the inhibitor. Results are reported in Table 3. The mandatory
requirement of a basic nitrogen atom within this subseries was corroborated by the modest
biochemical potency against PARPꢀ1 of amide derivatives such as 12aꢀc. Linear or branched
alkyl substituted piperidine (13aꢀe) delivered potent unselective compounds with flat SAR, while
the presence of an oxygen atom within the aliphatic chain (13f) depressed the ability in inhibiting
the target. More SAR information was gathered by playing around the initially selected benzyl
substituent (10u). Thus, the introduction of a simple (electron donating) methyl substituent in the
ortho, meta and para position of the phenyl ring, respectively, afforded both biochemically
potent and selective compounds (13gꢀi) whose enzymatic activity (13i = 13h > 13g) did not
stringently mirror the cellular one (13i > 13g > 13h). Moderately deactivating groups such as
bromine (13jꢀl) again confirmed the positive impact of a para substitution (13l) on both
biochemical potency and selectivity, however with no significant activity in cells. Strong electron
withdrawing (13mꢀo) or donating (13p) groups generally worsened the inhibitor profile by
delivering poorly active (13m), poorly selective (13o) or compounds lacking cellular activity
(13n and 13p). The replacement of the phenyl ring with a pyridine (13qꢀs) produced potent and
selective compounds and, in the case of the pyridinꢀ2ꢀylmethyl derivative (13s), also with good
11
ACS Paragon Plus Environment

Page 13 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
despite still suboptimal) cellular activity. Biochemically potent PARPꢀ1 inhibitors also emerged
by replacing the phenyl ring with a fiveꢀmembered heterocycle such as thiophene (13tꢀu), furan
13vꢀz) and pyrrole (13aaꢀab). However, the pair of compounds (13u and 13ab) which proved
(
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
also to be selective against PARPꢀ2 according to our screening criteria delivered disappointing
results in PAR assay. Indole regioisomers (13acꢀad) behaved quite differently. While 13ac was
not selective enough for further profiling, the potent and selective 5ꢀsubstituted regioisomer
(13ad) displayed an excellent cellular activity. Finally, few isoindolinones decorated with nonꢀ
aromatic rings directly linked to the piperidine moiety (13aeꢀag) were investigated. Despite being
4
0
both potent and exceedingly selective, the 4,4ꢀdifluorocyclohexyl substituted isoindolinone
(13af) proved to be four times more active in cellular assay then the corresponding tetrahydroꢀ
2Hꢀpyranꢀ4ꢀyl derivative (13ae). On the contrary, 1,3ꢀdioxolaneꢀcontaining compound 13ag
showed only modest inhibitory activity against PARPꢀ1. Having identified three compounds (13i,
13ad and 13af) possessing the suitable biochemical potency, selectivity and cellular activity, they
were further profiled by analyzing their in vitro ADME and pharmacokinetic properties. Whereas
permeability, plasma protein binding and HLM clearance data were, on average, quite similar
between the compounds, isoindolinone 13af showed a definitely higher solubility (Table 4). The
PK profiles in mouse allowed selecting 13af as the lead compound of this isoindolinones subset
(Table 5). In fact, while 13i demonstrated to possess low oral bioavailability (15%) and 13ad,
despite being orally available (80%), it is rapidly cleared from the body (110 mL/min/kg), 13af
.
proved to be suitable for oral administration (80%) with high exposure (21 µM h) and low
clearance (17 mL/min/kg). With the proper arrangement of the right end portion of the inhibitor
in hand, a final medicinal chemistry refinement was undertaken in the attempt to further improve
1
3af cellular activity. The potency of 13af in PAR assay was indeed suboptimal (IC = 0.15
50
µM). The planned optimization was focused on the preparation of a handful of compounds
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 88
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
bearing small substituents on the phenyl ring of the isoindolinone bicyclic core, as the features of
the back wall of PARPꢀ1 hosting the nicotinamide moiety prevents the introduction of large
4
groups. The biochemical and cellular activities of this set of inhibitors are reported in Table 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
As none of the substituents introduced in the different positions of the aromatic ring are expected
to deliver any steric clash within the catalytic site of the protein, SAR rationalization has to be
found in the substituents’ aptitude in modulating the stereoelectronic properties of the inhibitors.
Thus, the tremendous drop in biochemical activity of 20ay against PARPꢀ1 was attributed to the
loosening of the intramolecular hydrogen bond between the primary amide and the carbonyl
oxygen by the fluorine atom in position 5. This effect can be appreciated observing the small
1
chemical shift differences between the two amide protons in the H NMR spectra of 20ay (Table
3
7,38
7 and Supplementary Table 1),
and is reasonably due to electron pair repulsion leading to
conformational destabilization where the fluorine atom and the carbonyl are coplanar. A
reduction in the intramolecular hydrogen bond strength, again measured as a decrease in the ꢁδ
chemical shift between the two amide protons might also account for the lower inhibitory activity
displayed by the 7ꢀfluoro isoindolinone 20cy compared to 13af (Table 7 and Supplementary
Table 1). On the other hand, in order to rationalize the high biochemical potency observed for
both the 6ꢀfluoro 20by and the corresponding 6ꢀchloro derivative 20dy, additional interactions
within the PARPꢀ1 protein binding site were presumed (and later demonstrated, see below) to
counterbalance the substituent’s negative impact on hydrogen bond strength (Table 7 and
Supplementary Table 1). Since 20by was the most potent compound in PAR assay (IC = 0.04
5
0
4
1
µM), the replacement of fluorineꢀ with chlorine atoms onto the cyclohexyl moiety was then
investigated by synthesizing compound 20bx (Table 6). Isoindolinone 20bx proved to be
biochemically potent, selective against PARPꢀ2 (as were all tested examples belonging to this
subseries) and displayed an excellent activity in cells, thus demonstrating the importance of a
13
ACS Paragon Plus Environment

Page 15 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
properly 4ꢀsubstituted cyclohexyl moiety to deliver PARPꢀ1 inhibitors effective at a cellular
level. In order to gain a deeper insight into 20by selectivity profile, the compound was tested
2
1
against PARPꢀ3 FL (aka ARTDꢀ3; K = 0.69 µM), TNKSꢀ1 catalytic domain (aka PARPꢀ5a,
D
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
21
ARTDꢀ5; K > 10 µM), sirtuin 1 (SIRT1; IC > 10 µM). Since clinical PARP inhibitors, such
D
50
4
2
as rucaparib and veliparib, show micromolar inhibitory activity also on some kinases,
compound 20by was also tested on a panel of 56 different kinases to assess its selectivity (IC >
5
0
1
0 µM; see experimental section for details). Furthermore, to confirm the fluorescence
polarization displacement assay data, 20by, along with 13af and the (unselective) 13c were also
tested against PARPꢀ1 and PARPꢀ2 catalytic domain by Surface Plasmon Resonance (SPR), an
2
1
independent binding assay. For comparative purposes, also the clinical candidates olaparib and
2
1
veliparib (Figure 1) were tested in this assay. SPR results are reported in Table 8. While 13c, as
expected (Table 2), showed similar potency against both isoforms, isoindolinone 13af exhibits a
PARPꢀ1 K = 0.016 ꢀM and has approximately 75ꢀfold less affinity for PARPꢀ2. Finally,
D
compound 20by proved to be a single digit nanomolar PARPꢀ1 inhibitor (K = 0.009 ꢀM), with
D
exquisite selectivity versus PARPꢀ2 (KD = 1.39 ꢀM). Detailed kinetic parameters
(
Supplementary Table 2) and kinetic analysis (Supplementary Figure 1) of this assay can be
found in Supporting Information. The approximately 30% decrease in mere efficiency of binding
(MW = 395 Da, BEI = 20.4, LE = 0.39), compared to the initial hits’ efficiency (vide supra),
underscores the fact that late stage optimization of 20by favoured modifications contributing to
the overall developability and bioactivity, not just focusing on molecular binding interactions
alone.
Compound 20by was tested along with 13af for the ability to inhibit proliferation of MDAꢀMBꢀ
4
36, a triple negative (HER2/ER/PR –ve) breast cancer cell line carrying BRCA1 gene mutation.
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 88
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
The isoindolinone 20by resulted about 4 times more potent compared to 13af, with IC of 0.14
5
0
and 0.6 ꢀM respectively, reflecting the higher target inhibition in PAR assay. Cellular models
such as Mcf7 and MIAꢀPACA that do not have BRCA mutations or HR defects were resistant to
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
both compounds with IC s >10 ꢀM (Table 9). Aiming at rationalizing the high selectivity
50
against PARPꢀ2 showed by 20by, its Xꢀray coꢀcrystal structures within the catalytic domains of
hPARPꢀ1 (Figure 6A) and hPARPꢀ2 (Figure 6B) were solved. As expected, the contacts
established by the isoindolinoneꢀ4ꢀcarboxamide core of 20by mirrored those of 10b (see Figure
3A and 3B). Additionally, the fluorine atom at position 6 fills a small cavity present in the
binding sites of both proteins (defined by Phe897, Ala898, Lys903 and Glu988 residues in
PARPꢀ1 and by Phe463, Ala464, Lys469 and Glu558 residues in PARPꢀ2) and interacts with the
backbone carbonyl of Phe897 in PARPꢀ1 and Phe463 in PARPꢀ2. The (4,4ꢀ
difluorocyclohexyl)piperidinylꢀ moiety points towards the donor loop and the αꢀhelical bundle
domain, thus allowing the 4,4ꢀdifluorocyclohexyl ring to fill an induced pocket, to trigger the
rearrangement of αꢀhelix 5, and to establish favorable hydrophobic interactions with a tyrosine
residue of the donor loop (Tyr889 in PARPꢀ1 and Tyr455 in PARPꢀ2). The slightly dissimilar
orientation of αꢀhelix 5 in the two proteins, together with the presence of different residues
(Gln759 and Val762 in PARPꢀ1 and Ile328 and Ser331 in PARPꢀ2) may account for the
selectivity observed, as the induced pocket is larger in PARPꢀ1 (Figure 6A) than in PARPꢀ2
(Figure 6B). As a result, the 4,4ꢀdifluorocyclohexyl substituent is better accommodated in PARPꢀ
1. The observation that α helix 5 is intrinsically closer to the active site in PARPꢀ2 than in
4
3
PARPꢀ1, further supports the hypothesis that the aforementioned rearrangement is energetically
less favorable in PARPꢀ2. In order to better characterize 20by, experiments of cross species
metabolic stability in liver hepatocytes (Supplementary Table 3), cytochrome P450 inhibition
15
ACS Paragon Plus Environment

Page 17 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Supplementary Table 4) and in vitro myelotoxicity across species (Supplementary Table 5), the
latter in comparison with olaparib (Figure 1), were performed. Compound 20by proved to be
metabolically stable, it modestly inhibited two cytochrome P450 family members (CYPꢀ2B6
IC : 8.15 µM; CYPꢀ2D6 IC : 9.51 µM) out of eight isoforms tested. Its ability in hampering the
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
50
proliferation of bone marrow cells was from 5 to > 60 times lower then olaparib according to the
species.
Overall, compound 20by appeared the most promising and was further progressed by evaluating
its ADME and pharmacokinetic parameters in mouse, which proved to be excellent (Table 10).
The compound showed high solubility and permeability, low in vitro and in vivo clearance and
complete oral bioavailability. The pharmacokinetic profile of 20by in rat dosed iv at 10 mg/kg
and orally at 10 and 100 mg/kg (Table 10), mirrored that observed the in mouse, with oral
bioavailability >65%, and linearity of exposure with dose.
To assess the in vivo antiꢀtumor activity of compound 20by when used as single agent, we treated
nude mice bearing established subcutaneous MDAꢀMBꢀ436 tumor xenografts (Figure 7). Oral
administration of compound 20by for 28 days (once a day dose of 150 mg/kg) as a methocel
suspension significantly inhibited tumor growth, inducing complete response (tumor impalpable)
in all treated mice with no signs of toxicity or body weight loss. Six out of seven mice were still
tumor free one month after the end of treatment.
We also examined the effect of combining 20by with temozolomide in vivo. Tumor growth of
subcutaneously implanted Capanꢀ1 pancreatic xenografts was significantly inhibited when
animals were treated by oral administration of 100 mg/kg 20by in combination with 50 mg/kg of
temozolomide. Even in the absence of significant anti tumor activity for the two compounds
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 88
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
given as single agent, 20by strongly potentiated temozolomide, resulting in complete tumor
regressions in 3 out of 6 animals (Figure 8). Only a slight increase in body weight loss was
observed compared to the group treated with temozolomide alone.
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
To assess the in vivo pharmacodynamics of 20by, Capanꢀ1 xenograftꢀbearing mice received a
single oral administration of 100 mg/kg of compound. Tumors were harvested at 1, 2, 6 and 24
hours after treatment, and intraꢀtumor PAR levels were determined by ELISA. 20by treatment
dramatically decreased intraꢀtumoral PAR levels at 1, 2 and 6 hours after administration,
confirming its expected mechanism of action also in vivo. Partial recovery of PAR levels was
observed at 24 hours (Figure 9).
Conclusion
PARP inhibitors currently undergoing clinical trials in oncology are unselective PARPꢀ1/ꢀ2
inhibitors. Poly(ADPꢀribose) polymeraseꢀ2 (PARPꢀ2), the closest isoform of PARPꢀ1, has a
number of reported physiological functions other than its subsidiary role in DNA repair. These
considerations inspired our efforts towards developing selective PARPꢀ1 inhibitors as potentially
better tolerated drugs, especially in view of potential drug/drug combination settings. An HTS
campaign on our proprietary collection was performed, followed by an intensive medicinal
chemistry exploration around the resulting moderately potent, but unselective isoindolinoneꢀ4ꢀ
carboxamide hits 10a and 10b. Investigation of the substituent on the nitrogen atom of the bicylic
core allowed the discovery of the selective 13af, whose cellular activity and pharmacokinetic
properties were however still suboptimal. To further gain insight into the SAR of isoindolinoneꢀ
4ꢀcarboxamides, a completely different synthetic route was then implemented to access
derivatives modified on the aromatic ring of the bicyclic scaffold. These efforts successfully
17
ACS Paragon Plus Environment

Page 19 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
culminated in the discovery of 20by (NMSꢀP118), a potent and exceedingly selective PARPꢀ1
inhibitor, whose unprecedented selectivity might reside in its distinct binding poses within the
catalytic domains of PARPꢀ1 and PARPꢀ2. NMSꢀP118 shows excellent ADME and
pharmacokinetic profiles, high oral availability in the mouse and rat and high efficacy both as a
single agent and in combination with temozolomide in BRCA1ꢀmutated MDAꢀMBꢀ436 and
BRCAꢀ2 deficient CAPANꢀ1 human tumor xenograft models, respectively. NMSꢀP118 was
found to be less myelotoxic in vitro than olaparib (now marketed as Lynparza), a dual PARPꢀ1/ꢀ2
inhibitor. NMSꢀP118 is, to our knowledge, the first PARPꢀ1 selective inhibitor with
demonstrated anticancer activity as single agent, as well as in combination, and thus we provide
compelling proofꢀofꢀconcept that the sole pharmacological inhibition of PARPꢀ1 vs. PARPꢀ2 is
sufficient for acheiving high antitumor efficacy in BRCA deficient tumor settings.
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
Experimental Section
1. Chemistry. All solvents and reagents, unless otherwise stated, were high grade, commercially
available and were used without further purification. All experiments dealing with moistureꢀ
sensitive compounds were conducted under dry nitrogen or argon. Organic solutions were
evaporated using a Heidolph WB 2001 rotary evaporator at 15ꢀ20 mmHg. Thinꢀlayer
chromatography was performed on Merck silica gel 60 F254 preꢀcoated plates. Flash
Chromatography was performed on silica gel (Merck grade 9395, 60A). All tested compounds,
reported in the present paper, possess a purity of at least 95% as determined by an HPLCꢀUV/MS
method. HPLC Analyses were performed on Waters XTerra RP18 (4,6 x 50 mm, 3.5 ꢁm) column
using a Waters 2790 HPLC system equipped with a 996 Waters PDA detector and Micromass
ZQ single quadrupole mass spectrometer, equipped with an electrospray (ESI) ion source. Mobile
phase A was ammonium acetate 5 mM buffer (pH 5.5 with AcOHꢀacetonitrile 95:5), and Mobile
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 88
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
phase B was waterꢀacetonitrile (5:95). Gradient from 10 to 90% B in 8 minutes, hold 90% B 2
minutes. UV detection at 220 nm and 254 nm. Flow rate 1 mL/min. Injection volume 10 ꢁL. Full
scan mass spectra were recorded in the mass range of 100ꢀ800 amu. Capillary voltage was 2.5
KV; source temperature was 120 °C; cone was 10 V. Mass are given as m/z ratio. When
necessary, compounds were purified by preparative HPLC on a Waters Symmetry C18 (19 x 50
mm, 5 ꢁm) column or on a Waters XTerra RP 18 (30 x 150 mm, 5 ꢁm) column using a Waters
preparative HPLC 600 equipped with a 996 Waters PDA detector and a Micromass ZMD single
quadrupole mass spectrometer, electrospray ionization (positive ion mode). Mobile phase A was
waterꢀ0.01% TFA, and mobile phase B was acetonitrile. Gradient from 10 to 90% B in 8
minutes, hold 90% B 2 minutes. Flow rate 20 mL/min. Alternatively, mobile phase A was waterꢀ
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
0.1% NH OH, and mobile phase B was acetonitrile. Gradient from 10 to 100% B in 8 minutes,
4
1
hold 100% B 2 minutes. Flow rate 20 mL/min. HꢀNMR spectra were recorded at 28 °C on a
1
Varian INOVA 400 spectrometer operating at 400.5 MHz for H and equipped with 5 mm
1
15
31
H{ N, P} z axis PFG Indirect Detection Probe, at 25 °C on a Varian INOVA 500
1
1
13
15
spectrometer operating at 499.75 MHz for H and equipped with 5 mm H{ C, N} z axis PFG
Indirect Detection Probe and on a Varian Mercury 300 spectrometer operating at 300.5 MHz for
1
1
19
13 31
H and equipped with 5mm PFG autoswitchable H, F, C P probe. Spectra were recorded in
DMSOꢀd or CDCl . Chemical shifts were referenced with respect to the residual solvent signal
6
3
(
DMSOꢀd : 2.50 ppm, CDCl : 7.27 ppm). Data are reported as follows: chemical shift (δ),
6 3
multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, bs = broad signal, td = triplet of
doublets, dd = doublet of doublets, ddd = doublet of doublets of doublets, m = multiplet),
4
4
coupling constants (Hz), and number of protons. As formerly reported, ESI(+) highꢀresolution
mass spectra (HRMS) were obtained on a QꢀTof Ultima (Waters, Manchester, UK) mass
spectrometer directly connected with an Agilent 1100 microꢀHPLC system (Palo Alto, US).
19
ACS Paragon Plus Environment

Page 21 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
-benzyl-N-(furan-2-ylmethyl)piperidin-4-amine (7u). An equimolar solution of furanꢀ2ꢀ
carbaldehyde (4.8 g, 50 mmol) and 1ꢀbenzylpiperidinꢀ4ꢀamine 6u (9.5 g, 50 mmol) in dioxane
30 mL) was heated to reflux for 8 hours by employing a DeanꢀStark apparatus. The reaction
(
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
mixture was concentrated under vacuum and rinsed with ethanol (1000 mL). Sodium
borohydride (2.08 g, 55 mmol) was added and the mixture was left overnight at room
temperature. Then it was diluted with water and extracted with diethyl ether. The organic phase
was dried over anhydrous sodium sulfate and concentrated under reduced pressure to give the
title compound, employed in the following steps without any further purification.
2-(1-benzylpiperidin-4-yl)-1-oxo-1,2,3,6,7,7a-hexahydro-3a,6-epoxyisoindole-7-carboxylic
acid (8u). To a solution of 7u (13.3 g, 50 mmol) in tetrahydrofuran (140 mL) maleic anhydride
(4.9 g, 50 mmol) was added. The reaction mixture was refluxed for 6 hours and stirred overnight
at room temperature. The precipitate solid obtained was filtered, washed with diethyl ether and
+
dried to give the desired compound as a white solid. ESI(+) MS: m/z 369 (MH ).
2-(1-benzylpiperidin-4-yl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxylic
acid
(9u).
Compound 8u (18.4 g, 50 mmol) was dissolved in 37% hydrochloric acid (150 mL) and the
solution was refluxed for 3 hours. The solvent was removed under reduced pressure to afford the
+
+
resulting crude as hydrochloride salt. HRMS (ESI+): calcd. for C H N O [M + H] 351.1703;
2
1
23
2
3
found 351.1706.
-(1-benzylpiperidin-4-yl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10u).
To a solution of 9u (19.3 g, 50 mmol) in N,Nꢀdimethylformamide (300 mL)
hydroxybenzotriazole ammonium salt (15.2 g, 100 mmol), 1ꢀethylꢀ3ꢀ(3’ꢀ
2
dimethylamino)carbodiimide hydrochloric acid salt (19.2 g, 100 mmol) and triethylamine (17
mL, 200 mmol) were added. The reaction mixture was stirred at room temperature overnight.
The solvent was removed under reduced pressure and the residue was dissolved in ethyl acetate.
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 88
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
The solution was washed with 15% ammonium hydroxide and the organic phase was dried over
anhydrous sodium sulfate and concentrated in vacuo. The crude was purified by flash
chromatography (dichloromethane/methanol 97:3) to afford the title compound (3.5 g, 20% over
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
four steps) as a white solid. H NMR (400 MHz, DMSOꢀd )
δ 10.72 (bs, 1H), 8.20 (dd, J = 7.4,
6
1.5 Hz, 1H), 7.76 (dd, J = 7.4, 1.5 Hz, 1H), 7.71 (dd, J = 7.4, 7.4 Hz, 1H), 7.66 (bs, 1H), 7.36 ꢀ
7.30 (m, 4H), 7.30 ꢀ 7.20 (m, 1H), 4.56 (s, 2H), 4.11 – 4.00 (m, 1H), 3.51 (s, 2H), 2.96 – 2.88 (m,
+
+
2
H), 2.15 – 2.03 (m, 2H), 1.90 ꢀ 1.70 (m, 4H). HRMS (ESI+): calcd. for C H N O [M + H]
21 24 3 2
350.1863; found 350.1874
Compounds 10a-t. By employment of the aboveꢀdescribed procedure and using the suitably
substituted amine 6a-t, compounds 10a-t were obtained.
1
3
-oxo-2-propyl-2,3-dihydro-1H-isoindole-4-carboxamide (10a). H NMR (400 MHz, DMSOꢀ
d )
6
δ
10.76 (bs, 1H), 8.20 (d, J = 7.6 Hz, 1H), 7.76 (d, J = 7.3 Hz, 1H), 7.72 (dd, J = 7.6, 7.3 Hz,
1H), 7.66 (bs, 1H), 4.57 (s, 2H), 3.56 – 3.51 (m, 2H), 1.71 – 1.61 (m, 2H), 0.89 (t, J = 7.3 Hz,
+
+
3
H). HRMS (ESI+): calcd. for C H N O [M + H] 219.1128; found 219.1122.
12 15 2 2
1
2
-(3-methoxypropyl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10b). H NMR (400
10.75 (bs, 1H), 8.21 (d, J = 7.7 Hz, 1H), 7.77 (d, J = 7.3 Hz, 1H), 7.72 (dd, J
7.7, 7.3 Hz, 1H), 7.67 (bs, 1H), 4.58 (s, 2H), 3.66 – 3.60 (m, 2H), 3.40 – 3.36 (m, 2H), 3.24 (s,
MHz, DMSOꢀd6)
δ
=
+
+
3H), 1.92 ꢀ 1.84 (m, 2H). HRMS (ESI+): calcd. for C H N O [M + H] 249.1234; found
1
3
17
2
3
2
49.1232.
-(2-methoxyethyl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10c). H NMR (400
10.68 (bs, 1H), 8.20 (dd, J = 7.5, 1.3 Hz, 1H), 7.77 (dd, J = 7.6, 1.3 Hz, 1H),
.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.67 (bs, 1H), 4.61 (s, 2H), 3.75 (t, J = 5.3 Hz, 2H), 3.60 (t, J =
1
2
MHz, DMSOꢀd6)
δ
7
21
ACS Paragon Plus Environment

Page 23 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
+
5
2
3
.3 Hz, 2H), 3.28 (s, 3H). HRMS (ESI+): calcd. for C H N O [M + H] 235.1077; found
12 15 2 3
35.1083.
-oxo-2-(tetrahydrofuran-2-ylmethyl)-2,3-dihydro-1H-isoindole-4-carboxamide (±-10d). H
10.66 (bs, 1H), 8.20 (dd, J = 7.5, 1.3 Hz, 1H), 7.77 (dd, J = 7.6,
1.3 Hz, 1H), 7.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.67 (bs, 1H), 4.71 – 4.58 (m, 2H), 4.14 – 4.07 (m,
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
NMR (400 MHz, DMSOꢀd6)
δ
1
H), 3.84 – 3.76 (m, 1H), 3.73 – 3.55 (m, 3H), 1.93 – 2.02 (m, 1H), 1.88 ꢀ 1.77 (m, 2H), 1.63 ꢀ
+
+
1
.53 (m, 1H). HRMS (ESI+): calcd. for C H N O [M + H] 261.1234; found 261.1242.
1
4
17
2
3
1
2-(2-hydroxyethyl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10e). H NMR (400
MHz, DMSOꢀd6) 10.77 (bs, 1H), 8.21 (dd, J = 7.4, 1.2 Hz, 1H), 7.77 (dd, J = 7.6, 1.2 Hz, 1H),
7.72 (dd, J = 7.6, 7.4 Hz, 1H), 7.66 (bs, 1H), 4.87 (t, J = 5.2 Hz, 1H), 4.63 (s, 2H), 3.70 – 3.60
δ
+
+
(
m, 4H). HRMS (ESI+): calcd. for C H N O [M + H] 221.0921; found 221.0919.
11 13 2 3
1
2
-benzyl-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10f). H NMR (400 MHz, DMSOꢀ
d )
δ
10.59 (bs, 1H), 8.22 ꢀ 8.17 (m, 1H), 7.75 ꢀ 7.69 (m, 3H), 7.40 ꢀ 7.28 (m, 5H), 4.79 (s, 2H),
6
+
+
4
.47 (s, 2H). HRMS (ESI+): calcd. for C H N O [M + H] 267.1128; found 267.1120.
16 15 2 2
1
3-oxo-2-(2-phenylethyl)-2,3-dihydro-1H-isoindole-4-carboxamide (10g). H NMR (400 MHz,
DMSOꢀd6) 10.68 (bs, 1H), 8.19 (dd, J = 7.3. 1.6 Hz, 1H), 7.74 (dd, J = 7.6, 1.6 Hz, 1H), 7.70
dd, J = 7.6, 7.3 Hz, 1H), 7.66 (bs, 1H), 7.32 ꢀ 7.25 (m, 4H), 7.23 ꢀ 7.18 (m, 1H), 4.49 (s, 2H),
δ
(
+
+
3
2
3
.85 – 3.80 (m, 2H), 3.00 ꢀ 2.94 (m, 2H). HRMS (ESI+): calcd. for C H N O [M + H]
17 17 2 2
81.1285; found 281.1295.
-oxo-2-(3-phenylpropyl)-2,3-dihydro-1H-isoindole-4-carboxamide (10h). H NMR (400
10.74 (bs, 1H), 8.21 (dd, J = 7.4. 1.5 Hz, 1H), 7.78 ꢀ 7.75 (m, 1H), 7.75 ꢀ
.70 (m, 1H), 7.67 (bs, 1H), 7.31 ꢀ 7.24 (m, 4H), 7.20 ꢀ 7.15 (m, 1H), 4.59 (s, 2H), 3.64 – 3.59
1
MHz, DMSOꢀd6)
δ
7
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 88
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
+
(
m, 2H), 2.67 ꢀ 2.60 (m, 2H), 2.01 – 1.92 (m, 2H). HRMS (ESI+): calcd. for C H N O [M +
18 19 2 2
+
H] 295.1441; found 295.1433.
-oxo-2-[2-(pyridin-2-yl)ethyl]-2,3-dihydro-1H-isoindole-4-carboxamide (10i). H NMR (400
MHz, DMSOꢀd6) 10.66 (bs, 1H), 8.48 (ddd, J = 4.9, 1.8, 1.0 Hz, 1H), 8.19 (dd, J = 7.4, 1.5 Hz,
1H), 7.75 (dd, J = 7.6, 1.5 Hz, 1H), 7.73 – 7.68 (m, 2H), 7.65 (bs, 1H), 7.32 (td, J = 7.8, 1.0 Hz,
H), 7.23 (ddd, J = 7.5, 4.9, 1.0 Hz, 1H), 4.52 (s, 2H), 3.98 – 3.93 (m, 2H), 3.15 – 3.10 (m, 2H).
1
3
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
δ
1
+
+
HRMS (ESI+): calcd. for C H N O [M + H] 282.1237; found 282.1243.
1
6
16
3
2
2-[2-(morpholin-4-yl)ethyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10j).
1
H NMR (400 MHz, DMSOꢀd6)
δ 10.73 (bs, 1H), 8.20 (dd, J = 7.5, 1.3 Hz, 1H), 7.78 (dd, J =
7.6, 1.3 Hz, 1H), 7.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.66 (bs, 1H), 4.64 (s, 2H), 3.71 (t, J = 6.2 Hz,
H), 3.57 – 3.52 (m, 4H), 2.59 (t, J = 6.2 Hz, 2H), 2.46 – 2.41 (m, 4H). HRMS (ESI+): calcd. for
2
+
+
C H N O [M + H] 290.1499; found 290.1507.
1
5
20
3
3
1
2-[3-(morpholin-4-yl)propyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10k).
NMR (400 MHz, DMSOꢀd6) 10.76 (bs, 1H), 8.20 (dd, J = 7.5, 1.2 Hz, 1H), 7.76 (dd, J = 7.6,
1.2 Hz, 1H), 7.71 (dd, J = 7.6, 7.5 Hz, 1H), 7.65 (bs, 1H), 4.58 (s, 2H), 3.61 (t, J = 7.1 Hz, 2H),
.53 – 3.47 (m, 4H), 2.37 – 2.29 (m, 6H), 1.85 – 1.76 (m, 2H). HRMS (ESI+): calcd. for
H
δ
3
+
+
C H N O [M + H] 304.1656; found 304.1664.
1
6
22
3
3
3-oxo-2-[2-(piperidin-1-yl)ethyl]-2,3-dihydro-1H-isoindole-4-carboxamide (10l).
1
H NMR (400 MHz, DMSOꢀd6)
δ 10.75 (bs, 1H), 8.20 (dd, J = 7.7, 1.2 Hz, 1H), 7.78 (dd, J =
7
.4. 1.2 Hz, 1H), 7.72 (dd, J = 7.7, 7.4 Hz, 1H), 7.66 (bs, 1H), 4.63 (s, 2H), 3.68 (t, J = 6.3 Hz,
H), 2.54 (t, J = 6.3 Hz, 2H), 2.43 ꢀ 2.36 (m, 4H), 1.50 ꢀ 1.42 (m, 4H), 1.41 – 1.32 (m, 2H).
2
+
+
HRMS (ESI+): calcd. for C H N O [M + H] 288.1707; found 288.1712.
1
6
22
3
2
23
ACS Paragon Plus Environment

Page 25 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
-oxo-2-[3-(piperidin-1-yl)propyl]-2,3-dihydro-1H-isoindole-4-carboxamide hydrochloride
1
(10m). H NMR (400 MHz, DMSOꢀd )
δ
10.58 (bs, 1H), 8.93 (bs, 1H), 8.21 (dd, J = 7.6, 1.1 Hz,
6
1
H), 7.80 (dd, J = 7.4, 1.1 Hz, 1H), 7.74 (dd, J = 7.6, 7.4 Hz, 1H), 7.71 (bs, 1H), 4.59 (s, 2H),
.66 (t, J = 6.6 Hz, 2H), 3.50 – 3.40 (m, 2H), 3.13 ꢀ 3.03 (m, 2H), 2.92 ꢀ 2.78 (m, 2H), 2.09 –
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
1.98 (m, 2H), 1.85 – 1.75 (m, 2H), 1.73 – 1.52 (m, 3H), 1.43 – 1.28 (m, 1H). HRMS (ESI+):
+
+
calcd. for C H N O [M + H] 302.1863; found 302.1865.
1
7
24
3
2
cis-2-[3-(2,6-dimethylpiperidin-1-yl)propyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
1
(
cis-10n). H NMR (400 MHz, DMSOꢀd )
δ
10.58 (bs, 1H), 8.72 (bs, 1H), 8.21 (dd, J = 7.5, 1.2
6
Hz, 1H), 7.80 (dd, J = 7.6, 1.2 Hz, 1H), 7.74 (dd, J = 7.6, 7.5 Hz, 1H), 7.70 (bs, 1H), 4.63 (s,
2H), 3.72 – 3.64 (m, 2H), 3.35 – 3.20 (m partially overlapped by water signal, 4 H), 2.06 – 1.93
(m, 2H), 1.89 – 1.80 (m, 2H), 1.54 – 1.43 (m, 4H), 1.25 (d, J = 6.3 Hz, 6H). HRMS (ESI+):
+
+
calcd. for C H N O [M + H] 330.2176; found 330.2176.
1
9
28
3
2
2-[3-(1,4'-bipiperidin-1'-yl)propyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
1
dihydrochloride (10o). H NMR (400 MHz, DMSOꢀd )
δ 10.58 (bs, 1H), 9.38 (bs, 2H), 8.21
6
(dd, J = 7.6, 1.2 Hz, 1H), 7.80 (d, J = 7.5, 1H), 7.75 (dd, J = 7.6, 7.5 Hz, 1H), 7.72 (bs, 1H), 4.59
+
(
s, 2H), 3.70 – 2.82 (m, 13H), 2.30 – 1.30 (m, 12H). HRMS (ESI+): calcd. for C H N O [M +
2
2
33
4
2
+
H] 385.2598; found 385.2611.
2-[3-(4-benzylpiperidin-1-yl)propyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (10p).
1
H NMR (400 MHz, DMSOꢀd6)
.5, 1.3 Hz, 1H), 7.72 (dd, J = 7.5, 7.4 Hz, 1H), 7.65 (bs, 1H), 7.28 – 7.23 (m, 2H), 7.19 – 7.13
(m, 1H), 7.12 – 7.09 (m, 2H), 4.56 (s, 2H), 3.62 – 3.56 (m, 2H), 2.87 – 2.74 (m, 2H), 2.41 (d, J =
.8 Hz, 2H), 2.33 – 2.22 (m, 2H), 1.82 – 1.70 (m, 4H), 1.52 – 1.37 (m, 3H), 1.11 – 0.99 (m, 2H).
δ 10.78 (bs, 1H), 8.21 (dd, J = 7.4, 1.3 Hz, 1H), 7.76 (dd, J =
7
6
+
+
HRMS (ESI+): calcd. for C H N O [M + H] 392.2333; found 392.2346.
24 30
3
2
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 88
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
-[2-(3,4-dihydroquinolin-1(2H)-yl)ethyl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
1
(10q). H NMR (400 MHz, DMSOꢀd )
δ 10.68 (bs, 1H), 8.20 (dd, J = 7.6, 1.3 Hz, 1H), 7.77 (dd,
6
J = 7.4, 1.3 Hz, 1H), 7.72 (dd, J = 7.6, 7.4 Hz, 1H), 7.69 (bs, 1H), 6.97 – 6.92 (m, 1H), 6.88 –
.85 (m, 1H), 6.75 ꢀ 6.70 (m, 1H), 6.49 – 6.44 (m, 1H), 4.65 (s, 2H), 3.78 – 3.73 (m, 2H), 3.58 –
3.53 (m, 2H), 3.34 ꢀ 3.28 (m overlapped by water signal, 2H), 2.69 – 2.65 (m, 2H), 1.88 ꢀ 1.80
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
6
+
+
(m, 2H). HRMS (ESI+): calcd. for C H N O [M + H] 336.1707; found 336.1692.
20 22 3 2
2
-[2-(3,4-dihydroisoquinolin-2(1H)-yl)ethyl]-3-oxo-2,3-dihydro-1H-isoindole-4-
carboxamide (10r).
1
H NMR (400 MHz, DMSOꢀd6)
δ 10.75 (bs, 1H), 8.19 (dd, J = 7.6, 1.2 Hz, 1H), 7.76 (dd, J =
7.7, 1.2 Hz, 1H), 7.69 (dd, J = 7.7, 7.6 Hz, 1H), 7.66 (bs, 1H), 7.11 – 7.00 (m, 4H), 4.65 (s, 2H),
.81 (t, J = 6.2 Hz, 2H), 3.65 (s, 2H), 2.82 – 2.73 (bs, 6H). HRMS (ESI+): calcd. for
3
+
+
C H N O [M + H] 336.1707; found 336.1722.
2
0
22
3
2
2-{[1-(4-methylbenzyl)piperidin-4-yl]methyl}-3-oxo-2,3-dihydro-1H-isoindole-4-
1
carboxamide (10s). H NMR (400 MHz, DMSOꢀd )
δ 10.72 (bs, 1H), 8.20 (dd, J = 7.4, 1.6 Hz,
6
1
3
–
H), 7.77 – 7.74 (m, 1H), 7.74 – 7.69 (m, 1H), 7.66 (bs, 1H), 7.20 – 7.07 (m, 4H), 4.57 (s, 2H),
.46 (d, J = 7.2 Hz, 2H), 3.39 (s, 2H), 2.82 ꢀ 2.72 (m, 2H), 2.27 (s, 3H), 1.95 – 1.82 (m, 2H), 1.82
+
1.68 (m, 1H), 1.65 ꢀ 1.53 (m, 2H), 1.30 – 1.16 (m, 2H). HRMS (ESI+): calcd. for C H N O
2
2
3
28
3
+
[M + H] 378.2176; found 378.2183.
2
-{[1-(3-methylbenzyl)piperidin-4-yl]methyl}-3-oxo-2,3-dihydro-1H-isoindole-4-
1
carboxamide (10t). H NMR (400 MHz, DMSOꢀd )
δ 10.72 (bs, 1H), 8.20 (dd, J = 7.3, 1.5 Hz,
6
1H), 7.77 – 7.73 (m, 1H), 7.73 – 7.69 (m, 1H), 7.66 (bs, 1H), 7.20 – 7.15 (m, 1H), 7.11 ꢀ 7.00 (m,
H), 4.57 (s, 2H), 3.46 (d, J = 7.3 Hz, 2H), 3.39 (s, 2H), 2.80 ꢀ 2.74 (m, 2H), 2.28 (s, 3H), 1.94 –
3
25
ACS Paragon Plus Environment

Page 27 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
.85 (m, 2H), 1.83 – 1.70 (m, 1H), 1.62 ꢀ 1.54 (m, 2H), 1.29 – 1.18 (m, 2H). HRMS (ESI+):
+
+
calcd. for C H N O [M + H] 378.2176; found 378.2188.
23 28
3
2
3
-oxo-2-(piperidin-4-yl)-2,3-dihydro-1H-isoindole-4-carboxamide hydrochloride (11). A
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
solution of 10u (1.3 g, 3.72 mmol) in acetic acid (100ml) in the presence of Pd/C 10% (260 mg)
was hydrogenated at 50 psi for 8 hours. The mixture was filtered over a pad of celite and the
1
solution was concentrated to afford the title compound (900 mg, 93%). H NMR (400 MHz,
DMSOꢀd ) δ 10.58 (bs, 1H), 8.82 (bs, 1H), 8.59 (bs, 1H), 8.21 (dd, J = 7.5, 1.1 Hz, 1H), 7.82
6
(dd, J = 7.6, 1.1 Hz 1H), 7.75 (dd, J = 7.6, 7.5 Hz, 1H), 7.71 (bs, 1H), 4.56 (s, 2H), 4.32 ꢀ 4.45
(m, 1H), 3.46 ꢀ 3.36 (m overlapped by water signal, 2H), 3.17 – 3.04 (m, 2H), 2.09 – 1.92 (m,
+
+
4
H). HRMS (ESI+): calcd. for C H N O [M + H] 260.1394; found 260.1398.
14 18 3 2
2-[1-(cyclopropylcarbonyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
(12b). To a solution of 11 (60 mg, 0.23 mmol) in pyridine (2 ml) cyclopropanecarbonyl chloride
(26.4 mg, 0.25 mmol) was added. After stirring at 60 °С for 8 hours the solvent was removed
under reduced pressure and the crude was purified by flash chromatography
1
(
dichloromethane/methanol 95:5) to give the desired compound (32 mg, 40%). H NMR (400
MHz, DMSOꢀd ) δ 10.70 (bs, 1H), 8.21 (dd, J = 7.2, 1.6 Hz, 1H), 7.78 ꢀ 7.75 (m, 1H), 7.75 ꢀ
6
7.71 (m, 1H), 7.69 (bs, 1H), 4.57 (s, 2H), 4.60 – 4.38 (m, 2H), 4.40 – 4.30 (m, 1H), 3.40 – 3.19
(m partially overlapped by water signal, 1H), 2.76 – 2.65 (m, 1H), 2.07 – 2.00 (m, 1H), 1.97 –
+
+
1
.54 (m, 4H), 0.83 – 0.67 (m, 4H). HRMS (ESI+): calcd. for C H N O [M + H] 328.1656;
18 22 3 3
found 328.1669.
Compounds 12a-12c. Operating in an analogous way but employing suitably substituted starting
material, the following compounds were obtained:
1
2
-(1-acetylpiperidin-4-yl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (12a). H NMR
(
400 MHz, DMSOꢀd ) δ 10.69 (bs, 1H), 8.21 (dd, J = 7.3, 1.5 Hz, 1H), 7.78 ꢀ 7.75 (m, 1H), 7.75
6
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 88
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
–
7.70 (m, 1H), 7.69 (bs, 1H), 4.55 (s, 2H), 4.58 ꢀ 4.49 (m, 1H), 4.36 ꢀ 4.26 (m, 1H), 3.99 ꢀ 3.91
(m, 1H), 3.15 ꢀ 3.25 (m, 1H), 2.70 ꢀ 2.60 (m, 1H), 2.04 (s, 3H), 1.90 – 1.70 (m., 3H), 1.68 ꢀ 1.55
+
+
(m, 1H). HRMS (ESI+): calcd. for C H N O [M + H] 302.1499; found 302.1506.
16 20 3 3
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
2
-[1-(cyclohexylcarbonyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
1
(12c). H NMR (400 MHz, DMSOꢀd ) δ 10.69 (bs, 1H), 8.21 (dd, J = 7.2, 1.7 Hz, 1H), 7.77 ꢀ
6
7
.70 (m, 2H), 7.69 (bs, 1H), 4.62 – 4.50 (m, 1H), 4.56 (s, 2H), 4.37 ꢀ 4.27 (m, 1H), 4.13 – 4.02
m, 1H), 3.24 – 3.10 (m, 1H), 2.71 ꢀ 2.57 (m, 2H), 1.93 – 1.50 (m, 8H), 1.47 – 1.09 (m, 6H).
(
+
+
HRMS (ESI+): calcd. for C H N O [M + H] 370.2125; found 370.2117.
2
1
28
3
3
2
-[1-(4,4-difluorocyclohexyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
13af). To a suspension of 11 (56 mg, 0.19 mmol) in dichloromethane (2 mL) 4,4ꢀ
difluorocyclohexanone ( 37.5 mg, 0.28 mmol) sodium acetate (32 mg, 0.38 mmol) and methanol
0.3 mL) were added. The resulted solution was stirred at room temperature for 5 hours. Then
(
(
sodium cyanoborohydride was added and the mixture was stirred overnight. Solvents were
removed under reduce pressure and the residue was dissolved in dichloromethane and washed
twice with water. The organic phase was dried over anhydrous sodium sulfate and concentrated
in vacuo and the residue was purified by flash chromatography (dichloromethane/methanol 95:5)
1
to give the title compound (33 mg, 45%). H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1 H), 8.20
6
(
dd, J = 7.4, 1.5 Hz, 1H), 7.77 ꢀ 7.74 (m, 1H), 7.74 ꢀ 7.69 (m, 1H), 7.66 (bs, 1H), 4.55 (s, 2H),
.09 – 3.95 (m., 1H), 3.00 – 2.90 (m, 2H), 2.66 – 2.44 (m overlapped by DMSO signal, 1H), 2.35
ꢀ 2.23 (m, 2H), 2.10 – 1.96 (m, 2H), 1.94 – 1.70 (m, 8H), 1.61 – 1.47 (m, 2H). HRMS (ESI+):
4
+
+
calcd. for C H F N O [M + H] 378.1988; found 378.1982.
2
0
26
2
3
2
Compounds 13a-13ag. Operating in an analogous way but employing suitably substituted
starting material, the following compounds were obtained:
27
ACS Paragon Plus Environment

Page 29 of 88
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
2
-(1-methylpiperidin-4-yl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13a). H NMR
(
400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.21 (dd, J = 7.5, 1.2 Hz, 1H), 7.77 (dd, J = 7.6, 1.2 Hz,
6
1
H), 7.73 (dd, J = 7.6, 7.5 Hz, 1H), 7.67 (bs, 1H), 4.56 (s, 2H), 4.06 – 3.98 (m, 1H), 2.91 ꢀ 2.85
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
(
m, 2H), 2.21 (s, 3H), 2.06 – 1.97 (m, 2H), 1.88 ꢀ 1.78 (m, 2H), 1.78 ꢀ 1.70 (m, 2H). HRMS
+
+
(ESI+): calcd. for C H N O [M + H] 274.1550; found 274.1557.
15 20 3 2
1
2
-(1-ethylpiperidin-4-yl)-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13b). H NMR
(400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.20 (dd, J = 7.5, 1.5 Hz, 1H), 7.77 (dd, J = 7.6, 1.5 Hz,
6
1H), 7.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.66 (bs, 1H), 4.55 (s, 2H), 4.09 – 3.98 (m, 1H), 3.01 ꢀ 2.95
(m, 2H), 2.36 (q, J = 7.1 Hz, 2H), 2.02 ꢀ 1.95 (m, 2H), 1.85 – 1.72 (m, 4H), 1.01 (t, J = 7.1 Hz,
+
+
3
H). HRMS (ESI+): calcd. for C H N O [M + H] 288.1707; found 288.1707.
16 22 3 2
1
3-oxo-2-[1-(propan-2-yl)piperidin-4-yl]-2,3-dihydro-1H-isoindole-4-carboxamide (13c). H
NMR (400 MHz, DMSOꢀd ) δ 10.74 (bs, 1H), 8.21 (dd, J = 7.5, 1.3 Hz, 1H), 7.77 (dd, J = 7.6,
6
1
.3 Hz, 1H), 7.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.67 (bs, 1H), 4.56 (s, 2H), 4.10 – 3.96 (m, 1H), 2.96
– 2.84 (m, 2H), 2.80 – 2.68 (m, 1H), 2.30 – 2.18 (m, 2H), 1.83 – 1.69 (m, 4H), 0.99 (d, J = 6.1
+
+
Hz, 6H). HRMS (ESI+): calcd. for C H N O [M + H] 302.1863; found 302.1862.
1
7
24
3
2
2
-[1-(2-methylpropyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13d).
1
H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.20 (dd, J = 7.4, 1.5 Hz, 1H), 7.76 (d, J = 7.5
6
Hz, 1H), 7.72 (dd, J = 7.5, 7.4 Hz, 1H), 7.66 (bs, 1H), 4.56 (s, 2H), 4.08 – 3.98 (m, 1H), 2.96 –
2
.89 (m, 2H), 2.06 (d, J = 6.2 Hz, 2H), 2.04 – 1.95 (m, 2H), 1.87 – 1.70 (m, 5H), 0.87 (d, J = 6.6
+
+
Hz, 6H). HRMS (ESI+): calcd. for C H N O [M + H] 316.2020; found 316.2020.
1
8
26
3
2
2
-[1-(cyclopropylmethyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide
1
(13e). H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H ), 8.20 (dd, J = 7.5, 1.2 Hz, 1H), 7.77 (dd,
6
J = 7.6, 1.2 Hz, 1H), 7.72 (dd, J = 7.6, 7.5 Hz, 1H), 7.66 (bs, 1H), 4.56 (s, 2H), 4.10 – 3.97 (m,
H), 3.13 ꢀ 3.04 (m, 2H), 2.21 (d, J = 6.3 Hz, 2 H), 2.10 ꢀ 2.00 (m, 2H), 1.90 ꢀ 1.71 (m, 4H), 0.90
1
28
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 30 of 88
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
+
ꢀ
[
0.80 (m, 1H), 0.50 ꢀ 0.44 (m, 2H), 0.12 ꢀ 0.06 (m, 2H). HRMS (ESI+): calcd. for C H N O
2
1
8
24
3
+
M + H] 314.1863; found 314.1860.
2
-[1-(2-methoxyethyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13f).
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
H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.20 (dd, J = 7.6, 1.3 Hz, 1H), 7.76 (dd, J =
6
7
.4, 1.3 Hz, 1H), 7.71 (dd, J = 7.6, 7.4 Hz, 1H), 7.66 (bs, 1H), 4.55 (s, 2H), 4.08 – 3.97 (m, 1H),
3
.44 (t, J = 5.8 Hz, 2H), 3.24 (s, 3H), 3.02 ꢀ 2.95 (m, 2H), 2.53 – 2.47 (m overlapped by DMSO
+
signal, 2H), 2.14 – 2.06 (m, 2), 1.86 – 1.69 (m, 4H). HRMS (ESI+): calcd. for C H N O [M +
1
7
24
3
2
+
H] 318.1812; found 318.1811.
2
-[1-(2-methylbenzyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13g).
1
H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.20 (dd, J = 7.3, 1.6 Hz, 1H), 7.78 ꢀ 7.75 (m,
6
1H), 7.75 ꢀ 7.70 (m, 1H), 7.67 (bs, 1 H), 7.28 – 7.23 (m, 1H), 7.19 – 7.12 (m, 3H), 4.57 (s, 2 H),
4
.14 ꢀ 4.02 (m, 1H), 3.47 (s, 2H), 2.96 ꢀ 2.88 (m, 2H), 2.35 (s, 3 H), 2.19 ꢀ 2.10 (m, 2 H), 1.88 –
+
+
1
.73 (m, 4H). HRMS (ESI+): calcd. for C H N O [M + H] 364.2020; found 364.2034.
2
2
26
3
2
2-[1-(3-methylbenzyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13h).
1
H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.21 (dd, J = 7.3, 1.4 Hz, 1H), 7.77 (dd, J =
6
7
.4, 1.4 Hz, 1H), 7.72 (dd, J = 7.4, 7.3 Hz, 1H), 7.67 (bs, 1H), 7.25 – 7.19 (m, 1H), 7.16 ꢀ 7.05
(m, 3H), 4.58 (s, 2H), 4.13 – 4.00 (m, 1H), 3.47 (s, 2H), 2.98 ꢀ 2.87 (m, 2H), 2.32 (s, 3H), 2.13 ꢀ
+
+
2
.04 (m, 2H), 1.90 ꢀ 1.72 (m, 4H). HRMS (ESI+): calcd. for C H N O [M + H] 364.2020;
22 26 3 2
found 364.2031.
2-[1-(4-methylbenzyl)piperidin-4-yl]-3-oxo-2,3-dihydro-1H-isoindole-4-carboxamide (13i).
1
H NMR (400 MHz, DMSOꢀd ) δ 10.73 (bs, 1H), 8.20 (dd, J = 7.3, 1.5 Hz, 1H), 7.77 (dd, J =
6
7
.4, 1.5 Hz, 1H), 7.72 (dd, J = 7.4, 7.3 Hz, 1H), 7.67 (bs, 1 H), 7.23 ꢀ 7.18 (m, 2H), 7.16 ꢀ 7.12
(m, 2H), 4.57 (s, 2H), 4.11 – 4.00 (m, 1H), 3.46 (s, 2H), 2.96 ꢀ 2.88 (m, 2H), 2.30 (s, 3H), 2.12 ꢀ
29
ACS Paragon Plus Environment