Identification of nicotinamide phosphoribosyltransferase (NAMPT) inhibitors with no evidence of CYP3A4 time-dependent inhibition and improved aqueous solubility

Article abstract of DOI:10.1016/j.bmcl.2014.12.026

Herein we report the optimization efforts to ameliorate the potent CYP3A4 time-dependent inhibition (TDI) and low aqueous solubility exhibited by a previously identified lead compound from our NAMPT inhibitor program (1, GNE-617). Metabolite identification studies pinpointed the imidazopyridine moiety present in 1 as the likely source of the TDI signal, and replacement with other bicyclic systems was found to reduce or eliminate the TDI finding. A strategy of reducing the number of aromatic rings and/or lowering c Log D_7.4 was then employed to significantly improve aqueous solubility. These efforts culminated in the discovery of 42, a compound with no evidence of TDI, improved aqueous solubility, and robust efficacy in tumor xenograft studies.

Full text of DOI:10.1016/j.bmcl.2014.12.026

Bioorganic & Medicinal Chemistry Letters 25 (2015) 529–541
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of nicotinamide phosphoribosyltransferase (NAMPT)
inhibitors with no evidence of CYP3A4 time-dependent inhibition
and improved aqueous solubility
a
a
c
b
Mark Zak a, , Bianca M. Liederer , Deepak Sampath , Po-wai Yuen , Kenneth W. Bair ,
Timm Baumeister b, Alexandre J. Buckmelter b, Karl H. Clodfelter b, Eric Cheng a, Lisa Crocker a,
Bang Fu c, Bingsong Han b, Guangkun Li c, Yen-Ching Ho b, Jian Lin b, Xiongcai Liu c, Justin Ly a,
Thomas O’Brien a, Dominic J. Reynolds b, Nicholas Skelton a, Chase C. Smith b, Suzanne Tay a,
Weiru Wang a, Zhongguo Wang b, Yang Xiao a, Lei Zhang c, Guiling Zhao a, Xiaozhang Zheng b,
Peter S. Dragovich a
⇑
a Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA
b Forma Therapeutics Inc., 500 Arsenal Street, Watertown, MA 02472, USA
c Pharmaron Beijing Co. Ltd, 6 Taihe Road, BDA, Beijing 100176, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein we report the optimization efforts to ameliorate the potent CYP3A4 time-dependent inhibition
(TDI) and low aqueous solubility exhibited by a previously identified lead compound from our NAMPT
inhibitor program (1, GNE-617). Metabolite identification studies pinpointed the imidazopyridine moiety
present in 1 as the likely source of the TDI signal, and replacement with other bicyclic systems was found
to reduce or eliminate the TDI finding. A strategy of reducing the number of aromatic rings and/or
lowering cLogD7.4 was then employed to significantly improve aqueous solubility. These efforts
culminated in the discovery of 42, a compound with no evidence of TDI, improved aqueous solubility,
and robust efficacy in tumor xenograft studies.
Received 28 October 2014
Accepted 9 December 2014
Available online 17 December 2014
Keywords:
Nicotinamide phosphoribosyltransferase
NAMPT
Cytochrome P450 time-dependent
inhibition
Ó 2014 Elsevier Ltd. All rights reserved.
CYP TDI
Aqueous solubility
Tumor metabolism
Nicotinamide adenine dinucleotide (NAD) is an important bio-
molecule implicated in a large number of cellular processes rang-
ing from cell metabolism and survival to calcium homeostasis.1
NAD carries out its biological functions by at least two distinct
modes of action. In one case NAD is an electron-transfer agent,
where it acts as a cofactor of enzymes involved in metabolism. In
the other case NAD is a substrate for enzymes such as the PARPs
and sirtuins, which cleave NAD by transferring its ADP-ribose unit
to acceptor molecules while liberating nicotinamide (NAM).1b The
constant consumption of NAD by enzymes such as the PARPs and
sirtuins, combined with its essential roles in multiple cellular func-
tions, requires cells to regenerate NAD in order to survive.2 Indeed,
multiple pathways for NAD biosynthesis have been characterized,3
including a pathway which recycles NAM back to NAD (Fig. 1).
Nicotinamide phosphoribosyltransferase (NAMPT) catalyzes the
first and rate-determining step of this sequence, namely the con-
densation of NAM and phosphoribosyl pyrophosphate (PRPP) to
produce the corresponding phosphoribose adduct, nicotinamide
mononucleotide (NMN).4 A subsequent reaction then converts
NMN to NAD under the catalysis of a second enzyme (nicotinate/
nicotinamide mononucleotide adenyltransferase, NMNAT). Due to
their high demand for NAD, rapidly proliferating cells such as those
found in cancers may be particularly sensitive to mechanisms that
reduce intracellular NAD concentrations.5 Blockade of the NAM-
NAD recycling pathway via NAMPT inhibition may, therefore, be
an appropriate therapeutic strategy for oncology indications.
Due to the compelling biological rationale, several groups have
targeted small molecule NAMPT inhibitors.6 Indeed, previous
efforts by our group led to the identification of compound 1
(GNE-617, Table 1), a highly potent inhibitor of NAMPT activity
containing a bicyclic imidazopyridine group in the left-hand
portion of the molecule.7 Despite multiple favorable attributes
including excellent in vivo preclinical pharmacokinetics and
⇑
Corresponding author. Tel.: +1 650 467 4533.
E-mail address: mzak@gene.com (M. Zak).
http://dx.doi.org/10.1016/j.bmcl.2014.12.026
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

530
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
To gain insight into the potential origin of the TDI signal, we
O
P O
O
H2N
O
N
O
P O
O
O
P
O
O
performed in vitro metabolite identification studies with com-
pound 1. Multiple metabolites were observed after human hepato-
cyte incubation, with the major products possessing molecular
masses and fragmentation patterns consistent with compounds 3
and 4. A potential mechanism for the formation of 3 and 4 from
1 is also delineated in Figure 2. A trace metabolite with a mass
and fragmentation pattern consistent with 2 was observed, which
could produce 3 via water opening of the epoxide. Ejection of a gly-
oxal moiety could then give rise to product 4. The small abundance
of 2 relative to 3 and 4 may indicate that the epoxide is highly reac-
tive and, once formed, may be rapidly consumed. The potentially
high reactivity of 2 led us to hypothesize that it could be the spe-
cies responsible for the strong TDI observed with compound 1.
With the imidazopyridine portion of 1 suspected as the source
of TDI activity, we conducted TDI screening on several of our pre-
viously reported7 bicyclic replacements (Table 2). Imidazopyrimi-
dine 5, also containing a bridgehead N atom in the bicycle,
likewise exhibited very potent TDI, and led us to focus on bicyclic
groups with all carbon bridgeheads. Azaindazole 6 offered an
improvement, however, weak but measurable TDI was still
observed. Giving further credence to this weak signal was the fact
that multiple other analogs containing the bicyclic azaindazole
system present in 6 also exhibited weak to moderate TDI signals
(data not shown). Satisfyingly, transposing the positions of the 5-
and 6-membered rings of the bicyclic system with respect to the
remainder of the inhibitor led to attenuation of the TDI signal.
Indeed, no TDI activity was detected with azabenzofuran 7, azain-
dole 8, or azabenzothiophene 9.
O
O
O
+
HO
OH
NAM
PRPP
ATP
NAMPT
PO43- + ADP
H2N
O
N
PARPs
Sirtuins
O
P
O
O
O
O
NMN
HO
OH
NMNAT
H2N
O
N
O
O
P
O
N
NH2
O
P O
O
O
O
O
N
N
N
HO
OH
HO
OH
NAD
Figure 1. The NAM-NAD recycling pathway.
As shown in Table 3, the three compounds with no evidence of
TDI (7, 8, 9) all possessed similar biochemical and cell-based
NAMPT potency to 1, although aqueous solubility remained low.
Thus, the focus of further SAR exploration became improving solu-
bility within the azabenzofuran, azaindole, and azabenzothiophene
sub-series while maintaining potency. Towards this goal, our gen-
eral strategy was to reduce the number of aromatic rings present in
the inhibitors and/or to lower cLogD7.4. Others have previously
defined the sum of cLogD7.4 and number of aromatic rings as the
‘solubility forecast index (SFI)’ and have demonstrated that reduc-
ing this parameter correlates positively to improved solubility.11
As will be discussed later in this disclosure, X-ray crystallography
indicated that the extreme right-hand side of the inhibitors pro-
truded into a relatively open and solvent-exposed region of the
outstanding efficacy in tumor xenograft studies, 1 possessed two
undesirable features we wished to address with additional optimi-
zation. Paramount among these was the potent time-dependent
inhibition (TDI) of CYP3A4 caused by 1. In addition to the potential
for drug–drug interactions due to irreversible CYP3A4 inactivation,
we were concerned this TDI signal was indicative of a reactive and
potentially toxic metabolite. A secondary concern associated with
1 was its low aqueous solubility, a potential issue for adequate
and consistent oral exposure with a solid dosage form. Notably,
compound 1 possessed aqueous solubility below the lower limit
of detection in both kinetic and thermodynamic solubility assay
formats.
Table 1
Selected data for previously reported NAMPT inhibitor 1 (GNE-617)
O
F
N
H
N
S
F
N
O
O
1 (GNE-617)
a
b
h
Entry
1
NAMPT IC50
(lM)
A2780 IC50
(l
M)
CYP3A4 TDI (%AUC shift)c
Aqueous solubility (
Kineticf
<1
l
M, pH 7.4)
cLogD7.4
Md
57
Te
Thermog
0.005
0.002
48
<2
3.1
All assay results are reported as the arithmetic mean of at least two separate runs. See Ref. 8 for experimental details.
a
NAMPT biochemical inhibition.
b
Antiproliferation activity determined in cell culture experiments using A2780 cell line. This inhibition can be reversed by addition of 0.33 lM of NMN, strongly
implicating NAMPT inhibition as the causative MOA.
c
Time-dependent inhibition of cytochrome P450 3A4. >15% AUC shift considered possible TDI risk. Experiments carried out as described in Ref. 9.
Substrate = midazolam.
Substrate = testosterone.
d
e
f
10 mM DMSO stock solution diluted to 200
l
M with pH 7.4 aqueous buffer. Final DMSO concentration: 2% (v/v). Solubility measured after shaking for 24 h.
Thermodynamic solubility. Solubility of crystalline powder after shaking for 24 h in aqueous pH 7.4 buffer.
See Ref. 10 for the method used to calculate cLogD7.4
g
h
.

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
531
O
F
Human
Hepatocytes
N
H
1
N
S
F
N
2
O
O
O
water addition
O
F
O
F
hemiaminal breakdown
and tautomerization
N
H
N
H
H2N
N
S
F
N
OH
S
F
O
N
HO
4
3
O
O
O
O
O
Figure 2. Major metabolites (3 and 4) after incubation of 1 with human hepatocytes and proposed mechanism of formation. For metabolite identification, 1 was incubated at
M with human hepatocytes for 3 h. Samples were analyzed using LC–MS/MS.
5
l
an aromatic ring and reducing cLogD7.4 produced the desired effect
of increasing solubility relative to the parent compounds 7, 8, and
9, albeit at the cost of reduced potency. Incorporation of polar
groups into the azaindole scaffold produced compounds with low
or low/moderate MDCK permeability (analogs 14 and 11), poten-
tially due to the presence of an extra hydrogen bond donor relative
to the azabenzofuran and azabenzothiophene series. Additionally,
the azabenzothiophene analogs (9 and 12) exhibited consistently
lower solubility and worse human liver microsome (HLM) meta-
bolic stability than the azaindole and azabenzofuran matched
pairs. Due to a superior balance of favorable solubility, high MDCK
permeability, and reasonable HLM metabolic stability, the majority
of the subsequent work targeting improved potency was carried
out in the azabenzofuran series. Only key representative analogs
were synthesized in the azaindole and azabenzothiophene series.
Several 4-piperidyl replacements (15–19) for the 4-THP moiety
present in 13 were then assessed. Excellent solubility was main-
tained with this set of compounds, however, compounds 15 and
16 showed no potency advantage over 13, while the analogs with
improved potency (17–19) exhibited markedly reduced MDCK
permeability.
Several related sulfoxides were subsequently synthesized, and
multiple noteworthy features were identified. Although calculated
to be more lipophilic and to possess lower polar surface area,
sulfoxides generally possessed lower MDCK permeability than
the corresponding sulfones (compare 15 vs 21, 16 vs 24, and 18
vs 26). When separated by chiral SFC chromatography, one sulfox-
ide enantiomer was always far more potent than the other (com-
pare 21 vs 22, and 24 vs 25), and in the matched pairs available,
the optimal sulfoxide isomer was generally more potent than
the corresponding sulfone (compare 15 vs 21, and 16 vs 24). These
observations on potency will be discussed in the context of an X-
ray structure of sulfoxide 24 in complex with NAMPT (see Fig. 3
and accompanying description).
Although compounds with a reduced aromatic ring count (10–
26) were found to exhibit consistently improved solubility relative
to 7, only analog 24 was considered to appropriately balance other
important parameters such as potency, metabolic stability, and
MDCK permeability. In an attempt to discover additional balanced
inhibitors, focus was shifted back to compounds containing four
aromatic rings. Reduced cLogD7.4 was targeted as the primary sol-
ubility enhancing strategy for such analogs (compounds 27–45).
Reduction of cLogD7.4 by incorporation of an unsubstituted phenyl
group in the right hand portion of the inhibitor (compound 27) led
to a modest solubility increase relative to 7. The sulfonimide
analogs 28 and 29 possessed reasonable levels of solubility, but
the more potent of the two (29) exhibited poor HLM metabolic
stability. Incorporation of heteroaryl groups in the right-hand
Table 2
TDI results for 1 and related analogs
O
F
Ar
N
H
S
F
O
O
Entry
Ar
CYP3A4 TDI (%AUC shift)a
Tc
Mb
57
1
5
48
65
N
N
N
N
N
50
N
6
<15
16
HN
N
O
7
8
9
<15
<15
<15
<15
<15
<15
N
N
N
NH
S
All assay results are reported as the arithmetic mean of at least two separate runs.
a
Time-dependent inhibition of cytochrome P450 3A4. >15% AUC shift considered
possible TDI risk. Experiments carried out as described in Ref. 9.
b
Substrate = midazolam.
Substrate = testosterone.
c
NAMPT protein that would likely accommodate saturated groups.
Thus, the initial work toward identifying more soluble analogs
commenced by replacing the 3,5-difluorophenyl group present in
7, 8, and 9 with non-aromatic moieties.12 Given its high through-
put nature, the kinetic solubility assay was used as the primary
screening assay (Table 3), with only select compounds advanced
to the more labor intensive thermodynamic solubility measure-
ment (Table 5).
The first analogs assessed, sulfonamides and sulfones 10–14,
highlighted several key trends. Reduction of SFI by both removing

532
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
Table 3
SAR of NAMPT inhibitors
Ar
O
Y
B =
C =
Ar
N
H
A =
N
S
O
X
NH
R
N
N
Entry
Ar
A
X
Y
R
NAMPT
A2780
Kinetic solubilityc
(lM)
cLogD7.4d, SFIe
MDCKf A:B, B:A
HLMg Clpred
(mL/min/kg)
(ꢀ10ꢁ6 cm/s)
a
b
IC50
(
l
M)
IC50
(l
M)
F
7
CH
CH
0.014
0.010
0.003
0.009
0.005
0.002
5.9
1.6
<1
3.0, 7.0
14, 15
13, 10
15, 13
11
8
S
S
F
F
F
O
O
O
O
F
F
8
9
B
C
CH
CH
CH
CH
2.2, 6.2
3.6, 7.6
O
14
S
S
S
S
S
S
O
O
O
O
O
O
O
N
O
10
11
12
13
14
A
B
C
A
B
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
0.032
0.088
0.021
0.047
0.088
0.132
0.037
0.017
0.087
0.047
101
42
0.7, 3.7
ꢁ0.1, 2.9
1.4, 4.4
1.3, 4.3
0.5, 3.5
15, 14
1.3, 6.5
13, 13
12, 14
0.9, 2.7
10
10
17
9
O
O
O
O
N
O
N
O
15
111
15
O
O
7
N
N
15
16
17
A
A
A
CH
CH
CH
CH
CH
CH
0.071
0.043
0.019
0.052
0.025
0.014
183
66
0.8, 3.8
1.7, 4.7
1.0, 4.0
8, 11
22, 17
2, 9
9
S
S
O
O
O
O
16
7
O
N
N
N
120
S
S
S
O
O
O
O
O
O
O
O
18
19
A
B
CH
CH
CH
CH
0.019
0.011
0.004
0.007
123
99
0.9, 3.9
0.1, 3.1
2, 12
10
0.4, 0.8
ND

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
533
Table 3 (continued)
Entry
Ar
X
Y
R
NAMPT
A2780
Kinetic solubilityc
M)
cLogD7.4d, SFIe
MDCKf A:B, B:A
HLMg Clpred
(mL/min/kg)
a
b
IC50
(
l
M)
IC50
(l
M)
(
l
(ꢀ10ꢁ6 cm/s)
N
N
20h
A
CH
CH
0.060
0.026
>2
0.020
0.010
>2
153
167
175
1.7, 4.7
ND
1, 5
ND
5
S
O
21i
22i
A
A
CH
CH
CH
CH
1.7, 4.7
1.7, 4.7
1
1
S
O
N
N
N
N
S
O
23h
24j
25k
A
A
A
CH
CH
CH
CH
CH
CH
0.022
0.008
1.6
0.004
0.002
>2
122
131
170
2.5, 5.5
2.5, 5.5
2.5, 5.5
ND
10
8
S
O
10, 10
ND
S
O
9
S
O
O
N
26h
A
CH
CH
0.019
0.006
156
1.7, 4.7
0.4, 5
7
S
O
27
28h
29h
30
31
32
33
34
A
A
A
A
A
A
A
A
CH
CH
CH
CH
CH
CH
N
CH
CH
CH
CH
CH
CH
N
0.015
0.049
0.005
0.011
0.010
0.010
0.170
0.150
0.023
0.084
0.037
0.005
0.004
0.003
0.355
0.139
31
94
80
47
12
73
72
54
2.6, 6.6
2.6, 6.6
3.6, 7.6
1.9, 5.9
1.9, 5.9
2.0, 6.0
0.7, 4.7
1.4, 5.4
19, 14
16, 13
ND
10
4
S
S
S
S
S
S
S
S
O
O
NH
N
O
O
O
O
O
O
O
16
6
N
N
ND
O
N
14, 13
19, 14
15, 16
15, 13
7
O
N
N
8
O
2
O
N
CH
6
O
(continued on next page)

534
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
Table 3 (continued)
Entry
Ar
X
Y
R
NAMPT
A2780
Kinetic solubilityc
M)
cLogD7.4d, SFIe
MDCKf A:B, B:A
HLMg Clpred
(mL/min/kg)
a
b
IC50
(
l
M)
IC50
(l
M)
(
l
(ꢀ10ꢁ6 cm/s)
35
A
N
CH
0.015
0.010
0.011
0.006
2.4
8.5
2.4, 6.4
19, 17
23, 12
10
8
S
S
S
CF3
O
O
O
O
F
F
F
36
37
A
B
N
N
CH
CH
1.8, 5.8
1.0, 5.0
F
O
O
0.039
0.036
0.036
0.030
11
70
8.8, 17
18, 15
3
F
38h
39i
40i
A
A
A
N
N
N
CH
CH
CH
1.1, 5.1
1.1, 5.1
1.1, 5.1
15
S
F
O
F
F
0.010
0.330
0.006
132
114
21, 9
14
S
F
F
O
>2
ND
ND
S
O
N
41
42
43
44
45
A
A
B
C
A
N
N
N
N
N
CH
CH
CH
CH
CH
0.034
0.039
0.084
0.007
0.008
0.027
0.011
0.122
0.020
0.010
59
0.7, 4.7
0.7, 4.7
-0.1, 3.9
1.3, 5.3
0.9, 4.9
14, 10
16, 17
2.3, 5.6
ND
2
S
S
S
S
S
O
O
O
O
O
O
O
O
O
O
N
N
109
46
4
N
N
N
N
6
N
42
11
8
N
113
8, 11
All assay results are reported as the arithmetic mean of at least two separate runs. See Ref. 8 for experimental details. ND = not determined.
a
NAMPT biochemical inhibition.
b
Antiproliferation activity determined in cell culture experiments using A2780 cell line. This inhibition can be reversed by addition of 0.33
lM of NMN, strongly
implicating NAMPT inhibition as the causative MOA.
c
10 mM DMSO stock solution diluted to 200
l
M with pH 7.4 aqueous buffer. Final DMSO concentration: 2% (v/v). Solubility measured after shaking for 24 h.
See Ref. 10 for the method used to calculate cLogD7.4
Solubility forecast index = cLogD7.4 + number of aromatic rings.11
Apparent permeability in MDCK transwell culture. Permeability categories are as follows (units of 10ꢁ6 cm/s): low <1, moderate = 1–10, high >10. A:B, apical-to-
d
e
f
.
basolateral. B:A, basolateral-to-apical.
g
Human hepatic clearance predicted from human liver microsomes (stable, moderate, labile = <6, 6–15, >15 mL/min/kg, respectively).
Racemic.
Single enantiomer. Absolute stereochemistry inferred by potency relationship between compounds 24 and 25.
Single enantiomer. Absolute stereochemistry established by X-ray crystallography in complex with NAMPT.
h
i
j
k
Single enantiomer. Absolute stereochemistry inferred from potency relationship with compound 24.

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
535
extreme of the inhibitors (30–32) improved solubility relative to 7
while improving potency relative to 10 and 13. Incorporation of a
pyrimidine (33) or pyridine (34) into the central ring led to analogs
with reasonable solubility, however, these compounds were not
sufficiently potent to be of further interest. Combination of the
central pyridine core with 3-trifluoromethyl (35), or 3,5-difluoro
(36) groups appended to the right-hand phenyl ring produced
compounds with improved potency, albeit with eroded solubility.
The azaindole 37 offered no advantage to the azabenzofuran 36
in this respect. The sulfoxide variants of sulfone 36 were found
to have notably improved solubility, with the optimal enantiomer
(39) maintaining outstanding levels of potency. Unfortunately,
sulfoxide 39 possessed reduced HLM metabolic stability relative
to sulfone 36, presumably due to facile metabolic oxidation of
the biaryl sulfoxide S. Finally, combining the pyridyl central ring
with heterocycles in the right-hand extreme of the inhibitors pro-
duced compounds 41–45. Notably, compounds 42 and 45 pos-
sessed an excellent balance of features including: high levels of
productive van der Waals contacts with I351 and V242, and an
edge to face pi interaction with H191. The sulfoxide O atom forms
a favorable hydrogen bonding interaction with water molecule B,
while the piperidine moiety extends into a relatively open and sol-
vent-exposed portion of the protein. Finally, the piperidine N atom
forms a hydrogen bond with water molecule C, while the lipophilic
portions of the piperidine ring and appended isobutyl group make
favorable van der Waals contacts with the sidechains of I309, P307,
and P273.
Figure 3b depicts the X-ray crystal structure of a previously
described sulfone analog from a related scaffold complexed with
NAMPT.15 Comparison of Figure 3a and b may offer a potential
explanation as to why one sulfoxide isomer was always far more
potent than the other (compare 21 vs 22, and 24 vs 25), and why
the preferred sulfoxide isomer was generally more potent than
the corresponding sulfone (compare 15 vs 21, and 16 vs 24). The
general binding modes of both the sulfoxide (Fig. 3a) and the sul-
fone (Fig. 3b) are virtually identical, with each inhibitor forming
similar interactions with the protein. For example, both the sulfox-
ide and the sulfone groups form a favorable hydrogen bonding
interaction with a similarly placed water molecule (water B). How-
ever, as depicted by the red arrows in Figure 3b, the additional O
atom of the sulfone moiety may experience unfavorable interac-
tions due to its proximity to two H-bond acceptors: the backbone
carbonyl of Y188, and water molecule D whose H-bond donating
groups are already satisfied by interactions with the protein. While
no crystal structures were generated for either the enantiomer of
sulfoxide 24 (compound 25) or the corresponding sulfone (com-
pound 16), 25 would likely suffer from the repulsive interactions
without the benefit of the favorable H-bond, while 16 would likely
experience both the favorable and disfavorable interactions
depicted in Figure 3. These observations may potentially explain
the differing magnitudes of potency loss of sulfone 16 (slightly less
potent) and the non-optimal sulfoxide 25 (far less potent) relative
to the preferred sulfoxide enantiomer 24.
aqueous solubility (>100 lM), excellent cell-based potency (IC50
ꢂ10 nM), moderate-high MDCK permeability, and moderate-low
predicted HLM clearance.
An X-ray crystal structure of the most potent sulfoxide pre-
sented in this work (24) in complex with NAMPT is depicted in Fig-
ure 3a. The structure of NMN, the phosphoribose adduct produced
by the NAMPT-catalyzed reaction between NAM and PRPP (see
Fig. 1), is also included for reference (yellow lines).13 Comparison
of these structures indicates that the bicyclic azabenzofuran moi-
ety of 24 occupies the NAM/NMN binding pocket, and further that
the pyridine N of 24 is appropriately positioned to react with PRPP
to generate the corresponding phosphoribose adduct. We believe
formation of such adducts is important to achieving potent cell-
based activity with NAMPT inhibitors.14 The azabenzofuran por-
tion of 24 further participates in pi stacking interactions with the
sidechains of residues F193 and Y180, while the appended carbox-
amide moiety forms hydrogen bonds directly with D219 and
through water A to R311 and S275. The benzylic portion of the
inhibitor extends through a hydrophobic tunnel, encountering
As shown in Table 4, several of the optimized compounds iden-
tified in this work were evaluated in both time-dependent and
Figure 3. X-ray crystal structures of sulfoxide 24 (a, ligand in green) and a structurally-related sulfone15 (b, ligand in mauve) in complex with NAMPT. The yellow lines depict
the position of NMN (from PDB code 3DHD13). Notable water molecules (A–D) are depicted as red spheres. Hydrogen bonds are indicated with dashed black lines, and the van
der Waals surface of the protein is shown in gray. NAMPT residues forming polar or lipophilic interactions with inhibitors are shown in gray. The red arrows in 3b highlight
potential disfavorable interactions between the ligand/protein complex. PDB codes (resolution): 3a = 4WQ6 (1.72 Å); 3b = 4KFP (1.84 Å).

536
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
reversible CYP inhibition assays. Optimized compounds generally
maintained the favorable lack of TDI originally observed with
compounds 7, 8, and 9. Notable exceptions included n-propyl pyr-
azoles 32 and 45 which both produced measurable CYP3A4 TDI.
Additionally, all compounds in Table 4 (except 16) were tested in
TDI assays with the following CYP isoforms: 1A2, 2C9, 2D6, 2C19
(data not shown). All compounds tested showed no evidence of
TDI against these isoforms, with the exception of 24, which showed
weak TDI against CYP2C9 (20% AUC shift) and CYP2D6 (23% AUC
shift). Finally, the majority of analogs in Table 4 showed weak or
unmeasurable reversible inhibition of all CYP isoforms except for
CYP2C9. Potent reversible CYP2C9 inhibition was common in both
the present and previously reported series16, although sulfoxides
typically ameliorated this effect to some extent (compare 16 vs
24, and 36 vs 39).
Several of the compounds with no evidence of CYP3A4 TDI
listed in Table 4 were then evaluated in thermodynamic solubility
and additional in vitro DMPK assays (Table 5). For the compounds
studied, similar trends were noted in the kinetic and thermody-
namic solubility assays. Compounds such as 39 and 42 with high
Table 4
Time-dependent and reversible CYP inhibition of selected compounds
CYP3A4 TDI (%AUC shift)a
Tc
CYP inhibition IC50
2D6
(
l
M)
d
Entry
Mb
3A4 (Mb)
3A4 (Tc)
1A2
2C19
2C9
16
18
19
24
31
32
36
39
42
45
ND
<15
<15
<15
<15
22
<15
<15
<15
21
ND
>10
>10
>10
>10
>10
5.7
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
2.6
0.19
0.3
0.2
<15
<15
<15
<15
32
<15
<15
<15
<15
1.9
0.07
0.07
0.2
1.0
0.16
0.07
2.9
>10
>10
>10
>10
All assay results are reported as the arithmetic mean of at least two separate runs. See Ref. 8 for experimental details. ND = not determined.
a
Time-dependent inhibition of cytochrome P450 3A4. >15% AUC shift considered possible TDI risk. Experiments carried out as described in Ref. 9.
Substrate = midazolam.
Substrate = testosterone.
Reversible cytochrome P450 inhibition.
b
c
d
Table 5
Thermodynamic solubility, in vitro hepatocyte stability and plasma protein binding of selected compounds
All assay results are reported as the arithmetic mean of at least two separate runs. See Ref. 8 for experimental details.
ND = not determined.
a Thermodynamic solubility. Solubility of crystalline powder after shaking for 24 h in aqueous pH 7.4 buffer.
b Mean of two individual measurements: <2 and 3.3
lM.
c Numerical values represent predicted hepatic clearance (mL/min/kg). Colors represent stability categories: green = stable
(Clpred <30% of liver blood flow), yellow = moderate (CLpred = 30–70% of liver blood flow), red = labile (CLpred >70% of liver
blood flow).
d Plasma protein binding; fraction unbound.
Table 6
In vivo PK parameters for selected compounds
Cla (mL/min/kg)
225
t1/2 (h)
Vss (L/kg)
AUCb
0.27
(lM h)
Cmax (lM)
Fb (%)
a
a
b
Entry
24
Species
Rat
0.93
17
0.07
31
36
Rat
Dog
40
6.7
1.0
4.7
2.0
1.6
3.3
33
1.0
4.7
61
115
39
42
Rat
105
0.6
3.0
1.2
1.0
62
Rat
Dog
55
9.8
2.3
1.5
3.2
1.0
3.0
28
1.5
8.6
74
141
All compounds were formulated in EtOH/PEG400/H2O (10:60:30) prior to dosing.
a
IV dosing (1 mg/kg).
PO dosing (5 mg/kg).
b

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
537
Figure 4. Plasma concentration–time profiles of compound 42 after intravenous and oral administration in male Sprague Dawley rats and male beagle dogs. Each time point
shows the mean plasma concentration ( standard deviation) from three separate animals.
kinetic solubility (>100
thermodynamic solubility than did compound 36 (kinetic
solubility = 8.5 M, Table 3). Compounds assessed in the plasma
lM, Table 3) exhibited significantly better
Table 7
Antiproliferative effects of compound 42 in human tumor cell linesa,b
l
Entry
HT1080
IC50 M)
PC3
IC50
HCT116
IC50 M)
protein binding assays typically exhibited moderate values, with
free fractions in the range of approximately 10–30%. Comparing
results in Tables 3 and 5, human hepatocytes generally predicted
similar levels of in vitro turnover relative to human liver
microsomes. Additionally, the stability in human hepatocytes
was typically consistent with that observed for rat and dog hepato-
cytes. However, the majority of compounds were less stable in
mouse hepatocyte assessments. Accordingly, rat was chosen as
(
l
(lM)
(l
42
0.035
0.042
0.025
a
CyQuant endpoint: Direct Cell Proliferation Assay (Invitrogen, Ltd) that mea-
sures nucleic acid in live cells; arithmetic mean of three separate runs (n = 3). See
Ref. 8 for experimental details.
b
This inhibition can be reversed by addition of 0.33
implicating NAMPT inhibition as the causative MOA.
lM of NMN, strongly
Figure 5. Plasma concentration–time profile of compound 42 after oral administration (50 mg/kg BID) in female NCR nude mice. Each time point shows the mean plasma
concentration ( standard deviation) from three separate animals. Compound 42 was formulated in EtOH/PEG400/H2O (10:60:30) prior to dosing.
Figure 6. In vivo efficacy and body weight change after treatment with vehicle (60% PEG400/30% H2O/10% EtOH) or compound 42 (50 mg/kg BID, PO) for 5 consecutive days
in the HT-1080 human fibrosarcoma xenograft model. 10 female NCR tumor-bearing mice were used per group. Data reflects the mean tumor volume or% body weight change
( standard error of mean). Rx denotes treatment period of 5 days. ⁄⁄⁄p <0.001 for compound 42 treated versus vehicle group based on Student’s t-test.

538
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
the primary screening species for in vivo PK studies, with dog used
as a second tier screen for the most promising compounds.
were less promising than for 36. It is, nonetheless, notable that
the primary metabolite observed after in vitro incubation of sulfox-
ide 39 with human hepatocytes was the corresponding sulfone
analog 36, raising the possibility of utilizing 39 as a soluble pro-
drug of 36 (metabolite ID data not shown). However, given the
increased challenges associated with development of a prodrug,
we instead focused on characterizing a soluble and metabolically
stable parent compound, namely sulfone 42. Gratifyingly, 42 bal-
anced excellent cell potency, lack of TDI, and high levels of aqueous
solubility, with rat and dog PK parameters similar to 36. The
plasma exposure profiles of 42 after IV and oral dosing in both
the rat and dog are shown in Figure 4, and were considered to be
acceptable for further progression of this compound.
The in vivo PK parameters of selected compounds are shown in
Table 6. Sulfoxide 24 exhibited extremely high clearance in the rat
after IV delivery, with correspondingly low levels of oral exposure
observed. Biaryl sulfone 36 exhibited markedly improved rat PK
parameters, showing lower clearance and much greater oral levels
than 24. Additionally, excellent dog PK was observed with low
clearance and high oral exposure noted. Although biaryl sulfone
35 showed no evidence of TDI and was highly cell potent, its solu-
bility was only modestly improved relative to compound 1. There-
fore, additional compounds with improved solubility were
advanced to in vivo PK. One such compound was the correspond-
ing biaryl sulfoxide 39. Unfortunately, consistent with the reduced
in vitro metabolic stability, the in vivo rat PK parameters of 39
Compound 42 was then advanced to additional studies to facil-
itate tumor xenograft efficacy experiments. In addition to the
O
Y
O
i
H2N
Y
Ar1-5
N
H
+
Ar1-5
OH
X
R1-24
Ar2
6
X
R1-24
46
47a - 47x
5
_ 45
Ar3
O
Ar1
Ar5
S
Ar groups:
Ar4
N
N
NH
N
N
N
N
N
HN
N
8, 11, 14,
19 37 43
Incorporated into
final product(s):
5
7, 10, 13, 15,
16-18, 20-36,
9, 44
,
,
38 42 45
-
,
Benzylic
Amine
Incorporated into
Final Product(s)
X
Y
R
R-Groups
F
i47a
5 9
-
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
N
N
N
N
N
N
N
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
N
CH
CH
CH
CH
CH
CH
CH
R1
ii47b
ii47c
ii47d
ii47e
ii47f
10 - 12
13, 14
15
16
17
R2
R1
R2
R3
R13
S
F
S
S
S
S
S
R3
O
O
O
N
R4
O
O
N
R5
R14
R15
N
O
S
S
O
O
O
O
O
O
O
O
R6
N
ii47g
iii47h
iii47i
iii47j
18 19
N
R7
,
R8
O
O
O
O
20 - 22
23 - 25
26
27
28
29
30
31
32
33
34
35
36, 37
38 - 40
41
42 - 44
45
N
N
N
N
R9
N
R4
R16
S
S
S
R10
R11
R12
R13
R14
R15
R16
R17
R18
R19
R20
R21
R22
R23
R24
O
O
O
ii47k
iii47l
iii47m
ii47n
ii47o
ii47p
iii47q
iii47r
iii47s
iii47t
iii47u
iii47v
iii47w
iii47x
R5
R6
R17
R18
O
O
O
O
S
S
S
O
N
N
N
R7
R8
R19
R20
S
CF3
F
F
F
O
O
O
O
O
O
S
O
R9
S
R21
R22
S
F
O
O
O
N
N
R10
S
S
S
S
O
O
O
O
O
O
O
N
N
i: Prepared as described in Reference 7.
ii: Prepared as described in Reference 12b.
iii: Prepared as described in Schemes 2 - 5.
R11
R12
R23
R24
S
S
O
O
O
N
N
NH
Scheme 1. Synthesis of final products 5–45. Reagents and conditions: (i) EDCI, HOBt, Et3N, DMF, 25 °C, 12 h, 10–84%.

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
539
A2780 cell line used for SAR screening (Table 3), compound 42
exhibited potent antiproliferative activity against several other
human tumor cell lines (Table 7). 42 also exhibited significant
plasma concentrations in the mouse after oral 50 mg/kg BID dosing
(Fig. 5). When the same oral 50 mg/kg dose of 42 was delivered
twice daily (BID) on 5 successive days to mice bearing HT1080
fibrosarcoma tumor xenografts, robust tumor growth inhibition
(83% vs vehicle control) was observed without significant body
weight loss over the course of a 21 day study (Fig. 6). Notably, effi-
cacy of compound 42 was maintained for 16 days even after dosing
ended on day 5.
Finally, the benzylic amines (47a–47x) were prepared as depicted
in Schemes 2–5, or as previously described.7,12b
As depicted in Scheme 2, the benzylic amines used to prepare
final compounds 20–26 were prepared by first reacting N-Boc pro-
tected 4-hydroxypiperidine with 4-bromobenzene-1-thiol under
Mitsunobu conditions. Partial oxidation of the resulting thioether
(49) then produced sulfoxide 50. Subsequent palladium catalyzed
cyanation, followed by acid promoted Boc cleavage, and reductive
aminations with the liberated piperidine produced intermediates
53. Finally, nitrile reduction produced the corresponding benzylic
amines 47h–47j.
The general coupling conditions to access the final products 5–
45 are depicted in Scheme 1. Aromatic acids 4617 were condensed
with benzylic amines 47a–47x under dehydrative conditions to
access the corresponding amides. The enantiopure sulfoxide ana-
logs 21, 22, 24, 25, 39, and 40 were accessed by preparative chiral
SFC separation of the corresponding racemates (20, 23, and 38).
Scheme 3 depicts the preparation of the benzylic amines 47l
and 47m used to prepare compounds 28 and 29. 4-Fluorobenzoni-
trile was condensed with benzenethiol under basic conditions to
afford the corresponding biphenyl sulfide 55, which was further
converted to cyanosulfinimide 56. Oxidation of 56 produced
cyanosulfonimide 57, which was then deprotected to liberate the
Boc
Br
Boc
Br
Boc
i
ii
N
N
N
N
HO
S
S
O
50
48
R1-3
R2
49
iii
N
R1-3
X
vi
H2N
N
R1
R3
=
=
S
S
O
O
1
47h
51
52
O
(R )
X = Boc
X = H
53e, 53g, 53j
=
iv
v
47i (R2)
47j (R3)
X = R1-3
Scheme 2. Synthesis of the amines 47h–47j incorporated into final compounds 20–26. Enantiopure final compounds 21, 22, 24, and 25 were isolated from chiral preparative
SFC separation of the corresponding racemates 20 and 23. Reagents and conditions: (i) 4-bromobenzene-1-thiol, PPh3, DEAD, THF, 25 °C, 16 h, 41%; (ii) H2O2 (30% in water),
H2SO4/i-PrOH (catalytic), CH2Cl2, 50 °C, 2 h, 43%; (iii) Zn(CN)2, Pd2(dba)3, dppf, DMF, 100 °C, 2 h, 50%; (iv) SOCl2/MeOH, 0–25 °C, 30 min, 100%; (v) 4-tetrahydropyran
carbaldehyde, or isobutyraldehyde, or 3-methylbutanal, Ti(i-PrO)4, AcOH, EtOH, 60 °C, 2 h, then NaBH3CN, 60 °C, 1 h, 43–68%; (vi) Raney Ni, H2, MeOH, 25 °C, 30 min, 54–81%.
N
N
N
i
ii
S
F
S
NCN
54
55
56
iii
R1 = H, R2 = CN
R
N
R2
58
47l
59
v
iv
1 = H, R2 = CH2NH2
R1 = Me, R2 = CN
vi
S
S
v
O
NR1
R
1 = Me, R2 = CH2NH2
O
47m
NCN
57
Scheme 3. Synthesis of the amines 47l and 47m incorporated into final compounds 28 and 29. All numbered intermediates except for 54 and 55 are racemic. Reagents and
conditions: (i) Benzenethiol, K2CO3, DMF, 120 °C, 16 h, 57%; (ii) NBS, t-BuOK, NH2CN, MeOH, 25 °C, 16 h, 46%; (iii) m-CPBA, MeOH, 0–25 °C, 16 h, 94%; (iv) TFAA, CH2Cl2, 0–
25 °C, 16 h, then K2CO3, MeOH, 0–25 °C, 2 h, 59%; (v) Raney-Ni, H2, MeOH, 25 °C, 2 h, 85–90%; (vi) MeI, NaH, DMF, 0–25 °C, 16 h, 94%.
R
N
R
N
R
N
ii
v
N
H
N
N
N
Br
S
S
O
O
62
63
64
R = CN
R = CH2NH2
R = CH2NHBoc
iii
iv
65
47q
R = Boc
R = H
60 R = Cl
61 R = CN
vi
i
Scheme 4. Synthesis of the amine 47q incorporated into final compound 33. Reagents and conditions: (i) KCN, DBU, H2O/DMSO, 25 °C, 16 h, 84%; (ii) Benzenethiol, Cs2CO3,
NMP, 100 °C, 2 h, 47%; (iii) Raney-Ni, H2, NH4OH (30% in water), MeOH, 25 °C, 4 h; (iv) Boc2O, Et3N, CH2Cl2, MeOH, 25 °C, 2 h, 36% (2 steps); (v) m-CPBA, CHCl3, 25 °C, 1 h, 41%;
(vi) HCl, CH2Cl2, 25 °C, 2 h, 83%.

540
M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
NC
N
R
N
Br
v
iv
Ar1,2 SH
NH
Ar1,2
Ar1
S
S
N
66
71
72
O
74t R = CN
47u R = CH2NH2
i
vii
Ar3 SO2Na
70
vi
iii
iii
Ar5,6 Br
67
ii
ii
N
NC
N
H2N
Ar1-6
S
O
viii
Ar4 Br
Ar4-6 SO2Li
Ar1-6
S
68
O
73
69
O
O
47r (Ar3) 47v (Ar4)
47s (Ar2) 47w (Ar5)
47t (Ar1) 47x (Ar6)
F
CF3
N
N
N
N
Ar6 =
Ar3 =
Ar5 =
Ar4 =
Ar1 =
Ar2 =
N
F
Scheme 5. Synthesis of the amines 47r–47x incorporated into final compounds 34–45. Enantiopure final compounds 39 and 40 were isolated from chiral preparative SFC
separation of the corresponding racemate 38. (i) Ar5: i-PrI, K2CO3, DMF, 80 °C, 12 h, 88%, Ar6: n-PrBr, KOH, EtOH, 80 °C, 12 h, 87%; (ii) n-BuLi, SO2, THF, ꢁ78 °C, 1 h, 84–91%;
(iii) 5-bromopicolinonitrile, DMF, 110 °C, 4 h, 30–58%; (iv) 5-bromopicolinonitrile, K2CO3, DMF, 120 °C, 12 h, 65–70%, (v) Ar1 only: m-CPBA, CH2Cl2, 25 °C, 2 h, 74%; (vi) m-
CPBA, CHCl3, 0 °C, 1.5 h, 63%; (vii) Raney-Ni, MeOH, H2, 25 °C, 1 h, 66%; (viii) Ar1–Ar2, Raney-Ni, MeOH, H2NNH2ꢃH2O, 0–25 °C, 10 min, 66–72%; Ar3–Ar6, Raney-Ni, H2, NH4OH
(30% in water), MeOH, 25 °C, 0.5–2 h, 45–70%.
free sulfonimide 58. Reduction of the nitrile group within interme-
diate 58 produced amine 47l, whereas methylation of the sulfoni-
mide moiety of 58, followed by nitrile reduction of 59 produced
amine 47m.
As summarized in Scheme 4, the first step in accessing the ben-
zylic amine 47q used to prepare final compound 33 was a cyano
displacement at the 2-position of 5-bromo-2-chloropyrimidine to
afford nitrile 61. SNAr displacement of the aromatic bromide with
benzenethiol gave sulfide 62, then reduction of the nitrile group,
followed by Boc protection of the resulting benzylic amine 63 pro-
duced biaryl sulfide 64, which was then oxidized to afford sulfone
65. Removal of the Boc protecting group gave the benzylic amine
47q.
The benzylic amines 47r–47x incorporated into final products
34–45 were prepared according to Scheme 5. 3-Bromopyrazole
(66) was alkylated with the appropriate electrophile to generate
the corresponding n-propyl or isopropyl analogs 67. Treatment of
intermediates 67 and 68 with n-BuLi promoted halogen-metal
exchange, and the resulting lithiated intermediates were
quenched with sulfur dioxide to produce the corresponding
sulfinates 69. Sulfones 73 were then accessed by either a two-step
condensation/oxidation sequence starting with thiophenols
(71 ? 72 ? 73), or a single step protocol using sulfinates 69 or
70. Reduction of nitriles 73 produced benzylic amines 47r–47t
and 47v–47x. Sulfoxide 74 was produced by partial oxidation of
the corresponding thioether 72, and the pendant nitrile group
was then reduced to afford benzylic amine 47u.
such as potency, MDCK permeability, or metabolic stability was
often sacrificed. During the course of the optimization campaign,
sulfoxides were found to have unique properties. For example, sul-
fone 36 was found to have low solubility, whereas the correspond-
ing sulfoxides improved this parameter (compounds 38 (racemate)
and 39/40 (single enantiomers)). Additionally, one sulfoxide enan-
tiomer was always far more potent than the other, and the favored
sulfoxide enantiomer was often more potent than the correspond-
ing sulfone (compare 15 vs 21 and 16 vs 24). These potency find-
ings were rationalized by comparison of the X-ray crystal
structures of sulfoxide 24 and a closely related sulfone15 in com-
plex with NAMPT (Fig. 3). Additionally, sulfoxides were found to
decrease, although not completely eliminate, the potent reversible
CYP2C9 inhibition noted with sulfone-containing analogs (Table 4).
Unfortunately, sulfoxides were generally found to possess poor
in vivo PK profiles, likely due to facile metabolism to the corre-
sponding sulfones. Ultimately, the compound with the best bal-
ance of attributes was found to be sulfone 42. Although it
contained four aromatic rings, solubility was significantly
enhanced by reducing cLogD7.4 via incorporation of heteroaromat-
ics. In addition to addressing the two key goals of our optimization
program, namely eliminating TDI and improving aqueous solubil-
ity, 42 combined potent in vitro NAMPT inhibition with favorable
PK properties to produce robust efficacy in a mouse tumor xeno-
graft model.
Acknowledgments
In summary, a lead NAMPT inhibitor previously described by
our group (1, GNE-617) was found to have two issues we wished
to address with additional optimization, namely, potent CYP3A4
TDI and poor aqueous solubility. Metabolite identification studies
revealed the imidazopyridine moiety within 1 to be the potential
source of the TDI signal and, indeed, replacement of this subunit
with multiple other bicyclic systems such as an azabenzofuran,
azaindole, and azabenzothiophene ameliorated the TDI liability
(compounds 7–9). Additional optimization was then carried out
with the goal of improving solubility. Our strategy focused on
The authors thank the Genentech DMPK group, especially Leslie
Wang, Emile Plise, and Hoa Le for generating ADME data; Jeffrey
Stults for XRPD and PLM measurements; Rashida Garcia-Dancey
and Shohini Ganguly for generating biochemical and cellular assay
data; and the Chempartner Structural Biology group for crystalliza-
tion and diffraction data collection for the X-ray structure of 24 in
complex with NAMPT.
Supplementary data
reducing the number of aromatic rings and/or lowering cLogD7.4
,
a metric previously defined as SFI, and shown to correlate with
improved aqueous solubility.11 While reducing aromatic ring count
was found to consistently improve solubility, such analogs were
typically less promising in other respects. One or more attributes
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.
12.026.

M. Zak et al. / Bioorg. Med. Chem. Lett. 25 (2015) 529–541
541
Wang, Z.; Dragovich, P. S.; Gunzner-Toste, J.; Liederer, B.; Ly, J.; O’Brien, T.; Oh, A.;
Wang, L.; Wang, W.; Xiao, Y.; Zak, M.; Zhao, G.; Yuen, P.-W.; Bair, K. W. J. Med.
Chem. 2013, 56, 6413; (b) Zheng, X.; Bair, K. W.; Bauer, P.; Baumeister, T.;
Bowman, K. K.; Buckmelter, A. J.; Caligiuri, M.; Clodfelter, K. H.; Feng, Y.; Han, B.;
Ho, Y.-C.; Kley, N.; Li, H.; Liang, X.; Liederer, B. M.; Lin, J.; Ly, J.; O’Brien, T.; Oeh, J.;
Oh, A.; Reynolds, D. J.; Sampath, D.; Sharma, G.; Skelton, N.; Smith, C. C.;
Tremayne, J.; Wang, L.; Wang, W.; Wang, Z.; Wu, H.; Wu, J.; Xiao, Y.; Yang, G.;
Yuen, P.-W.; Zak, M.; Dragovich, P. S. Bioorg. Med. Chem. Lett. 2013, 23, 5488.
8. Giannetti, A. M.; Zheng, X.; Skelton, N. J.; Wang, W.; Bravo, B. J.; Bair, K. W.;
Baumeister, T.; Cheng, E.; Crocker, L.; Feng, Y.; Gunzner-Toste, J.; Ho, Y.-C.; Hua,
R.; Liederer, B. M.; Liu, Y.; Ma, X.; O’Brien, T.; Oeh, J.; Sampath, D.; Shen, Y.;
Wang, C.; Wang, L.; Wu, H.; Xiao, Y.; Yuen, P.-w.; Zak, M.; Zhao, G.; Zhao, Q.;
Dragovich, P. S. J. Med. Chem. 2014, 57, 770.
References and notes
1. (a) Di Stefano, G.; Manerba, M.; Vettraino, M. Curr. Top. Med. Chem. 2013, 13,
2918; (b) Dölle, C.; Skoge, R. H.; Vanlinden, M. R.; Ziegler, M. Curr. Top. Med.
Chem. 2013, 13, 2907.
2. Galli, U.; Travelli, C.; Massarotti, A.; Fakhfouri, G.; Rahimian, R.; Tron, G. C.;
Genazzani, A. A. J. Med. Chem. 2013, 56, 6279.
3. (a) Bieganowski, P.; Brenner, C. Cell 2004, 117, 495; (b) Magni, G.; Orsomando,
G.; Raffelli, N.; Ruggieri, S. Front. Biosci. 2008, 13, 6135; (c) Chiarugi, A.; Dolle,
C.; Felici, R.; Ziegler, M. Nat. Rev. Cancer 2012, 12, 741.
4. (a) He, J.; Tu, C.; Li, M.; Wang, S.; Guan, X.; Lin, J.; Li, Z. Curr. Pharm. Des. 2012,
18, 6123; (b) Burgos, E. S. Curr. Med. Chem. 1947, 2011, 18; (c) Bi, T.-Q.; Che, X.-
M. Cancer Biol. Ther. 2010, 10, 119; (d) Garten, A.; Petzold, S.; Körner, A.; Imai,
S.-I.; Kiess, W. Trends Endocrinol. Metab. 2008, 20, 130.
5. (a) Zaremba, T.; Ketzer, P.; Cole, M.; Coulthard, S.; Plummer, E. R.; Curtin, N. J.
Br. J. Cancer 2009, 101, 256; (b) Khan, J. A.; Forouhar, F.; Tao, X.; Tong, L. Expert
Opin. Ther. Tar. 2007, 11, 695; (c) Koppenol, W. H.; Bounds, P. L.; Dang, C. V. Nat.
Rev. Cancer 2011, 11, 325; (d) Ward, P. S.; Thompson, C. B. Cancer Cell 2012, 21,
297; (e) Cerna, D.; Li, H.; Flaherty, S.; Takebe, N.; Coleman, C. N.; Yoo, S. S. J. Biol.
Chem. 2012, 287, 22408; (f) Vander Heiden, M. G.; Cantley, L. C.; Thompson, C.
B. Science 2009, 324, 1029.
6. A review on NAMPT inhibitors is presented in Ref. 2. Additional examples of
other groups’ work targeting NAMPT include: (a) Biedermann, E.; Hasmann,
M.; Loser, R.; Rattel, B.; Reiter, F.; Schein, B.; Seibel, K.; Vogt, K. WO 9748397,
1997.; (b) Holen, K.; Saltz, L. B.; Hollywood, E.; Burk, K.; Hanauske, A.-R. Invest.
New Drugs 2008, 26, 45; (c) Schou, C.; Ottosen, E. R.; Petersen, H. J.; Bjorkling, F.;
Latini, S.; Hjarnaa, P. V.; Bramm, E.; Binderup, L. Bioorg. Med. Chem. Lett. 1997, 7,
3095; (d) Ravaud, A.; Cerny, T.; Terret, C.; Wanders, J.; Bui, B. N.; Hess, D.; Droz,
J.-P.; Fumoleau, P.; Twelves, C. Eur. J. Cancer 2005, 41, 702; (e) Hovstadium, P.;
Larsson, R.; Jonsson, E.; Skov, T.; Kissmeyer, A.-M.; Krasilnikoff, K.; Bergh, J.;
Karlsson, M. O.; Lonnebo, A.; Ahlgren, J. Clin. Cancer Res. 2002, 8, 2843; (f) Galli,
U.; Ercolano, E.; Carraro, L.; BlasiRoman, C. R.; Sorba, G.; Canonico, P. L.;
Genazzani, A. A.; Tron, G. C.; Billington, R. A. ChemMedChem 2008, 3, 771; (g)
Willardsen, A. J.; Lockman, J. W.; Murphy, B. R.; Judd, W. R.; Kim, I. C.; Kim, S.-
H.; Zigar, D. F.; Yager, K. M.; Fleischer, T. C.; Terry-Lorenzo, R. T.; Boniface, J. J.;
Parker, D. P.; McAlexander, I. A.; Bursavich, M. G.; Dastrup, D. M. WO
2011109441, 2011.; (h) Lockman, J. W.; Murphy, B. R.; Zigar, D. F.; Judd, W.
R.; Slattum, P. M.; Gao, Z.-H.; Ostanin, K.; Green, J.; McKinnon, R.; Terry-
Lorenzo, R. T.; Fleischer, T. C.; Boniface, J. J.; Shenderovich, M.; Willardsen, J. A.
J. Med. Chem. 2010, 53, 8734; (i) Kumar, D. V.; Slattum, P. M.; Yager, K. M.;
Shenderovich, M. D.; Tangallapally, R.; Kim, S. WO 2012177782, 2012.; (j)
Bernhart, C.; Bouaboula, M.; Casellas, P.; Jegham, S.; Arigon, J; Combet, R.;
Hilairet, S.; Fraisse, P. US 20100222319, 2010.; (k) Bernhart, C.; Bouaboula, M.;
Casellas, P.; Duclos, O.; Bernhart, J. M.; Jegham, S.; McCort, G. WO 2010004198,
2010.; (l) Arigon, J.; Bernhart, C.; Bouaboula, M.; Combet, R.; Jegham, S.;
Hilairet, S. WO 2010109115, 2010.; (m) Arigon, J.; Bernhart, C.; Bouaboula, M.;
Combet, R.; Hilairet, S.; Jegham, S. WO 2010109112, 2010.; (n) Arigon, J.;
Bernhart, C.; Bouaboula, M.; Dimalta, A.; Nardi, F.; Jegham, S. WO 2012038904,
2012.; (o) Arigon, J.; Bernhart, C.; Bosch, M.; Bouaboula, M.; Nardi, F.; Jegham,
S.; Combet, R. WO 2012038905, 2012.; (p) Curtin, M. L.; Sorensen, B. K.;
Heyman, H. R.; Clark, R. F.; Woller, K. R.; Shah, O. J.; Michaelides, M.; Tse, C.;
Vasudevan, A.; Mack, H.; Hansen, T. M.; Sweis, R.; Pliushchev, M. A. US
20120122842, 2012.; (q) Curtin, M. L.; Sorensen, B. K.; Heyman, H. R.; Clark, R.
F.; Michaelides, M.; Tse, C. US 20120122924, 2012.; (r) Colombano, G.; Travelli,
C.; Galli, U.; Caldarelli, A.; Chini, M. G.; Canonico, P. L.; Sorba, G.; Bifulco, G.;
Tron, G. C.; Genazzani, A. A. J. Med. Chem. 2010, 53, 616; (s) You, H.; Youn, H.-S.;
Im, I.; Bae, M.-H.; Lee, S.-K.; Ko, H.; Eom, S. H.; Kim, Y.-C. Eur. J. Med. Chem. 2011,
46, 1153; (t) Christensen, M. K.; Erichsen, K. D.; Olesen, U. H.; Tjørnelund, J.;
Fristrup, P.; Thougaard, A.; Nielsen, S. J.; Sehested, M.; Jensen, P. B.; Loza, E.;
Kalvinsh, I.; Garten, A.; Kiess, W.; Björkling, F. J. Med. Chem. 2013, 56, 9071; (u)
Takeuchi, M.; Niimi, T.; Masumoto, M.; Orita, M.; Yokota, H.; Yamamoto, T. Biol.
Pharm. Bull. 2014, 37, 31.
9. Mukadam, S.; Tay, S.; Tran, D.; Wang, L.; Delarosa, E. M.; Khojasteh, S. C.;
Halladay, J. S.; Kenny, J. R. Drug Metab. Lett. 2012, 6, 43.
10. cLogDpH7.4
values
were
calculated
using
the
MoKa
sofware
(www.moldiscovery.com, (Ref. 10a)) and a custom pKa model (Ref. 10b) (a)
Milletti, F.; Storchi, L.; Sforna, G.; Cruciani, G. J. Chem. Inf. Model. 2007, 47, 2172;
(b) Milletti, F.; Storchi, L.; Goracci, L.; Bendels, S.; Wagner, B.; Kansy, M.;
Cruciani, G. Eur. J. Med. Chem. 2010, 45, 4270.
11. Hill, A. P.; Young, R. J. Drug Discovery Today 2010, 15, 648.
12. (a) We have previously described improving solubility with partially saturated
replacements for the left-hand imidazopyridine ring present in compound 1
(see Ref. 12b). However, the most highly optimized compound in that work
exhibited TDI (compound 18 in Ref. 12b). Thus, in the present work we wished
to find analogs with both improved solubility, and no evidence of TDI. (b)
Dragovich, P. S.; Bair, K. W.; Baumeister, T.; Ho, Y.-C.; Liederer, B. M.; Liu, X.;
Liu, Y.; O’Brien, T.; Oeh, J.; Sampath, D.; Skelton, N.; Wang, L.; Wang, W.; Wu,
H.; Xiao, Y.; Yuen, P.-W.; Zak, M.; Zhang, L.; Zheng, X. Bioorg. Med. Chem. Lett.
2013, 23, 4875.
13. The structure of NMN was obtained from PDB code 3DHD. This structure was
reported in: Burgos, E. S.; Ho, M. C.; Almo, S. C.; Schramm, V. L. Proc. Natl. Acad.
Sci. U.S.A. 2009, 106, 13748.
14. (a) Oh, A.; Ho, Y.-C.; Zak, M.; Liu, Y.; Chen, X.; Yuen, P.-W.; Zheng, X.; Liu, Y.;
Dragovich, P. S.; Wang, W. ChemBioChem 2014, 15, 1121; (b) Additional
discussion on the relationship between the ability of NAMPT inhibitors to form
phosphoribosylated adducts, and the cell-based NAMPT potency of such
compounds is presented in Ref. 8 and additional citations therein.
15. PDB code 4KFP. Reported in Ref. 12b The chemical structure of the ligand in
Figure 3b is as follows:
O
O
N
N
H
N
N
S
O
O
16. The lead compounds reported in multiple of our previous disclosures (see Refs.
7a,b,8,12b,16a,b) possessed potent reversible CYP 2C9 inhibition (IC50 <1 M).
l
Ref. 16c describes how this liability was addressed in a related urea series. (a)
Zheng, X.; Bauer, P.; Baumeister, T.; Buckmelter, A. J.; Caligiuri, M.; Clodfelter,
K. H.; Han, B.; Ho, Y.-C.; Kley, N.; Lin, J.; Reynolds, D. J.; Sharma, G.; Smith, C. C.;
Wang, Z.; Dragovich, P. S.; Oh, A.; Wang, W.; Zak, M.; Gunzner-Toste, J.; Zhao,
G.; Yuen, P.-W.; Bair, K. W. J. Med. Chem. 2013, 56, 4921; (b) Zheng, X.;
Baumeister, T.; Buckmelter, A. J.; Caligiuri, M.; Clodfelter, K. H.; Han, B.; Ho, Y.-
C.; Kley, N.; Lin, J.; Reynolds, D. J.; Sharma, G.; Smith, C. C.; Wang, Z.; Dragovich,
P. S.; Oh, A.; Wang, W.; Zak, M.; Wang, Y.; Yuen, P.-W.; Bair, K. W. Bioorg. Med.
Chem. Lett. 2014, 24, 337; (c) Gunzner-Toste, J.; Zhao, G.; Bauer, P.; Baumeister,
T.; Buckmelter, A. J.; Caligiuri, M.; Clodfelter, K. H.; Fu, B.; Han, B.; Ho, Y.-C.;
Kley, N.; Lian, X.; Liederer, B. M.; Lin, J.; Mukadam, S.; O’Brien, T.; Oh, A.;
Reynolds, D. J.; Sharma, G.; Skelton, N.; Smith, C. C.; Sodhi, J.; Wang, W.; Wang,
Z.; Xiao, Y.; Yuen, P.-w.; Zak, M.; Zhang, L.; Zheng, X.; Bair, K. W.; Dragovich, P.
S. Bioorg. Med. Chem. Lett. 2013, 23, 3531.
7. (a) Zheng, X.; Bauer, P.; Baumeister, T.; Buckmelter, A. J.; Caligiuri, M.; Clodfelter,
K. H.; Han, B.; Ho, Y.-C.; Kley, N.; Lin, J.; Reynolds, D. J.; Sharma, G.; Smith, C. C.;
17. Prepared as described in Ref. 7.