Minimizing CYP2C9 Inhibition of Exposed-Pyridine NAMPT (Nicotinamide Phosphoribosyltransferase) Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.6b00697

NAMPT inhibitors may show potential as therapeutics for oncology. Throughout our NAMPT inhibitor program, we found that exposed pyridines or related heterocyclic systems in the left-hand portion of the inhibitors are necessary pharmacophores for potent cellular NAMPT inhibition. However, when combined with a benzyl group in the center of the inhibitors, such pyridine-like moieties also led to consistent and potent inhibition of CYP2C9. In an attempt to reduce CYP2C9 inhibition, a parallel synthesis approach was used to identify central benzyl group replacements with increased Fsp3. A spirocyclic central motif was thus discovered that was combined with left-hand pyridines (or pyridine-like systems) to provide cellularly potent NAMPT inhibitors with minimal CYP2C9 inhibition. Further optimization of potency and ADME properties led to the discovery of compound 68, a highly potent NAMPT inhibitor with outstanding efficacy in a mouse tumor xenograft model and lacking measurable CYP2C9 inhibition at the concentrations tested.

Full text of DOI:10.1021/acs.jmedchem.6b00697

Subscriber access provided by Northern Illinois University
Article
Minimizing CYP2C9 Inhibition of Exposed-Pyridine NAMPT
Nicotinamide Phosphoribosyltransferase) Inhibitors
(
Mark Zak, Po-Wai Yuen, Xiongcai Liu, Snahel Patel, Deepak Sampath, Jason Oeh, Bianca M. Liederer,
Weiru Wang, Thomas O'Brien, Yang Xiao, Nicholas Skelton, Rongbao Hua, Jasleen Sodhi, Yunli Wang,
Lei Zhang, Guiling Zhao, Xiaozhang Zheng, Yen-Ching Ho, Kenneth W. Bair, and Peter S. Dragovich
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00697 • Publication Date (Web): 19 Aug 2016
Downloaded from http://pubs.acs.org on August 20, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Minimizing CYP2C9 Inhibition of ExposedꢀPyridine
NAMPT (Nicotinamide Phosphoribosyltransferase)
Inhibitors
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
a,*
c
c
a
a
a
Mark Zak , Poꢀwai Yuen , Xiongcai Liu , Snahel Patel , Deepak Sampath , Jason Oeh , Bianca M.
a
a
a
a
a
c
Liederer , Weiru Wang , Thomas O’Brien , Yang Xiao , Nicholas Skelton , Rongbao Hua , Jasleen
a
c
c
a
b
b
Sodhi , Yunli Wang , Lei Zhang , Guiling Zhao , Xiaozhang Zheng , YenꢀChing Ho , Kenneth W.
b
a
Bair , Peter S. Dragovich
a
b
c
*
Genentech Inc., 1 DNA Way, South San Francisco, California 94080, USA
FORMA Therapeutics Inc., 500 Arsenal Street, Watertown, Massachusetts 02472, USA
Pharmaron Beijing Co. Ltd., 6 Taihe Road, BDA, Beijing 100176, PR China
Corresponding Author. Eꢀmail address: mzak@gene.com. Phone: 650ꢀ467ꢀ4533
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 92
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
ABSTRACT:
NAMPT inhibitors may show potential as therapeutics for oncology. Throughout our NAMPT
inhibitor program, we found that exposed pyridines or related heterocyclic systems in the leftꢀhand
portion of the inhibitors are necessary pharmacophores for potent cellular NAMPT inhibition.
However, when combined with a benzyl group in the center of the inhibitors, such pyridineꢀlike
moieties also led to consistent and potent inhibition of CYP2C9. In an attempt to reduce CYP2C9
inhibition a parallel synthesis approach was used to identify central benzyl group replacements with
increased Fsp3. A spirocyclic central motif was thus discovered that was combined with leftꢀhand
pyridines (or pyridineꢀlike systems) to provide cellularly potent NAMPT inhibitors with minimal
CYP2C9 inhibition. Further optimization of potency and ADME properties led to the discovery of
compound 68, a highly potent NAMPT inhibitor with outstanding efficacy in a mouse tumor
xenograft model, and lacking measurable CYP2C9 inhibition at the concentrations tested.
INTRODUCTION:
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
Nicotinamide adenine dinucleotide (NAD) is a ubiquitously expressed biomolecule implicated in
multiple biological processes including cell metabolism, calcium homeostasis, cell survival, and
1
aging. NAD acts as both an electronꢀcarrying cofactor for oxidoreductases, and as an ADPꢀribose
1
donator in multiple enzymatic reactions. In two examples of the latter mode of action, NAD
2
participates in reactions catalyzed by PARPs (polyꢀADP ribosylation) and Sirtuins (deacylation and
3
ADP ribosylation). Such reactions consume NAD, while liberating nicotinamide (NAM) as a byꢀ
product (see Figure 1a). The wide ranging biological functions of NAD, coupled with its
consumption by enzymes such as the PARPs and Sirtuins requires cells to replenish NAD in order to
4
survive. Multiple pathways for NAD biosynthesis have been elucidated, including a mechanism
5
that recycles NAM back to NAD (see Figure 1a).
This pathway relies on nicotinamide
2
ACS Paragon Plus Environment

Page 3 of 92
Journal of Medicinal Chemistry
1
2
3
phosphoribosyltransferase (NAMPT) to catalyze its rateꢀdetermining step, a condensation reaction
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
between NAM and phosphoribosyl pyrophosphate (PRPP) to generate nicotinamide mononucleotide
6
(
NMN). NMN, the phosphoribosylated adduct of NAM, is then converted to NAD in a reaction
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
catalyzed by an additional enzyme, nicotinate/nicotinamide mononucleotide adenyltransferase
(NMNAT). Rapidly proliferating cells such as cancers possess a high metabolic load and a
corresponding high demand for NAD. Reduction of intracellular NAD concentrations via blockade
7
of the NAMꢀNAD recycling may, thus, be an appropriate mechanism to target such cells. As such,
4
,8
there is interest in identifying NAMPT inhibitors as potential therapies for oncology indications.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 92
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
Figure 1a. The NAMꢀNAD recycling pathway. Figure 1b. NMNꢀlike phosphoribose adducts
9
a,9c,20,21
generated by coꢀincubating NAMPT inhibitors 1, 2, 3, or 7 with NAMPT, ATP, and PRPP.
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 shown in Figure 2, multiple NAMPT inhibitors have been previously reported by both our own
9
10
11
group (compounds 1ꢀ6 ), as well as by other organizations (compounds 7 and 8 ). These
compounds share similar molecular features, and as exemplified by compound 1, can be divided into
four general regions. The leftꢀhand side is comprised of a pyridine or pyridineꢀlike heterocycle
containing an exposed and nucleophilic nitrogen atom. The linker group is an amide, urea, or
cyanoguanidine. The central region is a lipophilic group comprised of a benzyl group, a phenyl, or
an aliphatic chain. The rightꢀhand side may encompass a wide array of groups, including aromatic
and nonꢀaromatic cycles, with or without polar heteroatoms. One recurring issue with all these
molecules is persistent inhibition of CYP2C9. Indeed, as exemplified in Table 1, the majority of the
previously reported NAMPT inhibitors exhibited potent CYP2C9 inhibition, with IC50 values
typically less than 1 ꢁM ((S)ꢀwarfarin probe). Also notable is the CYP isoform selectivity exhibited
by certain classes of NAMPT inhibitors. Compounds containing bicyclic leftꢀhand moieties,
exemplified by 1, 3, 5, and 6, were typically potent inhibitors of only CYP2C9, and exhibited little
or no inhibition of the other CYP isoforms tested. Another notable finding was the importance of
the probe used to assess CYP2C9 inhibition. Indeed, as exemplified by compounds 1 and 3, and
1
2
consistent with what has previously been reported by others, much different inhibition profiles
were observed depending on whether (S)ꢀwarfarin or diclofenac was used. As (S)ꢀwarfarin was the
more sensitive of the two probes for this class of inhibitors, it was used to assess CYP2C9 inhibition
for the remainder of this work.
4
ACS Paragon Plus Environment

Page 5 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9,10,11
Figure 2. Structures of previously reported NAMPT inhibitors.
Table 1. NAMPT and CYP2C9 Inhibition Data for Previously Reported NAMPT
9,10,11
inhibitors.
NAMPT A2780
a
b
c
Cmpd. IC50
ꢁM)
IC50
CYP Inhibition IC50 (ꢁM)
(
(ꢁM)
2
C9
2C9
3A4
3A4
2D6
1A2
2C19
(
(S)ꢀWarfarin) (Diclofenac) (Testosterone) (Midazolam) (Dextromethorphan) (Tacrine) (Mephenytoin)
1
2
3
4
5
6
7
8
0.005
0.007
0.006
0.004
0.011
0.039
0.002
0.003
0.002
0.032
0.004
0.006
0.011
0.011
0.001
0.001
0.20
0.42
0.07
0.72
0.08
0.16
0.12
3.6
>10
>10
>10
7.9
>10
>10
9.0
>10
>10
>10
>10
>10
>10
0.19
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
6.5
4.8
0.09
>10
>10
0.22
1.5
0.68
>10
>10
0.32
0.54
>10
>10
>10
0.63
>10
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 92
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
All assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2). See reference 9d for
experimental details.
a
NAMPT biochemical inhibition. The average coefficient of variation (CV) for all compounds reported in this work is
0
.48. Individual CV values are reported in the Supporting Information.
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
b
Antiproliferation activity determined in cell culture experiments using A2780 cell line. This inhibition can be reversed
by addition of 0.33 ꢁM of NMN, strongly implicating NAMPT inhibition as the causative MOA. The average coefficient
of variation (CV) for all compounds reported in this work is 0.23. Individual CV values are reported in the Supporting
Information.
c
Reversible Cytochrome P450 2C9 inhibition. The probe used to assess inhibition is indicated in parentheses.
13
CYP2C9 inhibition is undesirable for a number of reasons.
CYP2C9 comprises a substantial
proportion of human liver microsomal P450 content, second only to CYP3A4 in abundance.
Additionally CYP2C9 is implicated in the metabolism of a significant percentage of drugs
1
4
undergoing phase 1 metabolism, with values of 15 – 20% reported in the literature.
Included
amongst these are several nonꢀsteroidal antiꢀinflammatory drugs, oral antidiabetics, and angiotensin
II receptor blockers. CYP2C9 is also the primary enzyme responsible for metabolism of the oral
anticoagulant (S)ꢀwarfarin, a drug with a low therapeutic index. Drugs metabolized by CYP2C9,
especially those with low therapeutic indexes, will be susceptible to drugꢀdrug interactions and
potential toxicity when coꢀadministered with potent CYP2C9 inhibitors. Finally the CYP2C9 gene
is highly polymorphic with common variants exhibiting poor metabolism phenotypes. Patients
carrying such genetic variants and expressing defective or partially defective forms of CYP2C9 are
especially susceptible to CYP2C9 inhibitors. As such, we wished to discover NAMPT inhibitors
with reduced CYP2C9 inhibition profiles.
6
ACS Paragon Plus Environment

Page 7 of 92
Journal of Medicinal Chemistry
1
2
3
Developing strategies to reduce CYP inhibition remains an emerging area of drug discovery
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
1
5
research. Although several computational approaches have been proposed, it remains challenging
1
6
to develop a general model to predict CYP2C9 inhibition due to several factors. The active sites of
CYPs are generally large and flexible, consistent with their role of accommodating and metabolizing
a wide array of xenobiotics and endogenous substrates. Multiple binding modes and a spectrum of
binding sites are possible, spanning direct interaction with the heme, to water mediated heme
binding, to binding in close proximity to the heme, or binding to allosteric sites distant from the
heme. Additionally, multiple ligands may bind simultaneously, and covalent attachment is also
possible. Despite not yet being routine, examples of structure based drug design influencing
CYP2C9 inhibition are also beginning to emerge in the literature. Indeed, a recent disclosure by
scientists at AstraZeneca and the University of Gothenburg has demonstrated the use of Xꢀray
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
16
crystal structures of modified forms of CYP2C9 and CYP3A4 in complex with inhibitors. While
two successful examples were presented, the authors also highlight the technical challenges
associated with routine collection of CYPꢀinhibitor complexed crystal structures and the fact that
iterative CYP structural support is not realistic at the present time. Physical chemical properties and
shape also play a role in CYP inhibition. Indeed, reducing the number of aromatic rings and
1
7
increasing the fraction of sp3 centers (Fsp3) may improve a number of drug developability
1
8
parameters, including reduced CYP inhibition. Additionally, a recent review summarizes common
strategies used to circumvent inhibition of CYP2C9 and other CYP isoforms from a medicinal
19
chemist’s perspective.
Although no general solutions are identified, successful examples are
presented of modulating properties such as lipophilicity and molecular weight, and blocking or
replacing common CYP inhibitory pharmacophores including exposed pyridines and related
aromatic heterocycles.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 92
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
Conspicuously, all of the compounds depicted in Figure 2 contain an archetypal CYP
pharmacophore in the leftꢀhand portion of the inhibitors, an exposed pyridine or related heterocycle
with an exposed and nucleophilic nitrogen atom. Such heterocycles are typically thought to cause
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
5f
inhibition by binding to the heme center of the CYP enzymes.
Unfortunately, these exposed and
nucleophilic leftꢀhand groups are also a necessary pharmacophore for potent NAMPT cell based
9
a,9c,20
21
inhibition. We
and others, have shown that such compounds mimic NAMPT’s endogenous
substrate, nicotinamide, by undergoing enzymeꢀcatalyzed condensation with PRPP as in step 1 of
Figure 1a. Indeed, as shown in Figure 1b, the NMNꢀlike phosphoribose adducts 1′, 2′, 3′, and 7′
have been generated and characterized by incubating inhibitors 1, 2, 3, and 7 with NAMPT, ATP,
9
a,9c,20,21
and PRPP.
Additionally, it has been demonstrated that the phosphoribose adduct 7′ generated
21
in HeLa cells was retained within the cell, whereas the parent compound 7 was readily washed out.
The formation and subsequent retention of highly impermeable phosphoribose adducts within the
cellular environment appears to be a critical factor associated with the cell based potency of NAMPT
inhibitors. Thus, the pharmacophores for CYP2C9 inhibition and potent cellular NAMPT inhibition
are closely linked.
A previous publication by our group has described a partial solution to the CYP2C9 inhibition issue
9
a,22
in the urea series of NAMPT inhibitors related to compound 2.
Indeed, as shown in Table 2,
compound 9, containing an exposed leftꢀhand pyridine moiety was both a relatively potent cellular
NAMPT inhibitor and an extremely potent inhibitor of CYP2C9. Consistent with the hypothesis that
the pyridine motif was the CYP2C9 pharmacophore, blocking the pyridine N with an ortho methyl
group eliminated measurable CYP inhibition at the concentrations measured (compound 10).
8
ACS Paragon Plus Environment

Page 9 of 92
Journal of Medicinal Chemistry
1
2
3
However, cellular NAMPT inhibition was also reduced dramatically. Multiple other groups were
screened in an attempt to reduce CYP inhibition while maintaining potent cellular NAMPT activity.
The only such modification to produce the desired result was an ortho amino group (compound 11).
The increased steric bulk adjacent to the pyridine center eliminated CYP2C9 inhibition, while the
strongly donating amino group presumably allowed the less accessible pyridine to maintain the
required nucleophilicity for phosphoribose adduct formation and cell activity. Unfortunately the
increased electron richness present in 11 also led to potent time dependent inhibition (TDI) of
CYP3A4, limiting the utility of this group. Interestingly, in the course of this work it was also found
that modification of the rightꢀhand side of the inhibitor, a site distal to the nucleophilic pyridine,
could dramatically reduce CYP2C9 inhibition. Regrettably, such modifications were also found to
introduce a new liability into the inhibitors. As exemplified by compound 12, the highly polar and
often basic groups that were effective in reducing CYP2C9 inhibition also led to extremely low
MDCK permeability and hampered oral bioavailability. One final improvement to CYP2C9
inhibition was previously noted in the optimization of bicyclic azabenzofuran NAMPT inhibitors,
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
9f
again at a site distal to the nucleophilic pyridine center. In this case moving from sulfone 13 to the
more polar sulfoxide 14 led to a minor decrease in CYP2C9 inhibition, but once again, this CYP2C9
improvement was accompanied by an additional negative outcome in the form of reduced metabolic
stability and increased in vivo clearance. To summarize our previous findings, certain modifications
in the immediate proximity of, as well as distal to, the leftꢀhand nucleophilic aromatic center of
NAMPT inhibitors were found to improve CYP2C9 inhibition. This observation implied a
cooperative effect between the leftꢀhand pyridine and the central or rightꢀhand portion of the
molecule in producing the CYP2C9 inhibition. However, all modifications to both the pyridines and
the center / rightꢀhand sides of previously reported molecules led to additional undesired liabilities.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 92
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
Thus, a solution to the CYP2C9 inhibition problem that did not introduce additional liabilities into
the inhibitors was still required.
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
Table 2. NAMPT and CYP2C9 Inhibition Data for Previously Reported NAMPT
9
a,9f,22
Inhibitors.
g
h
CYP3A4 TDId
MDCK
HLM
Clpred
(x10 cm/s) (mL/min/kg)
NAMPT
IC50
(ꢁM)
A2780
IC50
(ꢁM)
CYP2C9
(
%AUC shift)
A:B, B:A
a
a
b
Cmpd. Scaffold
X
H
Y
LogD7.4 Inhibition
ꢀ6
e
f
a,c
t
m
IC50 (ꢁM)
i
i
i
i
9
A
A
A
0.003
0.144
2.7
2.4
2.7
0.07
>10
>10
<15 <15
3.1 , 11
8
i
i
i
10
CH
3
0.059
0.003
>2
11
6
1
1
1
2
NH
H
2
0.070
45 34
3.0, 5.5
0
.07,
.43
i
A
0.006
0.076
<ꢀ0.5
>10
4
8
0
i
1
1
3
4
B
B
NA
NA
0.010
0.010
0.006
0.006
2.6
0.2
1.0
<15 <15
<15 <15
23, 12
21, 9
2.0
14
1
0
ACS Paragon Plus Environment

Page 11 of 92
Journal of Medicinal Chemistry
1
2
3
Unless otherwise indicated, all assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
Unless otherwise indicated, experimental details for each assay are described in reference 9b.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
See footnotes from Table 1 for assay descriptions.
b
Measured LogD at pH 7.4. Experimental details may be found in the Supporting Information of this work.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
(
S)ꢀWarfarin probe used.
d
e
f
Timeꢀdependent inhibition of cytochrome P450 3A4. >15% AUC shift considered possible TDI risk.
Substrate = testosterone.
Substrate = midazolam.
g
ꢀ6
Apparent permeability in MDCK transwell culture. Permeability categories are as follows (units of 10 cm/s): low < 1,
moderate = 1ꢀ10, high >10. A:B, apicalꢀtoꢀbasolateral. B:A, basolateralꢀtoꢀapical.
h
Human hepatic clearance predicted from human liver microsomes (stable, moderate, labile = <6, 6–15, >15 mL/min/kg,
respectively).
i
Single determination (n=1).
RESULTS AND DISCUSSION:
Similar to the urea series (e.g. 2, 9, and related compounds), our initial efforts at reducing CYP2C9
inhibition in the bicyclic series focused on blocking the nucleophilic leftꢀhand pyridine moieties in
9
b,9e
previously reported inhibitors 15, 17, and 19 . As shown in table 3, the novel amino substituted
pyridines 16, 18, and 20 all possessed notably reduced CYP2C9 inhibition compared to the
corresponding desꢀamino molecules, however, all suffered a significant loss in NAMPT potency.
Additionally, analog 16 was further found to display potent TDI against CYP3A4. Further work in
optimizing additional aminoꢀsubstituted pyridines and related heterocycles was, thus, abandoned.
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 92
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
Table 3. NAMPT and CYP2C9 Inhibition Data for Pyridine and Amino Pyridine Matched
Pairs.
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
CYP3A4 TDI
(%AUC shift)
NAMPT
IC50
A2780
IC50
(ꢁM)
CYP2C9
Inhibition
IC50 (ꢁM)
c
a
a
Cmpd. Scaffold
X
a,b
c
c
(
ꢁM)
t
m
15
16
A
A
B
B
C
C
H
NH
H
0.008
0.005
0.06
>10
0.05
>10
0.05
3.9
<15 <15
69 83
d
0.064
0.014
0.074
0.015
0.058
0.015
0.378
0.009
0.828
2
2
2
1
1
1
2
7
8
9
0
<15 <15
NH
H
<15 <15
d
NH
0.098
Unless otherwise indicated, all assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
a
See footnotes from Table 1 for assay descriptions.
b
(
S)ꢀWarfarin probe used.
c
See footnotes from Table 2 for assay descriptions.
Single determination (n=1).
d
Since modifying the inhibitors adjacent to the nucleophilic pyridine center did not appear to be a
viable strategy to reduce CYP2C9 inhibition without introducing additional liabilities, we further
explored alternate modifications in distal portions of the molecule. A number of inhibitors
1
2
ACS Paragon Plus Environment

Page 13 of 92
Journal of Medicinal Chemistry
1
2
3
containing benzylic central motifs, but truncated rightꢀhand portions were investigated (21 – 23,
Table 4). While NAMPT potency was drastically reduced in each case, an important difference in
CYP2C9 inhibition was observed. Indeed, sulfone 21 and sulfonamide 22 exhibited potent
inhibition, whereas the unsubstituted benzyl analog (23) showed no measurable CYP2C9 inhibition
at the concentrations tested. This observation appeared to suggest that the sulfone or sulfonamide
played a key role in CYP2C9 inhibition, thus, a number of related analogs containing either
sulfone/sulfonamide replacements or reverse sulfonamides were evaluated. Unfortunately, despite
surveying a diverse array of rightꢀhand side groups attached to the central benzylic moiety via O, N,
or C linkages (exemplified by 24 – 26) it was not possible to restore desired levels of NAMPT
potency while still maintaining lack of CYP2C9 inhibition. This suggested that the sulfone or
sulfonamide groups were not, themselves, key drivers of the CYP2C9 inhibition, and that any type
of substitution on the benzylic central motif leading to improved NAMPT potency would also likely
lead to CYP2C9 inhibition.
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Having concluded that it was likely not feasible to separate the overlapping NAMPT and CYP2C9
pharmacologies when benzylic groups were employed in the central region of the inhibitors, we
1
7,18
pursued an alternate strategy of increasing Fsp3.
Although increasing global Fsp3 in a molecule
9
e
containing a central benzylic group was not an effective strategy (exemplified by 29) , we were
pleased to see that compounds with simple aliphatic groups in their central regions (27 and 28)
exhibited little or no CYP2C9 inhibition. Interestingly, the reduction in CYP2C9 inhibition was not
associated with reduced lipophilicity. Indeed, compounds 27 and 28 possessed higher LogD7.4
values than the potent CYP2C9 inhibitors 21, 22, and 26. We then investigated whether related
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 92
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
molecules with saturated central regions would allow NAMPT potency to be restored while
maintaining the lack of CYP2C9 inhibition.
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
e
Table 4. NAMPT and CYP2C9 Inhibition Data for Analogs 21 – 29 .
NAMPT
IC50
A2780
IC50
(ꢁM)
CYP2C9
Inhibition
IC50 (ꢁM)
a
a
b
c
Cmpd.
R
LogD7.4
0.8
Fsp3
a,d
(
ꢁM)
2
1
0.402
>2
0.13
0.2
2
2
2
3
0.473
>2
>2
0.6
2.2
0.07
0.07
0.34
>10
e
2
2
4
5
0.021
0.014
0.016
0.027
2.5
2.5
0.16
0.30
0.40
0.36
e
N
2
2
6
7
0.423
>2
>2
1.6
3.1
0.25
0.47
0.45
>10
O
>2
>2
2
8
9
>2
1.7
2.0
0.33
0.43
>10
f
e
2
NA
0.002
0.013
0.08
1
4
ACS Paragon Plus Environment

Page 15 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
Unless otherwise indicated, all assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
See footnotes from Table 1 for assay descriptions.
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
b
See footnotes from Table 2 for assay description.
c
Fraction of sp3 centers. (number of sp3 hybridized carbons ÷ total carbon count)
d
(
S)ꢀWarfarin probe used.
e
f
Single determination (n=1).
Not applicable.
To aid further optimization, Xꢀray crystal structures of compounds 1 – 6 (or closely related analogs)
and compound 8 in complex with NAMPT were examined. These structures showed that the central
portions of the various molecules were not perfectly coincident, and indeed swept out a diverse
volume of space (see Figure 3). This observation suggested the NAMPT protein could be tolerant of
inhibitors with different shapes in this region. Additionally the interactions between the central
region of the inhibitors and the NAMPT protein were primarily hydrophobic in nature (see Figure 4a
for a discussion of the residues in this region). The lack of specific polar interactions with strict
requirements for directionality, coupled with the observation that a variety of shapes could be
tolerated in the central portion of the NAMPT inhibitors led us to focus on a parallel synthesis
approach as opposed to rigorous structure based design. We believed parallel synthesis could allow
us to rapidly explore an array of shapes extending into the central portion of the NAMPT protein.
The primary goal was to find high Fsp3 central moieties to improve potency relative to 27 and 28
while maintaining the favorable lack of CYP2C9 inhibition. Thus, we designed a 119ꢀmembered
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 92
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
library (see Table 5 for general structure and selected examples) heavily focused on maintaining a
2
3
high degree of saturation in the central and rightꢀhand portions of the inhibitors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Overlay of indicated ligands from previously reported Xꢀray structures in complex with
9
b
NAMPT. The proteins are omitted for clarity. Accession codes are as follows: 1: 4KFO, 2a:
9
a
24
9d
9e
9f
25 1
4JR5, 3: 4O13, 4: 4LVG, 5a: 4KFP, 6a: 4WQ6, 8: 2GVJ.
Xꢀray structures of compounds
2
, 5, and 6 in complex with NAMPT were not available at the time of this work. Thus, structures of
related analogs 2a, 5a, and 6a were used as surrogates.
After synthesis, all of the library compounds were initially tested at a single concentration of 1 ꢁM to
assess for NAMPT inhibition. Unfortunately, the majority were found to be virtually inactive, with
less than 50% inhibition at 1 M observed for 117 of the 119 analogs prepared. The two noteworthy
exceptions were spirocycles ( )ꢀ30 and ( )ꢀ31. These compounds showed significant NAMPT
inhibition at the 1 M screening concentration, which further translated to IC values of 830 nM
)ꢀ30) and 74 nM ((±)ꢀ31). Notably, the spirocyclic system was found to be crucial to NAMPT
ꢁ
±
±
ꢁ
5
0
((±
inhibition, as related open chain analogs such as 32 and (±)ꢀ33 were found to be far less potent
NAMPT inhibitors. The final significant finding was the greatly reduced CYP2C9 inhibition of the
1
6
ACS Paragon Plus Environment

Page 17 of 92
Journal of Medicinal Chemistry
1
2
3
compounds in Table 5. Importantly, the compound with the best NAMPT potency (compound (
±)ꢀ
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
3
1) also exhibited negligible CYP2C9 inhibition at the concentrations tested. These results validated
the hypothesis that a high degree of saturation in the central portion of the scaffold was a viable
approach to separate the overlapping NAMPT and CYP2C9 pharmacologies observed with earlier
inhibitors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 5. NAMPT and CYP2C9 Inhibition Data of Selected Compounds From a 119ꢀMembered
Parallel Synthesis Library.
c
%
NAMPT NAMPT
Fsp3
CYP2C9
Inhibition
IC50 (ꢁM)
Inhibition
at 1 ꢁM
IC50b
Cmpd.
R
a
b,d
(ꢁM)
(
±
±
)ꢀ30
)ꢀ31
64
0.830
0.61
>10
(
96
26
17
0.074
>1.7
>2
0.57
0.55
0.55
>10
>10
3.4
32
(±)ꢀ33
All assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
All compounds are racemic.
a
Biochemical NAMPT inhibition at a single inhibitor concentration of 1 ꢁM.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 92
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
b
c
d
See footnotes from Table 1 for assay descriptions.
See footnotes from Table 4 for description.
S)ꢀWarfarin probe used.
(
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 shown in Table 6, the most promising spirocyclic moiety identified from the parallel synthesis
2
6
approach was then appended to several additional leftꢀhand groups mimicking nicotinamide. The
,3ꢀdihydroꢀ1Hꢀpyrrolo[3,4ꢀc]pyridine group present in compound ( )ꢀ40 was found to combine best
with the spirocyclic moiety. Indeed ( )ꢀ40 was not only 10ꢀfold more potent than ( )ꢀ31 in the
biochemical assay, but was also found to be the most cell potent NAMPT inhibitor, with an IC50
value of 2 nM in the A2780 anti proliferation assay. The outstanding cell potency of ( )ꢀ40 did not
2
±
±
±
±
appear to be related to an increase in membrane permeability relative to the other spirocyclic
inhibitors tested. Indeed, all of the compounds in Table 6 tested in the MDCK assay were found to
be highly permeable, including compound (±)ꢀ34 which lacked measurable cell based NAMPT
inhibition. The difference in cell activity is likely related to the positioning of the nucleophilic
pyridine or related heterocycle in the leftꢀhand portion of the inhibitor. We believe the pyridine
present in (±)ꢀ40 is most ideally situated to form the phosphoribose adducts similar to NMN (see
2
0
Figure 1) previously shown to be a necessary prerequisite to NAMPT cell based potency. This
hypothesis is supported by the Xꢀray crystal structure of the enantiomerically pure form of
compound (±)ꢀ40 in complex with NAMPT (see Figure 4 and accompanying discussion).
Table 6. NAMPT and CYP2C9 Inhibition Data for Spirocyclic Analogs.
1
8
ACS Paragon Plus Environment

Page 19 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
c
MDCK
A:B, B:A
NAMPT
IC50
(ꢁM)
A2780
IC50
(ꢁM)
CYP2C9
Inhibition
IC50 (ꢁM)
a
a
Cmpd.
R
ꢀ6
a.b
(x10 cm/s)
O
(
±
±
)ꢀ31
)ꢀ34
0.074
0.101
>10
>10
15, 14
NH
N
(
0.059
>2
>2
17, 16
O
N
(±
(±
(±
)ꢀ35
)ꢀ36
)ꢀ37
HN
N
0.024
0.018
0.070
0.119
>10
>10
17, 18
O
O
S
(±
(±
(±
)ꢀ38
)ꢀ39
)ꢀ40
0.009
0.010
0.007
0.203
0.036
0.002
7.2
>10
>10
N
N
O
16, 15
15, 20
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 92
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
All assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
All compounds are racemic.
a
See footnotes from Table 1 for assay descriptions.
b
(
S)ꢀWarfarin probe used.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
See footnotes from Table 2 for assay descriptions.
As shown in Table 7, the individual enantiomers of (±)ꢀ40 were found to have differing levels of
NAMPT potency, with the (S) isomer (41) being more potent than the (R) (42). Both enantiomers
were found to maintain the favorable lack of CYP2C9 inhibition observed in the racemate, and both
had similar levels of stability in human liver microsomes.
Table 7. Characterization of Individual Enantiomers of (±)ꢀ40.
O
N
N
c
HLM
NAMPT
IC50
(ꢁM)
A2780
IC50
(ꢁM)
CYP2C9
Inhibition
IC50 (ꢁM)
a
a
Clpred
Cmpd.
R
a,b
(mL/min/kg)
41
0.013
0.001
>10
>10
17
16
4
2
0.035
0.034
All assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
2
0
ACS Paragon Plus Environment

Page 21 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
b
c
See footnotes from Table 1 for assay descriptions.
S)ꢀWarfarin probe used.
See footnotes from Table 2 for assay descriptions.
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4a shows the Xꢀray crystal structure of compound 41 in complex with NAMPT, and a
summary of the notable interactions is as follows. The bicyclic leftꢀhand side of the inhibitor
participates in face to face piꢀstacking interactions with Tyr18’ and Phe193. The pyridine N of 41
forms a favorable polar interaction with Arg196 and also engages in an Hꢀbond interaction with a
crystallographic water molecule. The pyrrolidine portion of the bicyclic ring system participates in a
favorable Van der Waals interaction with Arg311, while the pyridine ring makes a similar
interaction with the sidechain methylene of Asp219. The urea carbonyl of the linker forms an Hꢀ
bond interaction with the sidechain OH of Ser275, while the NH portion of the urea linker forms a
similar interaction with a tightly bound crystallographic water stabilized by Hꢀbonds with Asp219,
Val242, and Ser241. These linker interactions are a key pharmacophore for NAMPT inhibitors, and
often involve a bridging water molecule between the ligand and Ser275 or Asp219. In the central
region of 41 additional Van der Waals contacts are formed between the methylene group appended
to the cyclopropane and the sidechain CH of Ser275, while the cyclopropane itself makes further
2
Van der Waals contacts with sidechains of His191, Val242, and Ile351. The piperidine portion of
spirocyclic system makes favorable Van der Waals interactions with sidechains of Ile351 and
Ala379. Finally, the carbamate carbonyl in the rightꢀhand side of the inhibitor forms an Hꢀbond with
a crystallographic water, while the tertꢀbutyl group protrudes into a relatively open and solventꢀ
exposed region of the protein.
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A potential explanation for the potent NAMPT inhibition, yet lack of significant CYP2C9 inhibition
exhibited by compound 41 is as follows. Compounds 1, 8, and 41 were three of the most cellularly
potent NAMPT inhibitors profiled in our assays, each possessing A2780 IC values of 1–2 nM.
5
0
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
The ligands from each compound’s Xꢀray structure in complex with NAMPT are overlaid in Figure
4b. It is notable that the nucleophilic N in the leftꢀhand moiety of each inhibitor is placed in
virtually the same location within the active site of NAMPT (circled in Figure 4b). Thus, the
inhibitors are ideally poised to form the corresponding phosphoribose adducts, leading to the
outstanding cell based NAMPT potency observed in each case. Despite possessing similarly potent
NAMPT cellular activity, compounds 1, 8, and 41 displayed a range of CYP2C9 IC ’s spanning
5
0
potent (1, 0.2 µM) to moderate (8, 3.6 µM) to negligible (41, >10 µM) inhibition. As shown in
Figure 4b, notable differences were observed in the topology of the central and rightꢀhand regions,
which may contribute to each compound’s ability to bind to and inhibit CYP2C9. Additionally, the
spirocyclic moiety in 41 (negligible CYP2C9 inhibition: IC >10 µM) is much more rigid than the
50
saturated yet conformationally unrestrained central regions of compounds such as 7 and 8
(
significant CYP2C9 inhibition: IC ’s 0.12 µM and 3.6 µM, respectively). Thus, the unique shape
50
and rigidity of the spirocyclic central motif within 41 allow it to bind productively to NAMPT, but
may preclude it from binding effectively to CYP2C9.
4a
2
2
ACS Paragon Plus Environment

Page 23 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4b
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 92
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
Figure 4a. Xꢀray crystal structure of 41 in complex with NAMPT. The ligand 41 is shown in orange.
Protein sidechains forming interactions with the ligand are shown in teal. Hydrogen bonding
interactions are shown as dashed lines. Crystallographic waters forming Hꢀbonds with the ligand are
shown as red spheres. PDB code: 5KIT, resolution = 1.60 Å. Figure 4b. Overlay of the small
molecule ligand (41, orange) from PDB code 5KIT (37) with the ligands from previously reported
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
9
b
25
structures 1 (green): 4KFO, and 8 (white): 2GVJ. The location of the nucleophilic N in the leftꢀ
hand side of each ligand is circled.
As noted in Table 7, one issue associated with compound 41 is poor liver microsome stability. As
shown in Figure 4a, the tertꢀbutyl carbamate group protrudes into a relatively open region of the
protein and was expected to be tolerant of substitution. A strategy of reducing cLogD7.4 as a
potential means of improving metabolic stability was undertaken by modifying this portion of the
inhibitor. Additionally, since both enantiomers of (±)ꢀ40 possessed similar metabolic stability, the
optimization work was done with the racemic spirocyclic system, with only key analogs being
resolved into their constituent enantiomers.
As exemplified by the entries in Table 8, a number of tertꢀbutyl carbamate replacements were
prepared using either a subsequent round of parallel synthesis, or by targeted singleton synthesis for
more synthetically challenging analogs. Compounds (±)ꢀ43 – (±)ꢀ59 summarize the efforts at
replacing the tertꢀbutyl carbamate with amides. A wide array of aliphatic and aromatic amides were
prepared. Although reasonable to excellent NAMPT potency was maintained, it was very difficult to
substantively improve metabolic stability while maintaining high membrane permeability. Multiple
analogs such as (±)ꢀ47 – (±)ꢀ49, and (±)ꢀ56 – (±)ꢀ58 were found to be metabolically stable, yet
2
4
ACS Paragon Plus Environment

Page 25 of 92
Journal of Medicinal Chemistry
1
2
3
possessed poor MDCK permeability. By contrast, amides maintaining promising levels of MDCK
permeability were found to suffer from poor metabolic stability (analogs (±)ꢀ43, (±)ꢀ44, and (±)ꢀ51).
Compound (±)ꢀ60, the urea isostere of (±)ꢀ40, was found to moderately improve metabolic stability,
although cellꢀbased potency was reduced by over 10ꢀfold. Non carbonyl linked groups such as
alkylated analog (±)ꢀ61 suffered from a marked decrease in NAMPT inhibition. Having failed to
achieve the required balance of potency, metabolic stability, and membrane permeability with
carbamate replacements, we then investigated whether substituted carbamates ((±)ꢀ62 – (±)ꢀ66)
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
7
could provide the desired profile. These analogs maintained high levels of MDCK permeability,
and gratifyingly, the most polar example ((±)ꢀ66) also exhibited significantly improved metabolic
stability relative to the starting point (±)ꢀ40. Finally, an in vitro metabolite identification study in
human liver microsomes (data not shown) indicated the leftꢀhand side bicycle to be a site of
metabolism. Thus, the deuterated molecule (±)ꢀ67 was prepared in an attempt to further improve
metabolic stability, but was not found to be meaningfully different than the proto parent (±)ꢀ66.
Compound (±)ꢀ66 was, therefore, selected as the most promising compound for further
characterization and progression.
One final point to emphasize about the compounds in Table 8 is the consistent lack of CYP2C9
inhibition. The saturated and rigid spirocyclic central moiety offered a general solution to the
CYP2C9 inhibition issue.
Table 8. SAR of Carbamate Replacements and Substitutions.
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
d
d
Cmpd.
R
cLogD7.4
NAMPT
IC50
A2780
IC50
(ꢁM)
CYP2C9
HLM
Clpred
MDCK
A:B, B:A
b
b
Inhibition
b,c
ꢀ6
(
ꢁM)
IC50 (ꢁM) (mL/min/kg) (x10 cm/s)
(±)ꢀ40
(±)ꢀ43
(±)ꢀ44
2.2
2.1
1.8
0.0072
0.0212
0.0128
0.0016
>10
>10
>10
17
14
13
15, 20
9.2, 13
7.3, 14
0.0016
0.0044
(±)ꢀ45
(±)ꢀ46
(±)ꢀ47
1.4
1.6
1.1
0.0060
0.0161
0.0219
0.0004
0.0058
0.0069
>10
13
11
6
1, 9.1
e
>10
>10
1.6, 11
1.4, 7
0.2, 1
(±)ꢀ48
1
0.0589
0.0478
0.0265
0.0133
0.0112
0.0248
>10
>10
5
6
9
e
e
(±)ꢀ49
0.5
1.7
1.4 , 4
e
(±)ꢀ50
(±)ꢀ51
>10
e
e
1.6
0.6
0.0210
0.0061
0.0035
0.0053
>10
10
7
7.9 , 19
e
(±)ꢀ52
>10
0.2, 1
2
6
ACS Paragon Plus Environment

Page 27 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
O
O
N
S
e
e
e
e
(±)ꢀ53
ꢀ0.7
1.4
0.0410
0.0312
>10
>10
8
9
N
(±)ꢀ54
0.0142
0.0274
0.0038
0.0065
1.4, 11
0.3, 2.7
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
N
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
(±)ꢀ55
0.9
>10
>10
7
O
O
O
(±)ꢀ56
(±)ꢀ57
(±)ꢀ58
0.5
0.6
0.2
0.0204
0.0086
0.0412
0.0040
0.0062
0.0075
6
4
5
0.3, 2.6
0.2, 1.9
0.2, 1.1
N
N
e
>10
N
N
>10
N
N
(±)ꢀ59
0.9
0.0044
0.0016
>10
9
1.4, 9
e
(
±)ꢀ60
±)ꢀ61
1.6
1.2
0.0481
0.5554
0.0207
0.7535
>10
>10
11
15
e
(
(±)ꢀ62
(±)ꢀ63
(±)ꢀ64
(±)ꢀ65
1.8
1.5
0.0151
0.0387
0.0043
0.0294
>10
14
12
13, 14
e
>10
>10
>10
e
e
e
e
e
2.3
1.6
0.9
0.0222
0.0067
0.0171
0.0048
0.0016
0.0022
18
18
7
15 , 12
(±)ꢀ66
>10
11, 16
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 92
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
f
(
±)ꢀ67
NA
e
0
.9
0.0119
0.0010
>10
7
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
Unless otherwised indicated, all assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
All compounds are racemic.
a
b
c
d
e
f
28
Calculated LogD at pH 7.4.
See footnotes from Table 1 for assay descriptions.
S)ꢀWarfarin probe used.
(
See footnotes from Table 2 for assay descriptions.
Single determination (n=1).
Not applicable
The racemic compound with the best balance of potency, metabolic stability, and membrane
permeability was found to be (±)ꢀ66. The enantiopure versions (68 and 69) were, thus, isolated and
profiled. As shown in Table 9, the in vitro metabolic stability and membrane permeability was
virtually identical among each enantiomer (68 and 69) and the racemate ((±)ꢀ66). However, one
enantiomer (68) was notably more potent against NAMPT than the other, with a subꢀnM IC50
observed in the A2780 cell based antiproliferation assay. As a precursor to a tumor xenograft study,
the cell potency of the active enantiomer 68 was further evaluated in a relevant cell line (HT1080),
where a potent antiproliferative IC value of 1.5 nM was observed.
5
0
Table 9. Characterization of Single Enantiomers of (±)ꢀ66.
2
8
ACS Paragon Plus Environment

Page 29 of 92
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
N
N
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
HT1080
IC50
(ꢁM)
e
e
NAMPT A2780
HLM
Clpred
(mL/min/kg) (x10 cm/s)
MDCK
c
a
b
b
Cmpd.
R
Fsp3
IC50
ꢁM)
IC50
(ꢁM)
A:B, B:A
ꢀ6
(
d
6
8
0.67
0.67
0.0055 0.0006 0.0015
8
8.8, 12
f
f
69
0.075
7
11 , 15
All assay results are reported as the arithmetic mean of at least two separate runs (n ≥ 2).
a
See footnotes from Table 4 for description.
b
See footnotes from Table 1 for description.
c
Antiproliferation activity determined in cell culture experiments using HT1080 cell line. This inhibition can be reversed
by addition of 0.33 ꢁM of NMN, strongly implicating NAMPT inhibition as the causative MOA. See reference 9d for
experimental details.
d
The coefficient of variation for this value is 0.03 (n=3).
e
See footnotes from Table 2 for assay descriptions.
f
Single determination (n=1).
The in vitro ADME properties of compound 68 are shown in Table 10. Compound 68 did not
significantly inhibit any CYP isoform tested (including CYP2C9) at the concentrations evaluated.
Similarly, no time dependent CYP inhibition was observed against any isoform tested. 68 was found
to have moderate in vitro liver microsome stability in all species except for mouse, where poor
stability (high predicted clearance) was observed. Compound 68 was generally more metabolically
2
9
ACS Paragon Plus Environment