SAR and characterization of non-substrate isoindoline urea inhibitors of nicotinamide phosphoribosyltransferase (NAMPT)

Article abstract of DOI:10.1016/j.bmcl.2017.06.018

Herein we disclose SAR studies that led to a series of isoindoline ureas which we recently reported were first-in-class, non-substrate nicotinamide phosphoribosyltransferase (NAMPT) inhibitors. Modification of the isoindoline and/or the terminal functiona

Full text of DOI:10.1016/j.bmcl.2017.06.018

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
SAR and characterization of non-substrate isoindoline urea inhibitors
of nicotinamide phosphoribosyltransferase (NAMPT)
⇑
Michael L. Curtin , H. Robin Heyman, Richard F. Clark, Bryan K. Sorensen, George A. Doherty,
T. Matthew Hansen, Robin R. Frey, Kathy A. Sarris, Ana L. Aguirre, Anurupa Shrestha, Noah Tu, Kevin Woller,
Marina A. Pliushchev, Ramzi F. Sweis, Min Cheng, Julie L. Wilsbacher, Peter J. Kovar, Jun Guo, Dong Cheng,
a
Kenton L. Longenecker, Diana Raich , Alla V. Korepanova, Nirupama B. Soni, Mikkel A. Algire,
Paul L. Richardson, Violeta L. Marin, Ilaria Badagnani, Anil Vasudevan, F. Greg Buchanan, David Maag,
b
Gary G. Chiang , Chris Tse, Michael R. Michaelides
AbbVie Inc, 1 North Waukegan Rd., North Chicago, IL 60064, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein we disclose SAR studies that led to a series of isoindoline ureas which we recently reported were
first-in-class, non-substrate nicotinamide phosphoribosyltransferase (NAMPT) inhibitors. Modification of
the isoindoline and/or the terminal functionality of screening hit 5 provided inhibitors such as 52 and 58
with nanomolar antiproliferative activity and preclinical pharmacokinetics properties which enabled
potent antitumor activity when dosed orally in mouse xenograft models. X-ray crystal structures of
two inhibitors bound in the NAMPT active-site are discussed.
Received 24 April 2017
Revised 1 June 2017
Accepted 4 June 2017
Available online xxxx
Keywords:
Ó 2017 Elsevier Ltd. All rights reserved.
NAMPT inhibitors
Isoindoline ureas
Antitumor activity
Cancer
Nicotinamide adenine dinucleotide (NAD+) functions as a coen-
zyme for many metabolic processes and as a substrate for enzymes
involved in calcium signaling, DNA repair and regulation of gene
have a unique metabolism which includes an increased depen-
3
dence upon NAD+/NADH-driven processes (e.g., glycolysis) and
4
increased NAD+ consumption ; as a result, many cancer cells are
1
expression. In its role as a substrate for enzymes including the
highly dependent on NAMPT-driven NAD+ biosynthesis for prolif-
5
poly(ADP-ribose) polymerases and sirtuins, NAD+ is consumed
when its ADP-ribose unit is transferred to acceptor molecules, gen-
erating nicotinamide (NAM) in the process. While multiple path-
ways for NAD+ biosynthesis have been identified, NAM can be
efficiently recycled to NAD+ via a two-step process involving
nicotinamide phosphoribosyltransferase (NAMPT).2 This enzyme
catalyzes the first and rate-determining step of the sequence,
namely, the condensation of NAM with phosphoribosyl pyrophos-
phate (PRPP) to produce the phosphoribose adduct nicotinamide
mononucleotide (NMN) (Fig. 1). This phosphoribosylated product
is subsequently converted to NAD+ by a second enzyme, nicoti-
namide mononucleotide adenyltransferase (NMNAT). Cancer cells
eration and viability. Thus, it was hypothesized that blockade of
NAM-to-NAD+ recycling through NAMPT inhibition could repre-
sent an effective oncology treatment that would preferentially tar-
6
,7
get cancer cells.
The first generation small-molecule NAMPT inhibitors, FK8668
(1) and GMX1778 (2), were evaluated clinically almost ten years
9
1
0,11
ago
; these inhibitors had potent antitumor activity in mouse
xenograft models but lacked sufficient oral bioavailability and
failed to elicit efficacy in human cancer trials. In the past five years
12
a number of second generation inhibitors have appeared includ-
1
3
ing multiple series from Forma/Genentech [e.g., 3 and 4 (GNE-
617)]. The binding mode of many of these inhibitors has been ver-
1
4,15
ified through X-ray crystal structures
and, prior to the work
disclosed in this manuscript, all known NAMPT inhibitors with
potent cellular activity possessed a heterocyclic ‘cap group’ that
mimics the pyridyl ring of NAM and contains an sp nitrogen that
could potentially be phosphoribosylated by the enzyme. In fact,
phosphoribosylation of several of these inhibitors has been
observed experimentally using mass spectrometry or X-ray crystal
⇑
Corresponding author.
E-mail address: mike.curtin@abbvie.com (M.L. Curtin).
Current address: MPI Research, 54943 North Main Street, Mattawan, MI 49071,
2
a
United States.
b
Current address: eFFECTOR Therapeutics, 11180 Roselle Street, Suite A, San Diego,
CA 92121, United States.
http://dx.doi.org/10.1016/j.bmcl.2017.06.018
0
960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

2
M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Fig. 1. The phosphoribosylation of NAM by NAMPT to provide NMN.
structures.15–19 These observations, along with SAR that often
revealed steep losses in potency when the sp nitrogen of the
(subsite D). Our crystallography studies determined that: the isoin-
doline portion of 5 overlaps the pyridyl ring of bound NAM, engag-
ing in pi-stacking interactions with the same residues (i.e., Phe193
and Tyr18); the urea carbonyl oxygen and N-H of 5 make the same
hydrogen bond interactions with Ser275 and a protein-bound
water, respectively, as the carboxamide of NAM; the para-phenyl
linker of 5 displaces several protein-bound waters and nicely fills
the lipophilic subsite C; and the propylphenyl carboxamide termi-
nus of 5 makes hydrogen bonds and hydrophobic interactions in
subsite D. Comparison of this X-ray crystal structure with those
of the first- and second-generation inhibitors revealed that com-
pound 5 makes similar inhibitor-protein interactions and generally
fills the same active-site space. Furthermore, this binding mode
suggested to us that while certain structural features would likely
be critical to activity (e.g., urea hydrogen bonds, linker length/
regiochemistry), others would probably be amenable to modifica-
tion such as the subsite D functionality and the isoindoline aro-
matic ring.
The syntheses of most of our isoindoline ureas were straightfor-
ward and applicable to high-throughput synthesis (Scheme 1). The
commercially available isocyanate 6 was coupled with isoindoline
to give ester 7 which could be saponified and the resulting acid
coupled with a variety of primary and secondary amines to provide
amides such as 8. The use of isocyanate 9 gave the versatile aryl
bromide 10 which could be coupled using transition metal chem-
istry to a variety of reagents to provide compounds such as pyra-
zole 11. The coupling of 10 with alkenyl boronate 12 followed by
hydrogenolysis gave N-Boc-piperidine 13 which was deprotected
and coupled to a variety of carboxylic acids to give piperidinyl
amides such as 14. A more convergent approach to the piperidinyl
amides is exemplified by the urea coupling of isoindoline with
commercially available aniline 15 to give amide 14 after
subsequent deprotection and acid coupling; this route had greater
2
cap group was removed, led several groups to suggest that the abil-
ity of NAMPT inhibitors to form phosphoribosylated adducts may
15,17
be necessary to obtain potent cellular activity.
We recently disclosed19 that during a screening effort to simul-
taneously identify novel anti-cancer targets and small molecules
with activity against those targets, we observed a small cluster of
isoindoline ureas (e.g., 5, Fig. 2) that had potent antiproliferative
activity against five of the eight cancer cell lines tested. Inhibitor
affinity capture studies using a modified version of 5 as the probe
revealed NAMPT as the major biochemical target; this was con-
firmed by biochemical assays and X-ray crystallography using
human NAMPT. It was also shown that supplementation of cell cul-
ture medium with NMN, the product of NAMPT, fully restored the
viability of tumor cells in the presence of high concentrations of
compound 5 thus indicating that NAMPT inhibition was solely
responsible for the antiproliferative activity observed with this
compound. This experiment also demonstrated that an inhibitor
that cannot be phosphoribosylated by NAMPT could possess potent
activity against a tumor cell line. Herein we disclose the SAR and
physicochemical characterization of this series of first-in-class,
non-substrate, isoindoline urea inhibitors of NAMPT.
The overlaid X-ray crystal structures of isoindoline 5 (PDB ID:
5
UPE) and NAM (PDB ID: 2E5D) bound in the active site of NAMPT
is shown in Fig. 3. As summarized previously, the NAMPT active
site consists of four distinct subsites: the hydrophilic PRPP binding
site which is unoccupied by NAM (subsite A); the NAM binding
space in which NAM makes important pi-stacking and hydrogen
bond interactions (subsite B); the long, narrow and lipophilic tun-
nel which is not occupied by bound NAM (subsite C); and the distal
opening of the active site which provides a large surface for
hydrophobic interactions and several bound water molecules
Fig. 2. NAMPT inhibitors. The sp2 nitrogen that could potentially be phosphoribosylated by NAMPT is highlighted in red.
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Fig. 3. Crystal structure of NAMPT complex with 5 (PDB ID: 5UPE; 5 shown as yellow sticks) with aligned overlay of NAM (PDB ID: 2E5D; NAM shown as magenta sticks). The
NAMPT binding subsites (A–D) are highlighted (see text).
Scheme 1. Reagents: (a) isoindoline, THF, 0 °C then rt, 90–95%; (b) LiOH, THF/MeOH/H
4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-pyrazole, PdCl (dppf)-CH Cl , Na CO
4%; (f) H , 5% wet Pd/C, trifluoroethanol, 67%; (g) 4 M HCl in dioxane, THF, >95%; (h) 1-methylcyclopropanecarboxylic acid, EDC, HOBt, DIPEA, DMF, 95%; (i) isoindoline, bis
2,5-dioxopyrrolidin-1-yl) carbonate, pyridine, DIPEA, acetonitrile, 89%.
2
O, 96%; (c) 3-methylbutan-1-amine, EDC, HOBt, NMM, DMF, 84%; (d) 1-isopentyl-4-
(
2
2
2
2
3
, THF/MeOH/H O, 85 °C, 60%; (e) PdCl (dppf)-CH Cl , Na CO , THF/MeOH/H O, 80 °C,
2
2
2
2
2
3
2
7
2
(
versatility and was more amenable to the multigram preparation
of selected piperidinyl amide inhibitors.
while these compounds had nanomolar binding affinity to NAMPT
and potent, target-based antiproliferative activity (Table 1), these
lipophilic compounds had no measureable bioavailability when
dosed orally in mice and very low aqueous solubility (<5 mM).
Aside from establishing the SAR of this series, ensuing work
focused on reducing the lipophilicity of these compounds to
decrease the potential for toxicity, reduce metabolic liability and
increase the probability of robust in vivo activity.
The SAR work began with an examination of the alkyl carbox-
amide of inhibitors 5 and 8 with selected examples and two refer-
ence compounds shown in Table 1. Simplifying the terminus
demonstrated that while a para-methyl was essentially inactive
New analogs were assessed for human NAMPT binding potency
using a TR-FRET assay with an FK866- or isoindoline urea-based
Oregon green (488) probe and for antiproliferative activity using
the PC-3 (human prostate cancer) cell line. Compounds were also
evaluated in a primary ADME screen that included mouse, rat
and human microsomal clearance20 along with microsomal protein
binding, permeability and aqueous solubility. In addition, selected
compounds were tested for CYP3A4 inhibition, cardiovascular lia-
bilities, rodent pharmacokinetics (PK) and in vivo tumor inhibition
in mice. Initial assessment of screening hits 5 and 8 revealed that
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

4
M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Table 1
Initial SAR of terminal carboxamide modification and replacement.
R1
PC-3 cell viability IC50 (nM)
Mouse CLint,u,scaled (L/h/kg)
20
Human CLint,u,scaled (L/h/kg)
20
cLogP
Cmpd
i
NAMPT binding K (nM)
1
2
5
FK866
GMX1778
3
3
9
5.7
4.4
21
15.5
81.8
NV
5.5
34
1060
3.0
4.1
3.5
a
8
11
38
49
15
3.0
a
a
1
1
6
7
Me
2200
112
>10,000
2300
NV
NV
2.5
2.5
a
a
NV
NV
1
8
9
560
104
8000
880
8.9
4.3
1.2
3.4
0.8
2.1
1
2
2
0
1
14
21
8.7
13.0
120
3.4
3.0
160
1100
130
2
2
2
3
19
14
68
19
21
11
3.5
3.8
3.0
2.1
2
2
4
5
58
19
17
66
47
3.6
3.4
3.0
2.2
6.9
2
2
6
7
13
24
17
68
4.5
5.0
1.8
1.6
1.3
0.8
2
1
8
1
350
23
>10,000
32
265
NVa
26
3.1
4.0
NVa
a
Unbound intrinsic clearance could not be determined due to invalid microsomal protein binding data.
(
(
16), simple carbonyls such as methyl ester (17) and carboxamide
18) were active but significantly less potent, both in binding and
reversing the amide as in compound 24 provided a very potent
subseries of inhibitors. Importantly, we determined that not only
were termini bearing polar functionalities such as hydroxyl (25)
and ether (26, 27) well tolerated, they significantly increased the
polarity of these compounds (as measured by clogP or logD),
greatly improved oral bioavailability (e.g., see compound 58 in
Table 5) and provided some of the best compounds of the
isoindoline urea series (e.g., see Table 4). Lastly, replacement of
proliferation, than the screening hits. The n-propyl amide 19
regained some of the activity but the analogs with the larger alkyl
groups (5, 8, 20 and 22) provided the highest potencies. Interest-
ingly, tertiary amides such as 21 were significantly less potent than
secondary amides (e.g., 20) except if constrained in a ring as in 23,
perhaps suggesting a steric instead of an electronic explanation;
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5
Table 2
SAR of isoindoline substitution.
Cmpd R1
NAMPT binding K
nM)
PC-3 cell viability IC50
(nM)
Cmpd R1
NAMPT binding K
(nM)
PC-3 cell viability IC50
(nM)
i
i
(
8
11
38
33
4.7
>10,000
2
9
7.9
6.5
34
23
>10,000
3
3
0
1
13
10
35
36
16
>10,000
>10,000
4.9
1.4
>9000
3
2
17
640
37
7.5
8.4
carboxamide with sulfonamide (28) greatly diminished activity
while heterocyclic replacements such as pyrazole (11) gave potent
compounds that were highly lipophilic.
tunnel-like subsite C, the use of meta-substituted phenyl (38) elim-
inated binding and cellular activity as did non-linear tethers such
as 2,5-furanyl and 2,5-thienyl or bulky tethers such as 1,4-cyclo-
hexyl. The linking phenyl could be substituted with a fluoride
but nothing larger and, somewhat surprisingly, 2,5- and 2–6-pyri-
dyls were tolerated although the former (39) were generally more
active than the latter (40). In addition to carboxamides, a number
of terminal functionalities were well tolerated including alkyl
ether (41) and phenyl sulfone (42). In contrast to sulfone 42, the
Genentech sulfone 43 was found to be inactive in our PC-3 cell
viability assay; this is consistent with the Genentech antiprolifera-
tion data (IC50 > 2.0 mM, A2780 cell line) and exemplifies our expe-
rience that the nature of the tether linking the isoindoline urea to
the terminal functionality is a powerful determinant of potency.
Because an FK-866/Abbvie hybrid molecule (44) that combined
the alkylpiperidine benzamide with isoindoline urea proved to be
very potent, we envisioned the use of phenyl-linked heterocycles
(e.g., piperazine, piperidine) that would present the terminal ben-
zamide in the same proximity as that in compounds 1 and 44. As
modeling suggested, this idea proved fruitful and while the piper-
azinyl linker (45) was not optimal, the dihydropyridine- (46) and
piperidine-linked (47) benzamides were very potent inhibitors of
NAMPT.
A selected summary of the effect of isoindoline substitution on
activity is shown in Table 2 and demonstrates our experience that
most substitution on the aryl ring were not tolerated, at least based
on cellular cytotoxicity. Small substituents such as 3-fluoro (29)
and 3-methyl (30) provided potent NAMPT binders with significant
and robust improvement in antiproliferative activity while 3-
hydroxymethyl (31) also gave potent analogs but usually led to
inferior pharmacokinetics profiles (data not shown). 2-Chloro-sub-
stitution (32) attenuated cellular activity while 3-chloroisoindoline
was well tolerated depending on the context (see below). Other
substituents of similar size or larger such as 3-cyano (33), 3-triflu-
oromethyl (34) or 3-methoxyl (35) retained active-site binding but
abrogated activity in the proliferation assay; this was also true for
heterocycles at the 3-position including pyridyls and pyrazoles
2
1
(
data not shown). Compound 36 exemplifies the fact the 3,4-disub-
stitution negated binding and cellular activity. Given the discus-
2
sion above, it’s not surprising that addition of an sp nitrogen to
the isoindoline aryl ring as in 37 provided potent binders with
excellent antitproliferative activity. While a thorough comparison
of several of our isoindoline and azaisoindoline ureas has been pro-
1
9
vided elsewhere and will not be discussed further here, it is sig-
nificant that the non-substrate, 3-fluoroisoindoline ureas showed
equipotent cytotoxicity with their potentially processable aza-
isoindoline urea comparators (e.g., 29 versus 37). Regarding the
fluorine cellular ‘boost’ or the lack of cellular activity with many
of the isoindoline substituents, the reasons for the binding/prolif-
eration disconnects with isoindoline substitution remain unclear.
A broad examination of the tether and terminal functionality of
the isoindoline ureas was conducted and a few examples are
shown in Table 3. Consistent with the description of the NAMPT
Building out the SAR of the piperidine carboxamides exempli-
fied by 47 revealed a subseries of highly potent NAMPT inhibitors
with robust binding and antiproliferative actvity. The examples in
Table 4 demonstrate that a wide variety of carboxamides were tol-
erated, ranging from lipophilic (48, 49) to more polar (50–55), and
that sulfonamides (56) were accommodated as well. While isoin-
doline substitution was not always required for optimal activity,
3-fluoro-substitution (51, 53, 56 and 57) was often used to
enhance potency; we also found that 3-chloro-substitution (55)
was beneficial in some instances. Minor changes to the linker could
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

6
M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Table 3
SAR of tether and terminal functionality modification.
Cmpd Structure
NAMPT binding K
nM)
i
PC-3 cell viability IC50
(nM)
Mouse CLint,u,scaled20
(L/h/kg)
Human CLint,u,scaled20
(L/h/kg)
cLogP
3.0
(
3
8
9
>8900
>12,500
NVa
NVa
3
5.4
14
17
3.2
2.9
2.5
4
0
57
730
75
13
4
4
1
2
18
21
48
30
167
22
38
2.1
3.0
4.4
4
4
3
4
487
1.6
>10,000
8.4
220
280
39
19
3.0
3.6
4
5
19
84
NVa
7.8
2.6
4
4
6
7
3.0
1.0
8.1
2.8
NVa
13
11
2.9
3.0
59
a
Unbound intrinsic clearance could not be determined due to invalid microsomal protein binding data.
be made such as the use of azetidine as in compound 57. While no
broad link between calculated polarity (i.e., cLogP or AlogP) and
unbound in vitro intrinsic clearance could be readily observed, sev-
eral of these potent analogs had an optimal balance of polarity and
metabolic stability that provided promising in vivo pharmacoki-
netics and antitumor profiles.
The isoindoline fluorine is oriented in an ‘up’ position, as shown
here, which projects the halide towards Arg196 and the vacant
PRPP binding site. Interestingly, this fluorine is located where the
sp nitrogen of the azaindoles (e.g., 37) usually resides in the active
site (data not shown), although the reasons for this overlap are not
readily understood.
2
The X-ray crystal structure overlay of piperidine carboxamide
3 (PDB ID: 5UPF) and screening hit 5 bound in the NAMPT active
Initial analog screening based on potency, microsomal stability
and permeability provided numerous promising isoindoline ureas
from the various subseries that were assessed for off-target toxic-
ities and potential drug-drug interactions (e.g., hERG blockade,
in vivo rat cardiovascular effects and CYP inhibition). Selected
compounds were characterized in mouse or rat pharmacokinetics
studies and tested for their ability to inhibit tumor growth in a
5
site is shown in Fig. 4. As expected, the isoindoline urea of 53
makes the important pi-stacking and hydrogen bond interactions
as earlier inhibitors (e.g., 5) while the phenylpiperidine effi-
ciently fills the lipophilic tunnel with the tertiary alcohol of the
carboxamide making an interaction with a protein-bound water.
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
7
Table 4
SAR of piperidine carboxamide modification.
Cmpd
Structure
NAMPT binding
(nM)
PC-3 cell viability
IC50 (nM)
Mouse CLint,u,scaled20
(L/h/kg)
Human CLint,u,scaled20
(L/h/kg)
cLogP/AlogP
2.0/3.6
K
i
4
8
15
18
22
3.5
4
9
4.3
5.1
6.6
17
3.4
2.1/3.6
2.7/2.6
5
5
0
1
15
8.7
3.0
2.5
9.4
5.6
16
1.4/3.3
5
5
2
3
5.6
2.5
24
18
3.5
5.7
1.8/2.6
0.0/2.8
3.2
8.6
5
5
4
5
1.5
1.7
2.3
7.9
88
35
44
13
-0.2/3.2
1.1/3.4
5
6
3.9
6.5
30
3.5
2.5/2.9
(
continued on next page)
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

8
M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Table 4 (continued)
Cmpd
Structure
NAMPT binding
(nM)
PC-3 cell viability
IC50 (nM)
Mouse CLint,u,scaled20
(L/h/kg)
Human CLint,u,scaled20
(L/h/kg)
cLogP/AlogP
2.2/3.5
K
i
5
7
1.9
3.9
26
2.7
Fig. 4. Crystal structure of NAMPT complex with 5 (PDB ID: 5UPE; 5 shown as yellow sticks) with aligned overlay of 53 (PDB ID: 5UPF; 53 shown as blue sticks).
Table 5
Selected data for optimized analogs 52 and 58.
Cmpd Structure
PC-3 cell viability CYP3A4/2C9 Rat CLint,u,scaled20 Rat PK
%TGI
(15 mg/
kg/day)
Aqueous
solubility
(mM)
IC50 (nM)
inhibition
IC50, mM)
(L/h/kg)
CLint,u20
(L/h/kg) (L/h/kg)
t
(
1/2
CL
p
%F
e
(
h)
5
8
2
56
>20/>20
<2.8
3.0a 0.26a
1.7a
59b
78%
17
5
24
16/8.4
<3.4
4.8c 0.11c
1.7c
100d 90%
3.3
a
b
c
5
5
1
1
mg/kg IV dose.
mg/kg PO dose.
mg/kg IV dose.
mg/kg PO dose.
d
e
Tumor growth inhibition in a murine xenograft model using the HCT-116 (human colorectal) cell line; compound dosed daily (3on,4off) Â 3 at given dose (mg/kg/day).
mouse xenograft model using the HCT-116 (human colorectal) cell
line. Two examples, benzamide 58 (the more potent enantiomer of
racemate 27) and piperidinyl amide 52, are shown in Table 5 with
selected data. Both compounds have high oral bioavailability with
low total and unbound plasma clearance in rat and provide potent
antitumor activity when dosed orally in mice. Extensive experi-
mentation revealed that the optimal dosing schedule for the isoin-
doline ureas was three days on/four days off per cycle which
provided maximum efficacy while minimizing toxicity.19 Inhibitors
such as 58 exhibited broad in vivo anti-tumor activity across sev-
eral other human tumor lines including H345 (non-small cell lung),
A2780 (ovarian), ES-2 (ovarian), U87MG (glioblastoma) and
2
2
H460LM (non-small cell lung).
In summary, we have used crystallography-driven structure-
based drug design and high-throughput organic synthesis to effi-
ciently optimize screening hit 5, a novel non-substrate, isoindoline
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018

M.L. Curtin et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
9
urea NAMPT inhibitor. We established the SAR of these inhibitors
and identified multiple analogs with potent anti-proliferative
activity and vastly improved oral bioavailability. A significant
number of these analogs such as isoindolines 52 and 58 provide
robust antitumor activity in a mouse xenograft model after oral
dosing. Further in vivo characterization of the isoindoline urea
NAMPT inhibitors will be reported in the near future.
All authors are employees or former employees of AbbVie. This
study was sponsored by AbbVie. AbbVie contributed to the study
design, research, and interpretation of data, writing, reviewing,
and approving the manuscript. The authors would like to acknowl-
edge Stella Doktor, Patricia Stuart, Amanda Olson, Donald Oster-
ling, and DeAnne Stolarik from the DMPK-BA Department at
AbbVie for the in vitro ADME and in vivo pharmacokinetics mea-
surements. This research used resources of the Advanced Photon
Source, a U.S. Department of Energy (DOE) Office of Science User
Facility operated for the DOE Office of Science by Argonne National
Laboratory under Contract No. DE-AC02-06CH11357. Use of the
IMCA-CAT beamline 17-ID at the Advanced Photon Source was
supported by the companies of the Industrial Macromolecular
Crystallography Association through a contract with Hauptman-
Woodward Medical Research Institute.
6. Imai S. Nicotinamide phosphoribosyltransferase (Nampt): a link between NAD
biology, metabolism, and diseases. Curr Pharm Des. 2009;15:20.
7
.
.
Sampath D, Zabka TS, Misner DL, O’Brien T, Dragovich PS. Inhibition of
nicotinamide phosphoribosyltransferase (NAMPT) as a therapeutic strategy in
cancer. Pharmacol Ther. 2015;151:16.
Wosikowski K, Mattern K, Schemainda I, Hasmann M, Rattel B, Löser R. WK175,
a novel antitumor agent, decreases the intracellular nicotinamide adenine
dinucleotide concentration and induces the apoptotic cascade in human
leukemia cells. Cancer Res. 2002;62:1057.
8
9.
Beauparlant P, Bédard D, Bernier C, et al. Preclinical development of the
nicotinamide phosphoribosyl transferase inhibitor prodrug GMX1777.
Anticancer Drugs. 2009;20:346.
10. Holen K, Saltz LB, Hollywood E, Burk K, Hanauske A-R. Invest New Drugs.
008;26:45.
2
1
1. von Heideman A, Berglund A, Larsson R, Nygren P. Safety and efficacy of NAD
depleting cancer drugs: results of a phase I clinical trial of CHS 828 and
overview of published data. Cancer Chemother Pharmacol. 2010;65:1165.
2. Galli U, Travelli C, Massarotti A, et al. Medicinal chemistry of nicotinamide
phosphoribosyltransferase (NAMPT) inhibitors. J Med Chem. 2013;56:6279.
3. Zak M, Liederer BM, Sampath D, et al. Identification of nicotinamide
phosphoribosyltransferase (NAMPT) inhibitors with no evidence of CYP3A4
time-dependent inhibition and improved aqueous solubility. Bioorg Med Chem
Lett. 2015;25:529. and references cited therein.
1
1
14. Khan JA, Tao X, Tong L. Molecular basis for the inhibition of human NMPRTase,
a novel target for anticancer agents. Nat Struct Mol Biol. 2006;13:582.
15. Zheng X, Bauer P, Baumeister T, et al. Structure-based identification of ureas as
novel nicotinamide phosphoribosyltransferase (Nampt) inhibitors. J Med Chem.
2013;56:4921.
1
1
1
1
2
6. Zheng X, Bauer P, Baumeister T, et al. Structure-based discovery of novel
amide-containing nicotinamide phosphoribosyltransferase (Nampt) inhibitors.
J Med Chem. 2013;56:6413.
7. Oh A, Ho Y-C, Zak M, et al. Structural and biochemical analyses of the catalysis
and potency impact of inhibitor phosphoribosylation by human nicotinamide
phosphoribosyltransferase. ChemBioChem. 2014;15:1121.
8. Watson M, Roulston A, Bélec L, et al. The small molecule GMX1778 is a potent
inhibitor of NAD+ biosynthesis: strategy for enhanced therapy in nicotinic acid
phosphoribosyltransferase 1-deficient tumors. 2009;29:5872.
9. Wilsbacher JL, Cheng M, Cheng D, et al. Discovery and characterization of novel,
orally bioavailable non-substrate and substrate NAMPT inhibitors. Mol Cancer
Ther [in press].
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.06.
0
18.
References
0. Heinle L, Peterkin V, de Morais SM, Jenkins GJ, Badagnani I. A high throughput,
1.
2.
3.
4.
5.
Belenky P, Bogan KL, Brenner C. NAD+ metabolism in health and disease. Trends
Biochem Sci. 2007;32:12.
Burgos ES. NAMPT in regulated NAD biosynthesis and its pivotal role in human
metabolism. Curr Med Chem. 2011;18:1947.
Chiarugi A, Dolle C, Felici R, Ziegler M. The NAD metabolome—a key
determinant of cancer cell biology. Nat Rev Cancer. 2012;12:741.
Luo J, Soliminin NL, Elledge SJ. Principles of cancer therapy: oncogene and non-
oncogene addiction. Cell. 2009;136:823.
3
84-well, semi-automated, hepatocyte intrinsic clearance assay for screening
new molecular entities in drug discovery. Comb Chem High Throughput Screen.
015;18:442.
2
2
1. Dragovich PS, Bair KW, Baumeister T, et al. Identification of 2,3-dihydro-1H-
pyrrolo[3,4-c]pyridine-derived ureas as potent inhibitors of human
nicotinamide phosphoribosyltransferase (NAMPT). Bioorg Med Chem Lett.
2013;23:4875.
22. For a detailed summary of the in vivo antitumor activity of inhibitor 58 in a
Mei SC, Brenner C. NAD as
013;20:1307.
a genotype-specific drug target. Chem Biol.
broad range of tumor types, see: Buchanan, F.G. et al., in preparation.
2
Please cite this article in press as: Curtin M.L., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.06.018