Optimization of a urea-containing series of nicotinamide phosphoribosyltransferase (NAMPT) activators

Article abstract of DOI:10.1016/j.bmcl.2021.128007

NAD⁺ is a crucial cellular factor that plays multifaceted roles in wide ranging biological processes. Low levels of NAD⁺ have been linked to numerous diseases including metabolic disorders, cardiovascular disease, neurodegeneration,

Full text of DOI:10.1016/j.bmcl.2021.128007

Bioorg. Med. Chem. Lett. 41 (2021) 128007
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Optimization of a urea-containing series of nicotinamide
phosphoribosyltransferase (NAMPT) activators
,
Anthony B. Pinkertona *, E. Hampton Sessionsa, Paul Hershbergera, Patrick R. Maloneya,
Satyamaheshwar Peddibhotlaa, Meghan Hopfd, Eduard Sergienkoa, Chen-Ting Maa,
Layton H. Smitha, Michael R. Jacksona, Jun Tanakab, Takashi Tsujib, Mayuko Akiub,
Steven E. Cohenc, Tsuyoshi Nakamurab, Stephen J. Gardelld
a Conrad Prebys Center for Chemical Genomics, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, CA 92037, USA
b R&D Division, Daiichi Sankyo Co., Ltd., 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
c Daiichi Sankyo, Inc., Global Business Development, Basking Ridge, NJ 07920, USA
d Translational Research Institute. AdventHealth, Orlando, FL 32804, USA
A R T I C L E I N F O
A B S T R A C T
NAD+ is a crucial cellular factor that plays multifaceted roles in wide ranging biological processes. Low levels of
NAD+ have been linked to numerous diseases including metabolic disorders, cardiovascular disease, neuro-
degeneration, and muscle wasting disorders. A novel strategy to boost NAD+ is to activate nicotinamide phos-
phoribosyltransferase (NAMPT), the putative rate-limiting step in the NAD+ salvage pathway. We previously
showed that NAMPT activators increase NAD+ levels in vitro and in vivo. Herein we describe the optimization of
our NAMPT activator prototype (SBI-0797812) leading to the identification of 1-(4-((4-chlorophenyl)sulfonyl)
phenyl)-3-(oxazol-5-ylmethyl)urea (34) that showed far more potent NAMPT activation and improved oral
bioavailability.
Keywords:
NAMPT
NAMPT activators
NAD+ booster
Ureas
Nicotinamide adenine dinucleotide (NAD+) is involved in diverse
cellular processes that govern human health and disease.1 NAD+ has
long been known as a redox enzyme cofactor. More recently, NAD+ was
shown to be a co-substrate for sirtuins and poly-ADP-ribose polymerases
(PARPs) which has unveiled pivotal roles in cell signaling, DNA repair,
cell division, and epigenetics.2,3 Increased tissue levels of NAD+ have
been shown to mediate beneficial effects in a variety of preclinical dis-
ease models.4 These findings have spurred keen interest in pharmaco-
logical and nutraceutical strategies that boost intracellular NAD+
levels.5,6
(PRPP). In turn, NMNAT produces NAD+ from NMN.
NAMPT, a homodimeric type II phosphoribosyltransferase, is the
putative rate-limiting step in the NAM salvage pathway.8
Enzymology9–11 and X-ray crystallograph12–15 studies have yielded a
detailed understanding of NAMPT catalytic activity. NAMPT has also
been shown to be the dominant route for NAD+ synthesis in the heart
and kidney16, with preclinical data suggesting that NAD+ repletion
strategies have the potential to aid in the treatment of cardiac and
kidney injuries, possibly by maintaining sirtuin activity.5 Additionally,
early evidence may show beneficial effects of NAM supplementation in
some forms of eye disease.17,18 Conversely, potent and highly selective
NAMPT inhibitors have been developed and evaluated as a treatment for
cancer.19
NAD+ is consumed in cells due to the actions of sirtuins, PARPs and
NADases.7 Hence, cellular NAD+ biosynthetic pathways are obligatory
to preserve the NAD+ levels which are required for cell viability. The
dominant NAD+ synthetic route in most mammalian cells is the nico-
tinamide (NAM) salvage pathway involving sequential actions of nico-
We recently reported the discovery of SBI-0797812 (2), a small
molecule NAMPT activator which was generated from the HTS hit SBI-
0136892 (1; Fig. 1).20 Interestingly, SBI-0797812 (2) was recently
shown to exert an antiviral effect due to its ability to modulate innate
immunity.21 While SBI-0797812 displayed acceptable potency, its ro-
dent pharmacokinetic (PK) properties, including oral bioavailability
tinamide
phosphoribosyltransferase
(NAMPT)
and
NMN
adenylyltransferase (NMNAT). NAMPT forms nicotinamide mono-
nucleotide (NMN) and pyrophosphate (PP) from NAM (produced by
sirtuins and PARPs) and
α-D-5-phosphoribosyl-1-pyrophosphate
* Corresponding author.
E-mail address: apinkerton@sbpdiscovery.org (A.B. Pinkerton).
https://doi.org/10.1016/j.bmcl.2021.128007
Received 19 January 2021; Received in revised form 25 March 2021; Accepted 27 March 2021
Available online 31 March 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

A.B. Pinkerton et al.
Bioorganic & Medicinal Chemistry Letters 41 (2021) 128007
Table 2
SAR elucidation around the pyridyl group.
Fig. 1. Previously described NAMPT activators.
Cmpd
R1
NAMPT EC50
1.37
(
μ
M)a
Emaxa
10
ꢀ 36%
were inadequate. The results of our medicinal chemistry campaign
aimed at optimizing SBI-0797812 are described in this report.
The NAMPT activity assay measuring NMN production was used to
detect improved NAMPT activator potency.20 The results of our efforts
are summarized in Tables 1–4. Prior work had highlighted the impor-
tance of the 4-pyridyl group for NAMPT activation. We initially modi-
fied the urea linkage (Table 1) to ascertain the key elements required for
NAMPT activator activity. As shown in the table, only the urea moiety
was active. One amide linkage (3) was inactive, while the other isomer
(4) and the carbamate (5) were both weak NAMPT inhibitors. Methyl-
ation of the urea nitrogens (6, 7) abolished NAMPT activator activity.
Interestingly other linkers that have been previously used in NAMPT
inhibitors,22,23 including the thiourea (8) and cyanoguanidine (9) also
yielded weak NAMPT inhibitors. This SAR indicated that the structural
requirements for NAMPT activation appear to be narrower than for
NAMPT inhibition.
11
12
>100
ND
13.6
ꢀ 37%
13
14
2.28
1.85
ꢀ 44%
158%
15
16
69
ꢀ 23%
6.90
194%
We next turned our attention to the pyridyl group. Our previous
work showed that the 4-pyridyl ring mediated NAMPT activation
whereas the 2-pyridyl group yielded inactive compounds and the 3-pyr-
idyl group yielded NAMPT inhibitors.20 The 3-pyridyl group is the most
17
18
27.6
7.75
ꢀ 39%
ꢀ 26%
Table 1
a
Values are means of at least four experiments, standard deviations were
within 20% of reported values. ND = Not determined. Negative values represent
inhibition.
SAR elucidation of the urea linker.
widely used group in NAMPT inhibitor SAR, although a relatively wide
range of pyridine-like groups is tolerated. Further exploration of this
group is shown in Table 2. Phenyl (10), pyridazinyl (11) or pyrimidinyl
(12) gave inactive compounds or weak NAMPT inhibitors, indicating
relatively narrow SAR in this region. Indeed, adding a methyl group to
the methylene linker (13) switched the compound from an activator to a
weak inhibitor. We next explored a series of 5-membered heterocyclic
rings as potential substitutes for the 4-pyridyl ring. Interestingly, a 4-
pyrazoyl group (14) mediated strong NAMPT activation while a 3-pyra-
zoyl group (15) produced a weak inhibitor. Likewise, a 5-oxazolyl group
(16) gave strong NAMPT activation while a 4-oxazolyl (17) gave weak
NAMPT inhibition, as did a 4-triazolyl (18) group. The structural basis
for this toggling between NAMPT activation and inhibition is currently
under investigation.
Cmpd
A
hNAMPT EC50
3.25
(
μ
M)a
Emaxa
98%
1
3
4
5
6
>100
15.9
ND
ꢀ 13%
ꢀ 25%
ND
10.8
We next examined the SAR around a series of phenyl sulfonamide
analogs that deviated from the oxa-azabicyclooctane group in SBI-
0797812 (Table 3). Previous work on NAMPT inhibitors indicated a
very wide range of groups is tolerated in this position.24 This turns out to
be the case for NAMPT activators as well. In general, unsubstituted (19)
or methyl substituted analogs (20, 21) showed weaker potency. Step-
wise introduction of incrementally larger groups such as diethyl (22)
and disubstituted cyclopentyl (23, 24) moieties yielded improved
NAMPT activator potency. Lastly, while phenyl (25) showed little
improvement, benzyl (26), tetrahydroisoquinoline (27) and methyl
benzyl (28) gave potent NAMPT activators. Indeed, 28 displayed a > 10
fold improvement in potency. Both 27 and 28 were examined for their
rodent PK properties. While both compounds displayed reasonable
bioavailability, exposure after oral dosing was modest (vide infra).
Due to the modest exposure after oral dosing of the sulfonamides, we
pursued other linker groups in place of the sulfonamide. As previously
reported for NAMPT inhibitors25 a sulfone linker has been shown to
>100
7
>100
ND
8
9
9.77
41.4
ꢀ 15%
ꢀ 30%
a
Values are means of at least four experiments, standard deviations were
within 20% of reported values. ND = Not determined. Negative values represent
inhibition.
2

A.B. Pinkerton et al.
Bioorganic & Medicinal Chemistry Letters 41 (2021) 128007
Table 3
Table 4
SAR elucidation around the sulfonamide.
Additional SAR elucidation.
Cmpd
Ar
R3
NAMPT EC50
0.085
(μ
M)a
Emaxa
152%
Cmpd
R2
NAMPT EC50
0.279
(μ
M)a
Emaxa
173%
29
2
30
31
0.079
0.069
96%
90%
19
20
14.3
3.23
131%
138%
21
22
23
3.11
133%
179%
130%
0.384
0.123
32
33
0.126
0.028
102%
190%
34
35
0.023
0.077
214%
172%
24
25
0.060
0.410
160%
87%
36
0.070
208%
26
27
0.102
0.142
96%
130%
a
Values are means of at least four experiments, standard deviations were
within 20% of reported values.
28
0.023
153%
a
Table 5
Values are means of at least four experiments, standard deviations were
within 20% of reported values.
Selected mouse PK parameters for compounds 27, 28, 33, 34 and 36.a
Compds
Clpb(mL/
min/kg)
Vdb(L/
kg)
Cmaxc(ng/
mL)
AUCc (ng.
hr/mL)
t1/
%F
c(hr)
2
improve PK properties. We therefore examined a series of sulfone linked
analogs, while also incorporating the 4-pyridyl head-group re-
placements (Table 4). An unsubstituted phenyl (29) gave 4-fold
improved NAMPT activator potency (compare to 25). A wide range of
substitution were introduced for the phenyl group, with the 2-, 3-, and 4-
chloro (30, 31, 32) analogs shown in the table. In most cases, substi-
tution on this ring did not significantly affect NAMPT activator potency,
indicating that this group could be modified to improve other properties
(i.e. PK). We were delighted to find that incorporation of 4-pyrazoyl (33,
34) and 5-oxazolyl (35, 36) yielded potent NAMPT activators, which in
most cases showed significant improvements in exposure after oral
dosing (Table 5). In particular 34 showed low plasma clearance, good
exposure and excellent oral bioavailability (%F = 80).
27
28
33
34
36
20
43
8
0.60
1.37
1.12
0.61
0.58
7860
4953
2230
9550
15,313
4266
1.0
0.6
2.5
3.2
2.3
47
4508
>100
45
9344
3
54,463
45,203
80
4
>100
a
Compounds dosed 1 mpk iv and 10 mpk po.
b
c
iv
po.
The majority of the compounds described above were synthesized as
outlined in the general schemes below.26 A representative synthesis for 1
is shown in Scheme 1. Treatment of aniline 37 with phenyl-
chloroformate gave carbamate 38 in good yield. This was then reacted
with pyridin-4-ylmethanamine to give the desired product. Compounds
10–18 were synthesized in a similar fashion using the appropriate
amine.
Treatment of A549 lung epithelial cells with 34 resulted in a marked
increase of both NMN and NAD (Fig. 2). These effects were similar to
those observed previous with SBI-0797812.20
Compound 34 was mainly clean when evaluated in a Eurofins/
Ricerca Hit Profiling screen against 35 targets; however, 34 did show
The sulfonamides in Table 3 were synthesized as outlined in Scheme
2. 4-Nitrophenylsulfonyl chloride 39 was reacted with diethylamine to
give 40. 40 was reduced using hydrogen and Pd/C to give aniline 41.
This was then reacted with phenylchloroformate to give intermediate
42, which was reacted with pyridin-4-ylmethanamine to give 22.
Compounds 19–28 were in general synthesized using this approach.
Lastly, the sulfones in Table 4 were synthesized in general as outlined
in Scheme 3. 4-Chlorothiophenol (43) was reacted with 4-
moderate activity against CYP2C9 (0.060 μM), 2D6 (0.41 μM) and 2C19
(0.59 μM). CYP activity displayed by 34 was corroborated in internal
assays (see Table 6). Additional in vitro ADME screening (Table 6)
showed that 34 had moderate solubility, good stability in human liver
microsomes but poor permeability. Further optimization to obviate the
CYP activity and other limitations will be reported in subsequent
publications.
3

A.B. Pinkerton et al.
Bioorganic & Medicinal Chemistry Letters 41 (2021) 128007
Scheme 2. Synthesis of 22. i. Diethylamine, Et3N, CH2Cl2, 89% ii. H2, Pd/C,
EtOAc, 90% iii. Phenylchloroformate, Et3N, CH2Cl2, 82% iv. pyridin-4-
ylmethanamine, DMAP, CH3CN, 35%
Fig. 2. A549 Cells were treated with DMSO (control) or 34 (10 μM) for 4 hr.
Cells were harvested and assayed for NMN (left panel) or NAD+ (right panel) by
LC-MS/MS.17 Data is normalized to cell protein. *, P < 0.001 vs. control.
Table 6
Physicochemical and in vitro ADME properties of 34.
LogDa
PAMPA
Solubility (pH
Metabolic
stabilityb
(human, mouse)
(%)
CYP DIc
(1A2/
permeability (pH
5.0, pH 7.4) (nm/s)
1.2, 6.8) (
μg/
mL)
2C9/2D6/
3A4)
3.1
a
<0.1/0.4
110/8.5
87/18
65/96/
95/0
The distribution coefficient (Log D) was measured between 1-octanol and
phosphate buffered saline (pH 7.4).
b
Scheme 3. Synthesis of 34. i. 4-Chloronitrophenol, K2CO3, DMF, 89% ii. m-
CPBA, CH2Cl2, 92% iii. H2, Pd/C, MeOH, 90% iv. Phenylchloroformate, Et3N,
CH2Cl2, 78%. v. (1H-pyrazol-4-yl)methanamine, Et3N, CH3CN, 43%
The % remaining value at 1 μM concentration of compounds reacted with
human or mouse microsomes for 30 min.
c
The % inhibition value at 10 μM concentration of compounds reacted with
corresponding CYP isoforms for 10 min.
hit compounds. Our further investigation of this novel NAD+ booster
class across a broad range of disease models, as well as additional
compound optimization, will be disclosed in subsequent reports.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
This work was supported by funding provided by SBP, AdventHealth
and Daiichi Sankyo Co. Ltd.
Scheme 1. Synthesis of 1. i. Phenylchloroformate, Et3N, CH2Cl2, 20–93% ii.
pyridin-4-ylmethanamine, DMAP, CH3CN, 41–89%
References
chloronitrobenzene to give compound 44, which was oxidized using
mCPBA to sulfone 45. The nitro group of 45 was reduced via hydroge-
nation to provide 46, which was reacted as above with phenyl-
chloroformate to give intermediate 47. 47 was reacted with (1H-
pyrazol-4-yl)methanamine to provide 34. Compounds 29–36 were in
general synthesized using this approach.
1 Garten A, Schuster S, Penke M, Gorski T, de Giorgis T, Kiess W. Nat Rev Endocrinol.
2015;11:535–546.
2 Imai S, Guarente L. Trends Cell Biol. 2014;24:464–471.
3 Gibson BA, Kraus WL. Nat Rev Mol Cell Biol. 2012;13:411–424.
4 Rajman L, Chwalek K, Sinclair DA. Cell Metab. 2018;27:529–547.
5 Yoshino J, Baur JA, Imai SI. Cell Metab. 2018;27:513–528.
6 Liu L, Su X, Quinn 3rd WJ, et al. Cell Metab. 2018;27:1067–1080.
7 Petrelli R, Felczak K, Cappellacci L. Curr Med Chem. 2011;18:1973–1992.
8 Revollo JR, Grimm AA, Imai S. J Biol Chem. 2004;279:50754–50763.
9 Burgos ES, Ho MC, Almo SC, Schramm VL. PNAS. 2009;106:13748–13753.
10 Burgos ES, Schramm VL. Biochemistry. 2008;47:11086–11096.
11 Burgos ES, Vetticatt MJ, Schramm VL. J Am Chem Soc. 2013;135:3485–3493.
12 Wang T, Zhang X, Bheda P, Revollo JR, Imai S, Wolberger C. Nat Struct Mol Biol.
2006;13:661–662.
In conclusion, we have developed a potent series of NAMPT activa-
tors that produce a robust increase in NAD+ levels in cultured cells. A
systematic SAR examination around our urea-based scaffold elucidated
the key groups necessary for robust enzyme activation, including the
critical nature of the head group, wherein small structural changes be-
tween pyridyl or pyrazole isomers led to NAMPT inhibitors. Modifica-
tion of other areas of the scaffold, in particular incorporating a
sulfonamide or sulfone linker, gave compounds that displayed excellent
rodent PK properties, which is a significant improvement over the initial
13 Khan JA, Tao X, Tong L. Nat Struct Mol Biol. 2006;13:582–588.
14 Kim MK, Lee JH, Kim H, et al. J Mol Biol. 2006;362:66–77.
15 Oh A, Ho YC, Zak M, et al. ChemBioChem. 2014;15:1121–1130.
16 Hershberger KA, Martin AS, Hirschey MD. Nat Rev Nephrol. 2017;13:213–225.
4

A.B. Pinkerton et al.
Bioorganic & Medicinal Chemistry Letters 41 (2021) 128007
17 Klimova N. Kristian Neurochem Res. 2019;44:2280–2287.
18 Williams PA, Harder JM, Foxworth NE, Cardozo BH, Cochran KE, John SWM. Front
Neurosci. 2017;11:232.
21 Heer CD, Sanderson DJ, Voth LS, et al. J Biol Chem. 2020.
22 Zheng X, et al. J Med Chem. 2013;56:4921–4937.
23 Watson M, et al. Mol Cell Biol. 2009;29:5872–5888.
19 Sampath D, Zabka TS, Misner DL, O’Brien T, Dragovich PS. Pharmacol Ther. 2015;
151:16–31.
24 Zak M, et al. Bioorg Med Chem Lett. 2015;25:529–541.
25 Zak M, Yuen P, Liu X, et al. J Med Chem. 2016;59:8345–8368.
26 PCT WO 2018132372 Gardell, S.; Pinkerton, A. B.; Sergienko, E.; Sessions, H.
20 Gardell SJ, Hopf M, Khan A, et al. Nat Comm. 2019;10:3241.
5